Datasets:

gabrielaltay
/

pmcoa

Dataset card Data Studio Files Files and versions Community

Split (2)

train · 4.94M rows

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. Ind. Bus. Econ. Journal of Industrial and Business Economics 0391-2078 1972-4977 Springer International Publishing Cham 244 10.1007/s40812-022-00244-y Article Journal of industrial and business economics—Economia e politica industriale: a historical sketch of the first fifty years http://orcid.org/0000-0001-5743-4625 Mariotti Sergio [email protected] grid.4643.5 0000 0004 1937 0327 Department of Management, Economics and Industrial Engineering, Politecnico di Milano, Milan, Italy 8 12 2022 113 21 10 2022 27 11 2022 30 11 2022 © The Author(s) under exclusive licence to Associazione Amici di Economia e Politica Industriale 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 article offers a descriptive summary of the fifty years of the Journal of Industrial and Business Economics, with information and data on the protagonists, governance, issues and changes that have taken place in half a century. Reference is first made to the Italian context, in which the Journal was born and grew, and then to the international context in which it stands today. The non-conformist origin of the Journal and its focus on the major challenges of our changing world are emphasized, which we hope will be maintained in the future. Keywords Bibliometrics Historical analysis Industrial economics Journal impact JEL classification L00 N01 Y10 ==== Body pmcIntroduction In the 1950 and 1960 s, the world economy grew at twice the secular rate (Toniolo, 1998). In Italy, the uniqueness of those years caused a major misalignment between the economy, on the one hand, and institutions and the state, on the other hand, which were no longer able to support the growth imperatives of industrial capitalism (De Cecco, 2007). In the late 1960s, with the so-called “hot autumn”, workers’ wage claims exploded and the foundations were laid for a new form of trade unionism. Some of the largest groups in Italian capitalism sought to initiate a dialogue with the latter and the productive classes to reform the state and dissolve obstacles to growth1. In parallel, the exceptional increase in income had favored mass schooling, and the “1968 student movement” expressed the deep cultural and ideological unease of the new generations in the face of economic and social challenges (Salvati, 1981). In the early 1970s, the Italian question became intertwined with international dynamics. The world economy saw its linear and seemingly balanced growth come to a halt. Salient moments, such as the end of dollar convertibility and the 1973 oil crisis, accompanied the restructuring of relations among the Great Powers (Lewis, 1980). In this context of double crisis, but also of economic and social reforms, the most culturally and socially committed Italian scholars immersed themselves in the facts and sought interpretations that could help untangle the knots that were blocking growth. They tried to apply the method and models of industrial economics and policy to the Italian and European case. In 1973, under the stimulus of Sergio Vaccà, professor at the Bocconi University in Milan, the Bollettino di Economia e Politica Industriale was born, which the following year took on the title of Economia e Politica Industriale (EPI). Two years later, on the initiative of Romano Prodi, professor at the University of Trento and then Bologna, the Rivista di Economia e Politica Industriale (REPI) came into being, which, in 1980, took on the title L’Industria - Rivista di Economia e Politica Industriale, gaining access to the publishing rights of this historic journal, founded in 1886. The flourishing of these initiatives is hardly surprising, as the most influential Italian intellectuals had become aware of the need to contribute to the solution of the complex problems posed by the new economic and social phase. Perhaps it may come as a surprise that these two journals are still alive and well: the former has taken on an international dimension since 2015 and it is published in English under the title Journal of Industrial and Business Economics (JIBE), while the latter has retained its original title and publishes articles in both Italian and English. The two journals have remained firmly grounded in their history and have engaged in fair competition over time, thus creating a scientific and social space for the meeting and discussion between Italian scholars of industrial economics. However, many of JIBE-EPI’s current contributors and readers may be unaware of its origins and history. Therefore, on the occasion of its fiftieth anniversary, it seems appropriate to propose a historical sketch of its first fifty years. This should not only be seen as a grateful celebration of those who have devoted part of their intellectual efforts and time to the Journal. In our view, it is the best way to explain its identity, that is, the characteristics that have ensured JIBE-EPI’s uniqueness as a non-conformist and non-mainstream journal, open to interdisciplinary contributions complementary to industrial economics, which has remained its key pillar. We believe it is essential not to lose the memory of these events in order to keep a straight rudder in navigating today’s crowded sea of international journals, which are sometimes pure containers of articles, being blurry in their aims, methods and content. Below, the narrative will focus on JIBE-EPI, but some comparisons will be made both with its closest “rival” in the domestic market and, far more importantly today, with other international journals. Before moving forward, we must caution the reader. Although our arguments are supported by data, they can sometimes come across as “biased,“ given our emotional attachment to the Journal. The latter certainly has many strengths, but also some weaknesses, discussed in the text. But not all flaws should be blamed on the Journal on its 50th birthday! Data and trends The half-century of JIBE-EPI’s life can be divided into three periods: the first, from its origin until the end of 1993, when a systematic survey of the articles published up to then was drawn up (EPI Editorial Board, 1994); the second, until the end of 2014, when the transition to the new publisher and the English language took place; and the third, from 2015 until now. Table 1 shows the number of authors (who wrote one or more contributions) and the number of articles published in the three periods. The 1973–1993 period was characterized by a relatively low number of authors, compared to the total number of published articles (the ratio is 0.49). The EPI Editorial Board (1994) reported that the 16 most prolific out of a total of 417 authors (3.8%) wrote 28% of the articles, while the top 42 authors (4.1%) wrote 44% of the articles. This high concentration does not correspond to a process of “crystallization” and “closure”, but rather to the formation of a core group of scholars who, over time, have given the Journal a clear cultural and scientific identity. In fact, over the two decades, “new” authors (i.e., previously not contributors to the Journal) were on average twenty each year 1994. In this regard, the EPI Editorial Board (1994) stated that: «The presence of a stable and truly operational think tank, highly cohesive while fully respecting the individual opinions of its members, has given Economia e Politica Industriale […] the capability to: propose research questions and hypotheses; promote seminars and debates on key issues by involving scholars with different skills and points of view; and stimulate in-depth studies, critical reviews and updates, with contributions placed in time, but linked together in answering questions-often ‘intellectual provocations’-that arise from non-episodic reflection» (author’s translation). Table 1 Authors and articles in the succession of JIBE-EPI life stages 1973–1993 1994–2014 2015–2022 Authors (N.) 417 770 423 Articles (N.) 835 828 234 Authors/Articles 0.49 0.93 1.81 Authors/Year 19.90 36.67 52.88 Articles/Year 39.76 39.43 29.25 Source: EPI Editorial Board (1994) and our elaborations on Google Scholar The collective commitment of EPI’s pioneers led to an increasing involvement of other authors, whose number almost doubled in the following two decades, thus bringing the author/article ratio to 0.93, with the total number of articles almost similar between the two periods (about 40 per year). Based on our experience, it can be said that scholars were attracted to the Journal because of the cultural space created by its prerogatives to “be always on the ball”, attentive to major problems, fundamental research questions, and the generation of substantive ideas and approaches to address them, while avoiding the syndrome of elegant formalism and self-referentiality at the expense of the breadth and relevance of published research. In other words, EPI embraced what Krugman (2018) has also recently stated as: «the important point shouldn’t be “don’t formalize”; it should be that formalism is there to open your mind, not close it, and if the real world seems to be telling you something inconsistent with your model, the problem lies in the model, not in the world». Over time, EPI’s founder and inspirer Sergio Vaccà endeavored to figure out how to preserve the collegiality of the Journal when its catalytic role would cease. Thus, in 2007, the non-profit organization Associazione Amici di Economia e Politica Industriale (hereinafter EPI Association) was born with the aim of entrusting ownership of the Journal to an association of scholars, with the most loyal collaborators at the forefront2. The members of the Association have since grown in number, joining the mission and projects of the Journal. On the one hand, the EPI Association ensured the Journal’s survival and continuity; on the other hand, an institutional mechanism was intentionally activated to foster cohesion and a style of collective orientation toward a non-conformist journal of industrial economics. Times changed again in 2015, when the EPI Association decided to transform EPI into an international journal, with a predominantly European, but world-oriented perspective. The EPI Association was fully aware of how difficult it would be to position the Journal in the international market and maintain good quality and visibility together with the historical identity of EPI, which was henceforth associated with the title JIBE. A visible effect of this change was reflected in the number and variety of authors and articles. Table 1 shows how over the period 2015–2022, while the number of articles per year has decreased (29 compared to 40 in the past), the number of authors has increased considerably, reversing the author/article ratio, which rose to 1.81, underscoring the relevance of co-authored articles. Thus, the internationalization process of JIBE-EPI has been accompanied by alignment with the growing phenomenon of multi-authorship that has emerged in the postwar period in economics, as well as in other scientific areas (Hudson, 1996; McDowell & Melvin, 1983)3. The involvement of an increasing number of authors not affiliated with Italian institutions has positively enhanced the reputation of JIBE-EPI in the international scientific publication market, yielding important achievements in terms of journal ranking and bibliometric indicators (see Sect. 4). Table 2 shows the breakdown of articles by subject category over the three periods. Some aggregate trends deserve attention. Table 2 Article breakdown by subject category Subject category 1973–1993 1994–2014 2015–2022 Structure and dynamics of industrial capitalism 9.7 11.6 8.9 Public enterprises and institutions 7.3 2.3 2.3 Industrial organization and market structure 10.2 13.2 17.5 Industry studies 11.2 9.1 3.0 Districts, firm networks, environment 13.1 8.3 5.1 Technical change and innovation 10.6 8.2 14.2 Firm internationalization and multinational enterprises 8.7 10.7 6.6 Industrial policy, antitrust, and regulation 8.0 14.7 12.9 Economics and labour policy 6.4 6.3 4.6 Bank-business relationships and finance 3.5 3.7 13.5 Firm strategy and organization, corporate governance 11.3 11.9 11.4 Total 100.0 100.0 100.0 Source: EPI Editorial Board (1994) and our elaborations on Google Scholar (i) The share of articles devoted to large private and public companies and institutions has decreased over time. Above all, interest in the public sector has declined in connection with the privatization wave that swept through the world, including Italy. In fact, the share of “public enterprises and institutions” fell from 7.3% in the 1973–1993 period to 2.3% in the following two periods, while the share of “structure and dynamics of industrial capitalism” remained quite stable, with just physiological fluctuations (between 9% and 11%). (ii) The focal area closest to the study of industrial structures and markets has also shrunk, more markedly in the latest period (from 34.5 to 25.6% between the first and third periods). Especially, “industry studies”, so traditional in the industrial economics of the 1960-1980s, and articles on “districts, firm networks, environment”, which are quite specific to the Italian economy, have declined. In the latter case, the “internationalization effect”, with authors being less involved in these issues, is evident. (iii) The share of articles devoted to the most relevant changes in the world economy - internationalization and innovation - is stable (between 19% and 21%). However, there is a shift in the share from “firm internationalization and multinational enterprises” to “technical change and innovation”, with a greater focus on the latter. (iv) Over time, policies have captured more attention (share increased from 14.4 to 25.4% and 17.5%). The mirror effect of liberalization, namely the growing importance of competition and regulatory policies worldwide is undoubtedly the main causes. On the other hand, the share of “economic and labor policy” decreased from 6.4 to 4.6%. (v) Last but not least, mention should be made of the two subject categories closest to management and finance. While the share of articles devoted to “firm strategy and organization, corporate governance” has remained very stable (about 11%), “bank-business relationships and finance” is the subject category that has recorded the most dramatic increase in share in the transition from the first two periods to the last: 3.5% vs. 3.7% vs. 13.5%. This performance could be explained by many factors: the financialization of the world economy, the 2007–2008 global financial crisis, innovative finance instruments, and the financing of innovation and start-ups. Other explanations may be occasional in nature or due to editorial board preferences, compared to the past, when finance and financial market topics per se were less considered, if not in close connection with the industrial strategy of companies. In taking on this broad scope, JIBE-EPI has made use of Forums, Special Issues, and Special Sections to stimulate and engage scholars in making timely contributions on hot topics and/or deemed relevant. With reference to the last period, Table 3 shows the time sequence of publishing initiatives, by type and title. Table 3 Forum, special issues, and special section, 2015–2022 Year Type Title 2015 Special Issue Offshoring and innovation 2015 Special Section Offshoring, immigration and the labour market: A micro-level perspective 2016 Forum Perspectives on industrial policies in Italy and in Europe 2016 Special Issue Resources (mis)allocation, innovation and the competitiveness of Europe 2016 Special Section Public sector entrepreneurship 2017 Special Issue Entrepreneurial finance: New trends, theories and methods 2017 Special Issue Giacomo Becattini, industrial economics, and local development 2018 Special Issue Public procurement: New theoretical and empirical developments 2019 Special Issue Digitalizing industries? Labor, technology and work organization 2019 Special Issue Grand challanges and new avenues for corporate governance research 2020 Forum The socio-economic consequences of Covid-19 2020 Special Issue Cryptocurrencies: Market analysis and perspectives 2021 Special Section Economics-Engineering nexus 2021 Special Section Green economy and evironmental policies in oligopoly markets 2022 Special Issue Industrial dynamics in digital markets 2022 Special Section The Russian-Ukrainian war: Causes, implications and policies Source: Our elaborations Finally, the effectiveness of the Journal in influencing scientific debate should be assessed. We refer to Sect. 4 for an analysis in the present time, based on comparison with other similarly positioned journals in the international publishing market. Here we focus on a comparison between JIBE-EPI and L’Industria-REPI based on the number of citations over time4. Table 4 shows that, although the number of cited articles differed slightly between the two journals, the number of citations until the end of 2014 was the same (5,111 vs. 5,112). This parity in visibility breaks down starting in 2015, in connection with the internationalization of JIBE-EPI. In the 2015–2022 period, the latter’s citations are more than three and a half times those of L’Industria-REPI (3,354 vs. 901), an expected result, given the different geographic scope, but one that confirms the stifling that country-focused social science journals currently suffer from. The values of the indicators “citations/article” and “citations/year” are very telling in explaining the different dynamics of the two journals in the last period. It is worth noting the contribution of Special Issues and Forum to the total number of JIBE-EPI citations over the period 2015–2022. The timely Forum on Covid-19 has a 28% share; the top three initiatives (adding the Special Issues on digitization and cryptocurrencies) 45%; and the top six 60%. Table 4 A comparison between JIBE-EPI and L’Industria-REPI: Cited articles and citations (numbers) Cited articles Citations Citations/article Citations/year Journal Total 1988–2014 2015–2022* Total 1988–2014 2015–2022* Total 1988–2014 2015–2022* Total 1988–2014 2015–2022 JIBE-EPI 805 604 201 8,465 5,111 3,354 10.52 8.46 16.69 242 189 419 L’Industria-REPI 872 721 151 6,013 5,112 901 6.90 7.09 5.97 172 189 113 * End of November Source: Our elaborations on Google Scholar Actors Fifty years of life for a scholarly journal is primarily the result of the constant and converging efforts of scholars who give it continuity with their papers, but also of renewal choices that foster continuous new additions to the editorial board. Table A1 in Appendix is a very much due acknowledgment of those who were among the most committed authors in supporting the Journal with their scientific contributions. It is worth noting that 8 of the top 15 authors by number of published articles are still active in the Journal as of 2022. As far as the citations are concerned, Table A2 shows the major contributors to the Journal’s visibility. Credit goes to Sergio Vaccà and Enzo Rullani, whose contributions to the citations in the 1973–2014 period were sizeable, meaning that their articles have greatly influenced the debate in Italy among scholars of industrial economics, knowledge economics, and multinational enterprises. The 2015–2022 period marks the international debut of the Journal and the consolidation of its reputation on a global scale. Internationalization has enlarged the number of citations per article. Single articles exceeded 100 citations in just a short time, an event that rarely occurred in the past. Among the top 15 most cited authors for articles in JIBE-EPI during this period, the vast majority are foreign scholars (10) and Italian scholars affiliated with foreign institutions (2). Some authors who contribute to the Journal have outstanding scientific profiles. According to Google Scholar, David Audretsch has more than 100,000 citations (he was named Clarivate Citation Laureate in 2021), Giovanni Dosi 78,000, Bengt-Åke Lundvall 60,000, Tod Sandler 44,000, Ha-Joon Chang more than 33,000, Albert Link 30,000, Massimo Colombo and Rajneesh Narula more than 18,000, Cristiano Antonelli, Eric Lehmann and Roger Strange around 13,000. Present and future The current situation of the Journal is quite different from the past. As of 2019, it has taken on the title JIBE and the subtitle EPI. Today, JIBE-EPI is an international journal ranked by Scimago-Scopus in the first quartile of the subject categories Business and international management, Business, management and accounting, Economics, econometrics and finance. JIBE-EPI is indexed in ESCI Web of Science and ABS Academic Journal Quality Guide. Its worldwide diffusion is evidenced by the growth in annual downloads, which went from 1,779 to 2015, the year of its international debut, to 174,654 in 2021. To better assess its performance on the international scene, Table 5 compares JIBE-EPI with other journals, selected according to the following criteria: (i) disciplinary contiguity, in accordance with the academic positioning of the Journal and the Scimago similarity indicator5; (ii) less important and relevant to few cases, sharing of Italian origin6. The comparison is based on relevant bibliometric indicators, with reference to the most recent values (2021) made available by Scimago7. Table 5 A comparison between JIBE-EPI and similar journals: Bibliometric indicators, 2021 Title SJR Best quartile Total Cites (3years) Cites/Paper (2years) Industrial and Corporate Change 1.735 Q1 791 3.04 International Business Review 1.690 Q1 2256 8.44 Journal of Technology Transfer 1.609 Q1 1551 6.79 International Journal of Industrial Organization 1.298 Q1 381 1.70 Eurasian Business Review 1.190 Q1 343 4.74 Structural Change and Economic Dynamics 1.152 Q1 1417 5.04 Industry and Innovation 1.039 Q1 578 3.46 Journal of Industrial Economics 0.920 Q1 88 0.80 Journal of Industrial and Business Economics 0.846 Q1 366 5.43 Review of World Economics 0.832 Q1 173 1.82 Economics of Innovation and New Technology 0.804 Q1 419 2.57 World Economy 0.780 Q1 910 2.17 International Journal of the Economics of Business 0.469 Q2 93 1.10 Economia Politica 0.440 Q2 189 1.53 Journal of Industry, Competition and Trade 0.410 Q2 133 1.68 Journal of Entrepreneurship and Public Policy 0.295 Q2 111 1.61 L’Industria 0.295 Q2 51 0.67 Italian Economic Journal 0.290 Q2 63 1.05 Revue d’Economie Industrielle 0.153 Q4 18 0.26 Source: SCImago Journal & Country Ranking (Elsevier Scopus database) The table confirms the growth of the Journal. As for SJR and quartile, JIBE-EPI has left behind 10 journals out of the total 18 included in the list (among them, all those with a past national tradition). It is very close to the position of the Journal of Industrial Economics (JIE), founded in 1952 and recognized as a leading journal in the field. Moreover, by number of citations and citations per article in recent years, JIBE-EPI distances both JIE and other journals with a higher SJR, envisaging in the near future a further approach to the positions of well-renowned journals, such as Industrial and Corporate Change, International Business Review, Journal of Technology Transfer. In particular, it should be noted that, by number of citation/paper, JIBE-EPI (5.43) quite clearly distances almost all similar journals (average value equal to 2.1), with the only exceptions being International Business Review (8.44), and Journal of Technology Transfer (6.79). This performance can surely be considered an excellent achievement that rewards fifty years of commitment by three generations of scholars. The Journal has continued to innovate, reacting to, and sometimes anticipating the evolution (and revolution) of the real economy. Accordingly, the subjects, ideas and methods have changed over time. The challenge for the Journal’s internationalization has further encouraged such changes. But what about the near future? During the transition phase, JIBE-EPI may have suffered from a mismatch between expectations in line with its historical tradition and the submission of papers by foreign scholars not yet aware of its aim and scope. Having successfully overcome this phase, it is now up to the Journal’s editors to focus the topics so as to ensure high scholarly quality and effective positioning in the context of international publications. In pursuing these goals, they must not forget the non-conformist origin of the Journal and its orientation that favors ideas and interpretations addressing the great challenges of our changing world, rather than elegant and technically sophisticated models limited to specialized fields. Some of us may have come across this quote from John Maynard Keynes in the essay commemorating Alfred Marshall: the master-economist must possess a rare combination of gifts. He must reach a high standard in several different directions and must combine talents not often found together. He must be mathematician, historian, statesman, philosopher in some degree. He must understand symbols and speak in words. He must contemplate the particular in terms of the general, and touch abstract and concrete in the same flight of thought. He must study the present in the light of the past for the purposes of the future. No part of man’s nature or his institutions must lie entirely outside his regard (Keynes, 1924: 322). Keynes certainly had very high standards. The EPI Association is well aware of the fact that the intersection of all these attributes is most likely an empty set. However, in keeping with JIBE-EPI’s 50-year history, this still seems to be the ideal guide to address the Journal’s editorial policy. Appendix Table A1 Top15 authors by number of articles, 1973–2022 Author Articles (N.) Activity period Sergio Vaccà 60 1973–2007 Enzo Rullani 47 1974–2022 Sergio Mariotti 46 1982–2022 Gianni Cozzi 41 1973–2022 Antonello Zanfei 35 1985–2022 Francesco Silva 24 1982–2022 Marco Mutinelli 21 1997–2022 Augusto Ninni 20 1979–2022 Lorenzo Caselli 19 1973–2009 Luigi Frey 19 1974–2005 Massimo G. Colombo 19 1985–2022 Giorgio Lunghini 18 1985–1996 Giacomo Becattini 17 1984–2017 Giorgio Giorgetti 17 1973–1999 Graziella Fornengo 16 1976–2006 Source: Our elaborations Table A2 Top15 most cited authors, 1973–2014 and 2015–2022 (end of November) Author EPI citations 1973–2014 (N.) Author JIBE citations 2015–2022 (N.) Enzo Rullani 757 Rajneesh Narula 357 Sergio Vaccà 481 José Guimón 316 Sergio Mariotti 272 Jill Juergensen 316 Giacomo Becattini 207 Marta Fana 225 Marco Mutinelli 179 Giancarlo Giudici 224 Antonello Zanfei 163 Enrique Fernández-Macías 216 Gianni Lorenzoni 162 Sergio Torrejón Pérez 216 Riccardo Varaldo 152 Dmitri Vinogradov 161 Gianni Cozzi 117 Akistair Milne 154 Cristina Boari 112 Leopoldo Nascia 132 Luca Ferrucci 112 Roger Strange 116 Alessandro Grandi 110 Jayati Ghosh 109 Sebastiano Brusco 99 Mario Pianta 108 Roberto Schiattarella 91 Antonio Andreoni 104 Barbara Di Bernardo 87 Ha-Joon Chang 104 Source: Our elaborations on Google Scholar Declarations Conflict of interest The author states that there is no conflict of interest. 1 Perhaps the most emblematic event of the attempt at a capital-labor “alliance” was a conference on Italian economic development sponsored by Il Mulino and held in Bologna on April 14, 1973. In the introduction to the proceedings of the Conference (Associazione Il Mulino, 1973: 6), we can read: «Much has been said about this conference as an occasion for the “historic”- meeting between Agnelli and Amendola, that is, between an advanced capitalism available to an open discussion even with the Italian Communist Party (PCI), and a communist party ready to come to an agreement with the big capital on the skin of the working class» (author’s translation). The reference is to Umberto Agnelli, then CEO of Fiat, and Giorgio Amendola, a leading member of the reformist wing of PCI, who attended the conference along with other businessmen, trade unionists and politicians. 2 The Association was established on March 2, 2007 by Maria Rosa Battaggion, Gianni Cozzi, Luigi De Paoli, Sergio Mariotti, Augusto Ninni, Francesco Silva and Cesare Vaccà. The Association was chaired by Sergio Vaccà until his death. As of April 19, 2007, EPI was donated by Sergio Vaccà to the Association. 3 Multi-authorship can sometimes result from scientific misconduct in order to obtain more career credits (Liebowitz, 2014). As the new culture of “publish in a group or die alone” is clearly pervasive and appears to be here to stay, the JIBE-EPI editorial board must increasingly strive to safeguard the integrity of research. 4 Here and in the following we use the Google Scholar database, which has a wider temporal coverage than others. The systematic counting of citations was initiated by Google Scholar in 1988 (the year refers to citations and not to the publication of the cited papers). The reliability of the count has grown over time, with major shortcomings in the first period. In Sect. 4, for the comparison between international journals in recent years, we use the Scopus-Scimago database instead. 5 According to Scimago, similarity is expressed as the percentage of referenced publications shared between the selected journal and others. We consider journals with a percentage of 50% or more (International Journal of the Economics of Business, Eurasian Business Review, Journal of Entrepreneurship and Public Policy). 6 I.e., Economia Politica, Italian Economic Journal (the official journal of the Italian Economic Association), L’Industria. 7 Among others, the SCImago Journal Rank (SJR), i.e., an impact factor based on the idea that all citations are not created equal. SJR is the result of a metric that weights citations differently, depending on the subject field, quality and reputation of the journal where the citation is. It uses Elsevier Scopus database citation data. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Associazione Il Mulino. (1973). Sistema industriale e sviluppo economico in Italia. Il Mulino. De Cecco M Italy’s dysfunctional political economy West European Politics 2007 30 4 763 783 10.1080/01402380701500280 EPI Editorial Board (1994). Venti anni di vita: qualche numero e qualche idea, Economia e Politica Industriale, Special Issue, 1973–1993. Indice sistematico: 3–22. Hudson J Trends in multi-authored papers in economics Journal of Economic Perspectives 1996 10 3 153 158 10.1257/jep.10.3.153 Keynes JM Alfred Marshall, 1842–1924 The Economic Journal 1924 34 135 311 372 10.2307/2222645 Krugman, P. (2018). Uses and abuses of economic formalism (wonkish and self-referential), The New York Times, June 27. https://www.nytimes.com/2018/06/27/opinion/uses-and-abuses-of-economic-formalism-wonkish-and-self-referential.html. Lewis WA The slowing down of the engine of growth The American Economic Review 1980 70 4 555 564 Liebowitz SJ Willful blindness: the inefficient reward structure in academic research Economic Inquiry 2014 52 4 1267 1283 10.1111/ecin.12039 McDowell JM Melvin M The determinants of co-authorship: an analysis of the economics literature The Review of Economics and Statistics 1983 65 1 155 160 10.2307/1924423 Salvati, M. (1981). May 1968 and the hot autumn of 1969: the responses of two ruling classes. In S. Berger (Ed.), Organizing interests in Western Europe (pp. 329–363). Cambridge University Press. Toniolo G Europe’s golden age, 1950–1973: speculations from a long-run perspective The Economic History Review 1998 51 2 252 267 10.1111/1468-0289.00090	0	PMC9734346	NO-CC CODE	2022-12-14 23:28:28	no	J. Ind. Bus. Econ.. 2022 Dec 8;:1-13	utf-8	null	null	null	oa_other
==== Front Biophys Rev Biophys Rev Biophysical Reviews 1867-2450 1867-2469 Springer Berlin Heidelberg Berlin/Heidelberg 1012 10.1007/s12551-022-01012-x Correction Correction to: Biophysical Reviews: Publishing short and critical reviews written by key figures in the field Hall Damien [email protected] [email protected] 12 1 grid.9707.9 0000 0001 2308 3329 WPI Nano Life Science Institute, Kanazawa University, Kakumamachi, Kanazawa, Ishikawa 920‑1164 Japan 2 grid.5373.2 0000000108389418 Department of Applied Physics, Aalto University, 00076 Aalto, Finland 9 12 2022 12 © International Union for Pure and Applied Biophysics (IUPAB) 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. ==== Body pmc Correction to: Biophysical Reviews (2022) 14:1067–1074 10.1007/s12551-022-01009-6 In the original version of the article, the following references were provided without appropriate indexing Anashkina AA, Rubin AB, Gudimchuk NB, Vanin AF, Tsygankov AA, Orlov YL (2022) An open call for contributions to a special issue of Biophysical Reviews highlighting the research themes of VII Congress of Russian Biophysicists 2023. Biophys Rev. 10.1007/s12551-022-00998-8 Azbukina N, Zharikova A, Ramensky V (2022) Intragenic compensation through the lens of deep mutational scanning. Biophys Rev. 10.1007/s12551-022-01005-w Bhattacharjya S (2022) The structural basis of β2-integrin intra-cellular multi-protein complexes. Biophys Rev. 10.1007/s12551-022-00995-x Daniel Peluffo R, del V. Alonso S, Itri R, González Flecha FL, Barbosa LRS (2022) Biophysical Reviews special issue call: LAFeBS—highlighting biophysics in Latin America. Biophys Rev. 10.1007/s12551-022-00996-w Dornas W (2022) Nuclear factor erythroid 2-related factor 2 and autophagy regulation in cancer development. Biophys Rev. 10.1007/s12551-022-00992-0 Dos Remedios CG (2022) Vale Jean Garnier (1929–2022). Biophys Rev. 10.1007/s12551-022-01007-8 Ho JWK, Chen X, He M, Huang Y, Mar JC, Shih DJH, Wu AR (2022) Biophysical Reviews special issue call: quantitative methods to decipher cellular heterogeneity — from single-cell to spatial omic methods. Biophys Rev 14, Current Issue. 10.1007/s12551-022-00994-y Hofmann L, Mandato A, Saxena S, Ruthstein S (2022) The use of EPR spectroscopy to study transcription mechanisms. Biophys Rev. 10.1007/s12551-022-01004-x Negi G, Sharma A, Dey M, Dhanawat G, Parveen N (2022) Membrane attachment and fusion of HIV-1, influenza A, and SARS-CoV-2: resolving the mechanisms with biophysical methods. Biophys Rev. 10.1007/s12551-022-00999-7 Olson WK, He R, Benedetto A, Iskratsch T, Shaitan K, Hall D (2022) Editors’ roundup: October 2022. Biophys Rev. 10.1007/s12551-022-01002-z Robson B (2022) Obituary for Jean Garnier. Biophys Rev. 10.1007/s12551-022-01006-9 Su Z, Chen Z, Ma K, Chen H, Ho JWK (2022) Molecular determinants of intrinsic cellular stiffness in health and disease. Biophys Rev. 10.1007/s12551-022-00997-9 Yanagisawa M (2022) Cell size space effects on phase separation of binary polymer blends. Biophys Rev. 10.1007/s12551-022-01001-0 The original article has been corrected by including the correct indexing information as follows: Anashkina AA, Rubin AB, Gudimchuk NB, Vanin AF, Tsygankov AA, Orlov YL (2022) An open call for contributions to a special issue of Biophysical Reviews highlighting the research themes of VII Congress of Russian Biophysicists 2023. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-00998-8 Azbukina N, Zharikova A, Ramensky V (2022) Intragenic compensation through the lens of deep mutational scanning. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-01005-w Bhattacharjya S (2022) The structural basis of β2-integrin intra-cellular multi-protein complexes. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-00995-x Daniel Peluffo R, del V. Alonso S, Itri R, González Flecha FL, Barbosa LRS (2022) Biophysical Reviews special issue call: LAFeBS—highlighting biophysics in Latin America. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-00996-w Dornas W (2022) Nuclear factor erythroid 2-related factor 2 and autophagy regulation in cancer development. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-00992-0 Dos Remedios CG (2022) Vale Jean Garnier (1929–2022). Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-01007-8 Ho JWK, Chen X, He M, Huang Y, Mar JC, Shih DJH, Wu AR (2022) Biophysical Reviews special issue call: quantitative methods to decipher cellular heterogeneity — from single-cell to spatial omic methods. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-00994-y Hofmann L, Mandato A, Saxena S, Ruthstein S (2022) The use of EPR spectroscopy to study transcription mechanisms. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-01004-x Negi G, Sharma A, Dey M, Dhanawat G, Parveen N (2022) Membrane attachment and fusion of HIV-1, influenza A, and SARS-CoV-2: resolving the mechanisms with biophysical methods. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-00999-7 Olson WK, He R, Benedetto A, Iskratsch T, Shaitan K, Hall D (2022) Editors’ roundup: October 2022. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-01002-z Robson B (2022) Obituary for Jean Garnier. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-01006-9 Su Z, Chen Z, Ma K, Chen H, Ho JWK (2022) Molecular determinants of intrinsic cellular stiffness in health and disease. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-00997-9 Yanagisawa M (2022) Cell size space effects on phase separation of binary polymer blends. Biophys Rev. 14(5): (Current Issue) 10.1007/s12551-022-01001-0 Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.	0	PMC9734347	NO-CC CODE	2022-12-14 23:28:28	no	Biophys Rev. 2022 Dec 9;:1-2	utf-8	Biophys Rev	2,022	10.1007/s12551-022-01012-x	oa_other
==== Front Int J Child Maltreat Int J Child Maltreat International Journal on Child Maltreatment 2524-5236 2524-5244 Springer International Publishing Cham 141 10.1007/s42448-022-00141-w Research Article The COVID-19 Pandemic and Quality of Life: Experiences Contributing to and Harming the Well-Being of Canadian Children and Adolescents http://orcid.org/0000-0001-5695-9358 Gervais Christine [email protected] 1 Côté Isabel 2 Lampron-deSouza Sophie 3 Barrette Flavy 2 Tourigny Sarah 4 Pierce Tamarha 5 Lafantaisie Vicky 4 1 grid.265705.3 0000 0001 2112 1125 Nursing Department, Université du Québec en Outaouais, 5 Rue Saint-Joseph, Saint-Jérôme, Québec J7Z 0B7 Canada 2 grid.265705.3 0000 0001 2112 1125 Social Work Department, Université du Québec en Outaouais, 283 Boulevard Alexandre-Taché, C.P. 1250, Succursale Hull, Gatineau, Québec J8X 3X7 Canada 3 grid.14848.31 0000 0001 2292 3357 School of Psychoeducation, Université de Montréal, 90 Av, Vincent-d’Indy, Montréal, Québec H2V 2S9 Canada 4 grid.265705.3 0000 0001 2112 1125 Psychoeducation and Psychology Department, Université du Québec en Outaouais, 5 Rue Saint-Joseph, Saint-Jérôme, Québec J7Z 0B7 Canada 5 grid.23856.3a 0000 0004 1936 8390 School of Psychology, Université Laval, 2325, Rue Des Bibliothèques, Québec, G1V 0A6 Canada 7 12 2022 123 24 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 pandemic’s restrictive measures such as lockdowns, social distancing, and the wearing of masks transformed young people’s daily lives and brought up major concerns regarding children’s and adolescents’ well-being. This longitudinal mixed study aims to identify how different experiences contributed to children’s and adolescents’ well-being through different stages of the pandemic. The sample comprises 149 Canadian youth from Quebec who shared their experiences of the COVID-19 pandemic. Children and adolescents were met virtually for semi-directed interviews about their well-being at three measurement time (T1: May 2020 lockdown, T2: July 2020 progressive reopening, and T3: beginning of the second wave). At T3, they also completed a questionnaire measuring their quality of life. Our findings indicated that 22% reported a low level of well-being (N: 32), 66% a normal level of well-being (N: 90), and 18% a high level of well-being (N: 27). The comparative thematic analysis of the discourse of these three groups allows us to identify experiences that are favorable and unfavorable to the well-being of young people and to distinguish two configurations of interactions between children and their environment over the first year of the pandemic, namely that of young people who report a high level of well-being and that of those who report a worrying level of well-being. Results highlight the importance of activities, relationships, support, and representations of children and adolescents for their well-being in the pandemic context. Interventions and social measures to better support their well-being are discussed. Keywords Well-being Children Adolescent Pandemic COVID-19 Mixed study Social Sciences and Humanities Research Council1008-2020-1028 Gervais Christine Ministère de la famillePartenariat Familles en mouvanceÉquipe de recherche Paternité, famille et société ==== Body pmcIntroduction Professionals, researchers, and decision-makers are showing a growing interest in the well-being of children (Ben-Arieh, 2010) as children themselves represent and experience it (Amerijckx & Humblet, 2014; Gorza & Bolter, 2012) and are recognizing children’s right to define what well-being means to them (UNICEF Canada, 2019). In response to this interest, the past 20 years have seen a growth in studies looking at the subjective well-being of children (Fattore et al., 2019). A multidimensional concept, subjective well-being includes a cognitive dimension, which refers to children’s overall assessment of their lives; an emotional dimension, which looks at children’s moods and feelings; and a psychological dimension, which includes a sense of fulfillment, a positive vision of the future, and satisfaction with people’s responses to their psychological needs (Casas, 2011; Kaye-Tzadok et al., 2019). Since 2020, the restrictions imposed all over the world in response to the COVID-19 pandemic have brought up major concerns regarding the well-being of young people within the scientific community and among political decision-makers, workers, parents, and young people themselves. Indeed, restrictive measures such as lockdowns, social distancing, the closing of schools and shops, bans on gatherings and sports activities, and the wearing of masks are transforming young people’s everyday lives, depriving them of certain kinds of stimuli that are essential to their development (interactions with their peers, physical proximity to their friends, and access to school) (Gervais et al., 2020; Stoecklin et al., 2021) while exposing them to new stresses and unexpected circumstances (concerns with regard to illness, contamination risk, and so on) (Coyne et al., 2020; Fegert et al., 2020). Various studies have shown a deterioration in children’s and adolescents’ mental health since the pandemic began. Mostly carried out in the West during the first weeks of pandemic-related lockdowns, these studies identify a greater prevalence of depression and anxiety symptoms (Elharake et al., 2022; Hussong et al., 2021; Luijten et al., 2021), increases in sleep disturbances (Luijten et al., 2021) and behaviors and feelings related to loneliness and boredom (Shoshani & Kor, 2021). These studies also identify certain risk and protection factors to explain the variations in well-being among young people in the context of the pandemic. In keeping with the multi-level framework of child well-being developed by UNICEF (2020), these factors are internal to children themselves but are also found in the systems in which they interact. The multi-level framework of a child’s well-being, developed to allow international comparisons, identifies four levels of factors influencing children’s well-being. First, individual indicators represent physical and mental health antecedents, as well as emotional skills such as coping strategies. Second, the world of the child includes activities, for example, outdoor activities and screen time. It also comprises support provided by peers, school staff, and family members as well as children’s participation in decision-making at school and at home. Third, the world around the child consists of community support provided to parents, work-family balance, school resources, and access to facilities to play. Finally, characteristics of the world at large, such as the economic, environmental, and societal context, as well as health, education, and health policies, may shape children’s well-being. In the context of the pandemic, those factors may be deployed differently compared to other times. Children’s Individual Characteristics Contributing to Their Well-Being During the Pandemic Regarding young people’s characteristics, fear of the COVID-19 virus (Engel de Abreu et al., 2021), fragility or mental health problems that predate the pandemic (Shoshani & Kor, 2021), and the adoption of disengagement and emotion-focused engaged coping strategies (Hussong et al., 2021) all act as risk factors for children’s well-being. In contrast, having a high sense of self-efficacy (Hussong et al., 2021), feeling useful and able to face the situation (Shukla et al., 2022), and choosing problem-focused engaged coping strategies (Hussong et al., 2021) seem to act as buffers to limit the negative effects of the pandemic on children’s and adolescents’ well-being. Dimensions of the World of the Child Contributing to Their Well-Being During the Pandemic Studies highlighted activities and relationships that are essential for young people’s well-being in the context of the pandemic. Regarding activities, the suspension of extracurricular sports and cultural activities (Kutsar & Kurvet-Käosaar, 2021; Stoecklin et al., 2021), the perception of lacking freedom (Engel de Abreu et al., 2021), a high volume of homework, and the fact of not being able to attend school in an ongoing way (Ravens-Sieberer et al. 2021) are associated with alterations to well-being. In contrast, access to outdoor spaces, discovering new creative activities (Berasategi et al., 2021), regular routines during lockdown (Shoshani & Kor, 2021), less screen time (McArthur et al., 2021), and an appreciation of a less busy schedule and a slower pace of life (Stoecklin et al., 2021) seem favorable to well-being. Studies have also highlighted the essential role of young people’s relationships for their well-being during the pandemic. Indeed, connectedness to caregivers (McArthur et al., 2021), children’s attachment security with regard to their parents (Dubois-Comptois et al., 2021), the positive climate in the family (Ravens-Sieberer et al., 2021), adolescents’ satisfaction with the way adults listen to them (Engel de Abreu et al., 2021), and perceived social support (Ravens-Sieberer et al., 2021; Shoshani & Kor, 2021) are associated with better well-being, while family conflicts (Ravens-Sieberer et al., 2021) and separation from or loss of contact with friends represented significant difficulty during lockdown (Stoecklin et al., 2021). Contributions of the World Around the Child to Children’s Well-Being During the Pandemic Studies examining the world around the children also identified some risk and protective factors for their well-being in the context of the pandemic. More specifically, living in a single-parent family, in a family with three or more children (Luijten et al., 2021), in a family with less educated parents (Ravens-Sieberer et al., 2021), and dealing with precarious conditions (Engel de Abreu et al., 2021) are identified as risk factors. Moreover, parental strain and stress (Essler et al., 2021), negative changes in a parent’s work situation (Luijten et al., 2021), and parental mental health problems (Gervais et al., 2021; Kiss et al., 2022) are associated with an exacerbation of the pandemic’s negative effects on young people’s well-being. Contributions of the World at Large to Children’s Well-Being During the Pandemic While the pandemic profoundly transformed the economic, social, and environmental contexts in which children lived, the influence of these factors on children’s well-being has been less studied. Some studies have nevertheless highlighted how the transformation of child protection services to respect lockdown and social distancing hampered the protection of children, increasing the risk for child maltreatment and family violence (Katz et al., 2021). Moreover, insufficient governmental programs to support families and economic hardship were also identified as important risk factors for children’s well-being during the pandemic (Brooks et al., 2020; Thibodeau-Nielsen et al., 2021). While these results help provide an understanding of the factors that affect young people’s well-being, they mostly rely on data gathered in the first weeks of the pandemic using cross-sectional studies (Elharake et al., 2022) as well as on quantitative data gathered from adults who are part of children’s and adolescents’ lives rather than from young people themselves. As the pandemic has extended over time, many have underscored the need to take a longitudinal approach to studying its effects on children (Elharake et al., 2022; Essler et al., 2021; Holmes et al., 2020) and to include the voices of children and adolescents regarding what they perceive as being important for their well-being (Axford et al., 2014) in the context of the pandemic (Andrés et al., 2022; Berasategi et al., 2021). The Current Study This is the context in which this longitudinal study, inspired by the Children’s Understandings of Well-Being (CUWB) research protocol (Fattore et al., 2007, 2019), aims to identify how different experiences contribute to children’s and adolescents’ well-being through different stages of the pandemic. Based on the narratives of 149 Canadian children and adolescents regarding their experiences of the COVID-19 pandemic, it compares the statements of children who report a high, average, or poor quality of life in order to answer the research question: which experiences during the pandemic contributed to young people’s well-being over time? Methods This study relies on a mixed concurrent longitudinal design with a qualitative preponderance (Creswell et al., 2011) including three measurement times (T1: May 2020 lockdown, T2: July 2020 progressive reopening, and T3: November 2020 second wave) to capture children’s experience of the pivotal moments of the pandemic: the strict sanitary measures of the first lockdown, the progressive reopening following the first wave of the pandemic, and the return of strict sanitary measures in the second wave as we collectively realized that the pandemic would last much longer than envisioned in the early months. Anchored in a child-centered approach, this study looks at the experiences and perspectives of young people, considering their point of view on pandemic-related issues to be essential to our understanding of these issues (Côté et al., 2020a; Greene & Hill, 2005). It rests on research methods and tools developed in collaboration with a group of eight young people who acted as expert advisors throughout the study. Their involvement ensured that the questions were relevant and tailored to the realities and concerns of children and adolescents and that they were conducive to young people’s participation (Mayne et al., 2018) while respecting COVID-19 restrictions. Participants The sample is composed of 149 young people, 92 girls, and 57 boys from the province of Quebec, Canada, who took part along with their parents at the study’s three measurement points. They were aged seven to 17 (mean = 11 years old, SD = 2.52). They mainly lived with two parents (85% of them) and with siblings (54% have a brother or sister, and 43% have two or more). They lived in relatively favorable socioeconomic conditions; 82% had a parent who had completed university studies, and 62% benefited from a family income above US $95,000 per year. Study Design and Procedure The participating families were recruited during the first lockdown (April 2020), while schools, services, and businesses were closed, and going out was forbidden except for essential service workers (INSPQ, 2022). Recruitment was done through social media networks (Facebook) and through the newsletters of family-focused community organizations. To take part in the study, families had to have at least one child aged between seven and 17, have access to an internet connection, and be able to understand and communicate in French. Parents interested in the study first filled out an online questionnaire about their general situation (sociodemographic) and their state of health and functioning, and they consented for us to speak with their child. The children and adolescents were quickly contacted by a team member, and appointments were set for semi-directed Zoom interviews. At key moments of the pandemic, an email was sent to the parents to ask them and their children to participate once again. The data for the second measurement point were gathered in the month of July 2020 while most safety restrictions in Québec were loosened. Small gatherings were permitted at the time, and businesses and day camps were opened. The data at the third measurement point were gathered during the pandemic’s second wave, marked by the return of a number of restrictive measures. Extracurricular activities and sports were prohibited; students aged 14 and up were attending school one out of every 2 days; many classes and schools had to close for a week or two due to outbreaks, and restaurants and businesses were closed. To favor participant retention, which is always a challenge in longitudinal studies (Gifford et al., 2007), various strategies were put into place, including individualized thanks to each parent after the interviews with their children, the assignment of a specific assistant to each child participant in order to foster the development of a trusting relationship and coherence within the data gathered, and sending parents the journal articles published using the results. One $10-value reward per measurement point was given to each child at the end of their study participation (after the third measurement point) in the form of a gift card. Through a random draw, we also distributed four $50 gift cards per measurement point to the parents. Research Tools Qualitative Data The semi-directed interviews were done at each measurement time using the Zoom app. Lasting about 45 min each, they focused on scaffolding situations that promoted the children’s active involvement and the production of an explanatory discourse (Horgan, 2017), maximizing their contribution to knowledge building (Kirk, 2007). Young people’s knowledge and experience regarding the pandemic were discussed at each measurement point, along with their school experience, their family and social relationships, their coping strategies, and their vision of the future. Children were first asked a general prompt, “Has your daily life changed related to COVID-19 and the newly implemented measures? How did it change?” Interviewers asked questions related to family members (“How is it going with your mom, father, sisters or brothers?”), their friends (“How is it going with your friends?”), the school (“Can you explain how things are going with school?”), activities (“What kinds of activities have you been doing for the past weeks?”), and coping strategies (“When you face difficulties during the pandemic such as the ones you described earlier [interviewers named the difficulties], what do you do to feel better?”). They also asked questions related to the expected end of the pandemic (“How do you imagine your life when the pandemic is over?”). At each subsequent time measurement point, interviewers asked the same questions and added questions about how it had changed from the previous interviews, reminding the children which COVID-19 measures have changed compared to the other time measurement point (for example, since the lockdown ended, since you went back to school, or since you have been doing school online). Interviewers asked follow-up questions to encourage children to expand their answers, such as “Can you tell me more about that?” or “What do you consider more difficult or positive in that situation?” Interviewers also used techniques to encourage children to talk, such as saying “uh-uh” and “interesting.” Quantitative Data We assessed young people’s well-being with the KIDSCREEN 10 at the end of the third interview. A research assistant shared their screen through the Zoom platform to facilitate children’s understanding. Children indicated to the research assistant the answer that best represented their life. KIDSCREEN, a widely used questionnaire, measures an individual’s perception and subjective evaluation of their health and well-being within their unique cultural environment (Ravens-Sieberer et al., 2014). Items refer to the last week and are answered on a five-point Likert-type scale assessing frequency or intensity (i.e., have you felt lonely?) (a = 0.61). To allow comparison with international norms and to categorize the children, T-scores were calculated in accordance with the recommended KIDSCREEN 10 general health-related quality of life (HRQoL) scoring procedure (The KIDSCREEN Group Europe, 2006). Ethical Consideration The study was approved by the Université du Québec en Outaouais's ethical board. The parents who consented to take part in the study and to have their child take part, filled out a consent form on LimeSurvey at each measurement point. During the interviews with the young people, a tailored assent form developed by Côté et al (2018) was used to obtain their assent. Pictograms explaining the main consent-related issues (confidentiality and its limits, the right to withdraw, the benefits and risks of taking part, and free participation) were projected and discussed with each young person. The children age 7–13 years old gave their verbal consent to take part, while adolescents aged 14 and up filled out a consent form on LimeSurvey. In keeping with the child-centered approach, the research assistants also received training on the ethical issues of research with children in order to become aware of the power and authority issues inherent in the researcher-child relationship and to limit these by taking a transparent and benevolent approach (Côté et al. 2020b; Gallagher, 2009). Data Analysis Considering the scope of the material gathered and the restricted resources of the study, we prioritized data analysis that combined field notes and interview transcripts (Tessier, 2012). Immediately after each interview and using an outline developed for this study, the interviewer wrote a summary of the child’s related experiences, maximizing the reliability of the information gathered. These summaries (N = 192 at T1, 165 at T2, and 154 at T3) served as field notes and were subjected to a content analysis (Bardin, 2013) in order to categorize them based on the diversity of the participants’ personal situations (age, sex, region, time of return to school), the experiences described by the participants, and the richness of their statements. The interviews presenting contrasting experiences were transcribed word for word (N = 64 at T1, 70 at T2, 68 at T3). A coding tree was then developed in an inductive way based on the summaries. This was discussed by the team, then refined when the first transcripts were coded. The material as a whole was then subjected to a thematic analysis (Paillé & Mucchielli, 2016) using the N’vivo program and an expert coding strategy. As such, three teams (each including two research assistants and a researcher) were formed, each one in charge of coding two mother nodes and their sub-nodes, with each interview thus being read by three research assistants (one from each team). Coding comparisons were made regularly throughout the process (about 10% of the interviews) to ensure the coherence and reliability of the coding. The team then discussed the coding along with the meaning of certain excerpts, their belonging to certain codes, and the links between the different codes. The quantitative data were processed using the SPSS Statistics version 28 (IBM Corp., 2021). The scores obtained on KIDSCREEN were then compared to population averages (The KIDSCREEN Group Europe, 2006), and the young people were categorized based on what they reported for the third measurement point (November 2020, during the pandemic’s second wave) about their quality of life related to health, whether it was low, normal, or high; this was considered as their level of well-being. Crosstab in N’vivo made it possible to compare the experiences of these three groups of young people at the different times of the pandemic and to distinguish between the experiences that were favorable and unfavorable to their well-being. Result The presentation of the results is organized based on UNICEF’s multilevel framework of child well-being (UNICEF Innocenti, 2020). As such, we first present the well-being of participating children and adolescents, conceptualized as the result of their experience throughout the first months of the pandemic. Then, we describe the participants’ experiences related to the world of the child, the world around the child, and the world at large. For each of these spheres, we first set out common experiences, then describe the differences discerned within the children’s discourses when they reported a high level of well-being (group A) and a low level of well-being (group C). Young people from group B, who report a “normal” level of well-being, shared some experiences from each of these other groups. Since no specific experiences of this group could be identified, their narratives are included in the description of common experiences of children that introduce each theme, but their experiences are not contrasted with the ones of the two other groups of children. Children’s Well-Being The average level of well-being reported by youth participants was 51.83/100 (35.6–83.1, SD = 8.34), which is slightly higher (0,18 standard deviations) than the European norm data for children and adolescents (T = 50, SD = 10,0) (The KIDSCREEN Group Europe, 2006). Participants were categorized into three groups based on whether they reported a high level of well-being (group A, N = 27, 18%), one that was similar to the population average which is considered normal (group B, N = 90, 60%), or one that was low (group C, N = 32, 22%) during the third measurement period. World of the Child: Activities, Relationships, and Support Investing in Solitary Activities or Shared Activities The activities in which the young people invested time took up a lot of space in their experience of the first lockdown (T1). This period—synonymous with a sense of respite and freedom for some and of emptiness and boredom for others—gave them an opportunity to do activities other than the ones they normally would. For most of the young people, games and activities seemed to be a major source of well-being during the pandemic (T1, T2, and T3). Groups A and C invested in different types of activities. The young people of group A mainly engaged in activities with members of their immediate families, particularly with their siblings. Among others, they practiced outdoor sports and played board games as a family to stay busy and get their mind off things throughout the pandemic (T1, T2, and T3).“To feel better, I think it’s really about spending time with people, my family, playing board games. Because it seems like focusing on a specific activity with other people, and being in the moment with other people, everyone forgets a bit about what’s going on, you get the feeling like you’re living a normal life, and it feels good.” (Hayden_T1_17 years old) To deal with the limits imposed by the lockdown (T1 and T3), young people in group A also organized online activities with their friends and extended family members, for example, cooking on FaceTime with their grandmother or playing online games with friends, which helped them feel better. In contrast, the young people from group C reported that they mostly did solitary activities throughout the pandemic (T1, T2, and T3). Walking was the activity this group named most frequently, which allowed them to get outside, escape the family home, reduce the intensity of negative emotions, and find a better sense of well-being.“With the lockdown I realized that if I didn’t do anything, sometimes I felt worse, and going outside, like for example taking a walk outside every day, well, I felt better, y’know it was like I always needed to get some fresh air.” (Blanche_T1_12 years old) When they were going through more negative experiences during the pandemic, many young people in this group attempted to distract themselves by listening to music or watching videos or TV. Others spoke about finding comfort with their pets, expressing themselves through artistic creations, or playing alone, such as with Lego. Maintaining Relationships with Loved Ones The public health measures imposed to prevent COVID-19 have forced young people to redefine their interactions with their social networks. They were confined to their homes in the first months of the pandemic (T1), thus increasing the time spent with their immediate families while their contacts with friends and members of their extended families were rare or suspended, as explained by Laura–Marie (T1, 17 years old): “Honestly, they’re the only people I can have fun with right now and talk with. The only social interactions I have are with my family here.”. The majority of young people recounted that this period was a source of closeness with members of their immediate family, which they explained in particular by their parents’ greater availability due to the suspension of their work or to the requirement that they work from home. The scope and intensity of these family moments also generated tensions and conflicts with siblings and parents, mainly due to the lack of personal space during the lockdown. Similar interactions were reported during the second lockdown (T3). Young people used various instant messaging apps to maintain their ties with friends and extended family members during the first and second lockdowns (T1 and T3). However, these contacts appear to have been insufficient, as many children said they missed their loved ones during the pandemic’s first and second waves, as well as the fear to lose some of their extended family members who might contract and possibly die from COVID-19. Beyond these common experiences, young people from groups A and C reported certain differences in regard to their relationships with the people around them. As such, children in group A insisted that during the first lockdown (T1), it was important to support their grandparents despite the distance, as Alexane explained: “We still need to support and give love to the people who need it. Like to our grandparents, we need to call them and make sure they’re not, like, all alone in this” (Alexane_T1_12 years old). This concern led them to contact these loved ones regularly by telephone or to find alternative methods for seeing them in a safe way. However, the importance of giving support back to their grandparents was less present in the youths’ discourse during the second lockdown (T3). During the period of gradual reopening (T2), these young people expressed their relief and joy at once again seeing their loved ones “for real” in the summertime. Many of them chose to stop respecting social distancing in order to hug their friends and extended family members, while still applying other safety measures (hand-washing, coughing into their elbows, and so on). As children were starting to go out again and spending less time with their families, many of them said that tensions with their immediate families were significantly reduced. However, they kept up the outdoor activities they shared with their family members which began during the lockdown and which they still enjoyed. Others instead reported that their parents became less available as they went back to work, which was disappointing to them even if they said they understood their parents’ commitments. The children of group C, however, sacrificed their desire for closeness with their grandparents in order to protect them, both during the lockdown period (T1) and during the gradual reopening (T2) as well as the second lockdown (T3). Many see themselves as a danger to their relatives. Their narratives insisted on their fear of transmitting the virus to their grandparents as well as on their worries about the possible consequences of the virus on their loved one’s health and lives.“I try to hug them as little as possible, especially my grandmother. And my grandmother comes to give us hugs, but my mother said not to give any to her. It hurt that I couldn’t hug her, but I was careful because I really didn’t want to give it (COVID) to her.” (Amber_T3_10 years old) During the gradual reopening period (T2), the young people in this group expressed greater ambivalence about their parents’ return to work, as they were less available, which amplified the young people’s sense of solitude. Benefiting from Reliable and Diversified Support Regardless of their level of well-being, the young people described being supported by their parents during stressful times throughout the pandemic (T1, T2, and T3). Parents listened to them attentively; young people felt they could talk about the emotions they were feeling about the pandemic, and their parents shared hugs with them; these things consoled and reassured the children in more emotionally difficult moments. When necessary, the parents also offered advice to the children to help them solve problems. Friends were also another major source of support, helping young people to get their minds off things and be entertained. Beyond these general observations, some differences were observed with regard to the scope and forms of support that young people received. Young people from group A talked more about the emotional support provided by their parents in the form of kind words and markers of affection, such as hugs, which contributed to their sense of being loved and soothed during the lockdown period (T1) and the reopening (T2), as explained by Michèle (T1_11 years old): “My mother shared a lot of love. We were there for her, and she was there for us. We gave her hugs and kisses every day when she got home from the clinic.” Young people in this group were also distinguished with regard to having space for family conversations to talk about the COVID-19 pandemic. They described talking about their worries about COVID-19 with their parents, who listened to them, added nuance or filled out missing information, and helped them readjust their perception of the virus if needed, thus reducing their anxieties. Many children reported as well that their parents reminded them about safety measures to apply throughout the pandemic (T1, T2, and T3), which made them feel protected. Lastly, some young people in this group also mentioned the emotional support they received from their friends during the pandemic, in particular in the form of listening and making space for their emotions. While children from group C also said they reached out to their parents when they needed support, they spoke much less about the support they received during the first lockdown (T1), the gradual reopening (T2), and the second lockdown (T3). As well, very few of them reported talking about COVID-19 with their parents during the pandemic. Particularly during the second wave (T3), young people in this group explained that their parents were busy with their work, which led them to seek out solutions by themselves in order to feel better.“When I felt bad, I thought, ‘OK, I’ll stop working and take a moment to listen to some music.’ These last few months, I didn’t have many friends, so I couldn’t call anyone. My parents were working, so it was kinda just me for myself.” (Charlie_T3_15 years old) The emotional support provided by friends was also largely absent from these children’s responses. World Around the Child: Anticipating and Appreciating the Return to School Despite a great diversity in the stories children told about their school experience, the majority of participants were impatient to return to in-person schooling. Many of them emphasized the importance of the social aspect of their school experience: they missed their friends and felt a lack of social interactions during the lockdown periods (T1 and T3). The return to class, for many of them, meant reconnecting with their friends and starting to return to normal, as explained by two respondents: “It let me socialize, because for me, my parents wouldn’t have forced me to go to school, but I’m the one who wanted to.” (Sonya_T2_8 years old), “I was happy to go back to school, I was tired of staying home and not seeing anyone.” (Victor_T2_11 years old). However, the young people envisioned and experienced the transition back to in-person school in different ways. The young people from group A had a very positive expectations of their return to the classroom during the first lockdown (T1). Their responses included no fear of returning to school, which was described as a positive experience. They appreciated that the number of students in each class was reduced, which let them participate more, receive more support from their teachers, and accelerate their pace of learning. Excitement and relief were key elements of these young people’s experiences when returning to school (T2), as one explained: “I was happy and relieved to go back. When I have to miss school for two weeks, I miss it, so not going for six months was very long” (Talia_T2_10 years old). The children from group C reported more difficult experiences in returning to school. Many of them spoke about having understood there would be changes required by safety measures at their school, as well as the impacts of remote schooling on their learning during the first lockdown (T1). Stress and anxiety characterized the memories many children had of this period, such as those of Léon (T3_14 years old): “I wasn’t feeling it. It was a huge anxious period, and still is now. It was crazy, it was really a handicap in my life.” Some young people mentioned their fear of contracting COVID-19 and thus presenting a risk to their more vulnerable peers. After several months during which contact was forbidden, the children and adolescents of this group seemed to perceive social interactions as being dangerous during the gradual reopening (T2). While some of them had a harder time adapting to the social aspect of returning to the classroom, the rules put into place in the schools gave these young people a sense of safety and trust at the time of the second wave (T3).“Introducing masks and hygiene procedures went well. (...) If we hadn’t respected the measures, there may have been a lot more cases of COVID-19. They did it above all for our benefit.” (Inès_T3_16 years old) Some participants in this group also decried the impacts of the lockdown periods as well as distancing measures at school on their pleasure at school and on the quality of their friendships. They complained of classroom bubbles and their school’s approach to organizing recreations, which determined and limited the friends they were allowed to play and interact with during the second wave (T3). The majority of them still reported having quickly learned to live with the virus in the school setting and having gradually gotten used to the new measures in place.“I felt a little disoriented, but I didn’t really know what to do because there were tons of other rules, but I ended up getting used to the zones, and I ended up complaining a bit too, it kinda sucked.” (Milo_T3_9 years old) World at Large: Perceiving the Consequences of the Pandemic Throughout the three measurement times, the participants’ stories bore witness to the major effects on their lives of the public health policies and safety regulations implemented in response to COVID-19, particularly when they spoke about their experiences of the first lockdown (T1). As such, participants spoke at length about the lack of contact, both physical and social, the accumulation of deaths, the positive impact of the lockdown on the environment, and job losses. Beyond these global effects noticed by the young people, children in group A were distinct in terms of their nuanced view of the consequences of social and health measures, while those in group C shared a more negative understanding of them. The young people from group A also noted the positive consequences of the pandemic for the population and for individuals. They highlighted how the lockdown gave them an opportunity to discover new interests and spend time on activities they enjoyed but had previously neglected: “It changes people’s morale, we weren’t doing the same activities, we were using the car less and we got out for walks more” (Tyler_T1_13 years old). Other young people in this group noted that the pandemic helped people realize the importance of their families.“You realize that your family is important, for example right now we can only really see our close family. So in normal times, it’s important to spend time with your (extended) family because in times like this (lockdown) you can’t always see them.” (Eden_T1_13 years old) In contrast, the participants of group C identified more negative impacts of the pandemic for the whole population. They were concerned about the consequences of the lockdown on the population’s mental health, explaining that “It can create kinds of mental problems. Not problems, maybe, but it has an effect on the mental state” (Léon_T1_14 years old) and saying, “It can make people unhappier, people are losing their loved ones and some people might die” (Lara_T1_9 years old). Others underscored how the pandemic led to changes in routine that threw off their balance: “So, well, it changes pretty much everything. Your routine, you can no longer go to the grocery store with your parents. We don’t leave home anymore. Yeah. It’s a little destabilizing. Well, not just a little, it’s really destabilizing” (Céleste_T1_15 years old). They also shared their worries about elderly and dying people: “Some elderly people, sometimes it might worry them or stress them out… They might really feel bored because they can’t see their families” (Sasha_T1_10 years old). “It prevents families from seeing their dying loved ones one last time” (Charles-Antoine_T1_12 years old). During the gradual reopening and the second lockdown, children and adolescents talked less about the consequences of COVID-19 on society. Discussion A good childhood is often defined as one in which children have a positive experience with themselves and their environment, as well as the prospect of a good future (Ben-Arieh & Frønes, 2011; UNICEF Innocenti, 2020). Since the beginning of the pandemic, many scholars and health professionals have raised concerns about the threat that the state of health emergency represents for these two dimensions (Gervais et al., 2022; Prime et al., 2020) while children’s experience of it has still not been widely studied. In this sense, the results of this study are valuable because they provide a better understanding of the positive and negative effects on children’s well-being over the course of the pandemic. Although the research design did not make it possible to identify causal relationships between the children’s experiences and their level of well-being, our analyses shed light on two configurations of interactions between children and their environment that contributed to their well-being over the first year of the pandemic. On the one hand, the world of the children who presented a high level of well-being in the second wave of the pandemic is characterized by their investment in shared activities with their family members and their friends, thanks to communication technologies, which were present and available from the first major lockdown. We also note that these young people maintained close relationships with their loved ones despite the health measures in place during the different measurement periods: their responses were structured around the importance of supporting their grandparents and maintaining their ties with them despite restricted contact, and they noted their pleasure in seeing them again when health measures were relaxed. Their narratives reflect the readily available support that was facilitated by better work conditions for parents during the pandemic: they spoke of receiving strong support from their parents and having space to talk about COVID-19 and its issues within their families. With regard to the world around the children, the young people in this group looked forward to returning to school in a very positive way and appreciated it despite the restrictions enforced in the school setting and the higher risk of COVID-19 exposure. Lastly, these children’s experience of the world at large is distinguished by their nuanced understanding of the pandemic’s consequences; these young people underlined the things they learned and the closeness they developed because of the health measures just as much as the losses and boredom they experienced. On the other hand, the world of the children that showed lower well-being was characterized by involvement in mostly solitary activities, although we cannot determine whether this was due to preference or to their loved ones’ lack of availability for shared activities. They also expressed a fear of infecting others or being infected by others that contributed to weaken their bonds, especially with their grandparents and their social interactions. The support received from their loved ones was much more limited in these children’s experience during the first months of the pandemic, especially from their parents. The young people report that they needed to ask for support and perceived their parent’s availability as limited due to their work commitments so they mainly tried to manage by themselves. Apprehension and lack of pleasure regarding the return to school as well as disappointment and lack of control regarding the prevailing measures are at the heart of their experience of the world around them. Lastly, these children’s experience of the world at large was permeated by their concerns about the negative impacts of the pandemic, namely for the survival of the elderly and the mental health of the population. Briefly, the personal and family boundaries of these young people seemed to be more permeable, with the pandemic having a greater effect on the way they saw the world around them. While comparing the perspectives of children with high and low levels of well-being, the results of this study allowed us to identify individual and social factors that are intertwined and contribute both to well-being and to experiences of the pandemic. Indeed, the results underscore the resilience of the families we interviewed, meaning their capacity “to adapt successfully to significant challenges that threaten the function, viability, or development of the system” (Masten, 2018, p. 16). The fact that 78% of the children and adolescents we interviewed reported a high level of well-being, comparable to or higher than population averages, after several months of health measures that transformed their everyday lives and deprived them of many interactions and activities shines a light on the agency of children and the resilience of families (Masten & Motti-Stefanidi, 2020). The experiences they recounted in our study also emphasize the richness that families can produce in terms of shared experiences, the creation of meaning, and mutual help, which are processes favorable to family resilience (Maurović et al., 2020) and act as protective factors for the well-being and development of all family members (Prime et al., 2020). The children’s viewpoints described in this study also highlight the importance of the social support children received but also that which they perceived. Indeed, in the context of the pandemic, parents’ involvement in shared activities with their children was a form of social support, limiting boredom, and promoting enjoyment. Moreover, parents’ ability to be available was noticeably present in children’s discourse, which suggests that it created a reassuring environment for them (Gervais et al., submitted) despite the uncertainty and unpredictability of the world around them and despite the pandemic’s many impacts on their own mental health (Gassman-Pines et al., 2020; Russell et al., 2020). Some children also pointed out the interinfluence of their parents’ availability and their working conditions. Combining work, family and homeschooling added additional challenges for parents, which might have impacted their own well-being and consequently their availability to reassure children who felt insecure during the pandemic (Cusinato et al., 2020). These findings also provide a better understanding of coping strategies that may contribute to children’s experiences and well-being during the pandemic. Contrary to the study by Vera et al. (2012), which concluded that conventional coping strategies, such as seeking social support, may not buffer adolescents from the effects of stress on subjective well-being, children and adolescents reporting higher well-being in our study largely reported seeking social support to feel better during the pandemic. It seems that despite the uncontrollable nature of pandemic-related stress, solution-oriented coping strategies—such as asking for help, seeking information about the pandemic by talking with parents, and supporting grandparents from afar—acted as buffers and protected young people’s well-being, as suggested by other studies (Hussong et al., 2021). Moreover, while the study by Dominguez-Alvarez et al., (2020) concludes that avoidance and distraction strategies are associated with different indicators of maladjustment (2020), our results indicate that the majority of the young people we interviewed adopted avoidance and distraction strategies, which they describe as being activities that helped entertain them. Shared family activities, such as games and sports, seemed to provide a beneficial distraction to the young people who reported a high level of well-being, as proposed by Orgilés et al. (2021), while solitary distraction activities, such as watching TV and playing with Lego, were observed for children with lower well-being. Lastly, cognitive restructuring more frequently used by children who were minimally impacted by the pandemic also appears to be favorable to their well-being, as consistent with the study by Orgilés et al. (2021) and discerned in the capacity of the young people in group A to look forward to returning to school and perceiving positive impacts of the pandemic. Lastly, these data suggest that many dimensions of the subjective well-being of children and adolescents continue to be articulated in a similar way in the context of the pandemic and of restrictive health measures. In keeping with the study by Gonzalez-Carrasco et al. (2019), which identified that feeling afraid of being alone at home or of death was associated with lower levels of well-being among adolescents, the solitude experienced at various moments in the pandemic and the idea that social contacts were now a threat to themselves and their relatives were harmful to young people’s well-being. As well, the social dimension is key to the different experiences of the two groups of children. In this study, care, gratitude, reciprocity, and support within the parent–child relationship were more strongly present in the experiences of young people who reported greater well-being, which is consistent with prior studies of child well-being (Fattore & Mason, 2017; Newland et al., 2015; Tadić Vujčić et al., 2019). Our study outlines different stress experiences regarding the absence of contact during school closure and the increase of contact due to the return to in-person classes. In keeping with the studies by Schwartz et al. (2021) and Viner et al. (2022), showing that pandemic-related stress during the school closure and reopening was negatively related to children’s well-being, our results suggest that school experiences are indeed central to children’s well-being (Newland, 2015). Our study contributes to knowledge by setting out the ways in which children’s representations of global phenomena that change the world around them, such as the pandemic, are also associated with their well-being. Managing to perceive the consequences of this event in a nuanced way by also noting the positive effects of health measures seems to be a protective factor for young people’s well-being. Implications for Practice This study’s findings suggest several practical implications for social, education, and health professionals working with children, adolescents, and families. First, the importance of family interactions, shared moments, and support received from parents underscores parents’ role in young people’s well-being in the context of the pandemic. With this in mind, we encourage professionals to raise parents’ awareness about the listening, acceptance, and emotional validation they can provide to their children, as well as about the importance of creating spaces and times for talking about COVID-19 to foster young people’s understanding of the issues related to the pandemic and create nuanced family meanings about them. These data also underscore the need for school workers to broaden their definition of vulnerability in order to discern children whose well-being has been more severely compromised by the pandemic and to better support them. The experiences related by the young people in this study illustrate that beyond well-documented vulnerability factors (such as poverty, housing instability, family violence, and so on), the experiences young people have had during the pandemic may contribute to their vulnerability, including children living in privileged contexts (Lafantaisie et al., 2021). Given the central aspect of social relations in the experiences that contribute to children’s well-being, particular attention should be paid to children and adolescents whose lives have already been marked by relationship breakdowns and who receive little support from their immediate families. We also call upon decision-makers to put actions into place to render the world of children and the world around children favorable to their well-being. In particular, parents’ availability for their children, essential to their well-being in the unstable period of this pandemic, should be encouraged. Without supportive measures, we fear that parents’ adaptive resources, heavily solicited over the last 2 years, may run out, and their parenting practices may deteriorate. Lastly, creating more spaces and opportunities for children’s participation is essential. Children’s participation in the decisions that affect them, which was already insufficient, has dropped even further in reaction to the lockdown measures that limited their access to the spaces and people that supported their participation. Many of them experienced this lack of participation as a lack of consideration and interest on the part of leaders and society toward young people, who were heavily impacted by health measures even though the virus presented a very low risk to their health (Gervais et al., 2022). To dismantle this impression, young people should be included in all decisions about the services and activities that are aimed at them and that will be put into place to foster post-pandemic reopening and to orient what “normal life” will look like. Strengths and Limitations Despite the large sample size and the diversity of age and location of the study’s participants, the privileged conditions of the young people we interviewed constitute a limit in this study. Indeed, our respondents had low exposure to COVID-19, lived mostly in fairly well-off two-parent families, and generally had parents with a high level of schooling who were able to work from home and thus be present for them during the months when they were not at school. The level of well-being reported by the young participants was higher than that of young people living in Europe during the same time period according to the study by Ravens-Sieberer et al. (2021), and the experiences they shared all bear witness to their privileged context. Their experiences are doubtless different from those of children who faced great precarity during the pandemic, for whom family activities were probably less frequent, and who saw fewer positive impacts of the pandemic. The combined analysis of the field notes and transcribed interviews is also a limitation of the study, as the adult interviewer’s perspective may have seeped into the field notes, blurring the children’s perspective on some of the experiences. As well, the relationship developed between the young people and their interviewer over the course of their meetings may also have introduced a social desirability bias, with young people wishing to give a good impression of themselves and largely recounting their positive experiences during the pandemic. This relationship, however, also constitutes one of the study’s strengths, having contributed to a high retention rate over the course of the various measuring periods, helping the children feel safe and listened to, and helping them express themselves. The mixed longitudinal design is also one of the study’s strengths, as it allowed for the comparative analysis of the specific experiences of different groups of children during the different periods of the pandemic. However, it does not allow us to establish a causal link between youths’ level of well-being and their experiences during the early months of the pandemic. Indeed, because their well-being was only measured at the third measurement time, we cannot rule out that differences in well-being levels were already present before or at the very beginning of the pandemic and influenced young people’s experiences, perceptions, and ways of engaging or not engaging with their surroundings. Conclusion This study’s results provide a nuanced look at the experiences children lived through during the pandemic and their contribution to their well-being from their own perspective. The results highlight the complex interactions between children and their environment, revealing how children’s agency, relationships, resources, and context contribute to both their experiences of the pandemic and their well-being. The results also underline the importance of listening to what children have to say rather than solely measuring the negative impacts of the pandemic on their mental health, as well as implementing social measures to support children’s positive experiences and well-being in the context of a pandemic. Acknowledgements The authors would like to thank the participating children and adolescents and the research assistants who contributed to the data collection for the study. Funding This study was supported by grants from the Social Sciences and Humanities Research Council, the Ministère de la famille du Québec, the Partenariat de recherche Familles en Mouvance, and the research team Paternité, famille et société. Data Availability The datasets generated and analyzed during the current study are not publicly available due to the fact that they constitute an excerpt of research in progress but are available from the corresponding author on reasonable 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 Amerijckx G Humblet PC Child well-being: What does it mean? Children and Society 2014 28 5 404 415 10.1111/chso.12003 Andrés, M. L., Galli, J. I., Valle, M., Vernucci, S., Morales, H. L., & Trudo, R. G. (2022). Parental perceptions of child and adolescent mental health during the COVID ‑ 19 pandemic in Argentina. In Child & Youth Care Forum. 10.1007/s10566-021-09663-9 Axford N Jodrell D Hobbs T Objective or subjective well-being? Handbook of Child Well-Being: Theories, Methods and Policies in Global Perspective 2014 Springer 2699 2738 Bardin L L’analyse de contenu 2013 Presses Universitaires de France Ben-Arieh A Rose W McAuley C Developing indicators for child well-being in a changing contex Well-Being : Understanding children’s lives 2010 Jessica Kingsley Publishers 129 142 Ben-Arieh A Frønes I Taxonomy for child well-being indicators: A framework for the analysis of the well-being of children Childhood 2011 18 4 460 476 10.1177/0907568211398159 Berasategi N Idoiaga N Dosil M The well-being of children in lock-down : Physical, emotional, social and academic impact Children and Youth Services Review 2021 127 106085 10.1016/j.childyouth.2021.106085 34728873 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 In The Lancet 2020 395 10227 912 920 10.1016/S0140-6736(20)30460-8 UNICEF Canada. (2019). Where does Canada stand? The Canadian index of child and youth well-being. oneyouth.unicef.ca. https://oneyouth.unicef.ca/sites/default/files/2019-08/2019_Baseline_Report_Canadian_Index_of_Child_and_Youth_Well-being.pdf Casas F Subjective social indicators and child and adolescent well-being Child Indicators Research 2011 4 555 575 10.1007/s12187-010-9093-z Côté I Trottier-Cyr R-P Lavoie K Pagé G Marin A Eysermann B Giroux MT « Veux-tu participer à ma recherche? » : Principes, enjeux et stratégies concernant l’assentiment des enfants dans le processus de recherche Recrutement et consentement à la recherche : réalités et défis éthiques 2018 Éditions de l’Université de Sherbrooke 127 145 Côté I Lavoie K Trottier-Cyr R-P La méthodologie de recherche centrée sur l’enfant : Recueillir et restituer la parole des enfants et des adolescent·e·s 2020 Les Presse de l’Université Lava Côté, Isabel., Pagé, G., & Lavoie, K. (2020b). La participation de l’enfant en recherche sociale : Considérations éthiques et méthodologiques. In Isabel Côté, K. Lavoie, & R.-P. Trottier-Cyr (Eds.), La recherche centrée sur l’enfant. Défis éthiques et innovations méthodologiques (p. 17-38). Presses de l’Université Laval. Coyne, L. W., Gould, E. R., Grimaldi, M., Wilson, K. G., Baffuto, G., & Biglan, A. (2021). First things first: Parent psychological flexibility and self-compassion during COVID-19. Bahavior Analysis in Practice, 14, 1092–1098. 10.1007/s40617-020-00435-w Cusinato M Iannattone S Spoto A Poli M Moretti C Gatta M Miscioscia M Stress, resilience, and well-being in Italian children and their parents during the COVID-19 pandemic International Journal of Environmental Research and Public Health 2020 17 22 1 17 10.3390/ijerph17228297 Creswell J Klassen A Plano Clark V Smith K Best practices for mixed methods research in the health sciences 2011 National Institutes of Health Engel de Abreu PMJE Neumann S Wealer C Abreu N Macedo EC Kirsch C Subjective well-being of adolescents in Luxembourg, Germany, and Brazil during the COVID-19 pandemic Journal of Adolescent Health 2021 69 2 211 218 10.1016/j.jadohealth.2021.04.028 Domínguez-Álvarez B López-Romero L Isdahl-Troye A Gómez-Fraguela JA Romero E Children coping, contextual risk and their interplay during the COVID-19 pandemic: A Spanish case Frontiers in Psychology 2020 11 577763 10.3389/fpsyg.2020.577763 33391095 Dubois-Comptois K Suffren S St-Laurent D Milot T Lemelin J Child psychological functioning during the COVID-19 lockdown: An ecological, family -centered approach Journal of Behavioral Pediatric 2021 42 7 532 539 10.1097/DBP.0000000000000935 Elharake, J. A., Akbar, F., Malik, A. A., Gilliam, W., & Omer, S. B. (2022). Mental health impact of COVID ‑ 19 among children and college students : A systematic review. Child Psychiatry & Human Development, 1–13, Advance online publication. 10.1007/s10578-021-01297-1 Essler S Christner N Paulus M Longitudinal relations between parental strain, parent–child relationship quality, and child well-being during the unfolding COVID-19 pandemic Child Psychiatry & Human Development 2021 52 6 995 1011 10.1007/s10578-021-01232-4 34426893 Fattore T Mason J The significance of the social for child well-being Children and Society 2017 31 4 276 289 10.1111/chso.12205 Fattore T Mason J Watson E Children’s conceptualisation(s) of their well-being Social Indicators Research 2007 80 1 5 29 10.1007/s11205-006-9019-9 Fattore T Fegter S Hunner-Kreisel C Children’s understandings of well-being in global and local contexts: Theoretical and methodological considerations for a multinational qualitative study Child Indicators Research 2019 12 2 385 407 10.1007/s12187-018-9594-8 Fegert JM Vitiello B Plener PL Clemens V Challenges and burden of the coronavirus 2019 (COVID-19) pandemic for child and adolescent mental health: A narrative review to highlight clinical and research needs in the acute phase and the long return to normality Children Adolescescent Psychiatry Mental Health 2020 14 20 10.1186/s13034-020-00329-3 Gallagher M Tisdall EK Davis M Gallagher M Ethics Researching with children and young people: Research design, methods and analysis 2009 Sage 11 27 Gassman-Pines A Asanat EO Fitz-Henley J COVID-19 and parent-child psychological well-being Pediatrics 2020 146 4 e2020007294 10.1542/peds.2020-007294 32764151 Gervais, C., Côté, I., & Lalande, D. (2020). Étude Réactions: Récits d’Enfants et d’Adolescents sur la COVID 19. Université du Québec en Outaouais. http://www.partenariat-familles.inrs.ca/wp-content/uploads/2020/07/R%C3%A9actions_Premier-rapport-d%C3%A9tape.pdf. Accessed 15 Mar 2022 Gervais, C., Côté, I., Tourigny, S., Pierce, T., de Montigny, F., & Lafantaisie, V. (2021). Family well-being and child functioning during the first COVID-19 pandemic lockdown: A cross-sectional study. Journal of Child & Family Studies, Submitted. Gervais, C., Côté, I., & Miljus, M. (2022). Étude Réactions : Récits d’enfants et d’adolescents sur la COVID19. Rapport final. Université du Québec en Outaouais. http://www.partenariat-familles.inrs.ca/wp-content/uploads/2022/04/Reactions_Rapport_Final_GervaisCote-compressed.pdf. Accessed 13 Jan 2022 Gifford SM Bakopanos C Kaplan I Correa-Velez I Meaning or measurement? Researching the social contexts of health and settlement among newly-arrived refugee youth in Melbourne Australia Journal of Refugee Studies 2007 20 3 414 440 10.1093/jrs/fem004 González-Carrasco M Vaqué C Malo S Crous G Casas F Figuer C A qualitative longitudinal study on the well-being of children and adolescents Child Indicators Research 2019 12 2 479 499 10.1007/s12187-018-9534-7 Gorza M Bolter F Indicateurs de bien-être de l’enfant, une déclinaison en protection de l’enfance est-elle possible ? Journal Du Droit Des Jeunes 2012 312 2 26 36 10.3917/jdj.312.0026 Greene SM Hill M Greene SM Hogan DM Researching children’s experiences: Methods and methodological issues Researching Children’s Experiences: Approaches and Methods 2005 Sage 1 21 Holmes, E. A., O’Connor, R. C., Perry, V. H., Tracey, I., Wessely, S., Arseneault, L., Ballard, C., Christensen, H., Cohen Silver, R., Everall, I., Ford, T., John, A., Kabir, T., King, K., Madan, I., Michie, S., Przybylski, A. K., Shafran, R., Sweeney, A., … Bullmore, E. (2020). Multidisciplinary research priorities for the COVID-19 pandemic: A call for action for mental health science. The Lancet. Psychiatry, 7(6), 547–560. 10.1016/S2215-0366(20)30168-1 Horgan D Child participatory research methods: Attempts to go ‘deeper’ Childhood 2017 24 2 245 259 10.1177/0907568216647787 Hussong AM Midgette AJ Thomas TE Coffman JL Cho S Coping and mental health in early adolescence during COVID-19 Research on Child and Adolescent Psychopathology 2021 49 9 1113 1123 10.1007/s10802-021-00821-0 33856611 IBM Corp. (2021). IBM SPSS statistics for windows, Version 28.0. Innocenti, U. N. I. C. E. F. (2020). Worlds of influence: Understanding what shapes child well-being in riche countries. UNICEF Office of Research. https://www.unicef-irc.org/publications/pdf/Report-Card-16-Worlds-of-Influence-child-wellbeing.pdf. Accessed 15 Mar 2022 INSPQ. (2022). Ligne du temps de la COVID-19 au Québec. https://www.inspq.qc.ca/covid-19/donnees/ligne-du-temps. Accessed 20 Mar 2022 Katz I Katz C Andresen S Bérubé A Collin-Vezina D Fallon B Fouché A Haffejee S Masrawa N Muñoz P Priolo Filho SR Tarabulsy G Truter E Varela N Wekerle C Child maltreatment reports and child protection service responses during COVID-19: Knowledge exchange among Australia, Brazil, Canada, Colombia, Germany, Israel, and South Africa Child Abuse & Neglect 2021 116 105078 10.1016/J.CHIABU.2021.105078 33931238 Kaye-Tzadok A Ben-Arieh A Kosher H Hope, material resources, and subjective well-being of 8- to 12- year old children in Israel Child Development 2019 90 2 344 358 10.1111/cdev.13130 30125932 Kirk S Methodological and ethical issues in conducting qualitative research with children and young people: A literature review International Journal of Nursing Studies 2007 44 7 1250 1260 10.1016/j.ijnurstu.2006.08.015 17027985 Kiss, O., Alzueta, E., Yuksel, D., Pohl, K. M., de Zambotti, M., Muller-Oehring, E. M., Prouty, D., Durley, I., Pelham, W. E., Mccabe, C. J., Gonzalez, M. R., Brown, S. A., Wade, N. E., Marshall, A. T., Sowell, E. R., Breslin, F. J., Lisdahl, K. M., Dick, A. S., Sheth, C. S., … Baker, F. C. (2022). The pandemic’s toll on young adolescents: Prevention and intervention targets to preserve their mental health. Journal of Adolescent Health, 70(3), 387–395. 10.1016/j.jadohealth.2021.11.023 Kutsar D Kurvet-Käosaar L The impact of the COVID-19 pandemic on families: Young people’s experiences in Estonia Frontiers in Sociology 2021 6 732984 10.3389/fsoc.2021.732984 34513980 Lafantaisie, V., Gervais, C., Côté, I., & Tourigny, S. (2021). Confinement et vulnérabilités ordinaires : L’expérience d’enfants et d’adolescents au temps de la COVID19. In C. Lacharité & T. Milot (Eds.), Vulnérabilité et famille (pp. 7–27). Éditions CEIDEF. https://oraprdnt.uqtr.uquebec.ca/pls/public/docs/GSC1910/O0003692570_LesCahiersDuCEIDEF_vol_8.pdf. Accessed 18 Jan 2022 Luijten MAJ Van Muilekom MM Teela L Polderman TJC Terwee CB Zijlmans J Klaufus L Popma A Oostrom KJ Van Oers HA Haverman L The impact of lockdown during the COVID - 19 pandemic on mental and social health of children and adolescents Quality of Life Research 2021 30 10 2795 2804 10.1007/s11136-021-02861-x 33991278 Masten AS Resilience theory and research on children and families: Past, present, and promise Journal of Family Theory & Review 2018 10 1 12 31 10.1111/jftr.12255 Masten AS Motti-Stefanidi F Multisystem resilience for children and youth in disaster: Reflections in the context of COVID-19 Adversity and Resilience Science 2020 1 2 95 10.1007/S42844-020-00010-W 32838305 Maurović I Liebenberg L Ferić M A review of family resilience: Understanding the concept and operationalization challenges to inform research and practice Child Care in Practice 2020 26 4 337 357 10.1080/13575279.2020.1792838 Mayne F Howitt C Rennie LJ A hierarchical model of children’s research participation rights based on information, understanding, voice, and influence European Early Childhood Education Research Journal 2018 26 5 644 656 10.1080/1350293X.2018.1522480 McArthur BA Racine N Mcdonald S Tough S Madigan S Child and family factors associated with child mental health and well - being during COVID - 19 European Child & Adolescent Psychiatry 2021 10.1007/s00787-021-01849-9 Newland LA Lawler MJ Giger JT Roh S Carr ER Predictors of children’s subjective well-being in rural communities of the United States Child Indicators Research 2015 8 1 177 198 10.1007/s12187-014-9287-x Orgilés M Morales A Delvecchio E Francisco R Mazzeschi C Pedro M Espada JP Coping behaviors and psychological disturbances in youth affected by the COVID-19 health crisis Frontiers in Psychology 2021 12 565657 10.3389/FPSYG.2021.565657 33828499 Paillé P Mucchielli A L’analyse qualitative en sciences humaines et sociales 2016 4e Armand Colin Prime H Wade M Browne DT Risk and resilience in family well-being during the COVID-19 pandemic American Psychologist 2020 75 5 631 643 10.1037/amp0000660 32437181 Ravens-Sieberer U Herdman M Devine J Otto C Bullinger M Rose M Klasen F The European KIDSCREEN approach to measure quality of life and well-being in children : Development, current application, and future advances Quality of Life Research 2014 23 791 803 10.1007/s11136-013-0428-3 23686556 Ravens-Sieberer, U., Kaman, A., Erhart, M., Otto, C., Devine, J., Löffler, C., Hurrelmann, K., Bullinger, M., Barkmann, C., Siegel, N. A., Simon, A. M., Wieler, L. H., Schlack, R., & Hölling, H. (2021). Quality of life and mental health in children and adolescents during the first year of the COVID-19 pandemic: Results of a two-wave nationwide population-based study, –14. Advance online publication. European Child & Adolescent Psychiatry. 10.1007/S00787-021-01889-1 Russell BS Hutchison M Tambling R Tomkunas AJ Horton AL Initial challenges of caregiving during COVID-19: Caregiver burden, mental health, and the parent–child relationship Child Psychiatry and Human Development 2020 51 5 671 682 10.1007/s10578-020-01037-x 32749568 Schwartz KD Exner-Cortens D McMorris CA Makarenko E Arnold P Van Bavel M Williams S Canfield R COVID-19 and student well-being: Stress and mental health during return-to-school Canadian Journal of School Psychology 2021 36 2 166 185 10.1177/08295735211001653 34040284 Shoshani A Kor A The mental health effects of the COVID-19 pandemic on children and adolescents : Risk and protective factors Psychological Trauma: Theory, Research, Practice, and Policy 2021 14 8 1365 1373 10.1037/tra0001188 34928689 Shukla M Wu AFW Lavi I Riddleston L Hutchinson T Lau JYF A network analysis of adolescent mental well-being during the coronavirus pandemic : Evidence for cross-cultural differences in central features Personality and Individual Differences 2022 186 111316 1 10 10.1016/j.paid.2021.111316 Stoecklin D Gervais C Kutsar D Heite C Lockdown and children’s well-being: Experience of children from Switzerland, Canada and Estonia Childhood Vulnerability Journal 2021 3 41 59 10.1007/s41255-021-00015-2 Tadić Vujčić M Brajša-Žganec A Franc R Children and young peoples ’ views on well-being : A qualitative study Child Indicators Research 2019 12 791 819 10.1007/s12187-018-9559-yChildren Tessier S From field notes, to transcripts, to tape recordings: Evolution or combination? International Journal of Qualitative Methods 2012 11 4 446 460 10.1177/160940691201100410 The KIDSCREEN Group Europe The KIDSCREEN group Europe 2006 Pabst Science Publishers Thibodeau-Nielsen RB Palermo F White RE Wilson A Dier S Child adjustment during COVID-19: The role of economic hardship, caregiver stress, and pandemic play Frontiers in Psychology 2021 12 716651 10.3389/FPSYG.2021.716651/FULL 34484078 Vera EM Vacek K Blackmon S Coyle L Gomez K Jorgenson K Luginbuhl P Moallem I Steele JC Subjective well-being in urban, ethnically diverse adolescents: The role of stress and coping Youth and Society 2012 44 3 331 347 10.1177/0044118X11401432 Viner R Russell S Saulle R Croker H Stansfield C Packer J Nicholls D Goddings AL Bonell C Hudson L Hope S Ward J Schwalbe N Morgan A Minozzi S School closures during scial lockdown and mental health, health behaviors, and well-being among children and adolescents during the first COVID-19 wave: A systematic review JAMA Pediatrics 2022 176 4 400 409 10.1001/jamapediatrics.2021.5840 35040870	0	PMC9734348	NO-CC CODE	2022-12-14 23:28:28	no	Int J Child Maltreat. 2022 Dec 7;:1-23	utf-8	Int J Child Maltreat	2,022	10.1007/s42448-022-00141-w	oa_other
==== Front Int J Child Maltreat Int J Child Maltreat International Journal on Child Maltreatment 2524-5236 2524-5244 Springer International Publishing Cham 141 10.1007/s42448-022-00141-w Research Article The COVID-19 Pandemic and Quality of Life: Experiences Contributing to and Harming the Well-Being of Canadian Children and Adolescents http://orcid.org/0000-0001-5695-9358 Gervais Christine [email protected] 1 Côté Isabel 2 Lampron-deSouza Sophie 3 Barrette Flavy 2 Tourigny Sarah 4 Pierce Tamarha 5 Lafantaisie Vicky 4 1 grid.265705.3 0000 0001 2112 1125 Nursing Department, Université du Québec en Outaouais, 5 Rue Saint-Joseph, Saint-Jérôme, Québec J7Z 0B7 Canada 2 grid.265705.3 0000 0001 2112 1125 Social Work Department, Université du Québec en Outaouais, 283 Boulevard Alexandre-Taché, C.P. 1250, Succursale Hull, Gatineau, Québec J8X 3X7 Canada 3 grid.14848.31 0000 0001 2292 3357 School of Psychoeducation, Université de Montréal, 90 Av, Vincent-d’Indy, Montréal, Québec H2V 2S9 Canada 4 grid.265705.3 0000 0001 2112 1125 Psychoeducation and Psychology Department, Université du Québec en Outaouais, 5 Rue Saint-Joseph, Saint-Jérôme, Québec J7Z 0B7 Canada 5 grid.23856.3a 0000 0004 1936 8390 School of Psychology, Université Laval, 2325, Rue Des Bibliothèques, Québec, G1V 0A6 Canada 7 12 2022 123 24 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 pandemic’s restrictive measures such as lockdowns, social distancing, and the wearing of masks transformed young people’s daily lives and brought up major concerns regarding children’s and adolescents’ well-being. This longitudinal mixed study aims to identify how different experiences contributed to children’s and adolescents’ well-being through different stages of the pandemic. The sample comprises 149 Canadian youth from Quebec who shared their experiences of the COVID-19 pandemic. Children and adolescents were met virtually for semi-directed interviews about their well-being at three measurement time (T1: May 2020 lockdown, T2: July 2020 progressive reopening, and T3: beginning of the second wave). At T3, they also completed a questionnaire measuring their quality of life. Our findings indicated that 22% reported a low level of well-being (N: 32), 66% a normal level of well-being (N: 90), and 18% a high level of well-being (N: 27). The comparative thematic analysis of the discourse of these three groups allows us to identify experiences that are favorable and unfavorable to the well-being of young people and to distinguish two configurations of interactions between children and their environment over the first year of the pandemic, namely that of young people who report a high level of well-being and that of those who report a worrying level of well-being. Results highlight the importance of activities, relationships, support, and representations of children and adolescents for their well-being in the pandemic context. Interventions and social measures to better support their well-being are discussed. Keywords Well-being Children Adolescent Pandemic COVID-19 Mixed study Social Sciences and Humanities Research Council1008-2020-1028 Gervais Christine Ministère de la famillePartenariat Familles en mouvanceÉquipe de recherche Paternité, famille et société ==== Body pmcIntroduction Professionals, researchers, and decision-makers are showing a growing interest in the well-being of children (Ben-Arieh, 2010) as children themselves represent and experience it (Amerijckx & Humblet, 2014; Gorza & Bolter, 2012) and are recognizing children’s right to define what well-being means to them (UNICEF Canada, 2019). In response to this interest, the past 20 years have seen a growth in studies looking at the subjective well-being of children (Fattore et al., 2019). A multidimensional concept, subjective well-being includes a cognitive dimension, which refers to children’s overall assessment of their lives; an emotional dimension, which looks at children’s moods and feelings; and a psychological dimension, which includes a sense of fulfillment, a positive vision of the future, and satisfaction with people’s responses to their psychological needs (Casas, 2011; Kaye-Tzadok et al., 2019). Since 2020, the restrictions imposed all over the world in response to the COVID-19 pandemic have brought up major concerns regarding the well-being of young people within the scientific community and among political decision-makers, workers, parents, and young people themselves. Indeed, restrictive measures such as lockdowns, social distancing, the closing of schools and shops, bans on gatherings and sports activities, and the wearing of masks are transforming young people’s everyday lives, depriving them of certain kinds of stimuli that are essential to their development (interactions with their peers, physical proximity to their friends, and access to school) (Gervais et al., 2020; Stoecklin et al., 2021) while exposing them to new stresses and unexpected circumstances (concerns with regard to illness, contamination risk, and so on) (Coyne et al., 2020; Fegert et al., 2020). Various studies have shown a deterioration in children’s and adolescents’ mental health since the pandemic began. Mostly carried out in the West during the first weeks of pandemic-related lockdowns, these studies identify a greater prevalence of depression and anxiety symptoms (Elharake et al., 2022; Hussong et al., 2021; Luijten et al., 2021), increases in sleep disturbances (Luijten et al., 2021) and behaviors and feelings related to loneliness and boredom (Shoshani & Kor, 2021). These studies also identify certain risk and protection factors to explain the variations in well-being among young people in the context of the pandemic. In keeping with the multi-level framework of child well-being developed by UNICEF (2020), these factors are internal to children themselves but are also found in the systems in which they interact. The multi-level framework of a child’s well-being, developed to allow international comparisons, identifies four levels of factors influencing children’s well-being. First, individual indicators represent physical and mental health antecedents, as well as emotional skills such as coping strategies. Second, the world of the child includes activities, for example, outdoor activities and screen time. It also comprises support provided by peers, school staff, and family members as well as children’s participation in decision-making at school and at home. Third, the world around the child consists of community support provided to parents, work-family balance, school resources, and access to facilities to play. Finally, characteristics of the world at large, such as the economic, environmental, and societal context, as well as health, education, and health policies, may shape children’s well-being. In the context of the pandemic, those factors may be deployed differently compared to other times. Children’s Individual Characteristics Contributing to Their Well-Being During the Pandemic Regarding young people’s characteristics, fear of the COVID-19 virus (Engel de Abreu et al., 2021), fragility or mental health problems that predate the pandemic (Shoshani & Kor, 2021), and the adoption of disengagement and emotion-focused engaged coping strategies (Hussong et al., 2021) all act as risk factors for children’s well-being. In contrast, having a high sense of self-efficacy (Hussong et al., 2021), feeling useful and able to face the situation (Shukla et al., 2022), and choosing problem-focused engaged coping strategies (Hussong et al., 2021) seem to act as buffers to limit the negative effects of the pandemic on children’s and adolescents’ well-being. Dimensions of the World of the Child Contributing to Their Well-Being During the Pandemic Studies highlighted activities and relationships that are essential for young people’s well-being in the context of the pandemic. Regarding activities, the suspension of extracurricular sports and cultural activities (Kutsar & Kurvet-Käosaar, 2021; Stoecklin et al., 2021), the perception of lacking freedom (Engel de Abreu et al., 2021), a high volume of homework, and the fact of not being able to attend school in an ongoing way (Ravens-Sieberer et al. 2021) are associated with alterations to well-being. In contrast, access to outdoor spaces, discovering new creative activities (Berasategi et al., 2021), regular routines during lockdown (Shoshani & Kor, 2021), less screen time (McArthur et al., 2021), and an appreciation of a less busy schedule and a slower pace of life (Stoecklin et al., 2021) seem favorable to well-being. Studies have also highlighted the essential role of young people’s relationships for their well-being during the pandemic. Indeed, connectedness to caregivers (McArthur et al., 2021), children’s attachment security with regard to their parents (Dubois-Comptois et al., 2021), the positive climate in the family (Ravens-Sieberer et al., 2021), adolescents’ satisfaction with the way adults listen to them (Engel de Abreu et al., 2021), and perceived social support (Ravens-Sieberer et al., 2021; Shoshani & Kor, 2021) are associated with better well-being, while family conflicts (Ravens-Sieberer et al., 2021) and separation from or loss of contact with friends represented significant difficulty during lockdown (Stoecklin et al., 2021). Contributions of the World Around the Child to Children’s Well-Being During the Pandemic Studies examining the world around the children also identified some risk and protective factors for their well-being in the context of the pandemic. More specifically, living in a single-parent family, in a family with three or more children (Luijten et al., 2021), in a family with less educated parents (Ravens-Sieberer et al., 2021), and dealing with precarious conditions (Engel de Abreu et al., 2021) are identified as risk factors. Moreover, parental strain and stress (Essler et al., 2021), negative changes in a parent’s work situation (Luijten et al., 2021), and parental mental health problems (Gervais et al., 2021; Kiss et al., 2022) are associated with an exacerbation of the pandemic’s negative effects on young people’s well-being. Contributions of the World at Large to Children’s Well-Being During the Pandemic While the pandemic profoundly transformed the economic, social, and environmental contexts in which children lived, the influence of these factors on children’s well-being has been less studied. Some studies have nevertheless highlighted how the transformation of child protection services to respect lockdown and social distancing hampered the protection of children, increasing the risk for child maltreatment and family violence (Katz et al., 2021). Moreover, insufficient governmental programs to support families and economic hardship were also identified as important risk factors for children’s well-being during the pandemic (Brooks et al., 2020; Thibodeau-Nielsen et al., 2021). While these results help provide an understanding of the factors that affect young people’s well-being, they mostly rely on data gathered in the first weeks of the pandemic using cross-sectional studies (Elharake et al., 2022) as well as on quantitative data gathered from adults who are part of children’s and adolescents’ lives rather than from young people themselves. As the pandemic has extended over time, many have underscored the need to take a longitudinal approach to studying its effects on children (Elharake et al., 2022; Essler et al., 2021; Holmes et al., 2020) and to include the voices of children and adolescents regarding what they perceive as being important for their well-being (Axford et al., 2014) in the context of the pandemic (Andrés et al., 2022; Berasategi et al., 2021). The Current Study This is the context in which this longitudinal study, inspired by the Children’s Understandings of Well-Being (CUWB) research protocol (Fattore et al., 2007, 2019), aims to identify how different experiences contribute to children’s and adolescents’ well-being through different stages of the pandemic. Based on the narratives of 149 Canadian children and adolescents regarding their experiences of the COVID-19 pandemic, it compares the statements of children who report a high, average, or poor quality of life in order to answer the research question: which experiences during the pandemic contributed to young people’s well-being over time? Methods This study relies on a mixed concurrent longitudinal design with a qualitative preponderance (Creswell et al., 2011) including three measurement times (T1: May 2020 lockdown, T2: July 2020 progressive reopening, and T3: November 2020 second wave) to capture children’s experience of the pivotal moments of the pandemic: the strict sanitary measures of the first lockdown, the progressive reopening following the first wave of the pandemic, and the return of strict sanitary measures in the second wave as we collectively realized that the pandemic would last much longer than envisioned in the early months. Anchored in a child-centered approach, this study looks at the experiences and perspectives of young people, considering their point of view on pandemic-related issues to be essential to our understanding of these issues (Côté et al., 2020a; Greene & Hill, 2005). It rests on research methods and tools developed in collaboration with a group of eight young people who acted as expert advisors throughout the study. Their involvement ensured that the questions were relevant and tailored to the realities and concerns of children and adolescents and that they were conducive to young people’s participation (Mayne et al., 2018) while respecting COVID-19 restrictions. Participants The sample is composed of 149 young people, 92 girls, and 57 boys from the province of Quebec, Canada, who took part along with their parents at the study’s three measurement points. They were aged seven to 17 (mean = 11 years old, SD = 2.52). They mainly lived with two parents (85% of them) and with siblings (54% have a brother or sister, and 43% have two or more). They lived in relatively favorable socioeconomic conditions; 82% had a parent who had completed university studies, and 62% benefited from a family income above US $95,000 per year. Study Design and Procedure The participating families were recruited during the first lockdown (April 2020), while schools, services, and businesses were closed, and going out was forbidden except for essential service workers (INSPQ, 2022). Recruitment was done through social media networks (Facebook) and through the newsletters of family-focused community organizations. To take part in the study, families had to have at least one child aged between seven and 17, have access to an internet connection, and be able to understand and communicate in French. Parents interested in the study first filled out an online questionnaire about their general situation (sociodemographic) and their state of health and functioning, and they consented for us to speak with their child. The children and adolescents were quickly contacted by a team member, and appointments were set for semi-directed Zoom interviews. At key moments of the pandemic, an email was sent to the parents to ask them and their children to participate once again. The data for the second measurement point were gathered in the month of July 2020 while most safety restrictions in Québec were loosened. Small gatherings were permitted at the time, and businesses and day camps were opened. The data at the third measurement point were gathered during the pandemic’s second wave, marked by the return of a number of restrictive measures. Extracurricular activities and sports were prohibited; students aged 14 and up were attending school one out of every 2 days; many classes and schools had to close for a week or two due to outbreaks, and restaurants and businesses were closed. To favor participant retention, which is always a challenge in longitudinal studies (Gifford et al., 2007), various strategies were put into place, including individualized thanks to each parent after the interviews with their children, the assignment of a specific assistant to each child participant in order to foster the development of a trusting relationship and coherence within the data gathered, and sending parents the journal articles published using the results. One $10-value reward per measurement point was given to each child at the end of their study participation (after the third measurement point) in the form of a gift card. Through a random draw, we also distributed four $50 gift cards per measurement point to the parents. Research Tools Qualitative Data The semi-directed interviews were done at each measurement time using the Zoom app. Lasting about 45 min each, they focused on scaffolding situations that promoted the children’s active involvement and the production of an explanatory discourse (Horgan, 2017), maximizing their contribution to knowledge building (Kirk, 2007). Young people’s knowledge and experience regarding the pandemic were discussed at each measurement point, along with their school experience, their family and social relationships, their coping strategies, and their vision of the future. Children were first asked a general prompt, “Has your daily life changed related to COVID-19 and the newly implemented measures? How did it change?” Interviewers asked questions related to family members (“How is it going with your mom, father, sisters or brothers?”), their friends (“How is it going with your friends?”), the school (“Can you explain how things are going with school?”), activities (“What kinds of activities have you been doing for the past weeks?”), and coping strategies (“When you face difficulties during the pandemic such as the ones you described earlier [interviewers named the difficulties], what do you do to feel better?”). They also asked questions related to the expected end of the pandemic (“How do you imagine your life when the pandemic is over?”). At each subsequent time measurement point, interviewers asked the same questions and added questions about how it had changed from the previous interviews, reminding the children which COVID-19 measures have changed compared to the other time measurement point (for example, since the lockdown ended, since you went back to school, or since you have been doing school online). Interviewers asked follow-up questions to encourage children to expand their answers, such as “Can you tell me more about that?” or “What do you consider more difficult or positive in that situation?” Interviewers also used techniques to encourage children to talk, such as saying “uh-uh” and “interesting.” Quantitative Data We assessed young people’s well-being with the KIDSCREEN 10 at the end of the third interview. A research assistant shared their screen through the Zoom platform to facilitate children’s understanding. Children indicated to the research assistant the answer that best represented their life. KIDSCREEN, a widely used questionnaire, measures an individual’s perception and subjective evaluation of their health and well-being within their unique cultural environment (Ravens-Sieberer et al., 2014). Items refer to the last week and are answered on a five-point Likert-type scale assessing frequency or intensity (i.e., have you felt lonely?) (a = 0.61). To allow comparison with international norms and to categorize the children, T-scores were calculated in accordance with the recommended KIDSCREEN 10 general health-related quality of life (HRQoL) scoring procedure (The KIDSCREEN Group Europe, 2006). Ethical Consideration The study was approved by the Université du Québec en Outaouais's ethical board. The parents who consented to take part in the study and to have their child take part, filled out a consent form on LimeSurvey at each measurement point. During the interviews with the young people, a tailored assent form developed by Côté et al (2018) was used to obtain their assent. Pictograms explaining the main consent-related issues (confidentiality and its limits, the right to withdraw, the benefits and risks of taking part, and free participation) were projected and discussed with each young person. The children age 7–13 years old gave their verbal consent to take part, while adolescents aged 14 and up filled out a consent form on LimeSurvey. In keeping with the child-centered approach, the research assistants also received training on the ethical issues of research with children in order to become aware of the power and authority issues inherent in the researcher-child relationship and to limit these by taking a transparent and benevolent approach (Côté et al. 2020b; Gallagher, 2009). Data Analysis Considering the scope of the material gathered and the restricted resources of the study, we prioritized data analysis that combined field notes and interview transcripts (Tessier, 2012). Immediately after each interview and using an outline developed for this study, the interviewer wrote a summary of the child’s related experiences, maximizing the reliability of the information gathered. These summaries (N = 192 at T1, 165 at T2, and 154 at T3) served as field notes and were subjected to a content analysis (Bardin, 2013) in order to categorize them based on the diversity of the participants’ personal situations (age, sex, region, time of return to school), the experiences described by the participants, and the richness of their statements. The interviews presenting contrasting experiences were transcribed word for word (N = 64 at T1, 70 at T2, 68 at T3). A coding tree was then developed in an inductive way based on the summaries. This was discussed by the team, then refined when the first transcripts were coded. The material as a whole was then subjected to a thematic analysis (Paillé & Mucchielli, 2016) using the N’vivo program and an expert coding strategy. As such, three teams (each including two research assistants and a researcher) were formed, each one in charge of coding two mother nodes and their sub-nodes, with each interview thus being read by three research assistants (one from each team). Coding comparisons were made regularly throughout the process (about 10% of the interviews) to ensure the coherence and reliability of the coding. The team then discussed the coding along with the meaning of certain excerpts, their belonging to certain codes, and the links between the different codes. The quantitative data were processed using the SPSS Statistics version 28 (IBM Corp., 2021). The scores obtained on KIDSCREEN were then compared to population averages (The KIDSCREEN Group Europe, 2006), and the young people were categorized based on what they reported for the third measurement point (November 2020, during the pandemic’s second wave) about their quality of life related to health, whether it was low, normal, or high; this was considered as their level of well-being. Crosstab in N’vivo made it possible to compare the experiences of these three groups of young people at the different times of the pandemic and to distinguish between the experiences that were favorable and unfavorable to their well-being. Result The presentation of the results is organized based on UNICEF’s multilevel framework of child well-being (UNICEF Innocenti, 2020). As such, we first present the well-being of participating children and adolescents, conceptualized as the result of their experience throughout the first months of the pandemic. Then, we describe the participants’ experiences related to the world of the child, the world around the child, and the world at large. For each of these spheres, we first set out common experiences, then describe the differences discerned within the children’s discourses when they reported a high level of well-being (group A) and a low level of well-being (group C). Young people from group B, who report a “normal” level of well-being, shared some experiences from each of these other groups. Since no specific experiences of this group could be identified, their narratives are included in the description of common experiences of children that introduce each theme, but their experiences are not contrasted with the ones of the two other groups of children. Children’s Well-Being The average level of well-being reported by youth participants was 51.83/100 (35.6–83.1, SD = 8.34), which is slightly higher (0,18 standard deviations) than the European norm data for children and adolescents (T = 50, SD = 10,0) (The KIDSCREEN Group Europe, 2006). Participants were categorized into three groups based on whether they reported a high level of well-being (group A, N = 27, 18%), one that was similar to the population average which is considered normal (group B, N = 90, 60%), or one that was low (group C, N = 32, 22%) during the third measurement period. World of the Child: Activities, Relationships, and Support Investing in Solitary Activities or Shared Activities The activities in which the young people invested time took up a lot of space in their experience of the first lockdown (T1). This period—synonymous with a sense of respite and freedom for some and of emptiness and boredom for others—gave them an opportunity to do activities other than the ones they normally would. For most of the young people, games and activities seemed to be a major source of well-being during the pandemic (T1, T2, and T3). Groups A and C invested in different types of activities. The young people of group A mainly engaged in activities with members of their immediate families, particularly with their siblings. Among others, they practiced outdoor sports and played board games as a family to stay busy and get their mind off things throughout the pandemic (T1, T2, and T3).“To feel better, I think it’s really about spending time with people, my family, playing board games. Because it seems like focusing on a specific activity with other people, and being in the moment with other people, everyone forgets a bit about what’s going on, you get the feeling like you’re living a normal life, and it feels good.” (Hayden_T1_17 years old) To deal with the limits imposed by the lockdown (T1 and T3), young people in group A also organized online activities with their friends and extended family members, for example, cooking on FaceTime with their grandmother or playing online games with friends, which helped them feel better. In contrast, the young people from group C reported that they mostly did solitary activities throughout the pandemic (T1, T2, and T3). Walking was the activity this group named most frequently, which allowed them to get outside, escape the family home, reduce the intensity of negative emotions, and find a better sense of well-being.“With the lockdown I realized that if I didn’t do anything, sometimes I felt worse, and going outside, like for example taking a walk outside every day, well, I felt better, y’know it was like I always needed to get some fresh air.” (Blanche_T1_12 years old) When they were going through more negative experiences during the pandemic, many young people in this group attempted to distract themselves by listening to music or watching videos or TV. Others spoke about finding comfort with their pets, expressing themselves through artistic creations, or playing alone, such as with Lego. Maintaining Relationships with Loved Ones The public health measures imposed to prevent COVID-19 have forced young people to redefine their interactions with their social networks. They were confined to their homes in the first months of the pandemic (T1), thus increasing the time spent with their immediate families while their contacts with friends and members of their extended families were rare or suspended, as explained by Laura–Marie (T1, 17 years old): “Honestly, they’re the only people I can have fun with right now and talk with. The only social interactions I have are with my family here.”. The majority of young people recounted that this period was a source of closeness with members of their immediate family, which they explained in particular by their parents’ greater availability due to the suspension of their work or to the requirement that they work from home. The scope and intensity of these family moments also generated tensions and conflicts with siblings and parents, mainly due to the lack of personal space during the lockdown. Similar interactions were reported during the second lockdown (T3). Young people used various instant messaging apps to maintain their ties with friends and extended family members during the first and second lockdowns (T1 and T3). However, these contacts appear to have been insufficient, as many children said they missed their loved ones during the pandemic’s first and second waves, as well as the fear to lose some of their extended family members who might contract and possibly die from COVID-19. Beyond these common experiences, young people from groups A and C reported certain differences in regard to their relationships with the people around them. As such, children in group A insisted that during the first lockdown (T1), it was important to support their grandparents despite the distance, as Alexane explained: “We still need to support and give love to the people who need it. Like to our grandparents, we need to call them and make sure they’re not, like, all alone in this” (Alexane_T1_12 years old). This concern led them to contact these loved ones regularly by telephone or to find alternative methods for seeing them in a safe way. However, the importance of giving support back to their grandparents was less present in the youths’ discourse during the second lockdown (T3). During the period of gradual reopening (T2), these young people expressed their relief and joy at once again seeing their loved ones “for real” in the summertime. Many of them chose to stop respecting social distancing in order to hug their friends and extended family members, while still applying other safety measures (hand-washing, coughing into their elbows, and so on). As children were starting to go out again and spending less time with their families, many of them said that tensions with their immediate families were significantly reduced. However, they kept up the outdoor activities they shared with their family members which began during the lockdown and which they still enjoyed. Others instead reported that their parents became less available as they went back to work, which was disappointing to them even if they said they understood their parents’ commitments. The children of group C, however, sacrificed their desire for closeness with their grandparents in order to protect them, both during the lockdown period (T1) and during the gradual reopening (T2) as well as the second lockdown (T3). Many see themselves as a danger to their relatives. Their narratives insisted on their fear of transmitting the virus to their grandparents as well as on their worries about the possible consequences of the virus on their loved one’s health and lives.“I try to hug them as little as possible, especially my grandmother. And my grandmother comes to give us hugs, but my mother said not to give any to her. It hurt that I couldn’t hug her, but I was careful because I really didn’t want to give it (COVID) to her.” (Amber_T3_10 years old) During the gradual reopening period (T2), the young people in this group expressed greater ambivalence about their parents’ return to work, as they were less available, which amplified the young people’s sense of solitude. Benefiting from Reliable and Diversified Support Regardless of their level of well-being, the young people described being supported by their parents during stressful times throughout the pandemic (T1, T2, and T3). Parents listened to them attentively; young people felt they could talk about the emotions they were feeling about the pandemic, and their parents shared hugs with them; these things consoled and reassured the children in more emotionally difficult moments. When necessary, the parents also offered advice to the children to help them solve problems. Friends were also another major source of support, helping young people to get their minds off things and be entertained. Beyond these general observations, some differences were observed with regard to the scope and forms of support that young people received. Young people from group A talked more about the emotional support provided by their parents in the form of kind words and markers of affection, such as hugs, which contributed to their sense of being loved and soothed during the lockdown period (T1) and the reopening (T2), as explained by Michèle (T1_11 years old): “My mother shared a lot of love. We were there for her, and she was there for us. We gave her hugs and kisses every day when she got home from the clinic.” Young people in this group were also distinguished with regard to having space for family conversations to talk about the COVID-19 pandemic. They described talking about their worries about COVID-19 with their parents, who listened to them, added nuance or filled out missing information, and helped them readjust their perception of the virus if needed, thus reducing their anxieties. Many children reported as well that their parents reminded them about safety measures to apply throughout the pandemic (T1, T2, and T3), which made them feel protected. Lastly, some young people in this group also mentioned the emotional support they received from their friends during the pandemic, in particular in the form of listening and making space for their emotions. While children from group C also said they reached out to their parents when they needed support, they spoke much less about the support they received during the first lockdown (T1), the gradual reopening (T2), and the second lockdown (T3). As well, very few of them reported talking about COVID-19 with their parents during the pandemic. Particularly during the second wave (T3), young people in this group explained that their parents were busy with their work, which led them to seek out solutions by themselves in order to feel better.“When I felt bad, I thought, ‘OK, I’ll stop working and take a moment to listen to some music.’ These last few months, I didn’t have many friends, so I couldn’t call anyone. My parents were working, so it was kinda just me for myself.” (Charlie_T3_15 years old) The emotional support provided by friends was also largely absent from these children’s responses. World Around the Child: Anticipating and Appreciating the Return to School Despite a great diversity in the stories children told about their school experience, the majority of participants were impatient to return to in-person schooling. Many of them emphasized the importance of the social aspect of their school experience: they missed their friends and felt a lack of social interactions during the lockdown periods (T1 and T3). The return to class, for many of them, meant reconnecting with their friends and starting to return to normal, as explained by two respondents: “It let me socialize, because for me, my parents wouldn’t have forced me to go to school, but I’m the one who wanted to.” (Sonya_T2_8 years old), “I was happy to go back to school, I was tired of staying home and not seeing anyone.” (Victor_T2_11 years old). However, the young people envisioned and experienced the transition back to in-person school in different ways. The young people from group A had a very positive expectations of their return to the classroom during the first lockdown (T1). Their responses included no fear of returning to school, which was described as a positive experience. They appreciated that the number of students in each class was reduced, which let them participate more, receive more support from their teachers, and accelerate their pace of learning. Excitement and relief were key elements of these young people’s experiences when returning to school (T2), as one explained: “I was happy and relieved to go back. When I have to miss school for two weeks, I miss it, so not going for six months was very long” (Talia_T2_10 years old). The children from group C reported more difficult experiences in returning to school. Many of them spoke about having understood there would be changes required by safety measures at their school, as well as the impacts of remote schooling on their learning during the first lockdown (T1). Stress and anxiety characterized the memories many children had of this period, such as those of Léon (T3_14 years old): “I wasn’t feeling it. It was a huge anxious period, and still is now. It was crazy, it was really a handicap in my life.” Some young people mentioned their fear of contracting COVID-19 and thus presenting a risk to their more vulnerable peers. After several months during which contact was forbidden, the children and adolescents of this group seemed to perceive social interactions as being dangerous during the gradual reopening (T2). While some of them had a harder time adapting to the social aspect of returning to the classroom, the rules put into place in the schools gave these young people a sense of safety and trust at the time of the second wave (T3).“Introducing masks and hygiene procedures went well. (...) If we hadn’t respected the measures, there may have been a lot more cases of COVID-19. They did it above all for our benefit.” (Inès_T3_16 years old) Some participants in this group also decried the impacts of the lockdown periods as well as distancing measures at school on their pleasure at school and on the quality of their friendships. They complained of classroom bubbles and their school’s approach to organizing recreations, which determined and limited the friends they were allowed to play and interact with during the second wave (T3). The majority of them still reported having quickly learned to live with the virus in the school setting and having gradually gotten used to the new measures in place.“I felt a little disoriented, but I didn’t really know what to do because there were tons of other rules, but I ended up getting used to the zones, and I ended up complaining a bit too, it kinda sucked.” (Milo_T3_9 years old) World at Large: Perceiving the Consequences of the Pandemic Throughout the three measurement times, the participants’ stories bore witness to the major effects on their lives of the public health policies and safety regulations implemented in response to COVID-19, particularly when they spoke about their experiences of the first lockdown (T1). As such, participants spoke at length about the lack of contact, both physical and social, the accumulation of deaths, the positive impact of the lockdown on the environment, and job losses. Beyond these global effects noticed by the young people, children in group A were distinct in terms of their nuanced view of the consequences of social and health measures, while those in group C shared a more negative understanding of them. The young people from group A also noted the positive consequences of the pandemic for the population and for individuals. They highlighted how the lockdown gave them an opportunity to discover new interests and spend time on activities they enjoyed but had previously neglected: “It changes people’s morale, we weren’t doing the same activities, we were using the car less and we got out for walks more” (Tyler_T1_13 years old). Other young people in this group noted that the pandemic helped people realize the importance of their families.“You realize that your family is important, for example right now we can only really see our close family. So in normal times, it’s important to spend time with your (extended) family because in times like this (lockdown) you can’t always see them.” (Eden_T1_13 years old) In contrast, the participants of group C identified more negative impacts of the pandemic for the whole population. They were concerned about the consequences of the lockdown on the population’s mental health, explaining that “It can create kinds of mental problems. Not problems, maybe, but it has an effect on the mental state” (Léon_T1_14 years old) and saying, “It can make people unhappier, people are losing their loved ones and some people might die” (Lara_T1_9 years old). Others underscored how the pandemic led to changes in routine that threw off their balance: “So, well, it changes pretty much everything. Your routine, you can no longer go to the grocery store with your parents. We don’t leave home anymore. Yeah. It’s a little destabilizing. Well, not just a little, it’s really destabilizing” (Céleste_T1_15 years old). They also shared their worries about elderly and dying people: “Some elderly people, sometimes it might worry them or stress them out… They might really feel bored because they can’t see their families” (Sasha_T1_10 years old). “It prevents families from seeing their dying loved ones one last time” (Charles-Antoine_T1_12 years old). During the gradual reopening and the second lockdown, children and adolescents talked less about the consequences of COVID-19 on society. Discussion A good childhood is often defined as one in which children have a positive experience with themselves and their environment, as well as the prospect of a good future (Ben-Arieh & Frønes, 2011; UNICEF Innocenti, 2020). Since the beginning of the pandemic, many scholars and health professionals have raised concerns about the threat that the state of health emergency represents for these two dimensions (Gervais et al., 2022; Prime et al., 2020) while children’s experience of it has still not been widely studied. In this sense, the results of this study are valuable because they provide a better understanding of the positive and negative effects on children’s well-being over the course of the pandemic. Although the research design did not make it possible to identify causal relationships between the children’s experiences and their level of well-being, our analyses shed light on two configurations of interactions between children and their environment that contributed to their well-being over the first year of the pandemic. On the one hand, the world of the children who presented a high level of well-being in the second wave of the pandemic is characterized by their investment in shared activities with their family members and their friends, thanks to communication technologies, which were present and available from the first major lockdown. We also note that these young people maintained close relationships with their loved ones despite the health measures in place during the different measurement periods: their responses were structured around the importance of supporting their grandparents and maintaining their ties with them despite restricted contact, and they noted their pleasure in seeing them again when health measures were relaxed. Their narratives reflect the readily available support that was facilitated by better work conditions for parents during the pandemic: they spoke of receiving strong support from their parents and having space to talk about COVID-19 and its issues within their families. With regard to the world around the children, the young people in this group looked forward to returning to school in a very positive way and appreciated it despite the restrictions enforced in the school setting and the higher risk of COVID-19 exposure. Lastly, these children’s experience of the world at large is distinguished by their nuanced understanding of the pandemic’s consequences; these young people underlined the things they learned and the closeness they developed because of the health measures just as much as the losses and boredom they experienced. On the other hand, the world of the children that showed lower well-being was characterized by involvement in mostly solitary activities, although we cannot determine whether this was due to preference or to their loved ones’ lack of availability for shared activities. They also expressed a fear of infecting others or being infected by others that contributed to weaken their bonds, especially with their grandparents and their social interactions. The support received from their loved ones was much more limited in these children’s experience during the first months of the pandemic, especially from their parents. The young people report that they needed to ask for support and perceived their parent’s availability as limited due to their work commitments so they mainly tried to manage by themselves. Apprehension and lack of pleasure regarding the return to school as well as disappointment and lack of control regarding the prevailing measures are at the heart of their experience of the world around them. Lastly, these children’s experience of the world at large was permeated by their concerns about the negative impacts of the pandemic, namely for the survival of the elderly and the mental health of the population. Briefly, the personal and family boundaries of these young people seemed to be more permeable, with the pandemic having a greater effect on the way they saw the world around them. While comparing the perspectives of children with high and low levels of well-being, the results of this study allowed us to identify individual and social factors that are intertwined and contribute both to well-being and to experiences of the pandemic. Indeed, the results underscore the resilience of the families we interviewed, meaning their capacity “to adapt successfully to significant challenges that threaten the function, viability, or development of the system” (Masten, 2018, p. 16). The fact that 78% of the children and adolescents we interviewed reported a high level of well-being, comparable to or higher than population averages, after several months of health measures that transformed their everyday lives and deprived them of many interactions and activities shines a light on the agency of children and the resilience of families (Masten & Motti-Stefanidi, 2020). The experiences they recounted in our study also emphasize the richness that families can produce in terms of shared experiences, the creation of meaning, and mutual help, which are processes favorable to family resilience (Maurović et al., 2020) and act as protective factors for the well-being and development of all family members (Prime et al., 2020). The children’s viewpoints described in this study also highlight the importance of the social support children received but also that which they perceived. Indeed, in the context of the pandemic, parents’ involvement in shared activities with their children was a form of social support, limiting boredom, and promoting enjoyment. Moreover, parents’ ability to be available was noticeably present in children’s discourse, which suggests that it created a reassuring environment for them (Gervais et al., submitted) despite the uncertainty and unpredictability of the world around them and despite the pandemic’s many impacts on their own mental health (Gassman-Pines et al., 2020; Russell et al., 2020). Some children also pointed out the interinfluence of their parents’ availability and their working conditions. Combining work, family and homeschooling added additional challenges for parents, which might have impacted their own well-being and consequently their availability to reassure children who felt insecure during the pandemic (Cusinato et al., 2020). These findings also provide a better understanding of coping strategies that may contribute to children’s experiences and well-being during the pandemic. Contrary to the study by Vera et al. (2012), which concluded that conventional coping strategies, such as seeking social support, may not buffer adolescents from the effects of stress on subjective well-being, children and adolescents reporting higher well-being in our study largely reported seeking social support to feel better during the pandemic. It seems that despite the uncontrollable nature of pandemic-related stress, solution-oriented coping strategies—such as asking for help, seeking information about the pandemic by talking with parents, and supporting grandparents from afar—acted as buffers and protected young people’s well-being, as suggested by other studies (Hussong et al., 2021). Moreover, while the study by Dominguez-Alvarez et al., (2020) concludes that avoidance and distraction strategies are associated with different indicators of maladjustment (2020), our results indicate that the majority of the young people we interviewed adopted avoidance and distraction strategies, which they describe as being activities that helped entertain them. Shared family activities, such as games and sports, seemed to provide a beneficial distraction to the young people who reported a high level of well-being, as proposed by Orgilés et al. (2021), while solitary distraction activities, such as watching TV and playing with Lego, were observed for children with lower well-being. Lastly, cognitive restructuring more frequently used by children who were minimally impacted by the pandemic also appears to be favorable to their well-being, as consistent with the study by Orgilés et al. (2021) and discerned in the capacity of the young people in group A to look forward to returning to school and perceiving positive impacts of the pandemic. Lastly, these data suggest that many dimensions of the subjective well-being of children and adolescents continue to be articulated in a similar way in the context of the pandemic and of restrictive health measures. In keeping with the study by Gonzalez-Carrasco et al. (2019), which identified that feeling afraid of being alone at home or of death was associated with lower levels of well-being among adolescents, the solitude experienced at various moments in the pandemic and the idea that social contacts were now a threat to themselves and their relatives were harmful to young people’s well-being. As well, the social dimension is key to the different experiences of the two groups of children. In this study, care, gratitude, reciprocity, and support within the parent–child relationship were more strongly present in the experiences of young people who reported greater well-being, which is consistent with prior studies of child well-being (Fattore & Mason, 2017; Newland et al., 2015; Tadić Vujčić et al., 2019). Our study outlines different stress experiences regarding the absence of contact during school closure and the increase of contact due to the return to in-person classes. In keeping with the studies by Schwartz et al. (2021) and Viner et al. (2022), showing that pandemic-related stress during the school closure and reopening was negatively related to children’s well-being, our results suggest that school experiences are indeed central to children’s well-being (Newland, 2015). Our study contributes to knowledge by setting out the ways in which children’s representations of global phenomena that change the world around them, such as the pandemic, are also associated with their well-being. Managing to perceive the consequences of this event in a nuanced way by also noting the positive effects of health measures seems to be a protective factor for young people’s well-being. Implications for Practice This study’s findings suggest several practical implications for social, education, and health professionals working with children, adolescents, and families. First, the importance of family interactions, shared moments, and support received from parents underscores parents’ role in young people’s well-being in the context of the pandemic. With this in mind, we encourage professionals to raise parents’ awareness about the listening, acceptance, and emotional validation they can provide to their children, as well as about the importance of creating spaces and times for talking about COVID-19 to foster young people’s understanding of the issues related to the pandemic and create nuanced family meanings about them. These data also underscore the need for school workers to broaden their definition of vulnerability in order to discern children whose well-being has been more severely compromised by the pandemic and to better support them. The experiences related by the young people in this study illustrate that beyond well-documented vulnerability factors (such as poverty, housing instability, family violence, and so on), the experiences young people have had during the pandemic may contribute to their vulnerability, including children living in privileged contexts (Lafantaisie et al., 2021). Given the central aspect of social relations in the experiences that contribute to children’s well-being, particular attention should be paid to children and adolescents whose lives have already been marked by relationship breakdowns and who receive little support from their immediate families. We also call upon decision-makers to put actions into place to render the world of children and the world around children favorable to their well-being. In particular, parents’ availability for their children, essential to their well-being in the unstable period of this pandemic, should be encouraged. Without supportive measures, we fear that parents’ adaptive resources, heavily solicited over the last 2 years, may run out, and their parenting practices may deteriorate. Lastly, creating more spaces and opportunities for children’s participation is essential. Children’s participation in the decisions that affect them, which was already insufficient, has dropped even further in reaction to the lockdown measures that limited their access to the spaces and people that supported their participation. Many of them experienced this lack of participation as a lack of consideration and interest on the part of leaders and society toward young people, who were heavily impacted by health measures even though the virus presented a very low risk to their health (Gervais et al., 2022). To dismantle this impression, young people should be included in all decisions about the services and activities that are aimed at them and that will be put into place to foster post-pandemic reopening and to orient what “normal life” will look like. Strengths and Limitations Despite the large sample size and the diversity of age and location of the study’s participants, the privileged conditions of the young people we interviewed constitute a limit in this study. Indeed, our respondents had low exposure to COVID-19, lived mostly in fairly well-off two-parent families, and generally had parents with a high level of schooling who were able to work from home and thus be present for them during the months when they were not at school. The level of well-being reported by the young participants was higher than that of young people living in Europe during the same time period according to the study by Ravens-Sieberer et al. (2021), and the experiences they shared all bear witness to their privileged context. Their experiences are doubtless different from those of children who faced great precarity during the pandemic, for whom family activities were probably less frequent, and who saw fewer positive impacts of the pandemic. The combined analysis of the field notes and transcribed interviews is also a limitation of the study, as the adult interviewer’s perspective may have seeped into the field notes, blurring the children’s perspective on some of the experiences. As well, the relationship developed between the young people and their interviewer over the course of their meetings may also have introduced a social desirability bias, with young people wishing to give a good impression of themselves and largely recounting their positive experiences during the pandemic. This relationship, however, also constitutes one of the study’s strengths, having contributed to a high retention rate over the course of the various measuring periods, helping the children feel safe and listened to, and helping them express themselves. The mixed longitudinal design is also one of the study’s strengths, as it allowed for the comparative analysis of the specific experiences of different groups of children during the different periods of the pandemic. However, it does not allow us to establish a causal link between youths’ level of well-being and their experiences during the early months of the pandemic. Indeed, because their well-being was only measured at the third measurement time, we cannot rule out that differences in well-being levels were already present before or at the very beginning of the pandemic and influenced young people’s experiences, perceptions, and ways of engaging or not engaging with their surroundings. Conclusion This study’s results provide a nuanced look at the experiences children lived through during the pandemic and their contribution to their well-being from their own perspective. The results highlight the complex interactions between children and their environment, revealing how children’s agency, relationships, resources, and context contribute to both their experiences of the pandemic and their well-being. The results also underline the importance of listening to what children have to say rather than solely measuring the negative impacts of the pandemic on their mental health, as well as implementing social measures to support children’s positive experiences and well-being in the context of a pandemic. Acknowledgements The authors would like to thank the participating children and adolescents and the research assistants who contributed to the data collection for the study. Funding This study was supported by grants from the Social Sciences and Humanities Research Council, the Ministère de la famille du Québec, the Partenariat de recherche Familles en Mouvance, and the research team Paternité, famille et société. Data Availability The datasets generated and analyzed during the current study are not publicly available due to the fact that they constitute an excerpt of research in progress but are available from the corresponding author on reasonable 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 Amerijckx G Humblet PC Child well-being: What does it mean? Children and Society 2014 28 5 404 415 10.1111/chso.12003 Andrés, M. L., Galli, J. I., Valle, M., Vernucci, S., Morales, H. L., & Trudo, R. G. (2022). Parental perceptions of child and adolescent mental health during the COVID ‑ 19 pandemic in Argentina. In Child & Youth Care Forum. 10.1007/s10566-021-09663-9 Axford N Jodrell D Hobbs T Objective or subjective well-being? Handbook of Child Well-Being: Theories, Methods and Policies in Global Perspective 2014 Springer 2699 2738 Bardin L L’analyse de contenu 2013 Presses Universitaires de France Ben-Arieh A Rose W McAuley C Developing indicators for child well-being in a changing contex Well-Being : Understanding children’s lives 2010 Jessica Kingsley Publishers 129 142 Ben-Arieh A Frønes I Taxonomy for child well-being indicators: A framework for the analysis of the well-being of children Childhood 2011 18 4 460 476 10.1177/0907568211398159 Berasategi N Idoiaga N Dosil M The well-being of children in lock-down : Physical, emotional, social and academic impact Children and Youth Services Review 2021 127 106085 10.1016/j.childyouth.2021.106085 34728873 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 In The Lancet 2020 395 10227 912 920 10.1016/S0140-6736(20)30460-8 UNICEF Canada. (2019). Where does Canada stand? The Canadian index of child and youth well-being. oneyouth.unicef.ca. https://oneyouth.unicef.ca/sites/default/files/2019-08/2019_Baseline_Report_Canadian_Index_of_Child_and_Youth_Well-being.pdf Casas F Subjective social indicators and child and adolescent well-being Child Indicators Research 2011 4 555 575 10.1007/s12187-010-9093-z Côté I Trottier-Cyr R-P Lavoie K Pagé G Marin A Eysermann B Giroux MT « Veux-tu participer à ma recherche? » : Principes, enjeux et stratégies concernant l’assentiment des enfants dans le processus de recherche Recrutement et consentement à la recherche : réalités et défis éthiques 2018 Éditions de l’Université de Sherbrooke 127 145 Côté I Lavoie K Trottier-Cyr R-P La méthodologie de recherche centrée sur l’enfant : Recueillir et restituer la parole des enfants et des adolescent·e·s 2020 Les Presse de l’Université Lava Côté, Isabel., Pagé, G., & Lavoie, K. (2020b). La participation de l’enfant en recherche sociale : Considérations éthiques et méthodologiques. In Isabel Côté, K. Lavoie, & R.-P. Trottier-Cyr (Eds.), La recherche centrée sur l’enfant. Défis éthiques et innovations méthodologiques (p. 17-38). Presses de l’Université Laval. Coyne, L. W., Gould, E. R., Grimaldi, M., Wilson, K. G., Baffuto, G., & Biglan, A. (2021). First things first: Parent psychological flexibility and self-compassion during COVID-19. Bahavior Analysis in Practice, 14, 1092–1098. 10.1007/s40617-020-00435-w Cusinato M Iannattone S Spoto A Poli M Moretti C Gatta M Miscioscia M Stress, resilience, and well-being in Italian children and their parents during the COVID-19 pandemic International Journal of Environmental Research and Public Health 2020 17 22 1 17 10.3390/ijerph17228297 Creswell J Klassen A Plano Clark V Smith K Best practices for mixed methods research in the health sciences 2011 National Institutes of Health Engel de Abreu PMJE Neumann S Wealer C Abreu N Macedo EC Kirsch C Subjective well-being of adolescents in Luxembourg, Germany, and Brazil during the COVID-19 pandemic Journal of Adolescent Health 2021 69 2 211 218 10.1016/j.jadohealth.2021.04.028 Domínguez-Álvarez B López-Romero L Isdahl-Troye A Gómez-Fraguela JA Romero E Children coping, contextual risk and their interplay during the COVID-19 pandemic: A Spanish case Frontiers in Psychology 2020 11 577763 10.3389/fpsyg.2020.577763 33391095 Dubois-Comptois K Suffren S St-Laurent D Milot T Lemelin J Child psychological functioning during the COVID-19 lockdown: An ecological, family -centered approach Journal of Behavioral Pediatric 2021 42 7 532 539 10.1097/DBP.0000000000000935 Elharake, J. A., Akbar, F., Malik, A. A., Gilliam, W., & Omer, S. B. (2022). Mental health impact of COVID ‑ 19 among children and college students : A systematic review. Child Psychiatry & Human Development, 1–13, Advance online publication. 10.1007/s10578-021-01297-1 Essler S Christner N Paulus M Longitudinal relations between parental strain, parent–child relationship quality, and child well-being during the unfolding COVID-19 pandemic Child Psychiatry & Human Development 2021 52 6 995 1011 10.1007/s10578-021-01232-4 34426893 Fattore T Mason J The significance of the social for child well-being Children and Society 2017 31 4 276 289 10.1111/chso.12205 Fattore T Mason J Watson E Children’s conceptualisation(s) of their well-being Social Indicators Research 2007 80 1 5 29 10.1007/s11205-006-9019-9 Fattore T Fegter S Hunner-Kreisel C Children’s understandings of well-being in global and local contexts: Theoretical and methodological considerations for a multinational qualitative study Child Indicators Research 2019 12 2 385 407 10.1007/s12187-018-9594-8 Fegert JM Vitiello B Plener PL Clemens V Challenges and burden of the coronavirus 2019 (COVID-19) pandemic for child and adolescent mental health: A narrative review to highlight clinical and research needs in the acute phase and the long return to normality Children Adolescescent Psychiatry Mental Health 2020 14 20 10.1186/s13034-020-00329-3 Gallagher M Tisdall EK Davis M Gallagher M Ethics Researching with children and young people: Research design, methods and analysis 2009 Sage 11 27 Gassman-Pines A Asanat EO Fitz-Henley J COVID-19 and parent-child psychological well-being Pediatrics 2020 146 4 e2020007294 10.1542/peds.2020-007294 32764151 Gervais, C., Côté, I., & Lalande, D. (2020). Étude Réactions: Récits d’Enfants et d’Adolescents sur la COVID 19. Université du Québec en Outaouais. http://www.partenariat-familles.inrs.ca/wp-content/uploads/2020/07/R%C3%A9actions_Premier-rapport-d%C3%A9tape.pdf. Accessed 15 Mar 2022 Gervais, C., Côté, I., Tourigny, S., Pierce, T., de Montigny, F., & Lafantaisie, V. (2021). Family well-being and child functioning during the first COVID-19 pandemic lockdown: A cross-sectional study. Journal of Child & Family Studies, Submitted. Gervais, C., Côté, I., & Miljus, M. (2022). Étude Réactions : Récits d’enfants et d’adolescents sur la COVID19. Rapport final. Université du Québec en Outaouais. http://www.partenariat-familles.inrs.ca/wp-content/uploads/2022/04/Reactions_Rapport_Final_GervaisCote-compressed.pdf. Accessed 13 Jan 2022 Gifford SM Bakopanos C Kaplan I Correa-Velez I Meaning or measurement? Researching the social contexts of health and settlement among newly-arrived refugee youth in Melbourne Australia Journal of Refugee Studies 2007 20 3 414 440 10.1093/jrs/fem004 González-Carrasco M Vaqué C Malo S Crous G Casas F Figuer C A qualitative longitudinal study on the well-being of children and adolescents Child Indicators Research 2019 12 2 479 499 10.1007/s12187-018-9534-7 Gorza M Bolter F Indicateurs de bien-être de l’enfant, une déclinaison en protection de l’enfance est-elle possible ? Journal Du Droit Des Jeunes 2012 312 2 26 36 10.3917/jdj.312.0026 Greene SM Hill M Greene SM Hogan DM Researching children’s experiences: Methods and methodological issues Researching Children’s Experiences: Approaches and Methods 2005 Sage 1 21 Holmes, E. A., O’Connor, R. C., Perry, V. H., Tracey, I., Wessely, S., Arseneault, L., Ballard, C., Christensen, H., Cohen Silver, R., Everall, I., Ford, T., John, A., Kabir, T., King, K., Madan, I., Michie, S., Przybylski, A. K., Shafran, R., Sweeney, A., … Bullmore, E. (2020). Multidisciplinary research priorities for the COVID-19 pandemic: A call for action for mental health science. The Lancet. Psychiatry, 7(6), 547–560. 10.1016/S2215-0366(20)30168-1 Horgan D Child participatory research methods: Attempts to go ‘deeper’ Childhood 2017 24 2 245 259 10.1177/0907568216647787 Hussong AM Midgette AJ Thomas TE Coffman JL Cho S Coping and mental health in early adolescence during COVID-19 Research on Child and Adolescent Psychopathology 2021 49 9 1113 1123 10.1007/s10802-021-00821-0 33856611 IBM Corp. (2021). IBM SPSS statistics for windows, Version 28.0. Innocenti, U. N. I. C. E. F. (2020). Worlds of influence: Understanding what shapes child well-being in riche countries. UNICEF Office of Research. https://www.unicef-irc.org/publications/pdf/Report-Card-16-Worlds-of-Influence-child-wellbeing.pdf. Accessed 15 Mar 2022 INSPQ. (2022). Ligne du temps de la COVID-19 au Québec. https://www.inspq.qc.ca/covid-19/donnees/ligne-du-temps. Accessed 20 Mar 2022 Katz I Katz C Andresen S Bérubé A Collin-Vezina D Fallon B Fouché A Haffejee S Masrawa N Muñoz P Priolo Filho SR Tarabulsy G Truter E Varela N Wekerle C Child maltreatment reports and child protection service responses during COVID-19: Knowledge exchange among Australia, Brazil, Canada, Colombia, Germany, Israel, and South Africa Child Abuse & Neglect 2021 116 105078 10.1016/J.CHIABU.2021.105078 33931238 Kaye-Tzadok A Ben-Arieh A Kosher H Hope, material resources, and subjective well-being of 8- to 12- year old children in Israel Child Development 2019 90 2 344 358 10.1111/cdev.13130 30125932 Kirk S Methodological and ethical issues in conducting qualitative research with children and young people: A literature review International Journal of Nursing Studies 2007 44 7 1250 1260 10.1016/j.ijnurstu.2006.08.015 17027985 Kiss, O., Alzueta, E., Yuksel, D., Pohl, K. M., de Zambotti, M., Muller-Oehring, E. M., Prouty, D., Durley, I., Pelham, W. E., Mccabe, C. J., Gonzalez, M. R., Brown, S. A., Wade, N. E., Marshall, A. T., Sowell, E. R., Breslin, F. J., Lisdahl, K. M., Dick, A. S., Sheth, C. S., … Baker, F. C. (2022). The pandemic’s toll on young adolescents: Prevention and intervention targets to preserve their mental health. Journal of Adolescent Health, 70(3), 387–395. 10.1016/j.jadohealth.2021.11.023 Kutsar D Kurvet-Käosaar L The impact of the COVID-19 pandemic on families: Young people’s experiences in Estonia Frontiers in Sociology 2021 6 732984 10.3389/fsoc.2021.732984 34513980 Lafantaisie, V., Gervais, C., Côté, I., & Tourigny, S. (2021). Confinement et vulnérabilités ordinaires : L’expérience d’enfants et d’adolescents au temps de la COVID19. In C. Lacharité & T. Milot (Eds.), Vulnérabilité et famille (pp. 7–27). Éditions CEIDEF. https://oraprdnt.uqtr.uquebec.ca/pls/public/docs/GSC1910/O0003692570_LesCahiersDuCEIDEF_vol_8.pdf. Accessed 18 Jan 2022 Luijten MAJ Van Muilekom MM Teela L Polderman TJC Terwee CB Zijlmans J Klaufus L Popma A Oostrom KJ Van Oers HA Haverman L The impact of lockdown during the COVID - 19 pandemic on mental and social health of children and adolescents Quality of Life Research 2021 30 10 2795 2804 10.1007/s11136-021-02861-x 33991278 Masten AS Resilience theory and research on children and families: Past, present, and promise Journal of Family Theory & Review 2018 10 1 12 31 10.1111/jftr.12255 Masten AS Motti-Stefanidi F Multisystem resilience for children and youth in disaster: Reflections in the context of COVID-19 Adversity and Resilience Science 2020 1 2 95 10.1007/S42844-020-00010-W 32838305 Maurović I Liebenberg L Ferić M A review of family resilience: Understanding the concept and operationalization challenges to inform research and practice Child Care in Practice 2020 26 4 337 357 10.1080/13575279.2020.1792838 Mayne F Howitt C Rennie LJ A hierarchical model of children’s research participation rights based on information, understanding, voice, and influence European Early Childhood Education Research Journal 2018 26 5 644 656 10.1080/1350293X.2018.1522480 McArthur BA Racine N Mcdonald S Tough S Madigan S Child and family factors associated with child mental health and well - being during COVID - 19 European Child & Adolescent Psychiatry 2021 10.1007/s00787-021-01849-9 Newland LA Lawler MJ Giger JT Roh S Carr ER Predictors of children’s subjective well-being in rural communities of the United States Child Indicators Research 2015 8 1 177 198 10.1007/s12187-014-9287-x Orgilés M Morales A Delvecchio E Francisco R Mazzeschi C Pedro M Espada JP Coping behaviors and psychological disturbances in youth affected by the COVID-19 health crisis Frontiers in Psychology 2021 12 565657 10.3389/FPSYG.2021.565657 33828499 Paillé P Mucchielli A L’analyse qualitative en sciences humaines et sociales 2016 4e Armand Colin Prime H Wade M Browne DT Risk and resilience in family well-being during the COVID-19 pandemic American Psychologist 2020 75 5 631 643 10.1037/amp0000660 32437181 Ravens-Sieberer U Herdman M Devine J Otto C Bullinger M Rose M Klasen F The European KIDSCREEN approach to measure quality of life and well-being in children : Development, current application, and future advances Quality of Life Research 2014 23 791 803 10.1007/s11136-013-0428-3 23686556 Ravens-Sieberer, U., Kaman, A., Erhart, M., Otto, C., Devine, J., Löffler, C., Hurrelmann, K., Bullinger, M., Barkmann, C., Siegel, N. A., Simon, A. M., Wieler, L. H., Schlack, R., & Hölling, H. (2021). Quality of life and mental health in children and adolescents during the first year of the COVID-19 pandemic: Results of a two-wave nationwide population-based study, –14. Advance online publication. European Child & Adolescent Psychiatry. 10.1007/S00787-021-01889-1 Russell BS Hutchison M Tambling R Tomkunas AJ Horton AL Initial challenges of caregiving during COVID-19: Caregiver burden, mental health, and the parent–child relationship Child Psychiatry and Human Development 2020 51 5 671 682 10.1007/s10578-020-01037-x 32749568 Schwartz KD Exner-Cortens D McMorris CA Makarenko E Arnold P Van Bavel M Williams S Canfield R COVID-19 and student well-being: Stress and mental health during return-to-school Canadian Journal of School Psychology 2021 36 2 166 185 10.1177/08295735211001653 34040284 Shoshani A Kor A The mental health effects of the COVID-19 pandemic on children and adolescents : Risk and protective factors Psychological Trauma: Theory, Research, Practice, and Policy 2021 14 8 1365 1373 10.1037/tra0001188 34928689 Shukla M Wu AFW Lavi I Riddleston L Hutchinson T Lau JYF A network analysis of adolescent mental well-being during the coronavirus pandemic : Evidence for cross-cultural differences in central features Personality and Individual Differences 2022 186 111316 1 10 10.1016/j.paid.2021.111316 Stoecklin D Gervais C Kutsar D Heite C Lockdown and children’s well-being: Experience of children from Switzerland, Canada and Estonia Childhood Vulnerability Journal 2021 3 41 59 10.1007/s41255-021-00015-2 Tadić Vujčić M Brajša-Žganec A Franc R Children and young peoples ’ views on well-being : A qualitative study Child Indicators Research 2019 12 791 819 10.1007/s12187-018-9559-yChildren Tessier S From field notes, to transcripts, to tape recordings: Evolution or combination? International Journal of Qualitative Methods 2012 11 4 446 460 10.1177/160940691201100410 The KIDSCREEN Group Europe The KIDSCREEN group Europe 2006 Pabst Science Publishers Thibodeau-Nielsen RB Palermo F White RE Wilson A Dier S Child adjustment during COVID-19: The role of economic hardship, caregiver stress, and pandemic play Frontiers in Psychology 2021 12 716651 10.3389/FPSYG.2021.716651/FULL 34484078 Vera EM Vacek K Blackmon S Coyle L Gomez K Jorgenson K Luginbuhl P Moallem I Steele JC Subjective well-being in urban, ethnically diverse adolescents: The role of stress and coping Youth and Society 2012 44 3 331 347 10.1177/0044118X11401432 Viner R Russell S Saulle R Croker H Stansfield C Packer J Nicholls D Goddings AL Bonell C Hudson L Hope S Ward J Schwalbe N Morgan A Minozzi S School closures during scial lockdown and mental health, health behaviors, and well-being among children and adolescents during the first COVID-19 wave: A systematic review JAMA Pediatrics 2022 176 4 400 409 10.1001/jamapediatrics.2021.5840 35040870	0	PMC9734349	NO-CC CODE	2022-12-14 23:28:28	no	BDJ In Pract. 2022 Dec 5; 35(12):31	latin-1	null	null	null	oa_other
==== Front Orth Unfallchir Orthopa¨die Und Unfallchirurgie 2193-5254 2193-5262 Springer Medizin Heidelberg 3733 10.1007/s41785-022-3733-7 Aus unserem Fach Zurück in starker Präsenz DGRh-Kongress 2022 in Berlin Gaulke Ralph grid.10423.34 0000 0000 9529 9877 Oberarzt, Unfallchirurgische Klinik der MHH, Carl-Neuberg-Str. 1, 30625 Hannover, Germany 6 12 2022 2022 12 6 4142 © DGOU und BVOU 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© DGOU und BVOU 2022 ==== Body pmcNach zwei virtuellen Rheumatologiekongressen, im Jahr 2020 in München (Präsident Dr. Martin Arbogast) und im Jahr 2021 in Nürnberg (Präsident Prof. Dr. Hans-Dieter Carl) sowie einem virtuellen DGORh-Intensivmeeting 2021 in Hannover (wissenschaftliche Leitung Prof. Dr. Ralph Gaulke) konnten sich die orthopädischen Rheumatologen in diesem Jahr endlich wieder persönlich auf Kongressebene treffen und austauschen. DGORh-Intensivmeeting 2022 Beim Intensivmeeting am 2. April 2022 in Nürnberg (wissenschaftliche Leitung Prof. Dr. Hans-Dieter Carl) folgten nach den Berichten aus den Kommissionen der Deutschen Gesellschaft für Orthopädische Rheumatologie (DGORh) Beiträge zur Bedeutung der JAK-Inhibitoren gegenüber den klassischen Biologika und zur sozialmedizinischen Begutachtung bei entzündlich-rheumatischen Erkrankungen. Im Block interdisziplinäre Rheumatologie gab es ein Update zur Hyperurikämie und Gicht und es wurden die Möglichkeiten der digitalen Rheumatologie beleuchtet. Nach dem stark philosophisch geprägten Gastvortrag von Dr. Thomas Grethlein, Aufsichtsratsvorsitzender des 1. FC Nürnberg, zur Selbsterkenntnis und freien Meinungsbildung folgte der operative Block, in dem die aktuelle S2k-Leitlinie Synovialektomie vorgestellt wurde. Im zweiten Vortrag wurden die Indikation und Möglichkeiten gelenkerhaltender Eingriffe bei entzündlich-rheumatischen Erkrankungen in Abhängigkeit von der Krankheitsaktivität diskutiert. Trotz dieser sehr interessanten Themen kam es leider aufgrund der COVID-19-Pandemie zu mehreren kurzfristigen Absagen. Dennoch war dieses Treffen als erste Präsenzveranstaltung nach fast zwei Jahren Zwangspause durchweg gelungen und wurde von allen Teilnehmern als sehr erfrischend und erhellend empfunden. Die zahlreichen Gespräche neben dem wissenschaftlichen Programm zeugten von der Lebendigkeit der Orthopädischen Rheumatologie und dem Engagement ihrer Protagonisten. Das DGORh-Intensivmeeting soll in der Zukunft stärker gewichtet werden. Neben der Ausweitung des Programms soll in diesem Rahmen auch die Mitgliederversammlung der DGORh dort stattfinden. Um pandemiebedingten Einschränkungen aus dem Weg zu gehen, wurde der Termin in die Sommermonate verlegt. Das nächste DGORh-Intensivmeeting wird am 16. und 17. Juni 2023 im Stadion des 1. FC Kaiserslautern stattfinden. Gastgeber wird Dr. Harald Dinges aus Kusel sein. DGRh-Kongress 2022 Nur fünf Monate später konnte der Deutsche Rheumatologenkongress vom 30. August bis zum 3. September 2022 erstmals seit dem Jahr 2019 ohne Einschränkungen durchgeführt werden. In Berlin konnte eine Rekordteilnehmerzahl von über 2.760 Besuchern verbucht werden. Der Kongresspräsident der DGORh, Prof. Dr. Andreas Niemeyer aus Reinbek, hatte zusammen mit Prof. Dr. Andreas Krause von der DGRh und Prof. Dr. Kirsten Minden von der GKJR ein sehr interessantes und vielseitiges Programm zusammengestellt. Die Sitzungen waren allesamt gut besucht. Die Themen reichten von konservativen Therapien wie der Schuhversorgung und der physikalischen Therapie über interdisziplinäre Themen wie verschiedene Formen der Osteopathie bei entzündlich-rheumatischen Erkrankungen bis hin zur operativen Versorgung und deren Indikationen.Die Karl-Tillmann-Gedächtnisvorlesung wurde von Prof. Dr. Wolfgang Rüther, Hamburg, zum Thema "The rise of sneaker culture" sehr lehrreich und humorvoll gehalten. Mitgliederversammlung Auf der Mitgliederversammlung beim DGRh-Kongress 2022 wurden Prof. Dr. Ralph Gaulke, Hannover, als Präsident, Dr. Ludwig Bause, Sendenhorst, als operativer Vizepräsident, Prof. Dr. Wolfgang Rüther als konservativer Vizepräsident, und Prof. Dr. Andreas Niemeier, Reinbek, als Finanzvorstand im Amt bestätigt. Prof. Dr. Erika Gromnica-Ihle, Berlin, wurde aufgrund ihrer langjährigen Verdienste um die Orthopädische Rheumatologie zum Ehrenmitglied 2024 gewählt. Neue Kommissionen Um das ganze Spektrum der Orthopädischen Rheumatologie in all seinen Facetten pointiert repräsentieren zu können, wurden drei weitere Kommissionen ins Leben gerufen: "Frauen in der Orthopädischen Rheumatologie", der Dr. Kathryn Hassel, Kassel, vorsteht, "Lehre in der Orthopädischen Rheumatologie" unter Leitung von Prof. Dr. Hans-Dieter Carl, Nürnberg, und "Kinder- und Jugendlichen Orthopädische Rheumatologie", geleitet von Dr. Martin Arbogast, Oberammergau. Die Anzahl der DGORh-zertifizierten Spezialzentren für operative Rheumatologie ist in Deutschland auf elf gestiegen. Die Zulassungskriterien sind unter www.dgorh.de für jedermann einsehbar. Alle interessierten Mitglieder sind dazu aufgerufen, entsprechende Anträge bei der Geschäftsstelle einzureichen. Die neue Zusatzweiterbildung Orthopädische Rheumatologie wurde nun auch in Bayern als letztem Bundesland ratifiziert. Die Weiterbildungszeit wurde in Verbindung mit der Anpassung an andere Zusatzweiterbildungen von drei auf zwei Jahre reduziert. Die Inhalte wurden grundlegend reformiert. Zur medikamentösen perioperativen Therapie bei entzündlich-rheumatischen Erkrankungen erschienen im Dezember 2021 neue Richtlinien, die von der DGRh unter Mitwirkung der DGORh konzipiert wurden. Die Inhalte wurden in der OUMN-Ausgabe 4/2022 aufgearbeitet und mit einem Algorithmus zum Heraustrennen publiziert. Der Arthur-Vick-Preis, mit 7.000 € dotiert, ging in diesem Jahr an die Homburger Arbeitsgruppe um Dr. Sophie Haberkamp für die Arbeit "Analysis of spatial osteochondral heterogeneity in advanced knee osteoarthritis exposes influence of joint alignment". Diese Arbeit setzte sich gegen acht weitere hochgradige Publikationen durch. Somit haben die Bewerbungen um diesen Preis einen neuen Höchststand erreicht. Das aus sieben internationalen Gutachtern bestehende Review-Board beurteilte alle Arbeiten nach festgelegten Kriterien und kam zu einem knappen Urteil zugunsten der Homburger Arbeitsgruppe. Die Verzahnung der DGORh mit den konservativen Sektionen der DGOU und den rheumatologischen Fachgesellschaften wird durch den neu initiierten Austausch von Beiratsmitgliedern weiter zu einem zukunftsfähigen Konzept ausgebaut. Alle Orthopäden und Unfallchirurgen, die Interesse an der Behandlung entzündlich-rheumatischer Systemerkrankungen haben, sind herzlich eingeladen, sich in der DGORh zu informieren und zu engagieren. Bei Interesse sprechen Sie bitte die Mitglieder des Vorstandes und des Vereins an. Mitgliedsanträge sind an die DGORh-Geschäftsstelle in der Straße des 17. Juni 106-108 in Berlin zu richten. Mehr Informationen: www.dgorh.de Prof. Dr. Ralph Gaulke Präsident der DGORh Selbstständige Sektion der DGOU	0	PMC9734350	NO-CC CODE	2022-12-14 23:28:28	no	Orth Unfallchir. 2022 Dec 6; 12(6):41-42	utf-8	null	null	null	oa_other
==== Front Curr Psychol Curr Psychol Current Psychology (New Brunswick, N.j.) 1046-1310 1936-4733 Springer US New York 4020 10.1007/s12144-022-04020-y Article Meaning in life and health behavior habits during the COVID-19 pandemic: Mediating role of health values and moderating role of conscientiousness http://orcid.org/0000-0003-4043-6047 Hu Yaqi [email protected] http://orcid.org/0000-0002-2343-7470 Lü Wei [email protected] grid.412498.2 0000 0004 1759 8395 Shaanxi Provincial Key Research Center of Child Mental and Behavioral Health, Shaanxi Key Laboratory of Behavior and Cognitive Neuroscience, School of Psychology, Shaanxi Normal University, 199 South Chang’an Road, 710062 Xi’an, China 6 12 2022 19 7 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. An increasing number of studies have explored health behavior changes since the COVID-19 outbreak, however, the potential mechanism leading to the acquisition of COVID-19-related health behavior habits remains largely underexplored. The current study aimed to investigate how meaning in life contributed to the Chinese general public’s acquisition of COVID-19-related health behavior habits, and whether health values would play a mediating role and conscientiousness would play a further moderating role in this relation. A total of 1024 Chinese participants (age range = 17–63 years; 67.29% females) were recruited by posting flyers on an open-access web forum. All participants voluntarily completed a series of online anonymous questionnaires assessing conscientiousness, meaning in life, health values and health behavior habits. Results showed that (1) the majority of the respondents reported the acquisition of COVID-19-related health behavior habits, and meaning in life positively predicted COVID-19-related health behavior habits; (2) health values mediated the link between meaning in life and health behavior habits; and (3) conscientiousness moderated the indirect effect, such that the indirect effect was stronger among individuals with low conscientiousness. These findings have important implications for revealing the reconstruction of the Chinese public’s health behavior habits and its potential mechanism that meaning in life influences health behavior habits through health values during the COVID-19 pandemic, particularly for individuals with low conscientiousness. Keywords COVID-19 pandemic Meaning in life Health values Conscientiousness Health behavior habits Social Science Foundation of Shaanxi Province (CN)2019Q014 Research Program Funds of the Collaborative Innovation Center of Assessment toward Basic Education Quality at Beijing Normal University (CN)2019-05-025-BZPK01 ==== Body pmcIntroduction As a health-threatening event, the novel coronavirus disease (COVID-19) pandemic brings stress and anxiety to the general public (DeAngelis et al., 2022; Vindegaard & Benros, 2020). People have to adopt health preventative measures to keep healthy. In this regard, the COVID-19 pandemic might change people’s health behavior habits in a positive way. The health behavior habits related to COVID-19 can be mainly categorized into proactive behaviors (e.g., washing hands frequently) and social distancing behaviors (e.g., wearing face masks) (Makhanova & Shepherd, 2020). The general public’s health behavior habits, such as maintaining social distancing and personal hygiene, have increased significantly during the COVID-19 pandemic (Bruine de Bruin & Bennett, 2020; DeAngelis et al., 2022; Schneider et al., 2021; Stangier et al., 2021; Wang et al., 2020). However, cultural differences remain to be found (Hong et al., 2021; Wang et al., 2021), particularly for the health behavior habits influenced by traditional cultures. For instance, the traditional Chinese eating habits of sharing the dishes in the serving plates might be changed or even replaced by individual dining. Although an increasing number of studies have explored health behavior changes since the COVID-19 outbreak, the potential mechanism leading to the acquisition of COVID-19-related health behavior habits remains largely underexplored. Therefore, the present study focuses on the potential mechanism that leads to the Chinese general public’s acquisition of COVID-19-related health behavior habits. The meaning and health model (Steger et al., 2014) highlights the importance of meaning in life and health orientations (including health knowledge acquisition, health attitudes and health beliefs) for understanding health-related behaviors and health status in non-clinical populations. Meaning in life might influence health-related behaviors directly or indirectly. Meaning in life, particularly the presence of meaning conceptualized as the coherent comprehension of one’s self and life experiences, has long been viewed as a precursor of health behaviors, including promoting favorable health behaviors and avoiding detrimental health behaviors (Brassai et al., 2015; Czekierda et al., 2017; Hooker & Masters, 2016). In addition, Frankl’s existential theory views meaning as the primary motivation for human behavior, which can be reflected in the attitudinal values held in difficult or long-suffering situations (Frankl, 1969). Meanwhile, health values are identified as a positive proximal predictor of health behaviors due to the health-specific internal principles and subjective attitudes (Brassai et al., 2015; Honka et al., 2019; Stapleton et al., 2020). Accordingly, meaning in life might exert indirect effect on health behaviors via health values (Steger et al., 2014). The COVID-19 pandemic has been found to reconstruct individuals’ perceptions of the meaning in life (Chen et al., 2020; Trzebiński et al., 2020). Perceived new meaning in life under the COVID-19 pandemic might foster more beneficial health values and COVID-19-related health behavior habits. However, no study has tested whether perceived meaning in life (especially the presence of meaning) would motivate individuals to acquire more beneficial health behavior habits through improving their health values under the COVID-19 pandemic. In addition to attitudinal factors, predisposing characteristics can also affect the acquisition, maintenance, or deterioration of health behavior habits (Hampson, et al., 2015; Ruiz-Palomino et al., 2018). Conscientiousness, characterized by orderliness, responsibility, impulse control, self-discipline, compliance with rules and striving for goals (Roberts et al., 2014), has a robust positive link with health behavior habits (Artese et al., 2017; Lunn et al., 2014). Recent studies have revealed that conscientious individuals are more likely to adhere to epidemic prevention guidelines and engage in more health preventative behaviors, while less conscientious individuals tend to be less involved in precautions to avoid COVID-19 infection (Aschwanden et al., 2020; Bogg & Milad, 2020; Han et al., 2021; Willroth et al., 2021). Given the individual differences in conscientiousness for shaping health values and health behavior habits, conscientiousness is assumed to play a moderating role in the mediation process involving meaning in life, health values and health behavior habits under the COVID-19 pandemic. Overall, the main purpose of this study was to investigate the potential mechanism that leads to the Chinese general public’s acquisition of COVID-19-related health behavior habits by testing a moderated mediation model. Based on previous studies, we hypothesized that (1) meaning in life would be positively related to health behavior habits; (2) health values would mediate the link between meaning in life and health behavior habits; (3) mediation paths would be moderated by conscientiousness. The hypothesized model is shown in Fig. 1. Fig. 1 The proposed model Method Participants and procedure A total of 1024 Chinese participants were recruited during the COVID-19 pandemic through an online survey link published in an open-access web forum. The sociodemographic characteristics of the whole sample are shown in Table 1. As more young people participated in this study, the age distribution skewed to the left, with 73.73% of the participants under 30 years of age. All participants aged between 17 and 63 years, and approximately 67.29% were females. They rated their socioeconomic status (SES) (1 = lowest status, 10 = highest status) slightly above the middle class (M = 5.97, SD = 1.83). The sample’s household registration distribution was village (33.11%), small town (16.31%), and large city (50.59%). Participants were required to complete a set of fully anonymous questionnaires assessing meaning in life, health values, and health behavior habits after the COVID-19 outbreak. In addition to exploring the feelings, thoughts and behaviors evoked by the ongoing COVID-19 pandemic, the present study also measured participants’ conscientiousness levels. They were instructed to honestly choose the answer that suits them best. Ethical approval was granted by the local ethics committee, and electronic informed consent was provided prior to the survey. Table 1 Sample characteristics for sociodemographic variables (N = 1024) n % / M(SD) Gender female 689 67.29% male 335 32.71% Age (in years) < 18 18 1.76% [18, 25] 617 60.25% [26, 30] 120 11.72% [31, 40] 61 5.96% [41, 50] 100 9.77% [51, 60] 86 8.40% > 60 22 2.15% Household registration village 339 33.11% small town 167 16.31% large city 518 50.59% SES — 5.97(1.83) Note. SES: socioeconomic status Measures Socioeconomic status Participants were asked to indicate their SES on a 10-point Likert scale ranging from 1 (lowest status) to 10 (highest status), in which the lowest status represents having the least money, the lowest education level, and the least respected job, while the highest status represents having the most money, the highest education level, and the most respected job (Adler et al., 2000). Conscientiousness Conscientiousness was assessed by the Chinese version of the Conscientiousness Subscale from the NEO Personality Inventory Revised (NEO-PI-R; Costa & McCrae, 1992), which was demonstrated to have good reliability and validity in the Chinese sample (Dai et al., 2004). This scale contains 48 items and 6 dimensions, including competence, order, dutifulness, achievement striving, self-discipline, and deliberation (e.g., “I’m pretty good about pacing myself so as to get things done on time”). Each item was rated on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Cronbach’s α for this scale was 0.92 in the present study. Meaning in life Meaning in life was assessed by Meaning in Life Questionnaire-Chinese version (MLQ-C; Wang et al., 2016), which was originally developed by Steger et al. (2006). Given that our study focuced on the presence of meaning in life, only the 5-item subscale measuring the presence of meaning was used (e.g., “I understand my life’s meaning”). Each item was answered on a 7-point Likert scale ranging from 1 (absolutely never true) to 7 (absolutely true). After the reverse scoring item was processed, total scores represent the level of meaning in life. This subscale showed good reliability and validity in the Chinese sample (Wang et al., 2016), and Cronbach’s α was 0.85 in this study. Health values Health values were assessed by the Health Values Questionnaire (Zhao, 2005), which showed good reliability and validity. This questionnaire consists of 45 items that measure values and beliefs about health goals and health means (e.g., “In my opinion, being energetic and not tired easily is a sign of health”). Respondents were asked to rate on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). After the reverse-scored items were reversed, higher total scores indicate more positive health values. In the present study, Cronbach’s α for this questionnaire was 0.96. Health behavior habits Health behavior habits were measured via a self-designed questionnaire. The original items were selected from the epidemic-related health and hygiene habits recommended by the World Health Organization (WHO, 2020) and the National Health Commission of China (2020). The final version of this scale was guaranteed to have sufficient content validity and construct validity prior to the formal investigation. First, item deletion and modification were determined by three experts in psychology after detailed discussion and multiple evaluations on the representativeness of each item, thus forming a 6-item preliminary test version. Second, exploratory factor analysis (EFA) on these items was conducted in a sample of 500 participants (age range = 18–60 years; 64.00% females): Kaisor-Meyer-Olkin (KMO) and Bartlett’s test of sphericity showed that the sample was adequate for factor analysis (KMO = 0.79, Bartlett’s sphericity test: χ2 = 1070.09, df = 15, p < .001) (Thompson, 2010); principal component factor analysis using Varimax rotation with Kaiser Normalization yielded 2 factors, both of which had eigenvalues greater than 1.00 and explained the 71.88% variance. The load of each item on the identified factor was above 0.74, ranging from 0.74 to 0.86. Finally, based on the EFA results, confirmatory factor analysis (CFA) on these items was further conducted in another sample of 500 participants (age range = 18–60 years; 70.80% females): a good fit was achieved in the model with 2 latent variables, χ2 (8) = 23.85 (p = .002), χ2/df ratio = 2.98, CFI = 0.988, TLI = 0.978, SRMR = 0.021, RMSEA = 0.063 (90% CI: 0.035 − 0.093) (Hu & Bentler, 1999); we also tried testing the model with 1 latent variable and found that it did not fit the data well, χ2 (9) = 125.21 (p < .001), χ2/df ratio = 13.91, CFI = 0.913, TLI = 0.855, SRMR = 0.050, RMSEA = 0.161 (90% CI: 0.136 − 0.186). The final 6-item scale assesses 6 kinds of COVID-19-related health behavior habits. Consistent with Makhanova and Shepherd (2020)’s study, its structure involves 2 factors: social distancing behavior habits (i.e., wearing a face mask when out, using serving spoons and chopsticks, and maintaining individual dining) and proactive behavior habits (i.e., covering mouth when coughing and sneezing, washing hands and disinfecting frequently, and refusing to eat wild animals). An example item is “after the COVID-19 outbreak, I got into the habit of washing hands and disinfecting frequently, and I will keep this habit in my life even if the epidemic is over”. Each item was rated on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Higher total scores indicate more beneficial health behavior habits. In the present study, this questionnaire also showed good reliability (Cronbach’s α = 0.85). Statistical analyses The statistical analyses in this study were performed by SPSS 25.0. First, descriptive analysis and Pearson correlation analysis were performed on the key variables. Second, PROCESS v3.4 macro (Hayes, 2018) was adopted for conditional process analysis to test the moderated mediation model. In specific, 95% percentile bootstrap confidence intervals (CIs) were calculated based on 5000 bootstrapped samples of the data to test the indirect effect. The grouping conditions were 1 SD below the mean conscientiousness, the mean conscientiousness, and 1 SD above the mean conscientiousness. All continuous variables were standardized in the inferential analysis. As illustrated in Fig. 1, we hypothesized that conscientiousness would moderate the first (the path from meaning in life to health values) and second (the path from health values to health behavior habits) stages of the indirect effect, as well as the residual direct effect of meaning in life on health behavior habits. Assessment of common method biases Since all questionnaires were self-rated, common method bias was assessed using the Harman single factor test (Podsakoff et al., 2003). A total of 104 items from the questionnaires were included in the exploratory factor analysis without rotation. The analysis produced 18 common factors with the first explaining the 26.68% variance, which was lower than the critical standard of 40% (Zhou & Long, 2004). This result showed that there was no common method bias in the present study. Results Preliminary analyses Since each item of the online survey was mandatory before submission, there was no missing data in the present study. Descriptive statistics for self-reported health behavior habits during the COVID-19 pandemic are shown in Table 2. The majority of the respondents clearly reported that they got into the COVID-19-related health behavior habits after the COVID-19 outbreak, including wearing a face mask when out (69.53%), using serving spoons and chopsticks (66.70%), maintaining individual dining (62.89%), covering mouth when coughing and sneezing (86.04%), washing hands and disinfecting frequently (82.71%), and refusing to eat wild animals (87.11%). They also reported that they would maintain such habits in the future. Table 2 Descriptive statistics for self-reported health behavior habits during the COVID-19 pandemic (N = 1024) M(SD) strongly disagree n (%) somewhat disagree n (%) not sure n (%) somewhat agree n (%) strongly agree n (%) Wearing a face mask when out 3.90(1.15) 50(4.88) 85(8.30) 177(17.29) 318(31.05) 394(38.48) Using serving spoons and chopsticks 3.85(1.14) 48(4.69) 90(8.79) 203(19.82) 314(30.66) 369(36.04) Maintaining individual dining 3.78(1.14) 44(4.30) 105(10.25) 231(22.56) 300(29.30) 344(33.59) Covering mouth when coughing and sneezing 4.44(0.99) 35(3.42) 31(3.03) 77(7.52) 188(18.36) 693(67.68) Washing hands and disinfecting frequently 4.30(0.99) 35(3.42) 22(2.15) 120(11.72) 275(26.86) 572(55.86) Refusing to eat wild animals 4.55(0.96) 35(3.42) 18(1.76) 79(7.71) 107(10.45) 785(76.66) The means, SDs, and bivariate correlations of study variables are shown in Table 3. Results showed that meaning in life was positively associated with health values and health behavior habits. Health values were found to be positively associated with health behavior habits. In addition, conscientiousness was positively related to meaning in life, health values and health behavior habits. Table 3 Zero-order correlations between all study variables (N = 1024) M SD 1 2 3 1. Presence of meaning 24.55 5.32 — 2. Health values 175.76 25.92 0.54* — 3. Health behavior habits 24.81 4.82 0.28* 0.46* — 4. Conscientiousness 165.88 22.28 0.56* 0.54* 0.36* Note.*p < .001 Conditional process analysis A collinearity diagnosis analysis was performed, showing no overlap between all study variables (tolerance values were 0.52–0.60). Then PROCESS v3.4 macro (Model 59; Hayes, 2018) was utilized to examine the moderated mediation model. Given that demographic variables are not predictors of the key variables in this study, we only present the analysis results without controlling for demographic variables. This study found that the pattern of findings was unchanged even after controlling for gender, age, household registration, and SES. As displayed in Table 4, meaning in life significantly predicted health values (β = 0.34, t = 11.44, p < .001), but the interaction of meaning in life and conscientiousness on health values was not significant (β = −0.03, t = − 1.59, p > .05). These predictors in Equation 1 accounted for 37% of the variance in health values (F = 202.91, p < .001, R2 = 0.37). Equation 2 was also significant (F = 62.60, p < .001, R2 = 0.24). Health values significantly predicted health behavior habits (β = 0.34, t = 9.15, p < .001), and the interaction of health values and conscientiousness on health behavior habits was significant (β = −0.09, t = − 2.61, p = .009). However, the residual direct effect of meaning in life on health behavior habits was not significant (β = −0.00, t = − 0.01, p > .05), nor moderated by conscientiousness (β = 0.05, t = 1.73, p > .05). Table 4 Testing the moderated mediation model (N = 1024) Equation 1 (Health values) Equation 2 (Health behavior habits) β t 95% CI β t 95% CI Meaning in life 0.34 11.44* [0.284, 0.402] −0.00 −0.01 [− 0.070, 0.078] C 0.35 11.61* [0.291, 0.409] 0.18 5.04* [0.111, 0.252] Meaning in life × C −0.03 −1.59 [− 0.075, 0.008] 0.05 1.73 [− 0.007, 0.114] Health values 0.34 9.15* [0.268, 0.414] Health values × C −0.09 −2.61 [− 0.153, − 0.022] R 2 0.37 0.24 F 202.91* 62.60* Note. C: conscientiousness. All continuous variables were standardized. The variables in parentheses are the outcome variables of each corresponding equation. p < .01; *p < .001 A simple slope test (Cohen et al., 2003) was used to further demonstrate the significance of the second path of the mediation process at 1 SD below and above the mean value of conscientiousness. As exhibited in Fig. 2, the effect of health values on health behavior habits was stronger among individuals with low conscientiousness (β = 0.43, t = 10.75, p < .001) than those with high conscientiousness (β = 0.25, t = 4.33, p < .001). Fig. 2 The significant interaction between health values and conscientiousness in predicting health behavior habits. Note. All continuous variables were standardized. Low and high conscientiousness represent 1 SD below and above the mean value of conscientiousness, respectively The percentile bootstrap method was applied to test the conditional indirect effect in the mediation process, which indicated that the indirect effect of meaning in life on health behavior habits through health values was more remarkable among individuals low in conscientiousness (conditional indirect effect = 0.16, SE = 0.03, 95% CI = [0.106, 0.230]) than those high in conscientiousness (conditional indirect effect = 0.08, SE = 0.03, 95% CI = [0.023, 0.133]). Discussion The present study advanced existing research by establishing a moderated mediation model to explore whether meaning in life would be indirectly associated with COVID-19-related health behavior habits through health values among the Chinese general public, and whether the indirect association would be moderated by conscientiousness. Results showed that meaning in life positively predicted COVID-19-related health behavior habits, and that health values mediated the association between meaning in life and health behavior habits. Moreover, the indirect effect of meaning in life on health behavior habits through health values was more remarkable among individuals with low conscientiousness than those with high conscientiousness. As expected, we found that the Chinese general public’s health behavior habits were reconstructed in a positive way after the COVID-19 outbreak, particularly the changes of traditional health behavior habits such as increased individual dining and using serving spoons and chopsticks. Adding empirical evidence to the well-established link between meaning in life and health behaviors (e.g., Brassai et al., 2015; Czekierda et al., 2017; Hooker & Masters, 2016; Steger et al., 2014), meaning in life was found to positively predict COVID-19-related health behavior habits. Meaning in life, as the primary motivation for human behavior (Frankl, 1969), motivated people to actively engage in rebuilding their behavior habits to stay healthy under the COVID-19 pandemic. Echoing the meaning and health model of Steger et al. (2014) and their findings, mediation results in this study suggested that health values acted as a psychological mechanism linking meaning in life with health behavior habits during the COVID-19 pandemic. In other words, perceived meaning in life might contribute to fostering more beneficial health values, which in turn facilitates the acquisition of better health behavior habits in the context of COVID-19 pandemic. Supporting Frankl’s view that meaning in life is reflected in the attitudinal values held in difficult or long-suffering situations (Frankl, 1969), we found that meaning in life was positively associated with health values. This finding is also consistent with the perspective of post-traumatic growth that individuals may find new meaning in life under a perceived life-threatening situation, reappraise their comprehension of life, and shift toward an endorsement of positive schematic changes, such as enhanced health consciousness (Dursun et al., 2016; Linley & Joseph, 2011; Roepke et al., 2014; Triplett et al., 2012). We also found health values were positively related to health behavior habits. This finding adds to existing research, indicating that health values serve as a positive proximal predictor of health behaviors (Brassai et al., 2015; Honka et al., 2019; Niu et al., 2021; Nowak et al., 2020; Stapleton et al., 2020). The improved health values are indicative of stronger health consciousness, more reasonable health goals, and more beneficial health means (Zhao, 2005), which can provide guidance and direction for individuals to foster health behavior habits that result in favorable outcomes. Furthermore, the indirect association between meaning in life and health behavior habits through health values was found to be partially moderated by conscientiousness. In specific, the influence of health values on health behavior habits was more remarkable among individuals with low conscientiousness. This result suggests that individuals with low conscientiousness are more likely to acquire better health behavior habits benefiting from positive health values than those with high conscientiousness. Individuals with low conscientiousness who lack orderliness and self-discipline tend to have negative health values and unhealthy personal hygiene habits (Armon, 2014; Franks et al., 2009), and are more susceptible to the fear-inducing pandemic (Zhang et al., 2021). Thus, they are more likely to develop beneficial health behavior habits by enhancing health consciousness and adjusting health goals and means after the COVID-19 outbreak. In conclusion, these findings extend previous studies and add new empirical evidence under the context of the COVID-19 pandemic. As a major public health emergency, the COVID-19 pandemic has reshaped the Chinese general public’s health behavior habits, with a potential mechanism that meaning in life fosters beneficial health behavior habits via improving health values, particularly among individuals with low conscientiousness. These findings also have important implications for public health services to guide the development of personalized recommendations during the long-term fight against the epidemic. Admittedly, there are several limitations that should be considered. First, all variables in this study were self-reported. Further empirical research can apply objective measures or multiple informants to reduce the shared method variance. Second, selection bias might exist because this research relied on a monocultural sample, and the survey sampling was online and voluntary. The over-sampling of educated populations and Internet users might limit the generalizability of current research findings to the general population throughout the country and even in other cultures. Future research could also distribute paper questionnaires offline to test different sample groups. Finally, due to the unpredictability and urgency of the pandemic, the cross-sectional results of this study cannot reveal the true changes and relationships of all the research variables before and after the COVID-19 outbreak. Although these findings provide some interesting insights, future studies can examine the dynamic changes in research variables and the long-term impact of the epidemic on the Chinese public through longitudinal follow-up. Acknowledgements We would like to thank Chunyu Xie, Ziyan Yao, Sheng Wang, Huayu Ji, Yu Zhang, and Wenke Zhu at Shaanxi Normal University for their help in data collection. We also thank all participants who contributed to this study. Authors’ contribution statements This research was mainly directed by Wei Lü. The study conception and design were proposed by Wei Lü. Material preparation, data collection and analysis were performed by Wei Lü and Yaqi Hu. The first draft of the manuscript was written by Yaqi Hu, and was reviewed by Wei Lü. Both authors commented on previous versions of the manuscript and approved the final manuscript. Funding This work was supported by the Social Science Foundation of Shaanxi Province of China (2019Q014) and the Research Program Funds of the Collaborative Innovation Center of Assessment toward Basic Education Quality at Beijing Normal University (2019-05-025-BZPK01) awarded to Wei Lü. Data availability The data that support the findings of this study are available from the corresponding author (Wei Lü) upon reasonable request. Declarations Conflict of interest The authors declare that they have no conflict of interest. Ethics approval All procedures performed in this study were in accordance with the principles of the Declaration of Helsinki and the ethical standards of the Academic Committee of Shaanxi Normal University (Xi’an, China). Consent to participate Informed consent was obtained from 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 Adler NE Epel ES Castellazzo G Ickovics JR Relationship of subjective and objective social status with psychological and physiological functioning: Preliminary data in healthy white women Health Psychology 2000 19 6 586 592 10.1037//0278-6133.19.6.586 11129362 Armon G Personality and serum lipids: Does lifestyle account for their concurrent and long-term relationships European Journal of Personality 2014 28 6 550 559 10.1002/per.1943 Artese A Ehley D Sutin AR Terracciano A Personality and actigraphy-measured physical activity in older adults Psychology and Aging 2017 32 2 131 138 10.1037/pag0000158 28287783 Aschwanden D Strickhouser JE Sesker AA Lee JH Luchetti M Stephan Y Sutin AR Terracciano A Back M Psychological and behavioural responses to coronavirus disease 2019: The role of personality European Journal of Personality 2020 35 1 51 66 10.1002/per.2281 Bogg T Milad E Demographic, personality, and social cognition correlates of coronavirus guideline adherence in a U.S. sample Health Psychology 2020 39 12 1026 1036 10.1037/hea0000891 33252928 Brassai L Piko BF Steger MF A reason to stay healthy: The role of meaning in life in relation to physical activity and healthy eating among adolescents Journal of Health Psychology 2015 20 5 473 482 10.1177/1359105315576604 25903235 de Bruine W Bennett D Relationships between initial COVID-19 risk perceptions and protective health behaviors: A national survey American Journal of Preventive Medicine 2020 59 2 157 167 10.1016/j.amepre.2020.05.001 32576418 Chen C Zhang Y Xu A Chen X Lin J Reconstruction of meaning in life: Meaning made during the pandemic of COVID-19 International Journal of Mental Health Promotion 2020 22 3 173 184 10.32604/ijmhp.2020.011509 Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral sciences (3rd ed.). Lawrence Erlbaum Associates, Inc. Costa, P. T. J., & McCrae, R. R. (1992). NEO-PI-R professional manual. Psychological Assessment Resources, Inc. Czekierda K Banik A Park CL Luszczynska A Meaning in life and physical health: Systematic review and meta-analysis Health Psychology Review 2017 11 4 387 418 10.1080/17437199.2017.1327325 28488471 Dai X Yao S Cai T Yang J Reliability and validity of the NEO-PI-R in Mainland China Chinese Mental Health Journal 2004 18 3 171 174 DeAngelis BN Salah AB al’Absi M Stress and COVID-19 related behaviours: The mediating role of delay discounting Stress and Health 2022 38 1 140 146 10.1002/smi.3060 33955660 Dursun P Steger MF Bentele C Schulenberg SE Meaning and posttraumatic growth among survivors of the September 2013 Colorado Floods Journal of Clinical Psychology 2016 72 12 1247 1263 10.1002/jclp.22344 27459242 Frankl, V. E. (1969). The will to meaning: Foundations and applications of logotherapy. New American Library. Franks P Chapman B Duberstein P Jerant A Five factor model personality factors moderated the effects of an intervention to enhance chronic disease management self-efficacy British Journal of Health Psychology 2009 14 3 473 487 10.1348/135910708X360700 18808733 Hampson SE Edmonds GW Goldberg LR Dubanoski JP Hillier TA A life-span behavioral mechanism relating childhood conscientiousness to adult clinical health Health Psychology 2015 34 9 887 895 10.1037/hea0000209 25622076 Han Y Jang J Cho E Choi K Investigating how individual differences influence responses to the COVID-19 crisis: The role of maladaptive and five-factor personality traits Personality and Individual Differences 2021 176 110786 10.1016/j.paid.2021.110786 33642660 Hayes, A. F. (2018). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach (2nd ed.). Guilford Press. Hong W Liu RD Ding Y Hwang J Wang J Yang Y Cross-country differences in stay-at-home behaviors during peaks in the COVID-19 pandemic in China and the United States: The roles of health beliefs and behavioral intention International Journal of Environmental Research and Public Health 2021 18 4 2104 10.3390/ijerph18042104 33670078 Honka AM Helander E Pavel M Jimison H Mustonen P Korhonen I Ermes M Exploring associations between the self-reported values, well-being, and health behaviors of Finnish citizens: Cross-sectional analysis of more than 100,000 Web-survey responses JMIR Mental Health 2019 6 4 e12170 10.2196/12170 31008710 Hu L Bentler PM Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives Structural Equation Modeling: A Multidisciplinary Journal 1999 6 1 1 55 10.1080/10705519909540118 Linley PA Joseph S Meaning in life and posttraumatic growth Journal of Loss and Trauma 2011 16 2 150 159 10.1080/15325024.2010.519287 Lunn TE Nowson CA Worsley A Torres SJ Does personality affect dietary intake? Nutrition (Burbank, Los Angeles County, Calif.) 2014 30 4 403 409 10.1016/j.nut.2013.08.012 24607300 Makhanovaa A Shepherdb MA Behavioral immune system linked to responses to the threat of COVID-19 Personality and Individual Differences 2020 167 110221 10.1016/j.paid.2020.110221 32834281 National Health Commission of the People’s Republic of China (2020, January 31). Guidelines for the protection of people at different risks of novel coronavirus (2019-nCoV) infection. http://www.nhc.gov.cn/xcs/zhengcwj/202001/a3a261dabfcf4c3fa365d4eb07ddab34.shtml Niu, Z., Qin, Z., Hu, P., & Wang, T. (2021). Health beliefs, trust in media sources, health literacy, and preventive behaviors among high-risk Chinese for COVID-19. Health Communication. Advance online publication. 10.1080/10410236.2021.1880684 Nowak B Brzoska P Piotrowski J Sedikides C Zemojtel-Piotrowska M Jonason PK Adaptive and maladaptive behavior during the COVID-19 pandemic: The roles of Dark Triad traits, collective narcissism, and health beliefs Personality and Individual Differences 2020 167 110232 10.1016/j.paid.2020.110232 32834282 Podsakoff PM MacKenzie SB Lee JY Podsakoff NP Common method biases in behavioral research: A critical review of the literature and recommended remedies Journal of Applied Psychology 2003 88 5 879 903 10.1037/0021-9010.88.5.879 14516251 Roberts BW Lejuez C Krueger RF Richards JM Hill PL What is conscientiousness and how can it be assessed? Developmental Psychology 2014 50 5 1315 1330 10.1037/a0031109 23276130 Roepke AM Jayawickreme E Riffle OM Meaning and health: A systematic review Applied Research in Quality of Life 2014 9 1055 1079 10.1007/s11482-013-9288-9 Ruiz-Palomino E Gimenez-Garcia C Ballester-Arnal R Gil-Llario MD Health promotion in young people: Identifying the predisposing factors of self-care health habits Journal of Health Psychology 2018 25 10–11 1410 1424 10.1177/1359105318758858 29468900 Schneider, C. R., Dryhurst, S., Kerr, J., Freeman, A. L. J., Recchia, G., Spiegelhalter, D., & van der Linden, S. (2021). COVID-19 risk perception: A longitudinal analysis of its predictors and associations with health protective behaviours in the United Kingdom. Journal of Risk Research, 24(3–4), 294–313. 10.1080/13669877.2021.1890637 Stangier, U., Kananian, S., & Schüller, J. (2021). Perceived vulnerability to disease, knowledge about COVID-19, and changes in preventive behavior during lockdown in a German convenience sample. Current Psychology. Advance online publication. 10.1007/s12144-021-01456-6 Stapleton A O’Connor M Feerick E Kerr J McHugh L Testing the relationship between health values consistent living and health-related behavior Journal of Contextual Behavioral Science 2020 17 17 22 10.1016/j.jcbs.2020.05.002 Steger MF Fitch-Martin AR Donnelly J Rickard KM Meaning in life and health: Proactive health orientation links meaning in life to health variables among American undergraduates Journal of Happiness Studies 2014 16 3 583 597 10.1007/s10902-014-9523-6 Steger MF Frazier P Oishi S Kaler M The meaning in life questionnaire: Assessing the presence of and search for meaning in life Journal of Counseling Psychology 2006 53 1 80 93 10.1037/0022-0167.53.1.80 Thompson, B. (2010). Exploratory and confirmatory factor analysis: Understanding concepts and applications. American Psychological Association. Triplett KN Tedeschi RG Cann A Calhoun LG Reeve CL Posttraumatic growth, meaning in life, and life satisfaction in response to trauma Psychological Trauma: Theory Research Practice and Policy 2012 4 4 400 410 10.1037/a0024204 Trzebiński, J., Cabański, M., & Czarnecka, J. Z. (2020). Reaction to the COVID-19 pandemic: The influence of meaning in life, life satisfaction, and assumptions on world orderliness and positivity. Journal of Loss and Trauma, 25(6–7), 544–557. 10.1080/15325024.2020.1765098 Vindegaard, N., & Benros, M. E. (2020). COVID-19 pandemic and mental health consequences: Systematic review of the current evidence. Brain Behavior and Immunity, 89, 531–542. 10.1016/j.bbi.2020.05.048 Wang C Pan R Wan X Tan Y Xu L Ho CS Ho RC Immediate psychological responses and associated factors during the initial stage of the 2019 coronavirus disease (COVID-19) epidemic among the general population in China International Journal of Environmental Research and Public Health 2020 17 5 1729 10.3390/ijerph17051729 32155789 Wang C Tripp C Sears SF Xu L Tan Y Zhou D Ma W Xu Z Chan NA Ho C Ho R The impact of the COVID-19 pandemic on physical and mental health in the two largest economies in the world: A comparison between the United States and China Journal of Behavior Medicine 2021 44 741 759 10.1007/s10865-021-00237-7 Wang X You Y Zhang D Psychometric properties of Meaning in Life Questionnaire-Chinese version (MLQ-C) in Chinese university students and its relations with psychological quality Journal of Southwest University (Natural Science Edition) 2016 38 10 161 167 Willroth E Smith AM Shallcross AJ Graham EK Mroczek DK Ford BQ The health behavior model of personality in the context of a public health crisis Psychosomatic Medicine 2021 83 4 363 367 10.1097/PSY.0000000000000937 33790198 World Health Organization (2020, March 15). Coronavirus disease (COVID-19) advice for the public. https://www.who.int/emergencies/diseases/novel-coronavirus-2019/advice-for-public Zhang X Wang Y Lyu H Zhang Y Liu Y Luo J The influence of COVID-19 on the well-being of people: Big data methods for capturing the well-being of working adults and protective factors nationwide Frontiers in Psychology 2021 12 681091 10.3389/fpsyg.2021.681091 34234720 Zhao, C. (2005). The designing of the Undergraduates’ Health Values Questionnaire [Unpublished master’s thesis]. Southwest University of China. Zhou H Long L Statistical remedies for common method biases Advances in Psychological Science 2004 12 6 942 950 10.1007/BF02911031 Hooker Stephanie A Masters Kevin S Purpose in life is associated with physical activity measured by accelerometer Journal of Health Psychology 2016 21 6 962 971 10.1177/1359105314542822 25104777	0	PMC9734351	NO-CC CODE	2022-12-14 23:28:28	no	Curr Psychol. 2022 Dec 6;:1-9	utf-8	Curr Psychol	2,022	10.1007/s12144-022-04020-y	oa_other
==== Front Indian J Otolaryngol Head Neck Surg Indian J Otolaryngol Head Neck Surg Indian Journal of Otolaryngology and Head & Neck Surgery 2231-3796 0973-7707 Springer India New Delhi 3177 10.1007/s12070-022-03177-z Original Article Dysphagia in post Covid-19 Patients- a Prospective Cohort Study http://orcid.org/0000-0002-7756-6681 Sharma Priyam [email protected] Khaund Gautam [email protected] Agarwal Vivek [email protected] Barman Surajit [email protected] Baruah Debika [email protected] grid.477599.1 0000 0004 1802 510X Nightingale Hospital, 781005 Guwahati, Assam India 6 12 2022 16 18 2 2022 23 9 2022 © Association of Otolaryngologists of India 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. Never in recent history has mankind been so severely and diversely affected by any disease like the COVID-19 infection. Many post-COVID complications have been mentioned in the literature and other platforms, of which post-COVID Dysphagia is a very distressing complaint. The severity of dysphagia may range from mild discomfort in swallowing to life-threatening aspiration. This paper aims to study post-COVID dysphagia, its various presentations, possible causative factors and diagnosis. Like any other new disease on the block, continuous study and research is the need of the hour, for us to be able to mitigate the damage already inflicted by this pandemic. Keywords Post COVID 19 Dysphagia Presentation Findings ==== Body pmcIntroduction The SARS CoV 2 virus has put a halt to the entire world across boundaries and affected people in different ways. Even after recovery, patients have been found to present with varying complaints like loss of smell, taste, weakness etc. among which difficulty in swallowing is emerging as a very distressing complaint. Different causes have been mentioned in the limited literature available, including cranial nerve involvement, damage to peripheral nerves involved in swallowing reflex, prolonged intubation, tracheostomy, poor lung function among others. This paper aims to study the presentation and findings in patients with dysphagia after recovering from Covid-19. AIM To study and document. Ways of presentation (chief complaints). Functional Oral Intake Scale (FOIS). Findings in Fiberoptic Endoscopic Evaluation of Swallowing (FEES) in patients complaining of difficulty in swallowing post recovery from Covid-19. Methodology This study is an ongoing Prospective study being conducted at Nightingale Hospital and Pratiksha Hospital, Guwahati which started in March 2021. Informed consent was taken regarding the endoscopic procedure and inclusion in the study. All patients were included who presented with onset of difficulty in swallowing after suffering from Covid-19(RT-PCR proven). Complete history was taken followed by general examination and then ENT Examination was done. The chief complaints were recorded and grading was done based on Functional Oral Intake Scale (FOIS). Subsequently, all patients were put up for Fiberoptic Endoscopic Evaluation of Swallowing (FEES) [Fig. 1]. With the patient in sitting position (Fig. 1), Topical decongestant drops (Xylometazoline) and topical anaesthetic (4% lignocaine) were applied in the nasal cavities (throat anesthesia was avoided so as not to interfere with swallowing reflex) and a fiberoptic laryngoscope was passed into the throat of the patients. After initial anatomical and physiological assessment, dyed liquids and solids were given to the patients and findings were noted in a proforma. Based on the findings, appropriate diet modifications, swallowing therapies and maneuvers were advised to the patients. Follow-up was done between 6 weeks to 8 weeks to after the initial assessment and a repeat FEES was done in all the patients who turned up for the review. Fig. 1 FEES in progress Results and Observation A total of 41 patients were included in the study, of which 27 were males and 14 were females. Patients presented with complaints of frank difficulty in swallowing (19), Choking episodes (11), lump sensation in throat after every swallow (5) and nasal regurgitation of food (6) [Fig. 2]. Out of 41 patients, 22 patients required hospital admission for covid-19, of which 13 required non-invasive ventilation, 7 required mechanical ventilation while 2 patients had to undergo tracheostomy [Fig. 3]. Fig. 2 Depicting distribution of chief complaints in percentage Fig. 3 Distribution of No. of patients based on treatment course during Covid -19 On presentation- The grading of food intake was done based on Functional Oral Intake Scale [Fig. 4]. The FOIS score was Level 1 in 8 patients, Level 2–3 in 12 patients, Level 4–6 in 11 patients and Level 7 in 10 patients [Fig. 5]. The FEES findings were inadequate velopharyngeal closure indicating palatal palsy (7), pre mature spillage (3), Residue in PFS (7) [Fig. 6], Penetration (that is entry of food in laryngeal inlet but remaining above the level of vocal cords) (6) [Fig. 7] and Aspiration (that is entry of food below the level of vocal cords) (9) [Fig. 8]. Findings within normal limit were seen in 9 patients [Fig. 9]. Fig. 4 Functional Oral Intake Scale Fig. 5 Distribution of patients (percentage) based on FOIS grading in initial assessment Fig. 6 Depicting PFS residue in a patient post mechanical ventilation and in dwelling Ryle’s tube for over a month Fig. 7 Penetration as seen by entry of dyed liquid into laryngeal inlet Fig. 8 Aspiration seen in post tracheostomized patient, almost no cough was seen in response to the massive aspiration Fig. 9 Different FEES findings in patients with dysphagia (Initial assessment) Follow-up, though advised in all the patients, could only be done in 33 who turned up for the review. On follow -up FOIS grading was Level 1 in 4 patients, Level 2–3 in 5 patients, Level 4–6 in 10 patients and Level 7 in 14 patients [Fig. 10]. FEES findings were inadequate velopharyngeal closure (3), pre mature spillage (0), Residue in PFS (4), penetration (4) and aspiration (3) [Figs. 11 and 12]. Findings within normal limit were seen in 19 patients. Fig. 10 FOIS grading in subsequent follow up Fig. 11 FEES finding in follow up (in a total of 33 patients) Fig. 12 Reduction in amount of residue in PFS following swallowing therapy Discussion Among all the diseases currently afflicting the world, Covid-19 is probably the most treacherous owing to its varying presentations and unpredictable nature. As with the disease, the treatment protocols are also changing shape as time progresses. Patients, however, are suffering not only during but also after recovering from this deadly disease. Difficulty in swallowing is one such complaint post covid which hampers the complete physical recovery of the patient. The most commonly postulated reason for post -COVID dysphagia is the involvement of the cranial nerves responsible for the swallowing reflex namely the lingual nerve, glossopharyngeal, vagus and hypoglossal nerves. Another very plausible reason is prolonged intubation of the patient which may alter the sensation of the lower airway, thus suppressing the protective cough reflex even in the presence of aspiration. Tracheostomized patients may also experience varying degree of dysphagia owing to insufficient laryngeal elevation and decreased sub-glottic pressure. Angela Cavalagli et al. [1] have put forth the fact that cranial nerve, particularly bulbar nerves could be involved as late complications. Ulrike Frank et al. [2] have also, in their paper, discussed about primary damage to the central and peripheral neuronal swallowing network as a cause of swallowing disorders in post covid patients. Another study by Julie Regan et al. [3] investigates post extubation dysphagia and dysphonia and emphasized prompt evaluation and intervention to minimize complications and inform rehabilitation planning. Timely intervention with dysphagia therapy is mandatory to mitigate the potentially negative consequences of prolonged intubation and long term use of cuffed tracheostomy tube and allow the patients to be able to eat what they want and speak with family. Athanasia Printza et al. [4] have also investigated the severity of dysphagia in Covid-19. Margareta Gonzalez Lindh et al. [5] have explored the swallowing function and risk factors associated with delayed recovery of swallowing in covid-19 patients post invasive mechanical ventilation using the Functional Oral Intake Scale. Maria Raffaella Marchese et al. [6], in their study, found that the prevalence of upper dysphagia after hospitalization for SARS-CoV2 is not anecdotal and that probably this long lasting sequela has psychogenic etiology. One very effective way of improving pharyngeal function via Pharyngeal Electrical Stimulation (PES) has been postulated by Marianna Traugott et al. [7] where they also state that PES can restore safe swallowing in orally intubated or tracheotomized ICU patients. Conclusion As Covid-19 keeps on changing forms with new presentations and features, as clinicians we too need to update ourselves with newer information regarding this subject. Complete patient recovery is not always achieved on the day the RT PCR report comes negative. Dysphagia as a post covid complaint can not and should not be taken lightly as it can severely impact the overall physical recovery of the patient if the nutritional needs of the patient are not met with adequately. Early diagnosis and intervention can not only ensure proper nutrition but also prevent life threatening aspiration. Since Covid-19 primarily affects the airway, a lot of patients may have a non-responsive airway paving way for the sinister silent aspiration. We, as otolaryngologists, should be on our toes and promptly investigate any patient who comes to us with any complaint related to swallowing after he/she has recovered from Covid-19. Only when we keep our minds open, we have a chance of mitigating the damage inflicted by this pandemic. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Cavalagli A et al (2020) Cranial nerves impairment in post-acute oropharyngeal dysphagia after COVID-19. Eur J Phys Rehabil Med 56(6):853–857 2. Frank U et al (2021) COVID-19-New challenges in dysphagia and respiratory therapy. Nervenarzt 93(2):167–174 3. Regan J et al (2021) Post-extubation dysphagia and dysphonia amongst adults with COVID-19 in the Republic of Ireland: A prospective multi-site observational cohort study. Clin Otolaryngol 46(6):1290–1299 4. Printza A et al (2021) Dysphagia severity and management in patients with COVID-19. Curr Health Sci J 47(2):147–156 5. Gonzalez Lindh M et al (2022) Swallowing function in COVID-19 patients after invasive mechanical ventilation. Arch Rehabil Res Clin Transl 4(1):100177 6. Marchese MR et al (2021) Oropharyngeal dysphagia after hospitalization for COVID-19 disease: our screening results. Dysphagia 37(2):447–453 7. Traugott M et al (2021) Successful treatment of intubation-induced severe neurogenic post -extubation dysphagia using pharyngeal electrical stimulation in a COVID-19 survivor: a case report. J Med Case Rep 15(1):148	0	PMC9734353	NO-CC CODE	2022-12-14 23:28:28	no	Indian J Otolaryngol Head Neck Surg. 2022 Dec 6;:1-6	utf-8	Indian J Otolaryngol Head Neck Surg	2,022	10.1007/s12070-022-03177-z	oa_other
==== Front Sci China Life Sci Sci China Life Sci Science China. Life Sciences 1674-7305 1869-1889 Science China Press Beijing 36469218 2230 10.1007/s11427-022-2230-4 Review Current progress in the development of prophylactic and therapeutic vaccines Li Tingting 123 Qian Ciying 123 Gu Ying 123 Zhang Jun 123 Li Shaowei [email protected] 123 Xia Ningshao [email protected] 1234 1 grid.12955.3a 0000 0001 2264 7233 State Key Laboratory of Molecular Vaccinology and Molecular Diagnostics, School of Life Sciences, School of Public Health, Xiamen University, Xiamen, 361102 China 2 grid.12955.3a 0000 0001 2264 7233 National Institute of Diagnostics and Vaccine Development in Infectious Diseases, Xiamen University, Xiamen, 361102 China 3 Xiang An Biomedicine Laboratory, Xiamen, 361102 China 4 The Research Unit of Frontier Technology of Structural Vaccinology of Chinese Academy of Medical Sciences, Xiamen, 361102 China 2 12 2022 132 8 6 2022 21 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. Vaccines are essential public health tools and play an important role in reducing the burden of infectious diseases in the population. Emerging infectious diseases and outbreaks pose new challenges for vaccine development, requiring the rapid design and production of safe and effective vaccines against diseases with limited resources. Here, we focus on the development of vaccines in broad fields ranging from conventional prophylactic vaccines against infectious diseases to therapeutic vaccines against chronic diseases and cancer providing a comprehensive overview of recent advances in eight different vaccine forms (live attenuated vaccines, inactivated vaccines, polysaccharide and polysaccharide conjugate vaccines, recombinant subunit vaccines, virus-like particle and nanoparticle vaccines, polypeptide vaccines, DNA vaccines, and mRNA vaccines) and the therapeutic vaccines against five solid tumors (lung cancer breast cancer colorectal cancer liver cancer and gastric cancer), three infectious diseases (human immunodeficiency virus, hepatitis B virus and human papillomavirus-induced diseases) and three common chronic diseases (hypertension, diabetes mellitus and dyslipidemia). We aim to provide new insights into vaccine technologies, platforms, applications and understanding of potential next-generation preventive and therapeutic vaccine technologies paving the way for the vaccines design in the future. Keywords prophylactic vaccine therapeutic vaccine immune response infectious diseases cancers chronic diseases ==== Body pmcAcknowledgements This work was supported by the National Key Research and Development Program of China (2021YFC2301404), the National Natural Science Foundation of China (81991490, 82001756), the Health Education Joint Project of Fujian Province (2019-WJ-05), the President Foundation of Xiamen University (20720200062), and CAMS Innovation Fund for Medical Sciences of China (2019RU022). Data Availability Statement All relevant data are within the manuscript. Compliance and ethics The author(s) declare that they have no conflict of interest. Contributed equally to this work ==== Refs References Akhatova A Chan CK Azizan A Aimagambetova G The efficacy of therapeutic DNA vaccines expressing the human papillomavirus E6 and E7 oncoproteins for treatment of cervical cancer: systematic review Vaccines 2021 10 53 10.3390/vaccines10010053 35062714 Al Mahtab M Akbar SMF Aguilar JC Guillen G Penton E Tuero A Yoshida O Hiasa Y Onji M Treatment of chronic hepatitis B naïve patients with a therapeutic vaccine containing HBs and HBc antigens (a randomized, open and treatment controlled phase III clinical trial) PLoS ONE 2018 13 e0201236 10.1371/journal.pone.0201236 30133478 Al-Batran SE Ajani JA Impact of chemotherapy on quality of life in patients with metastatic esophagogastric cancer Cancer 2010 116 2511 2518 20301114 Alexander J Sidney J Southwood S Ruppert J Oseroff C Maewal A Snoke K Serra HM Kubo RT Sette A Development of high potency universal DR-restricted helper epitopes by modification of high affinity DR-blocking peptides Immunity 1994 1 751 761 10.1016/S1074-7613(94)80017-0 7895164 Alfonso M Díaz A Hernández AM Pérez A Rodríguez E Bitton R Pérez R Vázquez AM An anti-idiotype vaccine elicits a specific response to N-glycolyl sialic acid residues of glycoconjugates in melanoma patients J Immunol 2002 168 2523 2529 10.4049/jimmunol.168.5.2523 11859147 Alfonso S Valdés-Zayas A Santiesteban ER Flores YI Areces F Hernández M Viada CE Mendoza IC Guerra PP García E A randomized, multicenter, placebo-controlled clinical trial of racotumomab-alum vaccine as switch maintenance therapy in advanced non-small cell lung cancer patients Clin Cancer Res 2014 20 3660 3671 10.1158/1078-0432.CCR-13-1674 24788102 Anish C Beurret M Poolman J Combined effects of glycan chain length and linkage type on the immunogenicity of glycoconjugate vaccines npj Vaccines 2021 6 150 10.1038/s41541-021-00409-1 34893630 Antonarakis ES Small EJ Petrylak DP Quinn DI Kibel AS Chang NN Dearstyne E Harmon M Campogan D Haynes H Antigen-specific CD8 lytic phenotype induced by sipuleucel-T in hormone-sensitive or castration-resistant prostate cancer and association with overall survival Clin Cancer Res 2018 24 4662 4671 10.1158/1078-0432.CCR-18-0638 29858218 Arab A Yazdian-Robati R Behravan J HER2-positive breast cancer immunotherapy: a focus on vaccine development Arch Immunol Ther Exp 2020 68 2 10.1007/s00005-019-00566-1 Atsmon J Kate-Ilovitz E Shaikevich D Singer Y Volokhov I Haim KY Ben-Yedidia T Safety and immunogenicity of multimeric-001—a novel universal influenza vaccine J Clin Immunol 2012 32 595 603 10.1007/s10875-011-9632-5 22318394 Baden LR El Sahly HM Essink B Kotloff K Frey S Novak R Diemert D Spector SA Rouphael N Creech CB Efficacy and safety of the mRNA-1273 SARS-CoV-2 vaccine N Engl J Med 2020 384 403 416 10.1056/NEJMoa2035389 33378609 Baldwin WR Livengood JA Giebler HA Stovall JL Boroughs K L Sonnberg S Bohning KJ Dietrich EA Ong YT Danh HK Purified inactivated zika vaccine candidates afford protection against lethal challenge in mice Sci Rep 2018 8 16509 10.1038/s41598-018-34735-7 30405178 Barve M Bender J Senzer N Cunningham C Greco FA McCune D Steis R Khong H Richards D Stephenson J Induction of immune responses and clinical efficacy in a phase II trial of IDM-2101, a 10-epitope cytotoxic T-lymphocyte vaccine, in metastatic non-small-cell lung cancer J Clin Oncol 2008 26 4418 4425 10.1200/JCO.2008.16.6462 18802154 Boni C Janssen HLA Rossi M Yoon SK Vecchi A Barili V Yoshida EM Trinh H Rodell TC Laccabue D Combined GS-4774 and tenofovir therapy can improve HBV-specific T-cell responses in patients with chronic hepatitis Gastroenterology 2019 157 227 241 10.1053/j.gastro.2019.03.044 30930022 Boyoglu-Barnum S Ellis D Gillespie RA Hutchinson GB Park Y J Moin SM Acton OJ Ravichandran R Murphy M Pettie D Quadrivalent influenza nanoparticle vaccines induce broad protection Nature 2021 592 623 628 10.1038/s41586-021-03365-x 33762730 Brouwer PJM Antanasijevic A Berndsen Z Yasmeen A Fiala B Bijl TPL Bontjer I Bale JB Sheffler W Allen JD Enhancing and shaping the immunogenicity of native-like HIV-1 envelope trimers with a two-component protein nanoparticle Nat Commun 2019 10 4272 10.1038/s41467-019-12080-1 31537780 Brouwer PJM Brinkkemper M Maisonnasse P Dereuddre-Bosquet N Grobben M Claireaux M de Gast M Marlin R Chesnais V Diry S Two-component spike nanoparticle vaccine protects macaques from SARS-CoV-2 infection Cell 2021 184 1188 1200 10.1016/j.cell.2021.01.035 33577765 Brown MJ Coltart J Gunewardena K Ritter JM Auton TR Glover JF Randomized double-blind placebo-controlled study of an angiotensin immunotherapeutic vaccine (PMD3117) in hypertensive subjects Clin Sci 2004 107 167 173 10.1042/CS20030381 Buonaguro, L., Mayer, A., Loeffler, M., Missel, S., and Singh, H. (2020). Abstract LB-094: hepavac-101 first-in-man clinical trial of a multi-peptide-based vaccine for hepatocellular carcinoma. In: Proceedings: AACR Annual Meeting 2020. Philadelphia, PA. Cafri G Gartner JJ Zaks T Hopson K Levin N Paria BC Parkhurst MR Yossef R Lowery FJ Jafferji MS mRNA vaccine-induced neoantigen-specific T cell immunity in patients with gastrointestinal cancer J Clin Invest 2020 130 5976 5988 10.1172/JCI134915 33016924 Cai Z Su X Qiu L Li Z Li X Dong X Wei F Zhou Y Luo L Chen G Personalized neoantigen vaccine prevents postoperative recurrence in hepatocellular carcinoma patients with vascular invasion Mol Cancer 2021 20 164 10.1186/s12943-021-01467-8 34903219 Carmichael MG Benavides LC Holmes JP Gates JD Mittendorf EA Ponniah S Peoples GE Results of the first phase 1 clinical trial of the HER-2/neu peptide (GP2) vaccine in disease-free breast cancer patients Cancer 2010 116 292 301 10.1002/cncr.24756 19924797 Cavelti-Weder C Timper K Seelig E Keller C Osranek M Lässing U Spohn G Maurer P Müller P Jennings GT Development of an Interleukin-1β Vaccine in Patients with Type 2 Diabetes Mol Ther 2016 24 1003 1012 10.1038/mt.2015.227 26686385 Cech PG Aebi T Abdallah MS Mpina M Machunda EB Westerfeld N Stoffel SA Zurbriggen R Pluschke G Tanner M Virosome-formulated Plasmodium falciparum AMA-1 & CSP Derived peptides as malaria vaccine: randomized Phase 1b trial in semi-immune adults & children PLoS ONE 2011 6 e22273 10.1371/journal.pone.0022273 21799810 Chen GL Li XF Dai XH Li N Cheng ML Huang Z Shen J Ge YH Shen ZW Deng YQ Safety and immunogenicity of the SARS-CoV-2 ARCoV mRNA vaccine in Chinese adults: a randomised, double-blind, placebo-controlled, phase 1 trial Lancet Microbe 2022 3 e193 e202 10.1016/S2666-5247(21)00280-9 35098177 Chen N Xia P Li S Zhang T Wang TT Zhu J RNA sensors of the innate immune system and their detection of pathogens IUBMB Life 2017 69 297 304 10.1002/iub.1625 28374903 Chen SMY Wong YC Yim LY Zhang H Wang H Lui GCY Li X Tang X Cheng L Du Y Enhanced cross-reactive and polyfunctional effector-memory T cell responses by ICVAX—a human PD1-based bivalent HIV-1 Gag-p41 mosaic DNA vaccine J Virol 2022 96 e0216121 10.1128/jvi.02161-21 35297660 Choi YJ Hur SY Kim TJ Hong SR Lee JK Cho CH Park K S Woo JW Sung YC Suh YS A phase II, prospective, randomized, multicenter, open-label study of GX-188E, an HPV DNA vaccine, in patients with cervical intraepithelial neoplasia 3 Clin Cancer Res 2020 26 1616 1623 10.1158/1078-0432.CCR-19-1513 31727676 Clifton GT Hale D Vreeland TJ Hickerson AT Litton JK Alatrash G Murthy RK Qiao N Philips AV Lukas JJ Results of a randomized phase IIb trial of nelipepimut-S + trastuzumab versus trastuzumab to prevent recurrences in patients with high-risk HER2 low-expressing breast cancer Clin Cancer Res 2020 26 2515 2523 10.1158/1078-0432.CCR-19-2741 32071118 Clifton GT Peoples GE Mittendorf EA The development and use of the E75 (HER2 369–377) peptide vaccine Future Oncol 2016 12 1321 1329 10.2217/fon-2015-0054 27044454 Corti C Giachetti PPMB Eggermont AMM Delaloge S Curigliano G Therapeutic vaccines for breast cancer: has the time finally come? Eur J Cancer 2022 160 150 174 10.1016/j.ejca.2021.10.027 34823982 Crank MC Ruckwardt TJ Chen M Morabito KM Phung E Costner PJ Holman LSA Hickman SP Berkowitz NM Gordon IJ A proof of concept for structure-based vaccine design targeting RSV in humans Science 2019 365 505 509 10.1126/science.aav9033 31371616 Da Silva DM Skeate JG Chavez-Juan E Lühen KP Wu JM Wu CM Kast WM Hwang KK Therapeutic efficacy of a human papillomavirus type 16 E7 bacterial exotoxin fusion protein adjuvanted with CpG or GPI-0100 in a preclinical mouse model for HPV-associated disease Vaccine 2019 37 2915 2924 10.1016/j.vaccine.2019.04.043 31010714 Dai L Zheng T Xu K Han Y Xu L Huang E An Y Cheng Y Li S Liu M A universal design of betacoronavirus vaccines against COVID-19, MERS, and SARS Cell 2020 182 722 733 10.1016/j.cell.2020.06.035 32645327 Dai Y Chen X Song X Chen X Ma W Lin J Wu H Hu X Zhou Y Zhang H Immunotherapy of endothelin-1 receptor type A for pulmonary arterial hypertension J Am College Cardiol 2019 73 2567 2580 10.1016/j.jacc.2019.02.067 Darrow JJ Kesselheim AS A new wave of vaccines for non-communicable diseases: what are the regulatory challenges? Food Drug Law J 2015 70 243 258 26302599 Díaz A Alfonso M Alonso R Saurez G Troche M Catalá M Díaz RM Pérez R Vázquez AM Immune responses in breast cancer patients immunized with an anti-idiotype antibody mimicking NeuGc-containing gangliosides Clin Immunol 2003 107 80 89 10.1016/S1521-6616(03)00036-6 12763476 Diaz-Arévalo D Zeng M Nanoparticle-based vaccines: opportunities and limitations Nanopharmaceuticals 2020 1 135 150 10.1016/B978-0-12-817778-5.00007-5 Dores GM Bryant-Genevier M Perez-Vilar S Adverse events associated with the use of sipuleucel-T reported to the US Food and Drug Administration’s Adverse Event Reporting System, 2010–2017 JAMA Netw Open 2019 2 e199249 10.1001/jamanetworkopen.2019.9249 31411714 Downham MR Auton TR Rosul A Sharp HL Sjöström L Rushton A Richards JP Mant TGK Gardiner SM Bennett T Evaluation of two carrier protein-angiotensin I conjugate vaccines to assess their future potential to control high blood pressure (hypertension) in man Br J Clin Pharmacol 2003 56 505 512 10.1046/j.1365-2125.2003.01926.x 14651724 Dunkle LM Kotloff KL Gay CL Áñez G Adelglass JM Barrat Hernández AQ Harper WL Duncanson DM McArthur MA Florescu DF Efficacy and safety of NVX-CoV2373 in adults in the United States and Mexico N. Engl J Med 2021 386 531 543 10.1056/NEJMoa2116185 34910859 Emens LA Asquith JM Leatherman JM Kobrin BJ Petrik S Laiko M Levi J Daphtary MM Biedrzycki B Wolff AC Timed sequential treatment with cyclophosphamide, doxorubicin, and an allogeneic granulocyte-macrophage colony-stimulating factor-secreting breast tumor vaccine: a chemotherapy dose-ranging factorial study of safety and immune activation J Clin Oncol 2009 27 5911 5918 10.1200/JCO.2009.23.3494 19805669 Enotarpi J Tontini M Balocchi C van der Es D Auberger L Balducci E Carboni F Proietti D Casini D Filippov DV A stabilized glycomimetic conjugate vaccine inducing protective antibodies against Neisseria meningitidis serogroup A Nat Commun 2020 11 4434 10.1038/s41467-020-18279-x 32895393 Espinar-Buitrago M Muñoz-Fernández MA New approaches to dendritic cell-based therapeutic vaccines against HIV-1 infection Front Immunol 2022 12 719664 10.3389/fimmu.2021.719664 35058917 Fan A Wang B Wang X Nie Y Fan D Zhao X Lu Y Immunotherapy in colorectal cancer: current achievements and future perspective Int J Biol Sci 2021 17 3837 3849 10.7150/ijbs.64077 34671202 Felip E Besse B Dziadziuszko R Dols MC Denis F García-Campelo MR Debieuvre D Catino A Moran Bueno MT Madroszyk Flandin AC ATALANTE-1 randomized phase III trial, OSE-2101 versus standard treatment as second or third-line in HLA-A2 positive advanced non-small cell lung cancer (NSCLC) patients Ann Oncol 2019 30 v658 10.1093/annonc/mdz260.114 Fontaine H Chazallon C Mancini-Bourgine M Lebray P Buffet C Kahi S Godon O Meritet JF Saldi Y Michel ML 1367 Relapse after analogues treatment discontinuation is not prevented by dna vaccination in chronic hepatitis b (ANRS HB 02 VAC-ADN) J Hepatol 2011 54 S542 10.1016/S0168-8278(11)61369-X Freeman-Keller M Goldman J Gray J Vaccine immunotherapy in lung cancer: clinical experience and future directions Pharmacol Ther 2015 153 1 9 10.1016/j.pharmthera.2015.05.004 25989231 Fukami H Morinaga J Nakagami H Hayashi H Okadome Y Matsunaga E Kadomatsu T Horiguchi H Sato M Sugizaki T Vaccine targeting ANGPTL3 ameliorates dyslipidemia and associated diseases in mouse models of obese dyslipidemia and familial hypercholesterolemia Cell Rep Med 2021 2 100446 10.1016/j.xcrm.2021.100446 34841293 Gajdosik Z Racotumomab: a novel anti-idiotype monoclonal antibody vaccine for the treatment of cancer Drugs Today 2014 50 301 10.1358/dot.2014.50.4.2116670 Gardner TA Elzey BD Hahn NM Sipuleucel-T (Provenge) autologous vaccine approved for treatment of men with asymptomatic or minimally symptomatic castrate-resistant metastatic prostate cancer Hum Vaccines Immunother 2012 8 534 539 10.4161/hv.19795 Gay CL DeBenedette MA Tcherepanova IY Gamble A Lewis W E Cope AB Kuruc JAD McGee KS Kearney MF Coffin J M Immunogenicity of AGS-004 dendritic cell therapy in patients treated during acute HIV infection AIDS Res Hum Retroviruses 2018 34 111 122 10.1089/aid.2017.0071 28636433 Gebre MS Brito LA Tostanoski LH Edwards DK Carfi A Barouch DH Novel approaches for vaccine development Cell 2021 184 1589 1603 10.1016/j.cell.2021.02.030 33740454 Ghattas M Dwivedi G Lavertu M Alameh MG Vaccine technologies and platforms for infectious diseases: current progress, challenges, and opportunities Vaccines 2021 9 1490 10.3390/vaccines9121490 34960236 Giaccone G Felip E Cobo M Campelo RG Denis F Quoix E Madroszyk A Debieuvre D Hilgers W Moran T 1260MO Activity of OSE-2101 in HLA-A2+ non-small cell lung cancer (NSCLC) patients after failure to immune checkpoint inhibitors (ICI): Step 1 results of phase III ATALANTE-1 randomised trial Ann Oncol 2020 31 S814 S815 10.1016/j.annonc.2020.08.1574 Gilliam AD Watson SA Henwood M McKenzie AJ Humphreys J E Elder J Iftikhar SY Welch N Fielding J Broome P A phase II study of G17DT in gastric carcinoma Eur J Surg Oncol 2004 30 536 543 10.1016/j.ejso.2004.03.009 15135483 Gong R Ding C Hu J Lu Y Liu F Mann E Xu F Cohen MB Luo M Role for the membrane receptor guanylyl cyclase-C in attention deficiency and hyperactive behavior Science 2011 333 1642 1646 10.1126/science.1207675 21835979 González G Crombet T Catalá M Mirabal V Hernández JC González Y Marinello P Guillén G Lage A A novel cancer vaccine composed of human-recombinant epidermal growth factor linked to a carrier protein: report of a pilot clinical trial Ann Oncol 1998 9 431 435 10.1023/A:1008261031034 9636835 Gonzalez G Crombet T Torres F Catala M Alfonso L Osorio M Neninger E Garcia B Mulet A Perez R Epidermal growth factor-based cancer vaccine for non-small-cell lung cancer therapy Ann Oncol 2003 14 461 466 10.1093/annonc/mdg102 12598354 González-Perera I Gutiérrez-Nicolás F Nazco-Casariego GJ Ramos-Díaz R Hernández-San Gil R Pérez-Pérez JA González García J González De La Fuente GA 5-fluorouracil toxicity in the treatment of colon cancer associated with the genetic polymorphism 2846 A>G (rs67376798) J Oncol Pharm Pract 2017 23 396 398 10.1177/1078155216647202 27122156 Groenke N Trimpert J Merz S Conradie AM Wyler E Zhang H Hazapis OG Rausch S Landthaler M Osterrieder N Mechanism of virus attenuation by codon pair deoptimization Cell Rep 2020 31 107586 10.1016/j.celrep.2020.107586 32348767 Halfmann PJ Castro A Loeffler K Frey SJ Chiba S Kawaoka Y Kane RS Potent neutralization of SARS-CoV-2 including variants of concern by vaccines presenting the receptor-binding domain multivalently from nanoscaffolds Bioeng Transl Med 2021 6 e10253 10.1002/btm2.10253 34589610 Hatz CFR Bally B Rohrer S Steffen R Kramme S Siegrist CA Wacker M Alaimo C Fonck VG Safety and immunogenicity of a candidate bioconjugate vaccine against Shigella dysenteriae type 1 administered to healthy adults: a single blind, partially randomized Phase I study Vaccine 2015 33 4594 4601 10.1016/j.vaccine.2015.06.102 26162850 Heitmann JS Bilich T Tandler C Nelde A Maringer Y Marconato M Reusch J Jäger S Denk M Richter M A COVID-19 peptide vaccine for the induction of SARS-CoV-2 T cell immunity Nature 2022 601 617 622 10.1038/s41586-021-04232-5 34814158 Hernández AM Toledo D Martínez D Griñán T Brito V Macías A Alfonso S Rondón T Suárez E Vázquez AM Characterization of the antibody response against NeuGcGM3 ganglioside elicited in non-small cell lung cancer patients immunized with an anti-idiotype antibody J Immunol 2008 181 6625 6634 10.4049/jimmunol.181.9.6625 18941253 Higano CS Armstrong AJ Sartor AO Vogelzang NJ Kantoff P W McLeod DG Pieczonka CM Penson DF Shore ND Vacirca J Real-world outcomes of sipuleucel-T treatment in PROCEED, a prospective registry of men with metastatic castration-resistant prostate cancer Cancer 2019 125 4172 4180 10.1002/cncr.32445 31483485 Hirsch FR Scagliotti GV Mulshine JL Kwon R Curran WJ Jr Wu YL Paz-Ares L Lung cancer: current therapies and new targeted treatments Lancet 2017 389 299 311 10.1016/S0140-6736(16)30958-8 27574741 Horng JH Lin WH Wu CR Lin YY Wu LL Chen DS Chen PJ HBV X protein-based therapeutic vaccine accelerates viral antigen clearance by mobilizing monocyte infiltration into the liver in HBV carrier mice J Biomed Sci 2020 27 70 10.1186/s12929-020-00662-x 32466788 Hu YM Guo M Li CG Chu K He WG Zhang J Gu JX Li J Zhao H Wu XH Immunogenicity noninferiority study of 2 doses and 3 doses of an Escherichia coli-produced HPV bivalent vaccine in girls vs. 3 doses in young women Sci China Life Sci 2020 63 582 591 10.1007/s11427-019-9547-7 31231780 Hubbard JM Cremolini C Graham RP Moretto R Mitchell JL Wessling J Toke ER Csiszovszki Z Lőrincz O Molnar L Evaluation of safety, immunogenicity, and preliminary efficacy of PolyPEPI1018 off-the-shelf vaccine with fluoropyrimidine/bevacizumab maintenance therapy in metastatic colorectal cancer (mCRC) patients J Clin Oncol 2020 38 4048 10.1200/JCO.2020.38.15_suppl.4048 Ikeda M Okusaka T Ohno I Mitsunaga S Kondo S Ueno H Morizane C Gemmoto K Suna H Ushida Y Phase I studies of peptide vaccine cocktails derived from GPC3, WDRPUH and NEIL3 for advanced hepatocellular carcinoma Immunotherapy 2021 13 371 385 10.2217/imt-2020-0278 33525928 Jardine J Julien JP Menis S Ota T Kalyuzhniy O McGuire A Sok D Huang PS MacPherson S Jones M Rational HIV immunogen design to target specific germline B cell receptors Science 2013 340 711 716 10.1126/science.1234150 23539181 Jemal A Siegel R Ward E Murray T Xu J Thun MJ Cancer statistics, 2007 CA-Cancer J Clin 2007 57 43 66 10.3322/canjclin.57.1.43 17237035 Jin H Li D Lin MH Li L Harrich D Tat-based therapies as an adjuvant for an HIV-1 functional cure Viruses 2020 12 415 10.3390/v12040415 32276443 Jones B Efficacy and safety of evolocumab in reducing lipids and cardiovascular events Ann Clin Biochem 2015 52 518 10.1177/0004563215586714 Joyce MG Chen WH Sankhala RS Hajduczki A Thomas PV Choe M Martinez EJ Chang WC Peterson CE Morrison EB SARS-CoV-2 ferritin nanoparticle vaccines elicit broad SARS coronavirus immunogenicity Cell Rep 2021 37 110143 10.1016/j.celrep.2021.110143 34919799 Kanehira M Harada Y Takata R Shuin T Miki T Fujioka T Nakamura Y Katagiri T Involvement of upregulation of DEPDC1 (DEP domain containing 1) in bladder carcinogenesis Oncogene 2007 26 6448 6455 10.1038/sj.onc.1210466 17452976 Kanekiyo M Wei CJ Yassine HM McTamney PM Boyington JC Whittle JRR Rao SS Kong WP Wang L Nabel GJ Self-assembling influenza nanoparticle vaccines elicit broadly neutralizing H1N1 antibodies Nature 2013 499 102 106 10.1038/nature12202 23698367 Kang YK Boku N Satoh T Ryu MH Chao Y Kato K Chung H C Chen JS Muro K Kang WK Nivolumab in patients with advanced gastric or gastro-oesophageal junction cancer refractory to, or intolerant of, at least two previous chemotherapy regimens (ONO-4538–12, ATTRACTION-2): a randomised, doubleblind, placebo-controlled, phase 3 trial Lancet 2017 390 2461 2471 10.1016/S0140-6736(17)31827-5 28993052 Kantoff PW Higano CS Shore ND Berger ER Small EJ Penson DF Redfern CH Ferrari AC Dreicer R Sims RB Sipuleucel-T immunotherapy for castration-resistant prostate cancer N Engl J Med 2010 363 411 422 10.1056/NEJMoa1001294 20818862 Karayiannis P Hepatitis B virus: virology, molecular biology, life cycle and intrahepatic spread Hepatol Int 2017 11 500 508 10.1007/s12072-017-9829-7 29098564 Karikó K Buckstein M Ni H Weissman D Suppression of RNA recognition by toll-like receptors: the impact of nucleoside modification and the evolutionary origin of RNA Immunity 2005 23 165 175 10.1016/j.immuni.2005.06.008 16111635 Karikó K Ni H Capodici J Lamphier M Weissman D mRNA is an endogenous ligand for toll-like receptor 3 J Biol Chem 2004 279 12542 12550 10.1074/jbc.M310175200 14729660 Kawakami R Nozato Y Nakagami H Ikeda Y Shimamura M Yoshida S Sun J Kawano T Takami Y Noma T Development of vaccine for dyslipidemia targeted to a proprotein convertase subtilisin/kexin type 9 (PCSK9) epitope in mice PLoS ONE 2018 13 e0191895 10.1371/journal.pone.0191895 29438441 Keskin DB Anandappa AJ Sun J Tirosh I Mathewson ND Li S Oliveira G Giobbie-Hurder A Felt K Gjini E Neoantigen vaccine generates intratumoral T cell responses in phase Ib glioblastoma trial Nature 2019 565 234 239 10.1038/s41586-018-0792-9 30568305 Khan ST Montroy J Forbes N Bastin D Kennedy MA Diallo JS Kekre N Fergusson DA Lalu M Auer RC Safety and efficacy of autologous tumour cell vaccines as a cancer therapeutic to treat solid tumours and haematological malignancies: a meta-analysis protocol for two systematic reviews BMJ Open 2020 10 e034714 10.1136/bmjopen-2019-034714 Kheirollahpour M Mehrabi M Dounighi NM Mohammadi M Masoudi A Nanoparticles and vaccine development PNT 2020 8 6 21 10.2174/2211738507666191024162042 Khobragade A Bhate S Ramaiah V Deshpande S Giri K Phophle H Supe P Godara I Revanna R Nagarkar R Efficacy, safety, and immunogenicity of the DNA SARS-CoV-2 vaccine (ZyCoV-D): the interim efficacy results of a phase 3, randomised, double-blind, placebo-controlled study in India Lancet 2022 399 1313 1321 10.1016/S0140-6736(22)00151-9 35367003 Kim C Kim JD Seo SU Nanoparticle and virus-like particle vaccine approaches against SARS-CoV-2 J Microbiol 2022 60 335 346 10.1007/s12275-022-1608-z 35089583 Koriyama H Nakagami H Nakagami F Osako MK Kyutoku M Shimamura M Kurinami H Katsuya T Rakugi H Morishita R Long-term reduction of high blood pressure by angiotensin II DNA vaccine in spontaneously hypertensive rats Hypertension 2015 66 167 174 10.1161/HYPERTENSIONAHA.114.04534 26015450 Kremsner PG Ahuad Guerrero RA Arana-Arri E Aroca Martinez G J Bonten M Chandler R Corral G De Block EJL Ecker L Gabor JJ Efficacy and safety of the CVnCoV SARS-CoV-2 mRNA vaccine candidate in ten countries in Europe and Latin America (HERALD): a randomised, observer-blinded, placebo-controlled, phase 2b/3 trial Lancet Infect Dis 2022 22 329 340 10.1016/S1473-3099(21)00677-0 34826381 Kristensen LS Andersen MS Stagsted LVW Ebbesen KK Hansen TB Kjems J The biogenesis, biology and characterization of circular RNAs Nat Rev Genet 2019 20 675 691 10.1038/s41576-019-0158-7 31395983 Kühtreiber WM Tran L Kim T Dybala M Nguyen B Plager S Huang D Janes S Defusco A Baum D Long-term reduction in hyperglycemia in advanced type 1 diabetes: the value of induced aerobic glycolysis with BCG vaccinations npj Vaccines 2018 3 23 10.1038/s41541-018-0062-8 29951281 Lal H Cunningham AL Godeaux O Chlibek R Diez-Domingo J Hwang SJ Levin MJ McElhaney JE Poder A Puig-Barberà J Efficacy of an adjuvanted herpes zoster subunit vaccine in older adults N. Engl J Med 2015 372 2087 2096 10.1056/NEJMoa1501184 25916341 Leal L Couto E Sánchez-Palomino S Climent N Fernández I Miralles L Romero Y González T Maleno MJ Paño B Effect of intranodally administered dendritic cell-based HIV vaccine in combination with pegylated interferon α-2a on viral control following ART discontinuation: a phase 2A randomized clinical trial Front Immunol 2021 12 767370 10.3389/fimmu.2021.767370 34858423 Lesinski GB Westerink MAJ Novel vaccine strategies to T-independent antigens J Microbiol Methods 2001 47 135 149 10.1016/S0167-7012(01)00290-1 11576678 Lévy Y Lacabaratz C Lhomme E Wiedemann A Bauduin C Fenwick C Foucat E Surenaud M Guillaumat L Boilet V A randomized placebo-controlled efficacy study of a prime boost therapeutic vaccination strategy in HIV-1-infected individuals: VRI02 ANRS 149 LIGHT phase II trial J Virol 2021 95 e02165 20 10.1128/JVI.02165-20 33568510 Li C Yan X Wu D Zhang K Liang X Pan Y Zhou Y Chen F Chen X Yang S Vaccine targeted alpha 1D-adrenergic receptor for hypertension Hypertension 2019 74 1551 1562 10.1161/HYPERTENSIONAHA.119.13700 31607175 Li H Zhang Y Li D Deng YQ Xu H Zhao C Liu J Wen D Zhao J Li Y Enhanced protective immunity against SARS-CoV-2 elicited by a VSV vector expressing a chimeric spike protein Sig Transduct Target Ther 2021 6 389 10.1038/s41392-021-00797-9 Li Y Cao H Li Y Li Z Wei X Jiao R Cheng P Liu X Ma Y Xing Y Construction of a novel vaccine by conjugating a B-cell epitope of DPP4 to peptide IA2(5)-P2–1 to significantly control type 1 diabetes in NOD mice Vaccine 2017 35 7187 7197 10.1016/j.vaccine.2017.10.035 29169891 Li Z Fang J Jiao R Wei X Ma Y Liu X Cheng P Li T A novel multi-epitope vaccine based on Dipeptidyl Peptidase 4 prevents streptozotocin-induced diabetes by producing anti-DPP4 antibody and immunomodulatory effect in C57BL/6J mice Biomed pharmacoTher 2017 89 1467 1475 10.1016/j.biopha.2017.01.089 28376584 Li Z Song S He M Wang D Shi J Liu X Li Y Chi X Wei S Yang Y Rational design of a triple-type human papillomavirus vaccine by compromising viral-type specificity Nat Commun 2018 9 5360 10.1038/s41467-018-07199-6 30560935 Liang F Lindgren G Lin A Thompson EA Ols S Röhss J John S Hassett K Yuzhakov O Bahl K Efficient targeting and activation of antigen-presenting cells in vivo after modified mRNA vaccine administration in rhesus macaques Mol Ther 2017 25 2635 2647 10.1016/j.ymthe.2017.08.006 28958578 Lim H In HJ Lee JA Sik Yoo J Lee SW Chung GT Choi YK Chung JK Cho SJ Lee JW The immunogenicity and protection effect of an inactivated coxsackievirus A6, A10, and A16 vaccine against hand, foot, and mouth disease Vaccine 2018 36 3445 3452 10.1016/j.vaccine.2018.05.005 29739716 Lindgren G Ols S Liang F Thompson EA Lin A Hellgren F Bahl K John S Yuzhakov O Hassett KJ Corrigendum: induction of robust B cell responses after influenza mRNA vaccination is accompanied by circulating hemagglutininspecific ICOS+ PD-1+ CXCR3+ T follicular helper cells Front Immunol 2017 10 614 10.3389/fimmu.2019.00614 Liu Y Li T Liu Y Hao Y Wang J Wu Y Liang H Ren L Hong K Xu J Development of HIV-1 vaccines based on replication competent Tiantan vaccinia vector in China Lancet 2015 386 S10 10.1016/S0140-6736(15)00588-7 Lu K Su B Meng X Recent advances in the development of vaccines for diabetes, hypertension, and atherosclerosis J Diabetes Res 2018 2018 1 8 Lu L Zhang H Zhan M Jiang J Yin H Dauphars DJ Li SY Li Y He YW Antibody response and therapy in COVID-19 patients: what can be learned for vaccine development Sci China Life Sci 2020 63 1833 1849 10.1007/s11427-020-1859-y 33355886 Malonis RJ Lai JR Vergnolle O Peptide-based vaccines: current progress and future challenges Chem Rev 2020 120 3210 3229 10.1021/acs.chemrev.9b00472 31804810 Marcandalli J Fiala B Ols S Perotti M de van der Schueren W Snijder J Hodge E Benhaim M Ravichandran R Carter L Induction of potent neutralizing antibody responses by a designed protein nanoparticle vaccine for respiratory syncytial virus Cell 2019 176 1420 1431 10.1016/j.cell.2019.01.046 30849373 Marquina G Waki H Fernandez LE Kon K Carr A Valiente O Perez R Ando S Gangliosides expressed in human breast cancer Cancer Res 1996 56 5165 5171 8912852 Masuzawa T Fujiwara Y Okada K Nakamura A Takiguchi S Nakajima K Miyata H Yamasaki M Kurokawa Y Osawa R Phase I/II study of S-1 plus cisplatin combined with peptide vaccines for human vascular endothelial growth factor receptor 1 and 2 in patients with advanced gastric cancer Int J Oncol 2012 41 1297 1304 10.3892/ijo.2012.1573 22842485 Matić Z Šantak M Current view on novel vaccine technologies to combat human infectious diseases Appl Microbiol Biotechnol 2022 106 25 56 10.1007/s00253-021-11713-0 34889981 Merlo V Longo M Novello S Scagliotti GV EGFR pathway in advanced non-small cell lung cancer Front Biosci 2011 S3 501 517 10.2741/s168 Mittendorf EA Ardavanis A Symanowski J Murray JL Shumway NM Litton JK Hale DF Perez SA Anastasopoulou EA Pistamaltzian NF Primary analysis of a prospective, randomized, single-blinded phase II trial evaluating the HER2 peptide AE37 vaccine in breast cancer patients to prevent recurrence Ann Oncol 2016 27 1241 1248 10.1093/annonc/mdw150 27029708 Momtazi-Borojeni AA Jaafari MR Badiee A Banach M Sahebkar A Therapeutic effect of nanoliposomal PCSK9 vaccine in a mouse model of atherosclerosis BMC Med 2019 17 223 10.1186/s12916-019-1457-8 31818299 Monteiro MP Obesity vaccines Hum Vaccines Immunother 2014 10 887 895 10.4161/hv.27537 Muro K Chung HC Shankaran V Geva R Catenacci D Gupta S Eder JP Golan T Le DT Burtness B Pembrolizumab for patients with PD-L1-positive advanced gastric cancer (KEYNOTE-012): a multicentre, open-label, phase 1b trial Lancet Oncol 2016 17 717 726 10.1016/S1470-2045(16)00175-3 27157491 Nakagami H Ishihama T Daikyoji Y Sasakura C Yamada E Morishita R Brief report on a phase I/IIa study to assess the safety, tolerability, and immune response of AGMG0201 in patients with essential hypertension Hypertens Res 2022 45 61 65 10.1038/s41440-021-00755-6 34657138 Nakagawa H Mizukoshi E Kobayashi E Tamai T Hamana H Ozawa T Kishi H Kitahara M Yamashita T Arai K Association between high-avidity T-cell receptors, induced by α-fetoprotein-derived peptides, and anti-tumor effects in patients with hepatocellular carcinoma Gastroenterology 2017 152 1395 1406 10.1053/j.gastro.2017.02.001 28188748 Neninger E Díaz RM de la Torre A Rives R Díaz A Saurez G Gabri MR Alonso DF Wilkinson B Alfonso AM Active immunotherapy with 1E10 anti-idiotype vaccine in patients with small cell lung cancer: report of a phase I trial Cancer Biol Ther 2007 6 145 150 10.4161/cbt.6.2.3574 17218777 Neninger Vinageras E de la Torre A Osorio Rodríguez M CataláFerrer M Bravo I Mendoza del Pino M Abreu Abreu D Acosta Brooks S Rives R del Castillo Carrillo C Phase II randomized controlled trial of an epidermal growth factor vaccine in advanced non-small-cell lung cancer J Clin Oncol 2008 26 1452 1458 10.1200/JCO.2007.11.5980 18349395 Ogura A Konishi T Cunningham C Garcia-Aguilar J Iversen H Toda S Lee IK Lee HX Uehara K Lee P Neoadjuvant (Chemo)radiotherapy with total mesorectal excision only is not sufficient to prevent lateral local recurrence in enlarged nodes: results of the multicenter lateral node study of patients with low cT3/4 rectal cancer J Clin Oncol 2019 37 33 43 10.1200/JCO.18.00032 30403572 Okitsu SL Silvie O Westerfeld N Curcic M Kammer AR Mueller MS Sauerwein RW Robinson JA Genton B Mazier D A virosomal malaria peptide vaccine elicits a long-lasting sporozoite-inhibitory antibody response in a phase 1a clinical trial PLoS ONE 2007 2 e1278 10.1371/journal.pone.0001278 18060072 Olotu A Fegan G Wambua J Nyangweso G Leach A Lievens M Kaslow DC Njuguna P Marsh K Bejon P Seven-year efficacy of RTS,S/AS01 malaria vaccine among young african children N Engl J Med 2016 374 2519 2529 10.1056/NEJMoa1515257 27355532 Page IH Helmer OM Kohlstaedt KG Fouts PJ Kempf GF Reduction of arterial blood pressure of hypertensive patients and animals with extracts of kidneys J Exp Med 1941 73 7 41 10.1084/jem.73.1.7 19871067 Pan Y Zhou Y Wu H Chen X Hu X Zhang H Zhou Z Qiu Z Liao Y A therapeutic peptide vaccine against PCSK9 Sci Rep 2017 7 12534 10.1038/s41598-017-13069-w 28970592 Park JW Melisko ME Esserman LJ Jones LA Wollan JB Sims R Treatment with autologous antigen-presenting cells activated with the HER-2-based antigen lapuleucel-T: results of a phase I study in immunologic and clinical activity in HER-2-overexpressing breast cancer J Clin Oncol 2007 25 3680 3687 10.1200/JCO.2006.10.5718 17704416 Pizza M Scarlato V Masignani V Giuliani MM Arico B Comanducci M Jennings GT Baldi L Bartolini E Capecchi B Identification of vaccine candidates against serogroup B meningococcus by whole-genome sequencing Science 2000 287 1816 1820 10.1126/science.287.5459.1816 10710308 Polack FP Thomas SJ Kitchin N Absalon J Gurtman A Lockhart S Perez JL Pérez Marc G Moreira ED Zerbini C Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine N. Engl J Med 2020 383 2603 2615 10.1056/NEJMoa2034577 33301246 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 Qian C Liu X Xu Q Wang Z Chen J Li T Zheng Q Yu H Gu Y Li S Recent progress on the versatility of virus-like particles Vaccines 2020 8 139 10.3390/vaccines8010139 32244935 Qian C Yang Y Xu Q Wang Z Chen J Chi X Yu M Gao F Xu Y Lu Y Characterization of an Escherichia coli-derived triple-type chimeric vaccine against human papillomavirus types 39, 68 and 70 npj Vaccines 2022 7 134 10.1038/s41541-022-00557-y 36316367 Qin S Bai Y Lim HY Thongprasert S Chao Y Fan J Yang TS Bhudhisawasdi V Kang WK Zhou Y Randomized, multicenter, open-label study of oxaliplatin plus fluorouracil/leucovorin versus doxorubicin as palliative chemotherapy in patients with advanced hepatocellular carcinoma from Asia J Clin Oncol 2013 31 3501 3508 10.1200/JCO.2012.44.5643 23980077 Qu L Yi Z Shen Y Lin L Chen F Xu Y Wu Z Tang H Zhang X Tian F Circular RNA vaccines against SARS-CoV-2 and emerging variants Cell 2022 185 1728 1744 10.1016/j.cell.2022.03.044 35460644 Rajoriya N Combet C Zoulim F Janssen HLA How viral genetic variants and genotypes influence disease and treatment outcome of chronic hepatitis B Time for an individualised approach? J Hepatol 2017 67 1281 1297 28736138 Ramos da Silva J Ramos Moreno AC Silva Sales N de Oliveira Silva M Aps LRMM Porchia BFMM Bitencourt Rodrigues K Cestari Moreno N Venceslau-Carvalho AA Menck CFM A therapeutic DNA vaccine and gemcitabine act synergistically to eradicate HPV-associated tumors in a preclinical model Oncoimmunology 2021 10 1949896 10.1080/2162402X.2021.1949896 34367730 Ramos TC Vinageras EN CataláFerrer M García Verdecia B Leonard Rupalé I Martínez Pérez L González Marinello G Pérez Rodríguez R Lage Dávila A Treatment of NSCLC patients with an EGF-based cancer vaccine: report of a Phase I trial Cancer Biol Ther 2006 5 145 149 10.4161/cbt.5.2.2334 16357522 Rappuoli R De Gregorio E Costantino P On the mechanisms of conjugate vaccines Proc Natl Acad Sci USA 2019 116 14 16 10.1073/pnas.1819612116 30578318 Riddle Mark S Kaminski Robert W Di Paolo C Porter Chad K Gutierrez Ramiro L Clarkson Kristen A Weerts Hailey E Duplessis C Castellano A Alaimo C Safety and immunogenicity of a candidate bioconjugate vaccine against shigella flexneri 2a administered to healthy adults: a single-blind, randomized phase I study Clin Vacc Immunol 2016 23 908 917 10.1128/CVI.00224-16 Roberts DM Kearney JB Johnson JH Rosenberg MP Kumar R Bautch VL The vascular endothelial growth factor (VEGF) receptor Flt-1 (VEGFR-1) modulates Flk-1 (VEGFR-2) signaling during blood vessel formation Am J Pathol 2004 164 1531 1535 10.1016/S0002-9440(10)63711-X 15111299 Robinson JG Farnier M Krempf M Bergeron J Luc G Averna M Stroes ES Langslet G Raal FJ El Shahawy M Efficacy and safety of alirocumab in reducing lipids and cardiovascular events N. Engl J Med 2015 372 1489 1499 10.1056/NEJMoa1501031 25773378 Rodrigues MQ Alves PM Roldão A Functionalizing ferritin nanoparticles for vaccine development Pharmaceutics 2021 13 1621 10.3390/pharmaceutics13101621 34683914 Rodriguez PC Neninger E García B Popa X Viada C Luaces P González G Lage A Montero E Crombet T Safety, immunogenicity and preliminary efficacy of multiple-site vaccination with an Epidermal Growth Factor (EGF) based cancer vaccine in advanced non small cell lung cancer (NSCLC) patients J Immune Based Ther Vaccines 2011 9 7 10.1186/1476-8518-9-7 22024351 Sanchez L Muchene L Lorenzo-Luaces P Viada C Rodriguez PC Alfonso S Crombet T Neninger E Shkedy Z Lage A Differential effects of two therapeutic cancer vaccines on short- and long-term survival populations among patients with advanced lung cancer Semin Oncol 2018 45 52 57 10.1053/j.seminoncol.2018.04.005 30318084 Sanders B Koldijk M Schuitemaker H Inactivated Viral Vaccines Vaccine Analysis: Strategies, Principles, and Control 2014 Berlin, Heidelberg Springer-Verlag 45 80 Saunders KO Lee E Parks R Martinez DR Li D Chen H Edwards RJ Gobeil S Barr M Mansouri K Neutralizing antibody vaccine for pandemic and pre-emergent coronaviruses Nature 2021 594 553 559 10.1038/s41586-021-03594-0 33971664 Saxena M van der Burg SH Melief CJM Bhardwaj N Therapeutic cancer vaccines Nat Rev Cancer 2021 21 360 378 10.1038/s41568-021-00346-0 33907315 Shen J Wang LF Zou ZY Kong WW Yan J Meng FY Chen F J Du J Shao J Xu QP Phase I clinical study of personalized peptide vaccination combined with radiotherapy for advanced hepatocellular carcinoma World J Gastroenterol 2017 23 5395 5404 10.3748/wjg.v23.i29.5395 28839440 Shimizu Y Suzuki T Yoshikawa T Endo I Nakatsura T Next-generation cancer immunotherapy targeting glypican-3 Front Oncol 2019 9 248 10.3389/fonc.2019.00248 31024850 Si L Xu H Zhou X Zhang Z Tian Z Wang Y Wu Y Zhang B Niu Z Zhang C Generation of influenza A viruses as live but replication-incompetent virus vaccines Science 2016 354 1170 1173 10.1126/science.aah5869 27934767 Siegel RL Miller KD Jemal A Cancer statistics, 2020 CA Cancer J Clin 2020 70 7 30 10.3322/caac.21590 31912902 Slamon DJ Godolphin W Jones LA Holt JA Wong SG Keith D E Levin WJ Stuart SG Udove J Ullrich A Studies of the HER-2/neu proto-oncogene in human breast and ovarian cancer Science 1989 244 707 712 10.1126/science.2470152 2470152 Smalley Rumfield C Pellom ST Morillon YM II Schlom J Jochems C Immunomodulation to enhance the efficacy of an HPV therapeutic vaccine J Immunother Cancer 2020 8 e000612 10.1136/jitc-2020-000612 32554612 Smalley Rumfield C Roller N Pellom ST Schlom J Jochems C Therapeutic vaccines for HPV-associated malignancies ITT 2020 9 167 200 10.2147/ITT.S273327 33117742 Snook AE Baybutt TR Xiang B Abraham TS Flickinger JC Hyslop T Zhan T Kraft WK Sato T Waldman SA Split tolerance permits safe Ad5-GUCY2C-PADRE vaccine-induced T-cell responses in colon cancer patients J Immunother Cancer 2019 7 104 10.1186/s40425-019-0576-2 31010434 Snook AE Magee MS Schulz S Waldman SA Selective antigen-specific CD4+ T-celf but not CD8+ T- or B-celf tolerance corrupts cancer immunotherapy Eur J Immunol 2014 44 1956 1966 10.1002/eji.201444539 24771148 Sundar R Rha SY Yamaue H Katsuda M Kono K Kim HS Kim C Mimura K Kua LF Yong WP A phase I/Ib study of OTSGC-A24 combined peptide vaccine in advanced gastric cancer BMC Cancer 2018 18 332 10.1186/s12885-018-4234-8 29587677 Sung H Ferlay J Siegel RL Laversanne M Soerjomataram I Jemal A Bray F Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries CA Cancer J Clin 2021 71 209 249 10.3322/caac.21660 33538338 Tait DR Hatherill M Van Der Meeren O Ginsberg AM Van Brakel E Salaun B Scriba TJ Akite EJ Ayles HM Bollaerts A Final analysis of a trial of M72/AS01E vaccine to prevent tuberculosis N., Engl J Med 2019 381 2429 2439 10.1056/NEJMoa1909953 31661198 Taniguchi M Mizuno S Yoshikawa T Fujinami N Sugimoto M Kobayashi S Takahashi S Konishi M Gotohda N Nakatsura T Peptide vaccine as an adjuvant therapy for glypican-3-positive hepatocellular carcinoma induces peptide-specific CTLs and improves long prognosis Cancer Sci 2020 111 2747 2759 10.1111/cas.14497 32449239 Teh MT Gemenetzidis E Chaplin T Young BD Philpott MP Upregulation of FOXM1 induces genomic instability in human epidermal keratinocytes Mol Cancer 2010 9 45 10.1186/1476-4598-9-45 20187950 Tissot AC Maurer P Nussberger J Sabat R Pfister T Ignatenko S Volk HD Stocker H Müller P Jennings GT Effect of immunisation against angiotensin II with CYT006-AngQb on ambulatory blood pressure: a double-blind, randomised, placebo-controlled phase IIa study Lancet 2008 371 821 827 10.1016/S0140-6736(08)60381-5 18328929 Tkachuk AP Bykonia EN Popova LI Kleymenov DA Semashko MA Chulanov VP Fitilev SB Maksimov SL Smolyarchuk E A Manuylov VA Safety and immunogenicity of the GamTBvac, the recombinant subunit tuberculosis vaccine candidate: a phase II, multi-center, double-blind, randomized, placebo-controlled study Vaccines 2020 8 652 10.3390/vaccines8040652 33153191 Tobias J Jasinska J Baier K Kundi M Ede N Zielinski C Wiedermann U Enhanced and long term immunogenicity of a Her-2/neu multi-epitope vaccine conjugated to the carrier CRM197 in conjunction with the adjuvant Montanide BMC Cancer 2017 17 118 10.1186/s12885-017-3098-7 28183282 Tovey SM Reeves JR Stanton P Ozanne BW Bartlett JMS Cooke TG Low expression of HER2 protein in breast cancer is biologically significant J Pathol 2006 210 358 362 10.1002/path.2057 16981252 Trimble CL Morrow MP Kraynyak KA Shen X Dallas M Yan J Edwards L Parker RL Denny L Giffear M Safety, efficacy, and immunogenicity of VGX-3100, a therapeutic synthetic DNA vaccine targeting human papillomavirus 16 and 18 E6 and E7 proteins for cervical intraepithelial neoplasia 2/3: a randomised, double-blind, placebo-controlled phase 2b trial Lancet 2015 386 2078 2088 10.1016/S0140-6736(15)00239-1 26386540 Trimpert J Adler JM Eschke K Abdelgawad A Firsching TC Ebert N Thao TTN Gruber AD Thiel V Osterrieder N Live attenuated virus vaccine protects against SARS-CoV-2 variants of concern B.1.1.7 (Alpha) and B.1.351 (Beta) Sci Adv 2021 7 eabk0172 10.1126/sciadv.abk0172 34851677 Tsuchiya N Hosono A Yoshikawa T Shoda K Nosaka K Shimomura M Hara J Nitani C Manabe A Yoshihara H Phase I study of glypican-3-derived peptide vaccine therapy for patients with refractory pediatric solid tumors Oncoimmunology 2017 7 e1377872 10.1080/2162402X.2017.1377872 29296538 Tsuchiya N Yoshikawa T Fujinami N Saito K Mizuno S Sawada Y Endo I Nakatsura T Immunological efficacy of glypican-3 peptide vaccine in patients with advanced hepatocellular carcinoma Oncoimmunology 2017 6 e1346764 10.1080/2162402X.2017.1346764 29123959 Uyl-de Groot CA Vermorken JB Hanna MG Verboom P Groot M T Bonsel GJ Meijer CJLM Pinedo HM Immunotherapy with autologous tumor cell-BCG vaccine in patients with colon cancer: a prospective study of medical and economic benefits Vaccine 2005 23 2379 2387 10.1016/j.vaccine.2005.01.015 15755632 van Cruijsen H Ruiz MG van der Valk P de Gruijl TD Giaccone G Tissue micro array analysis of ganglioside N-glycolyl GM3 expression and signal transducer and activator of transcription (STAT)-3 activation in relation to dendritic cell infiltration and microvessel density in non-small cell lung cancer BMC Cancer 2009 9 180 10.1186/1471-2407-9-180 19519895 Vermorken JB Claessen AM van Tinteren H Gall HE Ezinga R Meijer S Scheper RJ Meijer CJ Bloemena E Ransom JH Active specific immunotherapy for stage II and stage III human colon cancer: a randomised trial Lancet 1999 353 345 350 10.1016/S0140-6736(98)07186-4 9950438 Verstraeten T Fletcher MA Suaya JA Jackson S Hall-Murray CK Scott DA Schmöle-Thoma B Isturiz RE Gessner BD Diabetes mellitus as a vaccine-effect modifier: a review Expert Rev Vaccines 2020 19 445 453 10.1080/14760584.2020.1760098 32516066 Vignuzzi M Wendt E Andino R Engineering attenuated virus vaccines by controlling replication fidelity Nat Med 2008 14 154 161 10.1038/nm1726 18246077 Walls AC Miranda MC Schäfer A Pham MN Greaney A Arunachalam PS Navarro MJ Tortorici MA Rogers K O’Connor MA Elicitation of broadly protective sarbecovirus immunity by receptor-binding domain nanoparticle vaccines Cell 2021 184 5432 5447 10.1016/j.cell.2021.09.015 34619077 Wang W Pan D Fu W Development of a skin- and neuro-attenuated live vaccine for varicella Nat Commun 2022 13 824 10.1038/s41467-022-28329-1 35149692 Wang TY Wang L Alam SK Hoeppner LH Yang R ScanNeo: identifying indel-derived neoantigens using RNA-Seq data Bioinformatics 2019 35 4159 4161 10.1093/bioinformatics/btz193 30887025 Wang Y Yang C Song Y Coleman JR Stawowczyk M Tafrova J Tasker S Boltz D Baker R Garcia L Scalable live-attenuated SARS-CoV-2 vaccine candidate demonstrates preclinical safety and efficacy Proc Natl Acad Sci USA 2021 118 e2102775118 10.1073/pnas.2102775118 34193524 Ward BJ Gobeil P Séguin A Atkins J Boulay I Charbonneau PY Couture M D’Aoust MA Dhaliwall J Finkle C Phase 1 randomized trial of a plant-derived virus-like particle vaccine for COVID-19 Nat Med 2021 27 1071 1078 10.1038/s41591-021-01370-1 34007070 Watson SA Gilliam AD G17DT—a new weapon in the therapeutic armoury for gastrointestinal malignancy Expert Opin Biol Ther 2001 1 309 317 10.1517/14712598.1.2.309 11727538 Wei L Zhao T Zhang J Mao Q Gong G Sun Y Chen Y Wang M Tan D Gong Z Efficacy and safety of a nanoparticle therapeutic vaccine in patients with chronic hepatitis B: a randomized clinical trial Hepatology 2022 75 182 195 10.1002/hep.32109 34396571 Wiedermann U Garner-Spitzer E Chao Y Maglakelidze M Bulat I Dechaphunkul A Arpornwirat W Charoentum C Yen CJ Yau T C Clinical and immunologic responses to a B-cell epitope vaccine in patients with HER2/neu-overexpressing advanced gastric cancer—results from phase Ib trial IMU.ACS.001 Clin Cancer Res 2021 27 3649 3660 10.1158/1078-0432.CCR-20-3742 33879458 Willson, J. (2020). Vaccinology in reverse (Nature). Wrobel P Ahmed S Current status of immunotherapy in metastatic colorectal cancer Int J Colorectal Dis 2019 34 13 25 10.1007/s00384-018-3202-8 30465238 Wu H Wang Y Wang G Qiu Z Hu X Zhang H Yan X Ke F Zou A Wang M A bivalent antihypertensive vaccine targeting L-type calcium channels and angiotensin AT1 receptors Br J Pharmacol 2020 177 402 419 10.1111/bph.14875 31625597 Xu DZ Wang XY Shen XL Gong GZ Ren H Guo LM Sun A M Xu M Li LJ Guo XH Results of a phase III clinical trial with an HBsAg-HBIG immunogenic complex therapeutic vaccine for chronic hepatitis B patients: experiences and findings J Hepatol 2013 59 450 456 10.1016/j.jhep.2013.05.003 23669281 Xu K Acharya P Kong R Cheng C Chuang GY Liu K Louder MK O’Dell S Rawi R Sastry M Epitope-based vaccine design yields fusion peptide-directed antibodies that neutralize diverse strains of HIV-1 Nat Med 2018 24 857 867 10.1038/s41591-018-0042-6 29867235 Yarchoan M Johnson III BA Lutz ER Laheru DA Jaffee EM Targeting neoantigens to augment antitumour immunity Nat Rev Cancer 2017 17 209 222 10.1038/nrc.2016.154 28233802 Yeh MT Bujaki E Dolan PT Smith M Wahid R Konz J Weiner AJ Bandyopadhyay AS Van Damme P De Coster I Engineering the live-attenuated polio vaccine to prevent reversion to virulence Cell Host Microbe 2020 27 736 751 10.1016/j.chom.2020.04.003 32330425 Yoshino T Arnold D Taniguchi H Pentheroudakis G Yamazaki K Xu RH Kim TW Ismail F Tan IB Yeh KH Pan-Asian adapted ESMO consensus guidelines for the management of patients with metastatic colorectal cancer: a JSMO-ESMO initiative endorsed by CSCO, KACO, MOS, SSO and TOS Ann Oncol 2018 29 44 70 10.1093/annonc/mdx738 29155929 Yu M Chi X Huang S Wang Z Chen J Qian C Han F Cao L Li J Sun H A bacterially expressed triple-type chimeric vaccine against human papillomavirus types 51, 69, and 26 Vaccine 2022 40 6141 6152 10.1016/j.vaccine.2022.09.010 36117002 Zeitlinger M Bauer M Reindl-Schwaighofer R Stoekenbroek RM Lambert G Berger-Sieczkowski E Lagler H Oesterreicher Z Wulkersdorfer B Lührs P A phase I study assessing the safety, tolerability, immunogenicity, and low-density lipoprotein cholesterol-lowering activity of immunotherapeutics targeting PCSK9 Eur J Clin Pharmacol 2021 77 1473 1484 10.1007/s00228-021-03149-2 33969434 Zhang NN Li XF Deng YQ Zhao H Huang YJ Yang G Huang WJ Gao P Zhou C Zhang RR A thermostable mRNA vaccine against COVID-19 Cell 2020 182 1271 1283 10.1016/j.cell.2020.07.024 32795413	36469218	PMC9734355	NO-CC CODE	2022-12-14 23:28:28	no	Sci China Life Sci. 2022 Dec 2;:1-32	utf-8	Sci China Life Sci	2,022	10.1007/s11427-022-2230-4	oa_other
==== Front High Educ (Dordr) High Educ (Dordr) Higher Education 0018-1560 1573-174X Springer Netherlands Dordrecht 978 10.1007/s10734-022-00978-7 Article Foreign early career academics’ well-being profiles at workplaces in Japan: a person-oriented approach http://orcid.org/0000-0003-1150-3132 Sakurai Yusuke [email protected] 1 Mason Shannon 2 1 grid.257022.0 0000 0000 8711 3200 Center for Academic Practice and Resources/Research Institute of Higher Education, Hiroshima University, Higashi-Hiroshima City, 1-3-2 Kagamiyama, Hiroshima, 739-8511 Japan 2 grid.174567.6 0000 0000 8902 2273 Faculty of Education, Nagasaki University, 1-14 Bunkyo-Machi, Nagasaki, 852-2315 Japan 7 12 2022 119 25 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 well-being of foreign early career academics (FECAs) has been the subject of research attention in relation to present demanding academic milieux in general and to those unfamiliar workplace settings in particular. A traditional variable-oriented approach that focuses on mean scores can easily gloss over the diverse nature of the group under study. Our study, conducted in Japan, took a person-oriented approach and identified FECAs’ distinct well-being profiles and the associations of their personal attributes with the profiles. Most (64%) were classified as having the highest stress scores and moderate scores for sense of belonging, control of workload and career development engagement. The second-largest profile (29%) included FECAs characterised by the lowest stress score and a strong sense of belonging, control of workload and career development engagement. Those in the smallest profile (8%), who had moderate levels of workload control and stress, lacked a sufficient sense of belonging and career development engagement. Among FECAs’ personal attributes, contract type was significantly associated with their distribution across the three well-being profiles, whereas no attributes of FECAs’ unique nature significantly pertained to their distribution. Our results suggested that support for well-being may be important regardless of background. Our investigation, using multifaceted well-being subscales over a composite scale, offers analytical, strategic support for academics in globalised higher education. Keywords Foreign academics Early career academics Well-being Person-oriented approach Japan Japan Society for the Promotion of Science20K14025 Sakurai Yusuke ==== Body pmcIntroduction As globalisation is a key force impacting higher education institutions (HEIs) across much of the world, academics may find themselves, by choice or necessity, in positions at institutions away from their home countries. This is particularly true of early career academics (ECAs), who may be more likely to relocate internationally in order to establish their careers (Tzanakou, 2021). ECAs in general may experience specific challenges due to excessive external expectations, stressful conditions and a flawed concept of meritocracy that limits their well-being (e.g. Castelló et al., 2017; Derby-Davis, 2014; Sun et al., 2011). For those who are in unfamiliar workplaces abroad, termed “foreign early career academics” (FECAs) in this study, they likely face additional challenges as they navigate linguistic and cultural differences (e.g. Brown, 2019; Huang et al., 2019; Kim, 2016). Japan is one nation that has attempted to attract more foreign academics in recent decades, largely in response to broader government efforts to improve the country’s position in the global economy. In particular, foreign academics are seen to contribute to HEIs in Japan in three main ways: English-medium teaching, foreign language teaching and on-campus internationalisation (Brotherhood, 2021). A robust government-led attempt to attract foreign academics began in the early 2000s (Huang, 2021). This led to the Global 30 and Top Global University initiatives which encouraged chosen HEIs to hire more foreign academics and diversify their international activities (Brotherhood, 2021). There was an increasing growth of foreign academics around this time, indicating a quantitative success of the initiatives (Brotherhood, 2021; Huang, 2021). However, Japan still has few foreign academics relative to other major countries (Franzoni et al., 2012), at 5% of the total population of full-time faculty (Ministry of Education, Culture, Sports, Science and Technology (Ministry of Education, Culture, Sports, Science and Technology, 2021). Furthermore, efforts to recruit international academics have been in decline specifically during 2016–2020 (National Institute of Science and Technology Policy, 2021). As the presence and importance of foreign academics in Japan increase, it is in the interests of individual institutions, and the higher education system as a whole, to ensure positive well-being and engagement in FECA’s academic work (e.g. Sasao & Hatta, 2016). Nevertheless, FECAs in Japan have reported various challenges. For example, they may be treated as crowd pullers to increase the institutional international façade, while not being afforded access to the institutional decision-making processes largely retained by Japan-born academics (Brotherhood et al., 2019). Other challenges, which are detailed later in this paper, include short-term contracts and limited opportunities for tenure and difficulties in developing interpersonal relationships and developing a sense of belonging, as well as navigating cultural differences. Within this context of potential conflict and challenge for FECAs in Japan, there has been some work toward developing means to increase their well-being and engagement in work (e.g. Sasao & Hatta, 2016). However, a knowledge gap exists regarding cohort composition, as well-being profiles are distinct across individuals. Thus, adopting a person-oriented approach, this study examined types of FECAs with unique well-being conditions. It also explored whether the sub-cohorts were characterised by personal attributes, including gender, discipline, position, contract type, language used at work, origin country and language competence. This study contributes to the literature on FECAs by treating the case of Japan, whose scholarly communities are considered insular (Franzoni et al., 2012), and thus, the challenges faced by FECAs may be intense. Well-being as a multifaceted construct Although no single accepted definition of well-being exists, instances of the concept often exhibit two shared aspects: positive emotions, such as happiness and satisfaction (hedonic approach), and meaningful engagement in life, including connections with others and autonomy (eudaimonic approach) (Ruggeri et al., 2020; Ryan & Deci, 2001; Ryff, 2014). Snyder et al. (2011) proposed a concept with three dimensions sharing some components: emotional, psychological and social. The emotional dimension concerns life satisfaction, and the psychological dimension pertains to satisfaction with the self, including personal growth and the sense of the meaning of life. The social dimension includes interpersonal relatedness, reflected by a sense of belonging. Others have proposed other dimensions of well-being (Ruggeri et al., 2020; Ryff, 2014; Ryff & Keyes, 1995), although they have utilised similar major components, relying on hedonic and eudaimonic aspects. The multifaceted conceptualisations have a history of theoretical and practical application. Ryan and Deci (2001) argued that well-being is ‘best conceived as a multidimensional phenomenon that includes aspects of both the hedonic and eudaimonic conceptions’ (p. 148). Ruggeri et al. (2020) suggested that single measures do not represent the potential implications of people’s nuanced realities. For example, while ECAs experience competition in academia, they express moderate satisfaction with work (Bentley et al., 2013). High levels of work-related satisfaction are also reported among FECAs in Japan (Huang et al., 2019; Sasao & Hatta, 2016). These findings call for a multifaceted analytic perspective rather than reliance on a summative measure. The breakdown of a scale into submeasures can offer analytic insights into patterns of workplace well-being and help develop strategic interventions (Ruggeri et al., 2020). Ryff (2014), however, warns that many submeasures may produce insufficient differentiation. Well-being is key to academics’ development and productivity (Bentley et al., 2013; Kumar et al., 2020; Pace et al., 2019; Ruggeri et al., 2020). Promoting foreign academics’ positive affect contributes to their performance (Ghasemy et al., 2021), thereby promoting an institution’s global presence. However, while the well-being literature has developed some theoretical approaches, empirical research on researchers’ well-being has often overlooked them. Kumar et al. (2020) used stress to proxy for well-being, and Pace et al. (2019) used a questionnaire covering ‘general health’ and ‘psychological strain’. Seipel and Larson (2018, p. 9) focused on academics’ satisfaction with ‘teaching/service’ and ‘overall satisfaction with their position, department and institution’. Castelló et al. (2017) and others relied on engagement and burnout literature to underscore their approach to doctoral students’ well-being. Well-being as a socio-psychological phenomenon Well-being is considered a social phenomenon in occupational health literature, not merely an individual one. Stubb et al. (2011) used ‘socio-psychological well-being’ to refer to the interplay between doctoral students’ well-being and their scholarly communities. Well-being is conceived as dependence on negotiating sense-making processes between cognitive processing and life events, including the complexities of work and collegial climate (Ryff, 2014). Hence, cognitive (mis)fit with the work environment may regulate positive and negative states of well-being (Kneer & Haybron, 2020). For academics from abroad, inter alia, living in an unfamiliar environment entails challenges to well-being. They may feel disconnected and overwhelmed by unstated workplace norms. Referring to the literature, we operationalised this study by focusing on the relevant key dimensions of FECAs’ socio-psychological well-being, considering satisfying (hedonic) and engaging (eudaimonic) experiences at work: sense of belonging, workload control, career development and stress conditions. Positive interpersonal relationships are key to well-being (Huppert, 2009). Collegiality is a foundational element of sound and ethical scholarly communities (Sasao & Hatta, 2016). It is associated with job satisfaction and may reduce turnover intention (Daly & Dee, 2006). However, ECAs working on short-term contracts often fail to achieve a sense of belonging (Seipel & Larson, 2018). In particular, HEIs are expected to nurture FECAs’ sense of belonging and provide an inclusive scholarly community (Liu-Farrer, 2015; Munene, 2014), but some foreign academics still feel like outsiders (Kim, 2016; Munene, 2014). FECAs’ reluctance to request help from their local colleagues can drive them further apart (Bailey et al., 2021). Research in Japan has indicated FECAs’ frustration with their work environment owing to language barriers, unspoken work norms and perception of foreigners as outsiders, which contribute to their impaired sense of belonging (Brotherhood et al., 2019; Brown, 2019; Komisarof & Hua, 2016; Larson-Hall & Stewart, 2018; Liu-Farrer, 2015). They furthermore experience xenophobic and systemic unfair treatment, occurring more frequently to females than to males (Liu-Farrer, 2015; Nagatomo & Cook, 2018). The degree of control over one’s situation contributes to the sense of well-being (Huppert, 2009). The number of miscellaneous tasks associated with teaching and research has increased, and overwork has become more widespread, even coming to be encouraged in academia (Pace et al., 2019). Excessive workload and pressure are major adverse factors for researchers’ psychological health (Horta et al., 2019; Opstrup & Pihl-Thingvad, 2016; Pace et al., 2019; Sabagh et al., 2018; Sun et al., 2011), and it appears that Japanese academia barely focuses on academics’ work–life balance (Sasao & Hatta, 2016). Inappropriate workloads lessen researchers’ motivation to remain at their workplace (Derby-Davis, 2014). Furthermore, FECAs in Japan experience frustration regarding their autonomy, workload and access to institutional decision-making (Brotherhood et al., 2019; Brown, 2019). Their inferior status may limit their autonomy regarding their work (Kim, 2016). Maintaining well-being requires doing things considered worth doing by the doer (Huppert, 2009; Ryan & Deci, 2001). Ryff and Keyes (1995) argued that feelings of development are an important dimension of well-being. For ECAs, healthy engagement in career development is a vital component of well-being (Sasao & Hatta, 2016). Strategic ECR training is growing in importance for universities and researchers and is contributing to disciplinary knowledge (Pearce & Metcalfe, 2016). Hence, FECAs’ perceptions of their engagement in development opportunities should be investigated in their constantly evolving communities. Stressful academic environments are an international phenomenon (e.g. Horta et al., 2019; Opstrup & Pihl-Thingvad, 2016; Sabagh et al., 2018). They result from increasing degrees of administrative responsibility, demanding teaching duties, competitive research expectations and inability to balance among wide-ranging responsibilities (Bentley et al., 2013; Opstrup & Pihl-Thingvad, 2016; Pace et al., 2019; Sabagh et al., 2018). Stress in researchers is associated with their collegial relationships and their access to support (Castelló et al., 2017; Horta et al., 2019; Sabagh et al., 2018). Viewing writing as a burden rather than as an intellectual vocation is correlated with mental fatigue (Castelló et al., 2017). Younger researchers are under more stress and less satisfied than senior staff (Bentley et al., 2013; Sun et al., 2011). Foreign academics experience greater stress than their domestic counterparts. Many work in a foreign-language setting, thereby adding an additional burden (McAllum, 2017). Their stress also comes from their peripheral status as newcomers and the dismissive attitudes of some colleagues and students (Munene, 2014). FECAs experience their work in many ways due to diversity and different situations. Bentley et al. (2013) observed different patterns of well-being in a cross-national study. Huang et al. (2019) indicated variability in foreign academics’ work experience in Japan, including varied academic positions, disciplines and nationalities. Additionally, Ruggeri et al. (2020) demonstrated that researchers with lower well-being report greater variability scores. Nonetheless, little research has addressed what unique profiles are detected pertaining to their well-being status. This study Research goals This study (1) identifies several profile clusters of well-being among FECAs in Japan and (2) examines their characteristic attributes, including gender, discipline, position, contract type, language used at work, country of origin and Japanese language competence. The study was motivated by the intrinsic interest of identifying particular profiles of FECAs’ regarding well-being at the workplace, which could be beneficial for institutions concerned with talent retention and a diverse workplace (e.g. Kim, 2016). In turn, this facilitates performance and institutional productivity. Setting There are approximately 800 HEIs in Japan, counting 4-year universities and technical colleges (Ministry of Education, Culture, Sports, Science and Technology, 2021). Researchers also work at about 40 other research institutions (Ministry of Education, Culture, Sports, Science and Technology, n.d.). As noted earlier, full-time foreign HEI academics constitute around 5% of the total. In quantitative terms, they constitute 9526 of 190,448 researchers (Ministry of Education, Culture, Sports, Science and Technology, 2021). FECAs’ initial contracts in Japan are often signed for 3 to 5 years, with little chance of tenure (Larson-Hall & Stewart, 2018). Usually, these fixed-term researchers do not receive associated benefits, such as sabbatical leave or retirement bonuses. A four-tiered rank structure is common in Japan: starting at assistant professor, then lecturer, associate professor and up to full professor. These can have either permanent or fixed-term (including tenure-track) status. Part-time instructors have no academic title. No division exists between research and non-research academics except for part-time teaching staff. According to Takagi (2018), fixed-term social science ECAs are often hired to manage service and teaching commitments and are not expected to conduct research. However, constant output during their limited working time or even outside of work hours is mandatory for those seeking to advance. Those who actively contribute to disciplinary knowledge may experience much frustration in Japanese academia. Data collection and participants We recruited survey participants according to the following criteria: (1) considered themselves academics or were affiliated with Japanese universities or research institutes, (2) had non-Japanese nationality and (3) had earned their PhD less than 8 years ago or were 39 years old or younger without a PhD, adopting the definition of an ECA as stipulated by Japan Society for the Promotion of Science. Doctoral students were not included due to their specific position in the Japanese higher education system. While doctoral students may be ‘employed’ as research ‘workers’ in Europe (Shin et al., 2018, pp. 2–3) and thus considered one of the cohort of ECAs, in Japan, doctoral students are neither deemed ‘academics’ nor ‘employed’ by the institution. As students, they are charged tuition, and while some may attain a grant from their university or a competitive nation-wide scheme, there is no status conferred to them as academics or researchers. They will rarely be given teaching duties. These conditions, which are relevant to one’s well-being, mean that they cannot reliably be placed in the same group as ECAs in this study. We found FECAs in researcher rosters by identifying non-Japanese names. Recruitment by email or post was accomplished using institutional websites and the national J-Global researcher database (jglobal.jst.go.jp), which gathers and organises researcher information in Japan. Where possible, we sent email invitations; where this was not possible, we sent letters through the mail with a QR code providing a link to the survey. We also sent invitations via J-Global, which is equipped with a private messaging function. In response to our 1544 invitations (468 by email, 295 by post and 781 via J-Global), we received 333 completed questionnaires. Among these, two were eliminated as duplicates another two, and a third one appeared thrice. We used an instructional manipulation check item (Maniaci & Rogge, 2014) to detect inattentive participants: ‘This is a control question. Please choose “strongly agree” for this item’. A total of 305 respondents provided usable answers. About two-thirds of the participants were male (61.3%) (Table 1). More participants were working at national universities (51.1%) than at any other types of institution. The 30 s were the largest age group (75.7%). More than half were Asian-born (57.4%). Predictably, most had the title of assistant professor (52.4%). There were three respondents who were fixed-term professors who satisfied the criteria for ECAs based on their demographic information. Two-thirds were employed for a fixed term (62.3%). The survey collected information on participants’ disciplines, the most competent language, Japanese language proficiency and years of work experience in Japan. There were no missing values. Approval for this project was obtained from the Research Ethics Committee at Ochanomizu University.Table 1 Participant demographics Variable Category F Ratio Gender Male 187 61.3 Female 104 34.1 Prefer not to say 12 3.9 Other 2 0.7 Institutional type National university 156 51.1 Private university 96 31.5 Public university 28 9.2 Research institute 13 4.3 Colleges of technology 9 3 Other 3 0.9 Age 51 or older 2 0.7 41–50 39 12.8 31–40 231 75.7 30 or younger 33 10.8 Discipline Humanities 77 25.2 Social sciences 71 23.3 Engineering sciences 55 18.0 Informatics 23 7.5 Medicine, dentistry and pharmacy 19 6.2 Biological sciences 17 5.6 Chemistry 16 5.2 Mathematical and physical sciences 12 3.9 Agricultural and environmental sciences 12 3.9 Multidisciplinary 3 1.0 Academic rank Professor 3 1 Associate professor 33 10.8 Lecturer 74 24.3 Assistant professor 160 52.4 Researcher 32 10.5 Part-time teaching staff 2 0.7 Other 1 0.3 Contract Tenured position 65 21.3 Tenure-track position 47 15.4 Fixed-term position 190 62.3 No formal contract 2 0.7 Do not know 1 0.3 Most competent language English 156 51.7 Japanese 57 18.9 Others 89 29.5 Japanese language proficiency test (JLPT) level Native speaker level 62 20.3 JLPT N1 level (highest) 103 33.8 JLPT N2 level 38 12.5 JLPT N3 level 34 11.1 JLPT N4 level 17 5.6 JLPT N5 level 14 4.6 Lower than JLPT N5 13 4.3 Other or unsure 24 8.0 Birthplace Asia (outside Japan) 175 57.4 Europe 57 18.7 North America 36 11.8 South America 12 3.9 Japan 11 3.6 Oceania 9 3.0 Africa 5 1.6 Work experience in Japan 3 years or less 61 20 6 years or less 109 35.7 9 years or less 73 23.9 More than 9 years 62 20.3 Measurement This study used self-reports to measure well-being in the following perspectives: sense of belonging, career development, control of workload and stress. We used the five-item Sense of Belonging Survey (Rubin et al., 2019) to measure the sense of membership and support; sample items are ‘I receive good support from my co-worker’ and ‘I don’t feel like I fit in well at my institution’ (reverse scoring item). We selected this survey for its suitability for scholarly communities (content validity). We assessed levels of control of workload by adopting three measures from the Academic Work Environment Survey (Houston et al., 2006). These items measure how well academics handle their current workload, including teaching and research; sample items are ‘I am expected to teach and/or supervise a reasonable number of students’ and ‘I often need to work after hours to meet my work requirements’ (reverse scoring). Studies have indicated the psychometric adequacy of this survey in the university context (e.g. Pace et al., 2019). We measured FECAs’ engagement in career development using the nine-item Career Engagement Scale (Hirschi et al., 2014) (e.g. ‘I sincerely thought about my personal values, interests, abilities and weaknesses’ and ‘I collected information about employers, professional development opportunities or the job market in my desired area’). This scale is relevant to career development for researchers. Hirschi et al. (2014) found good psychometric properties for the single factorial structure. The FECAs responded to these three scales on a scale of 1 (strongly disagree) to 7 (strongly agree). The stress variable was identified with a single-item measure: stress refers to a person’s state of being tense, restless, nervous or sleepless because his/her mind is troubled all the time. Have you felt this kind of stress recently? (Elo et al., 2003). Responses were given on a 9-point scale (1 = not at all to 9 = very much). Studies in various occupational settings (Elo et al., 2003) and HEIs (Opstrup & Pihl-Thingvad, 2016) have validated the measure. Analysis FECAs’ responses were computed with a mean score for each scale. To classify the FECAs according to patterns in their well-being scores, we conducted model-based clustering with the mclust package (version 5.4.7) in the R environment. This person-centred approach assumes a multivariate distribution of data across sub-populations, that is, assumes a necessary heterogeneity of a sample containing more than a single subgroup. This analysis classifies individuals into latent subgroups through variety in their score patterns, considering intra- and interpersonal differences (Raufelder et al., 2013). Because this approach offers data-driven statistics that estimate the optimal number of clusters, it has an advantage over heuristic approaches, such as hierarchical and k-means clustering (Fraley & Raftery, 2007). In our analysis, we used the Bayesian information criterion (BIC) to identify the number of clusters where greater scores indicate a better fit. We named the clusters according to patterns seen in the score means. To identify differences among cluster profiles, we used Welch’s test, which is suitable for samples with unequal variances. Applying the chi-squared and Fisher’s exact tests (significance level at 5%) with an effect size (w), we examined clusters concerning the distribution of FECAs’ genders, disciplines, academic ranks, contract types, most proficient languages, Japanese language competence and birthplace (Table 1). Fisher’s exact test is suitable when more than 20% of cells have expected cell frequencies smaller than five. The survey collected information on FECAs’ major research disciplines, following the Japanese funding disciplinary categories (11 categories). However, as some categories had a small number of respondents, we created new categories, partly adopting Biglan’s classification (Biglan, 1973), resulting in larger categories that allowed meaningful statistical inference. Because the largest groups were researchers in the humanities and social sciences, we kept them as they were. Accordingly, the disciplines were categorised into four domains: hard-pure (medicine, dentistry and pharmacy/chemistry/mathematical and physical sciences/biological sciences/agricultural and environmental sciences), hard-applied (engineering sciences/informatics), social sciences and humanities. Only three researchers were professors (fixed term), and two were part-time teaching staff; these were removed from the analysis as outliers. We adopted three main contractual categories: fixed term, tenure track and tenured. We classified the FECAs into three categories based on their most proficient language: English, Japanese and others. We grouped respondents according to their Japanese skill. Foreign researchers who must conduct institutional tasks in Japanese need high levels of Japanese language proficiency. Therefore, we investigated whether they had the highest level of the Japanese Language Proficiency Test (N1 level). Those who were ignorant about the test were excluded from the analysis. Finally, FECAs from some non-Asian regions were too few for adequate analysis, so the distribution of Asian or non-Asian FECAs across clusters was examined. Those who were born in Japan were omitted. Multiple comparisons using the Benjamini–Hochberg adjustment identified cluster pairs with significantly different observed frequencies for the FECAs’ attributes. The adjustment was suitable for controlling type I error in multiple comparisons (Thissen et al., 2002). Results Descriptive results (Table 2) suggested that the FECAs were highly engaged in career development. The low degrees of correlations imply that these factors were moderately independent. As expected, stress was significantly negatively correlated with other scales, with the exception of career development.Table 2 Means, standard deviations, correlations and internal consistencies for well-being scales Scale M (SD) 1 2 3 4 Internal consistency α ω 1. Sense of belonging 4.90 (1.37) - 0.87 0.90 2. Control of workload 3.77 (1.32) 0.26** - 0.64 0.66 3. Career development 5.56 (0.92) 0.14* .02 - 0.88 0.91 4. Stress (single scale, 1–9) 5.58 (2.29) − 0.28 − .31 − .08 - - - p < 0.05, p < 0.01 Identifying FECA subgroups The clusters that the FECAs form in relation to their well-being conditions were investigated. An inspection of the BIC values for the clustering analysis best supported a three-cluster solution (Table 3 and Fig. 1). We labelled the clusters in terms of sets of means as (1) stressed ‘in-control’ academics (n = 195), (2) unstressed engaged academics (n = 87) and (3) disengaged academics (n = 23).Table 3 Mean scores of well-being profile clusters Scale 1. Stressed ‘in-control’ (195) 2. Unstressed engaged (87) 3. Disengaged (23) Welch’s test Post hoc Games-Howell pairwise comparison M (SD) M (SD) M (SD) 1. Sense of belonging 4.70 (1.29) 5.64 (0.95) 3.86 (2.01) F(2, 55.66) = 27.23* 2 > 1, 2 > 3 2. Control of workload 3.46 (1.21) 4.51 (1.17) 3.58 (1.69) F(2, 55.46) = 23.76 2 > 1, 2 > 3 3. Career development 5.67 (0.71) 5.85 (0.68) 3.63 (1.11) F(2, 54.71) = 41.18 1 > 3, 2 > 3 4. Stress (1–9) 6.86 (1.22) 2.74 (1.08) 5.52 (2.89) F(2, 53.56) = 397.46 1 > 2, 3 > 2 p < 0.01 Fig. 1 Standardised scores of well-being profiles Stressed ‘in-control’ academics’ scores for sense of belonging, control of workload and career development showed moderate levels of engagement in collegial community and work. However, the large value for stress is notable. The unstressed engaged academics profile was characterised by the least stress mean accompanied by the highest sense of belonging and control of workload scores. Their career development score was the highest but not significantly different from that of the stressed ‘in-control’ academics group. The disengaged academics showed an extremely low career development score, but their perception of their control of workload and stress levels on average were not strongly pessimistic. However, this cluster had the lowest sense of belonging to their institutions. FECAs’ attributes in relation to their well-being profiles The second task was to examine differences among the clusters in relation to FECA attributions. Table 4 indicates the distribution patterns of FECAs and their personal attributes for each cluster. Significant differences were found among contract types (χ2 (4) = 10.36, p = 0.035, Fisher’s exact test, p = 0.023), showing a possible small effect size (w = 0.185, 1-β = 0.733) (small, 0.1–0.3; medium, 0.3–0.5; or large, 0.5 +). However, multiple comparisons with the Benjamini–Hochberg adjustment did not present a significant difference between the results of the clusters. The adjusted standardised residuals (p = 0.08) indicated that those on the tenure-track contract were overrepresented in the stressed ‘in-control’ academics cluster (n = 37, z = 2.254) and underrepresented in the unstressed engaged academics cluster (n = 6, z = − 2.597).Table 4 Cross-tabulation of the distribution of FECAs’ attributes in the well-being clusters Stressed ‘in-control’ Unstressed engaged Disengaged Gender: χ2(2) = 0.21, p = 0.924, w = 0.027, 1-β = 0.066; Fisher’s exact test, p = 0.925 Female 65 31 8 Male 121 51 15 Discipline: χ2(6) = 2.17, p = 0.904, w = 0.085, 1-β = 0.156; Fisher’s exact test, p = 0.906 Hard-applied 45 26 7 Hard-pure 51 19 6 Social sciences 47 20 4 Humanities 49 22 6 Academic rank: χ2(6) = 3.96, p = 0.682, w = 0. 115, 1-β = 0.267; Fisher’s exact test, p = 0.681 Associate professor 18 13 2 Lecturer 46 23 5 Assistant professor 104 43 13 Researcher 23 6 3 Contract type: χ2 (4) = 10.36, p = 0.035, w = 0.185, 1-β = 0.733; Fisher’s exact test, p = 0.023 Tenure-track 37 6 4 Fixed-term 120 60 10 Tenured 37 20 8 Most competent language: χ2 (4) = 1.29, p = 0.863, w = 0.065, 1-β = 0.124; Fisher’s exact test, p = 0.885 English 97 46 13 Other 57 27 6 Japanese 39 14 4 Japanese competence: χ2 (2) = 2.53, p = 0.283, w = 0.095, 1-β = 0.276; Fisher’s exact test, p = 0.279 N1 or higher 112 40 13 Lower than N1 69 38 9 Birthplace: χ2 (2) = 4.01, p = 0.135, w = 0.117, 1-β = 0.416; Fisher’s exact test, p = 0.135 Asia 117 49 9 Not Asia 70 36 13 The analyses further suggested that the well-being clusters were independent of FECAs’ gender (χ2 (2) = 0.21, p = 0.924; Fisher’s exact test, p = 0.925), discipline (χ2 (6) = 2.17, p = 0.904; Fisher’s exact test, p = 0.906), academic rank (χ2(6) = 3.96, p = 0.628; Fisher’s exact test, p = 0.681), work experience in Japan (χ2(6) = 3.78, p = 0.707; Fisher’s exact test, p = 0.701), most competent language (χ2(4) = 1.29, p = 0.863; Fisher’s exact test, p = 0.885), Japanese language competence (χ2(2) = 2.45, p = 0.294; Fisher’s exact test, p = 0.299) and birthplace (χ2(2) = 4.01, p = 0.135; Fisher’s exact test, p = 0.135). Discussion Methodological reflections The unique context of this study may prevent its generalisability to other contexts, as conceptions of well-being are tied to culture (Ryan & Deci, 2001). Bentley et al. (2013) demonstrated that Japanese academics’ patterns of occupational satisfaction and distress differ from their UK and Australian counterparts. Moreover, future policy strategies in Japan may seek to drastically refurbish international academics’ environment, affecting their well-being. Some measurement issues should be considered. The study data are only self-reports. Consistency indices for the measure, control of workload, were below 0.70. This indicates that the measure should be improved. Our results resemble those of Ruggeri et al. (2020), who found greater score variations in lower well-being groups; lower well-being groups’ scores were concentrated around the scales’ midpoint, which prompted greater selection options in both directions for these respondents than those with greater well-being. It will be worth examining whether this pattern is a methodological coincidence or an authentic phenomenon. These limitations do not overshadow the strengths of the study. Traditional regression studies have identified overall trends in target cohorts, and our person-oriented approach presented a novel understanding for FECAs’ well-being profiles. We furthermore used multiple scales to obtain a nuanced understanding of FECAs’ well-being conditions (e.g. Ruggeri et al., 2020). A single scale could not identify these unique patterns. Reflections on findings Employing a person-oriented approach, this study fills a gap in knowledge, demonstrating that most FECAs (64%) moderately engaged in their career development possessed a sense of belonging to their institution and took control of their work but felt highly stressed at work. Our results support Bentley et al. (2013), who showed that many academics in Japan are generally satisfied with but stressed by their work. In addition, Bentley et al. (2013) found that their stress levels were not correlated with dissatisfaction with work. Shin et al. (2018) also identified a similar pattern for Japan. The results of our person-oriented and multifaceted approaches can help us understand why satisfaction and stress levels were not anticorrelated overall; that is, the coexistence of subgroups with unique well-being profiles fails to explain this simple association. Our results revealed that FECAs’ demographic attributes were seldom associated with their well-being profiles. This is surprising, given previous results suggesting associations with local language skills (e.g. Komisarof & Hua, 2016) and contract lengths (Seipel & Larson, 2018). We hypothesised that high Japanese proficiency gave respondents access to certain informational resources, and Asian researchers felt greater affinity with locals (e.g. Brotherhood et al., 2019; Komisarof & Hua, 2016; Liu-Farrer, 2015). Researchers with limited Japanese skills, a distinct background or a fixed contract did not cluster with respondents having low levels of well-being. Bentley et al. (2013) found weak or nonsignificant associations of academics’ demographics with their job satisfaction while demonstrating more conspicuous associations of environmental conditions. Gender was nonsignificant. Huppert (2009) observed that the effect of gender on well-being was ambiguous. Xu (2008) argued that gender effects can be confounded with productivity and employment contracts. Our study supports the claim that academics’ demographic profiles cannot sufficiently explain their job well-being. Environmental conditions and their attitude to these may indicate their well-being to a greater degree; or, as Ryan and Deci (2001) argued, their inherent personality features may do this. The only significant factor associated with the FECA clusters was contractual differences. However, the effect size remained small and post hoc comparisons suggested the probationary result that tenure-track researchers are more likely to experience demanding psychological conditions. We did not explore the reasons for this, but our results suggest greater industriousness among FECAs seeking tenure. Ruggeri et al. (2020) indicated that researchers in unstable employment reported poor well-being. In Japan, even a tenure-track contract takes 3–5 years to result in tenure. Tenure-track workers normally perform duties equivalent to those of tenured staff. Conversely, researchers with a fixed-term contract exclusively engage in specific projects, either educational or research, and are often exempt from institutional management and committee responsibilities. Some contracts do not even demand research outputs (Takagi, 2018). This exemption from responsibilities may enable better psychological well-being. We acknowledge that COVID-19 has been an unprecedented challenge for all academic staff, including FECAs, although investigating this factor was beyond the scope of the study. Academics’ increased distress was associated with the restrictions and demands stemming from the pandemic (e.g. Huang, 2022; Watermeyer et al., 2021). The reform of institutions due to the pandemic has impaired job security and autonomy (Watermeyer et al., 2021), adversely impacting well-being. In Japan, COVID-19 has also entailed challenges for foreign academics. For example, the travel ban prevented family reunions, and poorer information flow due to the language barrier and decreased interaction with colleagues were observed (e.g. Huang, 2021). However, some positive changes resulting in better productivity were also identified, including reduced commuting time and increasing availability of online learning opportunities (Huang, 2022). Thus, the interplay between these positive and negative conditions in institutional and private hemispheres may have affected academics’ well-being during the pandemic. Implications This study shows that no personal attributes are particularly pertinent to foreign researchers’ distribution over well-being clusters. This is of practical value, as intervention and support practices are not necessary to highlight the variation of their origins and may be effective regardless of their nationality, if it were adequately provided in an appropriate language(s). In a Japanese context, studies have suggested possible threat to the well-being of social sciences (Takagi, 2018) and female ECAs (Liu-Farrer, 2015; Nagatomo & Cook, 2018). Our results instead implied comprehensive support is valid regardless of their disciplines and genders, while more comparative research is needed in Japan. Others argue that targeted approaches to particular groups may be ineffective, as the proportion with a serious condition is very small, even in target groups (e.g. Huppert, 2009). Rather, Huppert (2009) argued that collective initiatives can effectively address the potential low level of well-being of individuals at risk in a larger cohort. Our results underscore the particular importance of support for well-being among tenure-track researchers. Stressful workplace conditions may challenge them to develop as more mature and adaptable researchers (Kumar et al., 2020), but some may need support. Although further study is necessary to corroborate the reliability of intervention practices, there may be two strategies. First, more tenure-track FECAs fall into the cluster where members show a poorer sense of belonging, control of workload and stress conditions than the unstressed engaged respondents. Deliberate attention to these dimensions should be included in faculty development seminars and training. For example, Juberg et al. (2019) presented a protocol for mindfulness workshops to reduce the psychological burden experienced by university staff. Kumar et al. (2020) demonstrated the significant positive effects of casual meetings to discuss researchers’ stress levels during the COVID-19 pandemic. Haines et al. (2007) suggested that regular walking activity promoted the mental conditions of the faculty, although modern researchers’ time constraints are a major obstacle (Juberg et al., 2019). Second, both researchers and administrative staff should consider the challenges that FECAs face and be aware of key experiences that may influence their well-being. Simple, tangible actions can create a mutually caring community where FECAs can flourish. Lyubomirsky and Layous (2013) suggested that small behavioural changes, such as deliberately expressing gratitude and helping others, enhance one’s own well-being, while the effect on international cohorts may be inconsistent owing to their linguistic or cultural barriers (Juberg et al., 2019). Although their proportion was small, a cluster of FECAs was identified with least engagement in career development opportunities and weakest sense of membership. This combination is problematic for both individuals and institutions. The results identified no personal attributes to distinguish this group, preventing immediate countermeasures. Given their poor sense of membership and engagement in career development, those in this cluster likely have weak emotional and professional connection with their colleagues and few developmental opportunities, although they properly carry out their duties. After they acknowledge this disconnection, they have difficulty recovering their engagement because they have already become emotionally distanced. In this regard, it is extremely important to create an inclusive community in the early stages of employment. HEI leaders must offer career development opportunities accessible for FECAs as institutional employees (Lyubomirsky & Layous, 2013) and for other colleagues to proactively develop collegiality. Van Waes et al. (2018) showed that convener’s attention to staff networking in their pedagogical training was valuable to support staff’s enduring relationships, even after the training. The UK initiative of Vitae Researcher Development Framework may be helpful for Japan as well. This allows researchers and institutions to discuss ECAs’ strategic development (Pearce & Metcalfe, 2016). Some Vitae training opportunities are available in Japan (Pearce & Metcalfe, 2016), but more evidence-based approaches are still needed. As Sakurai and Pyhältö (2021) identified the disciplinary patterns of researcher engagement in skill development, their disciplinary expectations should also be considered for programme contents. Conclusion Drawing on a multifaceted well-being construct, this study identified three major clusters of FECAs in Japan according to their response patterns. The largest showed the highest stress score, with moderate scores of sense of belonging, control of workload and career development engagement. The second included FECAs with the least stress and highest sense of belonging, control of workload and career development. The smallest group of FECAs showed roughly average scores for workload control and stress but inadequate sense of belonging and career development engagement. Our analysis demonstrated that FECAs in tenure-track positions faced more challenges than tenured and fixed-term FECAs. Surprisingly, no attributes particularly associated with foreigners’ unique characteristics, such as birthplace or language competence, explained the FECAs’ systematic distribution across well-being clusters. A larger sample would allow the data to be broken down into more personal attribute categories, which could offer novel insights, such as researchers’ birthplaces (in this study, Asia vs not Asia) and disciplines (in this study, only four disciplinary domains). This study focused on FECAs’ workplaces, but the impact of other non-occupational factors should be also taken into account, such as social inclusion and work-family interface. Moreover, as the studies of faculty well-being have identified distinct characteristics in several countries (Bentley et al., 2013; Ruggeri et al., 2020), our study should be replicated in other contexts. To this end, the content and construct validity of researchers’ well-being measure should be addressed. The critique of Stubb et al. (2011) still holds true—the theoretical constructs of well-being for a scholarly community remain inconsistent. Huppert’s (2009) review noted that negative emotion makes a person amenable. The positive consequences entailed by temporary low levels of well-being and the negative consequences of well-being should also be attended to. The person-oriented approach may produce additional insights for local researchers, as it is seldom employed for researchers’ well-being. Because one’s improved well-being can benefit others, we hope this work will support those who ‘aspire to be fully functioning and satisfied in this earthly life’ (Ryan & Deci, 2001, p. 161). Funding This work was supported by JSPS KAKENHI grant number 20K14025. Declarations Ethics approval All procedures performed in this study were approved by a research ethics committee at Ochanomizu University, Japan. 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 Bailey W Bordogna CM Harvey H Jones G Walton S Transformational, inclusive, and multicultural or empty rhetoric? Perceptions and experiences of international academic staff Journal of Further and Higher Education 2021 45 3 349 362 10.1080/0309877X.2020.1762848 Bentley, P. J., Coates, H., Dobson, I. R., Goedegebuure, L., & Meek, V. L. (2013). Academic job satisfaction from an international comparative perspective: Factors associated with satisfaction across 12 countries. In P. J. Bentley, H. Coates, I. R. Dobson, L. Goedegebuure, & V. L. Meek (Eds.), Job satisfaction around the academic world (p239–262). Springer. 10.1007/978-94-007-5434-8_13 Biglan A The characteristics of subject matter in different academic areas Journal of Applied Psychology 1973 57 3 195 203 10.1037/h0034701 Brotherhood T Hammond CD Kim Y Towards an actor-centered typology of internationalization: A study of junior international faculty in Japanese universities Higher Education 2019 79 3 497 514 10.1007/s10734-019-00420-5 Brotherhood, T. (2021). International faculty: Increasing mobility. In P. Snowden (Ed.), Handbook of Higher Education in Japan (261–274). MHM Limited. 10.2307/j.ctv1zckxc3.25 Brown CA Foreign faculty tokenism, English, and “internationalization” in a Japanese university Asia Pacific Journal of Education 2019 39 3 404 416 10.1080/02188791.2019.1598850 Castelló M McAlpine L Pyhältö K Spanish and UK post-PhD researchers: Writing perceptions, well-being and productivity Higher Education Research & Development 2017 36 6 1108 1122 10.1080/07294360.2017.1296412 Daly CJ Dee JR Greener pastures: Faculty turnover intent in urban public universities The Journal of Higher Education 2006 77 5 776 803 10.1080/00221546.2006.11778944 Thissen, D., Steinberg, L., & Kuang, D. (2002). Quick and easy implementation of the Benjamini-Hochberg procedure for controlling the false positive rate in multiple comparisons. Journal of educational and behavioral statistics, 27(1), 77–83. 10.3102/10769986027001077 Derby-Davis MJ Predictors of nursing faculty's job satisfaction and intent to stay in academe Journal of Professional Nursing 2014 30 1 19 25 10.1016/j.profnurs.2013.04.001 24503311 Elo AL Leppinen A Jahkola A Validity of a single-item measure of stress symptoms Scandinavian Journal of Work, Environment & Health 2003 29 6 444 451 10.5271/sjweh.752 Fraley C Raftery AE Bayesian regularization for normal mixture estimation and model-based clustering Journal of Classification 2007 24 2 155 181 10.1007/s00357-007-0004-5 Franzoni C Scellato G Stephan P Foreign-born scientists: Mobility patterns for 16 countries Nature Biotechnology 2012 30 12 1250 1253 10.1038/nbt.2449 Ghasemy M Muhammad F Jamali J Roldán JL Satisfaction and performance of the international faculty: To what extent emotional reactions and conflict matter? SAGE Open 2021 11 3 1 15 10.1177/21582440211030598 Haines DJ Davis L Rancour P Robinson M Neel-Wilson T Wagner S A pilot intervention to promote walking and wellness and to improve the health of college faculty and staff Journal of American College Health 2007 55 4 219 225 10.3200/JACH.55.4.219-225 17319328 Hirschi A Freund PA Herrmann A The career engagement scale: Development and validation of a measure of proactive career behaviors Journal of Career Assessment 2014 22 4 575 594 10.1177/1069072713514813 Horta H Jung J Zhang L-F Postiglione GA Academics’ job-related stress and institutional commitment in Hong Kong universities Tertiary Education and Management 2019 25 4 327 348 10.1007/s11233-019-09039-8 Houston D Meyer LH Paewai S Academic staff workloads and job satisfaction: Expectations and values in academe Journal of Higher Education Policy and Management 2006 28 1 17 30 10.1080/13600800500283734 Huang F Impacts of the COVID pandemic on international faculty's academic activities and life in Japan Higher Education Quarterly 2022 76 2 260 275 10.1111/hequ.12369 Huang F Daizen T Kim Y Challenges facing international faculty at Japanese universities: Main findings from the 2017 national survey International Journal of Educational Development 2019 71 1 8 10.1016/j.ijedudev.2019.102103 Huang, F. (2021). International faculty at Japanese universities: Main findings from national survey in 2017. In F. Huang & A. R. Welch (Eds.), International faculty in Asia: In comparative global perspective (21, 45–62). Springer. 10.1007/978-981-33-4980-3_4 Huppert FA Psychological well-being: Evidence regarding its causes and consequences Applied Psychology: Health and Well-Being 2009 1 2 137 164 10.1111/j.1758-0854.2009.01008.x Juberg M Spencer SD Martin TJ Vibell J da Costa Ferro A Kam B Masuda A A mindfulness-based intervention for college students, faculty, and staff: A preliminary investigation Clinical Case Studies 2019 18 3 185 199 10.1177/1534650119836166 Kim SK Western faculty ‘flight risk’ at a Korean university and the complexities of internationalisation in Asian higher education Comparative Education 2016 52 1 78 90 10.1080/03050068.2015.1125620 Kneer M Haybron DM Happiness and well-being: Is it all in your head? 2020 Evidence from the folk Komisarof A Hua Z Komisarof A Hua Z Making sense of transnational academics’ experiences Crossing boundaries and weaving intercultural work, life, and scholarship in globalizing universities 2016 Taylor and Francis 174 200 Kumar, S., Kodidela, S., Kumar, A., Gerth, K., & Zhi, K. (2020). Intervention and improved well-being of basic science researchers during the COVID 19 era: A case study. Frontiers in Psychology, 1110.3389/fpsyg.2020.574712 Larson-Hall J Stewart J Hale CC Wadden P Making a career of university teaching in Japan: Getting (and keeping) a full-time job Teaching English at Japanese universities 2018 Routledge 11 24 Liu-Farrer, G. (2015). From outside in: The organizational life of a Chinese immigrant in Japan. In A. Komisarof & Z. Hua (Eds.), Crossing boundaries and weaving intercultural work, life, and scholarship in globalizing universities (53–56). Lyubomirsky S Layous K How do simple positive activities increase well-being? Current Directions in Psychological Science 2013 22 1 57 62 10.1177/0963721412469809 Maniaci MR Rogge RD Caring about carelessness: Participant inattention and its effects on research Journal of Research in Personality 2014 48 61 83 10.1016/j.jrp.2013.09.008 McAllum K Language (in)competency, communication strategies, and the development of an international academic teacher identity: Becoming an audible minority Higher Education Research & Development 2017 36 4 763 776 10.1080/07294360.2016.1239614 Ministry of Education, Culture, Sports, Science and Technology. (2021). 高等教育機関《報告書掲載集計》. Retrieved June 6 from http://www.mext.go.jp/b_menu/toukei/chousa01/kihon/1267995.htm Ministry of Education, Culture, Sports, Science and Technology. (n.d.). 研究所等 : 文部科学省. Retrieved June 13 from https://www.mext.go.jp/b_menu/link/ken.htm Munene II Outsiders within: Isolation of international faculty in an American university Research in Post-Compulsory Education 2014 19 4 450 467 10.1080/13596748.2014.955637 Nagatomo DH Cook M Wadden P Hale CC “He said, she said”: Female and male dynamics in Japanese universities Teaching English at Japanese Universities 2018 Routledge 159 164 National Institute of Science and Technology Policy. (2021). 科学技術の状況に係る総合的意識調査. Opstrup N Pihl-Thingvad S Stressing academia? Stress-as-offence-to-self at Danish universities Journal of Higher Education Policy and Management 2016 38 1 39 52 10.1080/1360080X.2015.1126895 Pace, F., D’Urso, G., Zappulla, C., & Pace, U. (2019). The relation between workload and personal well-being among university professors. Current Psychology, 1-810.1007/s12144-019-00294-x Pearce, E., & Metcalfe, J. (2016). A national approach to realising the potential of researchers: The case of the UK. In The Science and Technology Labor Force (273–290). 10.1007/978-3-319-27210-8_12 Raufelder, D., Jagenow, D., Hoferichter, F., & Drury, K. M. (2013). The person-oriented approach in the field of educational psychology. Problems of Psychology in the 21st Century 5 (2013) 79–88. 10.33225/ppc/13.05.79 Rubin M Paolini S Subašić E Giacomini A A confirmatory study of the relations between workplace sexism, sense of belonging, mental health, and job satisfaction among women in male-dominated industries Journal of Applied Social Psychology 2019 49 5 267 282 10.1111/jasp.12577 Ruggeri K Garcia-Garzon E Maguire Á Matz S Huppert FA Well-being is more than happiness and life satisfaction: A multidimensional analysis of 21 countries Health and Quality of Life Outcomes 2020 18 192 1 16 10.1186/s12955-020-01423-y 31898546 Ryan RM Deci EL On happiness and human potentials: A review of research on hedonic and eudaimonic well-being Annual Review of Psychology 2001 52 1 141 166 10.1146/annurev.psych.52.1.141 Ryff CD Psychological well-being revisited: Advances in the science and practice of eudaimonia Psychotherapy and Psychosomatics 2014 83 1 10 28 10.1159/00035326310.1159/000353263 24281296 Ryff CD Keyes CLM The structure of psychological well-being revisited Journal of Personality and Social Psychology 1995 69 4 719 727 10.1037//0022-3514.69.4.719 7473027 Sabagh Z Hall NC Saroyan A Antecedents, correlates and consequences of faculty burnout Educational Research 2018 60 2 131 156 10.1080/00131881.2018.1461573 Sakurai Y Pyhältö K Disciplinary differences in doctoral student engagement in generic skills learning Studies in Graduate and Postdoctoral Education 2021 12 2 230 246 10.1108/SGPE-03-2020-0018 Sasao, T., & Hatta, N. (2016). Internationalization and faculty well-being in liberal arts colleges: An often neglected issue in East Asia. In T. Sasao & N. Hatta (Eds.), Liberal arts education and colleges in East Asia (pp. 165–177). Springer. 10.1007/978-981-10-0513-8_14 Seipel MT Larson LM Supporting non-tenure-track faculty well-being Journal of Career Assessment 2018 26 1 154 171 10.1177/1069072716680046 Shin, J. C., Kehm, B. M., & Jones, G. A. (2018). The increasing importance, growth, and evolution of doctoral education. In Doctoral education for the knowledge society (pp. 1–10). Springer. Snyder, C. R., Lopez, S. J., & Pedrotti, J. T. (2011). Positive psychology: The scientific and practical explorations of human strengths. SAGE Publications. https://books.google.co.jp/books?id=ZCE9Qx8VHAEC Stubb J Pyhältö K Lonka K Balancing between inspiration and exhaustion: PhD students' experienced socio-psychological well-being Studies in Continuing Education 2011 33 1 33 50 10.1080/0158037X.2010.515572 Sun, W., Wu, H., & Wang, L. (2011). Occupational stress and its related factors among university teachers in China. Journal of Occupational Health, 280-28610.1539/joh.10-0058-oa Takagi K Accommodating project-based professionals in higher education institutions in Japan Journal of Higher Education Policy and Management 2018 40 3 272 286 10.1080/1360080x.2018.1462434 Tzanakou C Stickiness in academic career (im)mobilities of STEM early career researchers: An insight from Greece Higher Educucation 2021 82 695 713 10.1007/s10734-020-00596-1 Van Waes S De Maeyer S Moolenaar NM Van Petegem P Van den Bossche P Strengthening networks: A social network intervention among higher education teachers Learning and Instruction 2018 53 34 49 10.1016/j.learninstruc.2017.07.005 Watermeyer R Shankar K Crick T Knight C McGaughey F Hardman J Suri VR Chung R Phelan D ‘Pandemia’: A reckoning of UK universities’ corporate response to COVID-19 and its academic fallout British Journal of Sociology of Education 2021 42 5–6 651 666 10.1080/01425692.2021.1937058 Xu YJ Gender disparity in STEM disciplines: A study of faculty attrition and turnover intentions Research in Higher Education 2008 49 7 607 624 10.1007/s11162-008-9097-4	0	PMC9734358	NO-CC CODE	2022-12-14 23:28:28	no	High Educ (Dordr). 2022 Dec 7;:1-19	utf-8	High Educ (Dordr)	2,022	10.1007/s10734-022-00978-7	oa_other
==== Front Environmental Sustainability Environmental Sustainability 2523-8922 Springer Nature Singapore Singapore 257 10.1007/s42398-022-00257-2 Editorial Progress of sustainable development goal 7: clean and green energy for all as the biggest challenge to combat climate crisis Arora Naveen Kumar [email protected] 1 Mishra Isha 23 1 grid.440550.0 0000 0004 0506 5997 Department of Environmental Science, School of Earth and Environmental Sciences, Babasaheb Bhimrao Ambedkar University, Lucknow, 226025 India 2 Directorate of Environment, Lucknow, 226010 India 3 Society for Environmental Sustainability, Lucknow, 226025 India 9 12 2022 2022 5 4 395399 © The Author(s) under exclusive licence to Society for Environmental Sustainability 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© Society for Environmental Sustainability 2022 ==== Body pmcEnergy is intrinsically related to many crucial aspects of human life and inadequacy in its supplies has always been a constraint to human and economic development. Of the 17 Sustainable Development Goals (SDGs), Goal 7 is mainly focussed on ensuring access to affordable, reliable, sustainable and modern energy for all by 2030.1 In past few years, demand for energy has dramatically increased mainly due to our reliance on technology, high living standards and ever growing population. To close the gaps of energy supplies by 2030, we need to check the use of fossil fuels as source of energy and urgently scale up production and use of renewable sources of energy to prevent long term planetary scale consequences, the most severe product of which is climate change. The first and foremost target of SDG 7 i.e. target 7.1 is to ensure universal access to affordable, reliable and modern energy services by 20301. According to the latest ‘Sustainable Development Goals Report 2022’, the rate of electricity access has increased from 83% in 2010 to 91% in 2020, worldwide, and those who lived without electricity reduced from 1.2 billion to 733 million. However, the report further adds that the rate of progress has slowed down in recent past and the complexities have increased mainly due to outbreak of coronavirus disease (COVID-19) pandemic2. Referencing to the context, the annual electricity access rates grew by 0.5% points in 2018–2020 in comparison to 0.8% points in 2010–2018. In 2020, 77% of the world population without electricity lived in sub-Saharan Africa. Hence, this region needs to be the focus of national and international organizations to achieve the target by 2030. Report suggests that only 92% of the global population will gain electricity access by 2030, leaving 670 million people behind.2 On the other hand, World health Organization (WHO) (2022) states that one third of world’s population (or 2.4 billion people) are still devoid of clean cooking and rely on polluting fuels, which impact their health. The report also claims that one third of the global population will keep using polluted cooking fuels in 2030, with majority living in sub-Saharan Africa.3 Use of inefficient and polluting fuels including wood, coal, charcoal, dung and crop waste generates huge levels of household air pollution (HAP). This can be a major health concern and can contribute to serious diseases and even deaths, particularly in women and children living in low and middle income countries (Stoner et al. 2021). A report by WHO on Household Air Pollution (2022) estimated that HAP caused 3.2 million deaths per year in 2020, which included over 237,000 deaths of children under 5 years. Not only this, ambient air pollution and household air pollution when combined is responsible for 6.7 million premature deaths per year. HAP exposure leads to non-communicable diseases, which include ischemic heart disease, stroke, chronic obstructive pulmonary disease and lung cancer.4 Additionally, the unprocessed biomass solid fuels used for domestic cooking are also major cause of air pollution causing up to 50 times more pollution than cooking gas. These fuels release many toxic pollutants in environment including carbon monoxide (CO), suspended particulate matter (SPM), hydrocarbons (HCs) and oxides of nitrogen (NOx). Of these, black carbon is of major concern, which is second strongest contributor to the global warming after emissions generated by carbon dioxide (CO2) (Ravindra 2019). Furthermore, coal, which is a major contributor of CO2 emissions, is also a huge source of electricity generation in present times creating a challenge in transitioning to low carbon energy systems. Demand for coal is strong and plays key role in fuelling economic development in emerging markets. World Coal Association reports that currently, coal power plants generate 37% of global electricity and estimates from International Energy Agency (IEA) suggest that even by 2040, coal will produce 22% of world’s electricity, maintaining its hold as the single largest source of electricity in the world.5 Coal, oil and gas being the main drivers of global warming, phasing out coal has been one of the main objectives of 26th Conference of the Parties (COP 26) to translate the goal of ‘securing global net zero by mid-century and keeping 1.5°C within reach’ into action.6 However, many countries expressed their disagreement and the final pact went from ‘phase out’ to ‘phase down’ giving setback to the sustainability goals (Arora and Mishra 2021). Recently held COP27 in Egypt also did not show any signs of progress in this regard, and the draft suggests to accelerate the measures towards ‘phase down’ of unabated coal power and phase out and rationalize inefficient fossil fuel subsidies.7 Hence, intensified efforts are needed in order to make electricity accessible and adoption of clean cooking solutions particularly in low and middle income countries to boost their growth at multiple levels. Target 7.2 is to increase substantially the share of renewable energy in the global energy mix by 20301. Ever since the industrial revolution, energy mix of most of the countries relied on fossil fuels (coal, oil and natural gas) boosting economies for over 150 years, and currently also supplying about 80% of world’s energy. Fossil fuels were formed millions of years ago from carbon rich animals and plant remains and hence when they are burned, the stored carbon and green-house gases (GHGs) are released into atmosphere.8 Report by Our World in Data on ‘renewable energy’ shows that energy generated by burning of fossil fuels produces three-quarters of GHG emissions.9 The unprecedented increase in GHGs emissions is causing global warming and extreme weather events, harming overall health of our planet and its sustainability. Intergovernmental Panel on Climate Change (IPCC) report claims that to limit the pace of global warming, we need to make transitions in the energy sector. This calls for a significant reduction in fossil fuels, ubiquitous electrification, improved energy efficiency and use of alternative fuels like hydrogen.10 The report explains that right policies, infrastructure and technology to facilitate changes in our daily lives can result in 40–70% reduction in GHGs emission by 2050. Another report by Tyndall Centre at The University of Manchester commissioned by International Institute for Sustainable Development (IISD) (2022) states that to limit the global warming up to 1.5°C, oil and natural gas must be phased out by 2034 for rich countries and by 2050 for poor.11 Hence, we need to make changes in the way we produce and consume energy. For this, it is important that a transition is made towards renewable sources of energy as they are cheaper, reliable and more efficient for everyday use. Some mainstream renewable technologies including hydropower, wind energy, solar energy, biomass energy, biofuels and geothermal energy are already contributing towards safety and sustainability of our planet (Gielen et al. 2019). The deployment of renewables is not only needful for climate change objectives but also to meet the global demands for energy. As per the estimates shown in Sustainable Development Goals Report 2022’ for Goal 7, the share of renewables in total energy consumption attained 17.7% in 2019 i.e. 1.6% higher than 2010, while total renewable energy consumption increased by a quarter during the same period2. In electricity sector, increment in renewables was witnessed from 19.7% in 2010 to 26.2% in 2019. On the other hand, in heat sector, the progress was insignificant with less than 2% gains in 2019 compared to 2010. In case of transport, renewable energy reached 3.6% in 2019, which was a jump up from 2.6% in 20102. Report further added that traditional uses of biomass remained stagnant and represented more than a third of total renewable energy use in 20192. On the other hand, Our World in Data estimates show that worldwide, hydropower occupies the largest modern renewable resource occupying 4274 terawatt-hours (TWh) in 2021, followed by wind and solar power with 1596 TWh and 846 TWh respectively9. In case of electricity mix, renewables accounted for one-quarter of electricity generation in 2021. Furthermore, total biofuel production worldwide was estimated to be 1084 TWh, where North America showed highest production with 412 TWh followed by South and Central America (273 TWh), Asia Pacific (210 TWh) and Europe (187 TWh) in 20219. The above data shows that boosting renewable energy is of utmost importance in order to decarbonize our energy systems in coming future and meeting the global demands. Target 7.3 aims at doubling the global rate of improvement in energy efficiency by 20301. Improving annual energy efficiency at a fast rate has become imperative to mitigate the adverse impacts caused by climate change that will otherwise undermine the energy security and overall global goals. Tracking the SDG 7: The Energy Progress Report (2022),’ published by the International Energy Agency (IEA), the International Renewable Energy Agency (IRENA), the UN Statistics Division (UNSD), the World Bank, and WHO reveals that Target 7.3 aims to improve primary energy intensity to 2.6 in 2010–30 versus 1990–2010. But from 2010 to 2019, annual energy efficiency improvement globally has attained the value of 1.9% which is below the target.12 For early 2020, substantial decrease in intensity improvement was witnessed due to COVID-19 pandemic. According to Sustainable Development Goals Report (2022), global primary energy intensity [ratio of total energy supply to gross domestic product (GDP)] has improved from 5.6 megajoules per US dollar (2017 purchasing power parity) in 2010 to 4.7 in 2019, with an average improvement rate of 1.9% annually. However, the report states that in order to actually meet the targets of Goal 7, improvement in average annual energy intensity will need to reach 3.2% by 2030. Eastern and South-Eastern Asia is the only region that has achieved the targets with an annual improvement of 2.7% in 2010–2019, strongly pushed by economic growth2. In fact, in order to achieve net-zero emissions by 2050, the world will need to scale up the rate of energy efficiency to 4% for the rest of the decade12. Among the last two targets, 7A focuses on enhancing international cooperation to facilitate access to clean energy research and technology, including renewable energy, energy efficiency and advanced and cleaner fossil-fuel technology, promote investment in energy infrastructure and clean energy technology by 2030; Target 7B is to expand infrastructure and upgrade technology for supplying modern and sustainable energy services for all in developing countries, in particular least developed countries, small island nations and landlocked developing countries by 20301. Both targets rely on international financial flows, for promoting access to research, technology and investments in clean energy as well as to expand or upgrade energy services for developing countries. However, despite the urgent needs of sustainable development and climate change crisis around the world, developing nations showed a decrease in financial flow for clean energy for the second consecutive year. The amount was reduced to $10.9 billion in 2019, down by approximately 24% from previous year. Recent pandemic further worsened the situation in 20202, 12. The reduction was mainly concentrated in Eastern and South-Eastern Asia, where international monetary flows fell off by 66.2%; in Latin America and the Caribbean decreased by 29.8%, and Central and Southern Asia where flows dropped by 24.5%. Oceana was the only region with an exception, where international flow grew by 72%12. In addition to all aforementioned constraints, the energy crisis arising from Russia-Ukraine conflict coupled with COVID-19 pandemic is a setback to targets of Goal 7, resulting in soaring energy prices and uncertainty in global oil and gas markets impacting economies around the globe. In order to make renewable energy and its technology available for all and to achieve climate goals, swift actions are required. With only eight years left to achieve universal access to affordable and sustainable energy for all, there is need to plan the implementation of the pacts discussed in UN Climate Change Conference (COPs). As marked in the last two summits (COP 26 and 27), emission of GHGs has to be brought down by using new technologies and renewable energy sources, and by making countries equipped for climate action by catalysing the flow of finance for clean and green technologies.13 Full-fledged industrial decarbonization, right from smartphones to aircrafts, is required to achieve climate stabilization and reaching net zero GHGs emissions. Countries must amalgamate multitude approaches to meet the energy needs in a sustainable way. Emerging and breakthrough technologies like marine energy, hydrogen power, grid batteries, cellulosic ethanol and concentrated solar photovoltaic is the new future to cut down world’s carbon footprint and are attracting the research communities (Hussain et al. 2017; Woolston and Ong 2022). A study by Rissman et al. (2020) reported that new advancing technologies specific to top-emitting industries including cement, iron, steel, chemicals and plastics have to come up quick and fast. Some successes in this direction include cement admixtures, alternative green chemistries and zero-carbon steel-making. Marine ecosystems are also being viewed as potential sources of novel bio-energy systems. Apart from the unlimited algal biomass, cutting edge technologies for energy conversion e.g. bio-solar cells and photosynthetic microbes as source of energy and hydrogen, can prove to be very important in future (Zhu et al. 2022). Novel biotechnologies such as bio-electrochemical cells (BEC) have lately gained attention regarding generation of clean energy. However, the challenges which have to be overcome regarding the use of ‘microbial factories’ is the production of energy in adequate amounts and at affordable price. Cynaobacteria have been employed as major biofuel producers. Recent studies have mapped the presence of polar cynaobacteria and linked it to bioenergy conservation. The concepts and mechanisms of microbes are currently being harnessed to design modern cost effective and eco-friendly energy generating cells. The pigments produced by extremophilic bacteria have been evaluated in efficiently trapping solar energy, and have found their application as dye sensitized solar cells (DSSCs) (Silva et al. 2019). Bacterial pigments are capable of generating electricity and their application in energy cells increases the efficiency of the system. Further research, characterizing novel pigments from microbes habituating in extreme environment can embark the application of biotechnology in generation of renewable sources sustainably in the near future. Renewable hydrogen can play an important role in alleviating carbon emissions from both light and heavy industries. Rapid scale up of renewable hydrogen, investments in hydrogen-based research and development (R&D) and infrastructure for its adoption will be needed to decarbonize industries in line with Paris Agreement (Rissman et al. 2020). To swiftly utilize these novel technologies at large scale, accelerated advancement in R&D, cutting on costs, reformed policy frameworks and business models, and enhanced international cooperation driven by governments, communities and organizations are required. The actual progress will come with support of investors and funding agencies as well as informed policy makers. People living in low-income countries are looking for cheaper alternatives. In this regard, strong market mechanisms should be developed to offer clean and affordable energy options to poor and marginalized populations. Domestic and local bankers must come forward and work in line to support the green technologies involved in providing clean energy. Renewable Energy and Jobs: Annual Report (2022) has confirmed that despite multiple crises, growth in renewable energy job has hit 12.7 million in 2021, up from 12 million in 2020. Almost two-thirds of all jobs are in Asia, and China alone holds 42% of the global total, followed by Eurpean Union (EU) and Brazil with 10% each, while USA and India account for 7% each.14 Among all renewables, the fastest growing is solar photovoltaic with 4.3 million jobs in 2021 accounting for more than a third of all renewable energy workforce. Next is wind power with 1.3 million jobs in 2021, followed by 2.4 million jobs in hydropower, and biofuels accounting for 2.4 million jobs14. United Nations Environment Programme (UNEP) is also working with different partners to improve energy access and strengthen business related to energy efficiency in developing and emerging economies, making these nations gain economic and environmental profits to reduce poverty.15 Setting up consistent policy reform agenda and strong national level targets for electrification, clean cooking and energy access by governments will help small enterprises and private sectors to operate easily. Local organizations must engage in innovative and regenerative solutions through educational and hands-on learning experiences to encourage the use of renewables in everyday life, which in turn will help to combat climate crisis. Moreover, integrated planning is important to promote decentralized energy systems alongside electricity grid improvements. This will help in balancing the shift towards clean energy and also in providing off-grid energy to rural people, helping in economic development of rural areas.16 Energy policies supporting low carbon transition still require considerable efforts and resources to uplift. Furthermore, attention should be paid to aging infrastructures including power generating equipment, which are another major cause for negative impact on energy generation. Subsidies on renewables must be given for large scale and small scale deployments. Limits on emissions should be mandated to reduce pollution from GHGs emissions. Carbon taxes must be levied on consumers to pay for energy based on the amount of CO2 produced. This will create awareness among people to shift towards products with lower carbon emissions. Access to clean and modern energy is an important target in the 2030 Agenda for SDGs, because without energy access, it will not be possible to mitigate poverty, end hunger, boost education, improve health, increase water supply and industrialization, and combat climate change. Resources are exhausting quickly thus creating a challenge for future, which can turn into ‘an era of energy crises’ for upcoming generations. For this a momentum towards a decarbonized energy generation and movement towards green and sustainable economy is required. 1 https://sdgs.un.org/goals/goal7 2 https://unstats.un.org/sdgs/report/2022/The-Sustainable-Development-Goals-Report-2022.pdf 3 https://www.who.int/news/item/20-01-2022-who-publishes-new-global-data-on-the-use-of-clean-and-polluting-fuels-for-cooking-by-fuel-type#:~:text=One%20third%20of%20the%20global,%2D%20and%20middle%2Dincome%20countries 4 https://www.who.int/news-room/fact-sheets/detail/household-air-pollution-and-health 5 https://www.worldcoal.org/coal-facts/coal-electricity/ 6 https://ukcop26.org/cop26-goals/ 7 https://www.carbonbrief.org/daily-brief/draft-cop27-agreement-fails-to-call-for-phase-down-of-all-fossil-fuels/ 8 https://www.eesi.org/topics/fossil-fuels/description 9 https://ourworldindata.org/renewable-energy 10 https://www.ipcc.ch/2022/04/04/ipcc-ar6-wgiii-pressrelease/ 11 https://www.iisd.org/articles/analysis/phase-out-oil-gas-production 12 https://www.worldbank.org/en/news/press-release/2022/06/01/report-covid-19-slows-progress-towards-universal-energy-access 13 https://news.un.org/en/story/2022/10/1129947 14 https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/documents/publication/wcms_856649.pdf 15 https://www.unep.org/explore-topics/energy/what-we-do/energy-efficiency 16 https://www.seforall.org/news/seven-steps-to-achieve-sustainable-energy-for-all Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Arora NK Mishra I COP26: more challenges than achievements Environ Sustain 2021 4 585 588 Gielen D Boshell F Saygin D Bazilian MD Wagner N Gorini R The role of renewable energy in the global energy transformation Energ Strategy Rev 2019 24 38 50 Hussain K Arif SM Aslam M Emerging renewable and sustainable energy technologies: State of the art Renew Sust Energ Rev 2017 71 12 21 Ravindra K Emission of black carbon from rural households kitchens and assessment of lifetime excess cancer risk in villages of North India Environ Int 2019 122 201 212 30522824 Rissman J Bataille C Masanet E Aden N Morrow WR Zhou N Technologies and policies to decarbonize global industry: Review and assessment of mitigation drivers through 2070 Appl Energy 2020 266 114848 Silva C Santos A Salazar R Lamilla C Pavez B Meza P Evaluation of dye sensitized solar cells based on a pigment obtained from Antarctic Streptomyces fldesensis J Sol Energy 2019 181 379 385 Stoner O Lewis J Martínez IL Gumy S Economou T Adair-Rohani H Household cooking fuel estimates at global and country level for 1990 to 2030 Nat Commun 2021 12 5793 34608147 Woolston C Ong S Three scientists at the cutting edge of new energy solutions Nature 2022 609 14 15 10.1038/d41586-022-02835-0 Zhu H Xu L Luan G Zhan T Kang Z Li C A miniaturized bionic ocean-battery mimicking the structure of marine microbial ecosystems Nat Commun 2022 13 5608 36153325	0	PMC9734359	NO-CC CODE	2022-12-14 23:49:02	no	Environmental Sustainability. 2022 Dec 9; 5(4):395-399	utf-8	null	null	null	oa_other
==== Front Nat Immunol Nat Immunol Nature Immunology 1529-2908 1529-2916 Nature Publishing Group US New York 36474117 1353 10.1038/s41590-022-01353-5 Research Briefing Impaired CD4+ T cell recognition of SARS-CoV-2 variants of concern 6 12 2022 2022 23 12 16711672 © Springer Nature America, Inc. 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. We isolated CD4+ T cell clones from healthcare workers infected with SARS-CoV-2 during the first COVID-19 wave and identified 21 epitopes across three viral proteins: spike, membrane and nucleoprotein. Focusing on spike protein, for seven of ten epitopes mutated in variants of concern, we found that T cell recognition was impaired. Subject terms Viral infection Viral infection issue-copyright-statement© The Author(s), under exclusive licence to Springer Nature America, Inc. 2022 ==== Body pmcThe problem Evolution of the coronavirus SARS-CoV-2 has led to the emergence of variants of concern (VOCs) that evade antibody responses. T cells restrict SARS-CoV-2 infection and limit COVID-19 severity1, although the impact of VOC mutations on T cell recognition is unknown. CD4+ T cells orchestrate antiviral immunity, but a lack of defined CD4+ T cell epitopes and the imprecision of in silico prediction of human leukocyte antigen (HLA) class II–restricted epitopes has limited understanding of the impact of SARS-CoV-2 mutations at the epitope level. Ex vivo assays using peptide mixtures that span whole antigens can be used to measure the overall frequency of CD4+ T cell responses to a particular antigen but do not reveal the number and identity of the constituent epitopes and their HLA class II restriction. The use of high peptide concentrations in such assays may also mask the effects of mutations on T cell recognition2. More knowledge of SARS-CoV-2-specific CD4+ T cell immunity at the level of defined epitopes is needed to understand the impact of viral mutations on CD4+ T cell surveillance. The solution We recruited healthcare workers who were infected with SARS-Cov2 during the first wave of the COVID-19 pandemic in the United Kingdom. Recruited participants had memory responses to the main CD4+ T cell targets: spike protein, membrane protein and nucleoprotein3. In peripheral blood mononuclear cell samples that had been depleted of CD8+ T cells, screening with 186 individual 20–amino acid peptides, covering the entire sequence of these proteins, revealed that every donor had CD4+ T cell responses to multiple epitopes. We generated 159 CD4+ T cell clones and used these to define the optimal peptide and HLA class II restriction of 21 epitopes within spike protein, membrane protein and nucleoprotein. Responses to the 17 epitopes identified within spike protein were detected in multiple donors with the appropriate HLA class II restriction allele. This work used donors infected with the ancestral wild-type (D614G) strain of SARS-CoV-2 and peptides based on this virus sequence. To explore whether viral evolution impaired T cell recognition, we exposed clones to varying concentrations of wild-type epitope peptide and the same peptide containing mutations found in VOCs (Fig. 1). Strikingly, we found that T cell recognition was impaired for seven of ten spike protein epitopes containing mutations in VOCs. A single mutation in some epitopes was sufficient to abolish recognition by CD4+ T cells, whereas for other epitopes, multiple mutations had no impact. Notably, for several epitopes, T cell stimulation was equivalent at the high concentration used in ex vivo screening assays but was impaired at the lower levels of peptide more likely to be present in vivo. Ex vivo testing of peripheral blood mononuclear cell samples that had been depleted of CD8+ T cells confirmed that the effects of epitope mutations were not limited to a single T cell antigen receptor but also affected the circulating polyclonal response comprising multiple different T cell antigen receptors.Fig. 1 Spike protein–specific recognition of wild-type and variant epitope peptides by CD4+ T cell clones. ELISAs measuring interferon-γ (IFN-γ) production by SARS-CoV-2 spike protein–specific CD4+ T cell clones co-cultured with autologous lymphoblastoid cell lines exposed to 20–amino acid peptides. T cell clones, identified (above plots) by the first four amino acids of the epitope, were tested in parallel with wild-type (WT) peptide and corresponding peptides from various SARS-CoV-2 VOCs (keys). Results are presented as a percentage of the maximal interferon-γ produced against the wild-type peptide. © 2022, Tye, E. X. C. et al. The implications The fact that each person convalescing from SARS-CoV-2 infection or vaccinated against SARS-CoV-2 had memory CD4+ T cell responses to multiple epitopes probably allowed CD4+ T cell recognition of the current VOCs. However, our data clearly show that certain spike protein epitopes in wild-type SARS-CoV-2 have already been lost in VOCs. Understanding the impact of the evolution of SARS-CoV-2 on T cell surveillance is important because further epitope loss is likely as the virus continues to evolve. Our work shows that a complete understanding will require a detailed map of experimentally verified T cell epitopes and the amino acids essential for their recognition. We highlight the fact that the number of amino acid mutations in an epitope is, by itself, an unreliable guide to epitope loss; furthermore, we tested the NetMHCIIpan-4.0 binding-prediction algorithm4 and found it did not predict several of the epitopes we identified. Whether the mutations we observed in the spike protein epitopes were driven by antibody-mediated immune pressure, selection to increase viral fitness or T cell surveillance is unknown. We focused on spike protein on the basis of its use in vaccination against COVID-19 and its high mutation rate. Other viral proteins targeted by T cells5 are less prone to mutation, given SARS-CoV-2 sequencing data and their high level of conservation with other human β-coronaviruses. Extending our T cell epitope mapping across the SARS-CoV-2 proteome will allow us to better understand the risks presented by emerging SARS-CoV-2 variants and the mechanisms that drive viral evolution. It will also provide a valuable resource for the rational design of future vaccines against COVID-19 and for assessing the risk of emerging SARS-CoV-2 isolates. Graham S. Taylor & Heather M. Long, University of Birmingham, Birmingham, UK. Expert opinion “This is a very timely report on CD4+ T cell responses against SARS-CoV-2 in convalescent healthcare workers. Strengths of the study are the high clinical and biological relevance, as well as the deep characterization of CD4+ T cell clones.” Marcus Buggert, Karolinska Institute, Stockholm, Sweden. Behind the paper Our research groups have a longstanding interest in T cell control of Epstein–Barr virus and CD4+ T cell immunity. Working under pandemic conditions, we applied our expertise and established assays to SARS-CoV-2. Although they are difficult to isolate, T cell clones have proven to be valuable tools for the study of anti-viral immunity. Using CD4+ T cell clones, we were able to identify and carefully delineate the T cell response to SARS-CoV-2 and the impact of VOC mutations. Going forward, we will use our clones as tools for studying the processing of SARS-CoV-2 antigens and T cell recognition of infected cells. For Epstein–Barr virus, our groups have successfully used T cell clones to generate novel insights into how the immune system controls viral infection and the relative importance of different viral proteins and epitopes as T cell targets. We anticipate the same will be true for SARS-CoV-2. G.S.T. and H.M.L. From the editor “CD4+ T cells specific for SARS-CoV-2 proteins have been reported in people with no history of SARS-CoV-2 infection. By identifying CD4+ T cell clones specific for distinct epitopes on the SARS-CoV-2 proteins, the authors can provide a focused characterization of CD4+ T cell cross-reactivity to spike proteins in various human β-coronaviruses, and also analyze how mutations in these epitopes in VOCs lead to loss of recognition by the CD4+ T cells elicited by natural infection or vaccination.” Ioana Visan, Senior Editor, Nature Immunology. This is a summary of: Tye, E. X. C. et al. Mutations in SARS-CoV-2 spike protein impair epitope-specific CD4+ T cell recognition. Nat. Immunol. 10.1038/s41590-022-01351-7 (2022). Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Moss P The T cell immune response against SARS-CoV-2 Nat. Immunol. 2022 23 186 193 10.1038/s41590-021-01122-w 35105982 2. de Silva TI The impact of viral mutations on recognition by SARS-CoV-2 specific T cells iScience 2021 24 103353 10.1016/j.isci.2021.103353 34729465 3. Grifoni A Targets of T cell responses to SARS-CoV-2 coronavirus in humans with COVID-19 disease and unexposed individuals Cell 2020 181 1489 1501 10.1016/j.cell.2020.05.015 32473127 4. Reynisson B Improved prediction of MHC II antigen presentation through integration and motif deconvolution of mass spectrometry MHC eluted ligand data J. Proteome Res. 2020 19 2304 2315 10.1021/acs.jproteome.9b00874 32308001 5. Swadling L Pre-existing polymerase-specific T cells expand in abortive seronegative SARS-CoV-2 Nature 2022 601 110 117 10.1038/s41586-021-04186-8 34758478	36474117	PMC9734360	NO-CC CODE	2022-12-14 23:47:20	no	Nat Immunol. 2022 Dec 6; 23(12):1671-1672	utf-8	Nat Immunol	2,022	10.1038/s41590-022-01353-5	oa_other
==== Front Empir Econ Empir Econ Empirical Economics 0377-7332 1435-8921 Springer Berlin Heidelberg Berlin/Heidelberg 2338 10.1007/s00181-022-02338-x Article Quantifying interconnectedness and centrality ranking among financial institutions with TVP-VAR framework http://orcid.org/0000-0002-4035-0318 Xu Hai-Chuan 12 Jawadi Fredj 3 Zhou Jie 1 http://orcid.org/0000-0002-8952-8228 Zhou Wei-Xing [email protected] 12 1 grid.28056.39 0000 0001 2163 4895 School of Business, East China University of Science and Technology, 130 Meilong Road, Shanghai, 200237 Shanghai China 2 grid.28056.39 0000 0001 2163 4895 Research Center for Econophysics, East China University of Science and Technology, 130 Meilong Road, Shanghai, 200237 Shanghai China 3 grid.503422.2 0000 0001 2242 6780 IAE Lille University School of Management, 104 Avenue, Peuple Belge, 59043 Lille, France 6 12 2022 118 29 12 2021 21 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. Financial risk is spread and amplified through the interconnectedness among financial institutions. We apply a time-varying parameter vector autoregression model to analyze the dynamic spillover effects in the Chinese financial system. We find that the 2017 house price control policies have significantly increased the risk of China’s financial system. Before 2017, with the prosperity of the real estate market, the interconnectedness of the Chinese financial system continued to decline, while after 2017, with the slowdown of house price growth and the downturn of the real estate market, the interconnectedness turned to increase. For different sectors, the trends and the magnitudes of the spillover effects are diverse, and any sector can contribute to systemic risk in a dynamic way. Finally, we rank 20 systemically important financial institutions according to two centrality measures. The stable institution ranking provides less noisy information for regulators to formulate a policy and intervene in the market effectively. Keywords TVP-VAR Spillover effect Systemic risk Systemically important financial institutions Ranking stability JEL Classification G10 http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China U1811462 71971081 Xu Hai-Chuan Zhou Wei-Xing ==== Body pmcIntroduction The interconnectedness among financial institutions features key aspects of systemic risk (Diebold and Yılmaz 2014).1 Financial institutions play a key role in macroeconomic and monetary policy transmission; hence, understanding their interconnectedness is essential for monetary and macro prudential policies (Rampini et al. 2020). Therefore, the analysis of systemic risk needs to start from the perspective of network (Barigozzi and Brownlees 2019; Barigozzi and Hallin 2017; Giudici et al. 2020; Hautsch et al. 2015; Yang and Zhou 2013). After the 2008 financial crisis, regulators began to realize that they should not only focus on the risk of a single financial institution, but should also focus on the risk of the entire financial system. Preventing systemic financial risk has become an important task for national and other relevant regulatory authorities. Systemic risk is often regarded as “hard to define but you know it when you see it,” it refers to “many market participants suffer losses at the same time and spread to the whole system” (Benoit et al. 2017). Billio et al. (2012) believe that systemic risk refers to a series of events that threaten the stability of the financial system or public confidence in it. Giglio et al. (2016) propose a systemic risk index which can predict tail risks to economic growth. To maintain macroeconomic stability, we need to identify the systemic risk and driving factors in the financial system, and at the same time, we should not ignore the impact of the financial cycle on the financial crisis and strengthen macro prudential supervision (Arnold et al. 2012). The most popular measures of risk, the value at risk (VaR) and the expected shortfall (ES), focus on the risk of an single institution in isolation. However, a single institution’s risk measure does not necessarily reflect systemic risk. Adrian and Brunnermeier (2016) propose an influential systemic risk measure, “CoVaR,” which uses quantile regression to estimate the lower tail quantile of market returns conditional on the tail return of a given institution. In addition, White et al. (2015) use multiple quantile regression method to measure tail dependence. Diebold and Yilmaz (2009) use variance decomposition to measure volatility spillover effect. For high-frequency stock data, Pelger (2020) uses principal component analysis, which was applied to local volatility and jump covariance matrix to estimate systematic factors. Another statistical method for measuring the correlation between financial institutions is the Granger causality test (Billio et al. 2012; Hong et al. 2009), but it can only determine the direction of interconnectedness, not the size. The interconnectedness among financial institutions may change drastically over time, and thus, the volatility spillover is time varying. Adams et al. (2014) and Adams et al. (2015) use a state-dependent sensitivity VaR model to study the spillover effects among four US financial sectors, i.e., commercial banks, investment banks, hedge funds, and insurance companies. Geraci and Gnabo (2018) propose a time-varying parameter vector autoregression model (TVP-VAR) to estimate the dynamic network of financial spillover effects. They find a gradual decrease in interconnectedness after the Long-Term Capital Management crisis and the 2008 financial crisis. Chan et al. (2020) develop a factor-like structure to estimate high-dimensional time-varying parameter structural vector autoregressive model (TVP-SVAR). The interconnectedness is also viewed as the driving factor of systemic importance (Drehmann and Tarashev 2013). Nucera et al. (2016) propose a principal component approach for systemic risk ranking. Acharya et al. (2017) estimate the systemic expected shortfall (SES) by integrating the leverage level (Nakajima and Omori 2012) and expected equity loss (Kupiec and Guntay 2016). Brownlees and Engle (2017) propose a similar measure called SRISK, which also computes the conditional capital shortfall, but using a different estimation approach. In addition, Nijskens and Wagner (2011) find that banks have a greater impact on systemic risk. Bernal et al. (2014) use ΔCoVaR to measure systemic risk and find banks contribute more to systemic risk, compared with insurance industry. The insurance industry is mainly the risk taker, which shows little impact on the increase in financial system vulnerability, and the risk in insurance industry is not the cause of macroeconomic recession, which lags behind the macroeconomic recession (Bierth et al. 2015). Differently, Jourde (2022) shows that the interconnectedness of the insurance industry with financial and non-financial companies has increased over the last decades for 16 developed countries. Some of our empirical results also suggest that after 2017 the outgoing connections from insurers to both brokers and real estates show an increasing trend. The proposed approaches above are widely applied in analyzing interconnectedness for developed markets. In this paper, we explore the spillover effects in the Chinese financial market. China has become the second largest economy in the world. In the past decade, China’s financial system has also developed rapidly, playing an important role in promoting China’s economic expansion. According to data released by the People’s Bank of China, the total assets of financial institutions reached 381.95 trillion yuan at the end of 2021.2 According to the Financial Stability Board, there are 4 of China’s commercial banks and 1 insurance company among the G-SIBs3 and the G-SIIs.4 It implies that China faces great regulatory challenges to prevent domestic financial risks from infecting the global market. Since 2017, in response to the surging price bubble in the real estate market, regulators have implemented the most stringent house price control policies in history. These policies include raising mortgage interest rates, restricting loans, restricting price, restricting the purchase of real estate, restricting the sale of real estate, adjusting land supply, and putting property taxes on the legislative agenda. These policies were targeted at containing the surge in house prices and regulating housing speculation. The starting point of these regulations is to maintain the stability of the real estate market, so as to curb the further expansion of the gap between the rich and the poor. However, the implementation of these policies brought unintended consequences in the financial system. It requires us to measure the systemic risk and identify the risk contribution of institutions from the macroprudential perspective. Several papers have studied the connectedness/spillover in the Chinese financial system based on various approaches, such as the Granger causality connectedness (Gong et al. 2019), CoVaR (Wang et al. 2018a), and variance decompositions (Wang et al. 2021, 2018b). In this paper, we choose the TVP-VAR framework as in Geraci and Gnabo (2018), which allows us to construct a continuously evolving spillover network, with connections changing gradually through time. Geraci and Gnabo (2018) show that importance rankings based on the TVP-VAR model are more stable than rankings based on other methods. Stable ranking is very useful for regulators to formulate a policy. More importantly, we contribute to the literature by including real estate sector into financial system. The existing literature basically considers three sectors: banks, broker–dealers, and insurers. While the real estate sector is not part of the traditional financial services industry, it has strong capital attributes. As stated above, the house price control policies can bring risks to the Chinese financial system by hitting the real estate market. Our empirical results show that before 2017, with the prosperity of the real estate market, the interconnectedness of the Chinese financial system continued to decline. However, after the house price control policies, with the slowdown of house price growth and the downturn of the real estate market, the interconnectedness turned to increase. The market-based interconnectedness measure suggests that systemic risk in the Chinese financial system has increased since 2017. In addition to the overall interconnectedness, we further analyze the interconnectedness at sectoral level and at institution level. We find the trends and the magnitudes of the spillover effects for different sectors are diverse, and any sector can contribute to systemic risk significantly. Then, we measure interconnectedness at the individual level among 20 systemically important financial institutions. The analyses at the institution level provide the importance rankings for financial institutions. Although connection strength is declining, Ping An insurance is still the most important risk taker (top 2) and the most important risk spreader. The paper is structured as follows. Section 2 introduces the TVP-VAR framework for modeling interconnectedness. Section 3 describes the data. Section 4 measures interconnectedness at the sectoral level. Section 5 measures interconnectedness at the individual institution level. Section 6 takes a robust test using an alternative measure. Section 7 draws the conclusions. The model To characterize interconnectedness among financial institutions, we estimate a time-varying VAR model following Geraci and Gnabo (2018):1 Rt=ct+BtRt-1+et≡Xt′θt+et,et∼N0,Ωt where Rt≡r1t,…,rNt′ is the vector of N stock returns, Bt is an N×N time-varying coefficient matrix, and Ωt is an N×N time-varying covariance matrix. We do not consider contemporaneous dependence because we focus on the spillover effect, which is directional and intertemporal dependence. Although contemporaneous dependence is also part of interconnectedness, it is not helpful to understand the spread of risks or fluctuations. Since the serial correlations decay with the lag order, the interconnectedness can be mostly characterized by the one lag model. We stack the matrix Bt in θt so that Eq. (1) can be viewed as the measurement equation of a state space model. The absolute value of Btji represents the strength of spillover effect from i to j at time t, defined as the time-varying marginal effect of the risk from a given institution to another institution (Hautsch et al. 2015). According to the setting of many macroeconomic literature (e.g., Cogley and Sargent (2005), Primiceri (2005), and Marco and Primiceri (2015)), we let coefficients evolve according to a driftless random walk. Thus, the state equation of the model is as follows:2 θt+1=θt+uθtut∼N(0,Σθ) where we assume that et and uθt are independent. The covariance matrix Ωt captures heteroscedastic volatility and time-varying correlation among errors, which reflects the contemporaneous spillovers among financial institutions. It can be decomposed as Ωt=At-1ΣtΣtAt′-1, where At is a lower triangular matrix where the diagonal elements equal to one, and Σt=diagσ1t,…,σNt. Thereby, the errors can be written as et=At-1Σtεt. We stack the lower triangular elements of At as at=a1t,…,aqt′ and define ht=h1t,…,hNt where hit=logσit2. These time-varying parameters follow the random walk process:3 at+1=at+uatht+1=ht+uht Then, the vector of errors [εt,uθt,uat,uht] is jointly normal with mean 0 and variance–covariance matrix defined as:4 εtuθtuatuht∼N0,IOOOOΣθOOOOΣaOOOOΣh with Σa and Σh being diagonal. The model is estimated by standard Bayesian methods as described in Nakajima (2011). It implements the Markov Chain Monte Carlo (MCMC) algorithm to generate sample from the posterior distribution of the TVP-VAR models. Data We collect the weekly closing prices of four sectoral indexes and 20 systemically important financial institutions listed in the Chinese stock market. The four sectoral indexes are SWS Banking II Index (801192), SWS Broker II Index (801193), SWS Insurance II Index (801194), and SWS Real Estate Development II Index (801181). The first three are typical financial industry indexes, and the real estate sector has close debt relationships with banks. The time series cover from July 2010 to December 2021.5 We do not use daily frequency since the daily price is updated much frequently with large amount of noise. We calculate the weekly stock returns for company i at week t as rit=logPit-logPi,t-1, where Pit is the stock price of company i at the end of week t. Next we proceed to analyze the interconnectedness for the Chinese financial system at two levels. First, we measure interconnectedness at the sectoral level. Then, we measure interconnectedness at the individual level among 20 systemically important financial institutions. Interconnectedness at the sectoral level In this section, we study the interconnectedness among four different sectors: banks, broker–dealers, insurers, and real estate companies, using a 4-variable TVP-VAR model with 1 lag. In order to measure the spillover effect among sectors, we introduce the time-varying network density, which reflects the average strength of a dependence at every period. It is given by5 DENSITYt=1N(N-1)∑i=1N∑j≠ii→tj×Bt(ji) where i,j∈{banks, brokers, insurers, real estate} and Bt(ji) is the cross coefficient between i and j at period t estimated by the TVP-VAR model with N=4. Here we use absolute beta because we are concerned with the average dependent strength. Both positive and negative spillovers represent interconnectedness. If we took the original beta, the positive and negative dependencies cancel each other out when calculating the average density across different pairs of stocks/sectors.Fig. 1 Overall density estimated by the time-varying parameter vector autoregression model Figure 1 displays the evolution of overall density for the Chinese financial system. The value of the overall density locates between 0.0646 and 0.0701; and the curve shows a downward trend first and then upward. Before 2017, the overall density continued to decline. Geraci and Gnabo (2018) find a similar decline in the US economy after the 2008 financial crisis. However, the overall density has turned to increase since April 2017. We attribute this to the most stringent house price control policies implemented by the regulatory authorities since 2017. These policies strained the capital flow of real estate companies. For example, the sale restriction policy affects the return of funds and the price limit policy reduces the profits of real estate companies. Meanwhile, the loan restriction policy reduces the bank’s profits. In addition, if a property loan defaults, banks will also take a hit. These policies reduce the growth rate of the economy, which in turn affects the performance of broker-dealers and insurance companies. Therefore, systemic risk starts to rise across the financial sectors. The COVID-19 has little effect on the upward trend in systemic risk. This is because the Chinese government has been very successful in controlling the epidemic. With the release of domestic systemically important banks (D-SIBs) and the strengthening of supervision over financial institutions, the upward trend of systemic risk was contained by the end of 2021.Fig. 2 Price index (left) and volume index (right) of listed houses in China. These two indexes were released on April 1, 2015, and the value on base period is 100 To illustrate that the evolution of interconnectedness is indeed affected by the house price control policies, we show the price index and volume index of listed houses in Fig. 2. We can see that the growth of house price index slowed down significantly after 2018. More significantly, the house listing volume presents an opposite trend to the interconnectedness of China’s financial system. In other words, the interconnectedness presents a V-shaped evolution pattern, while the volume of listed houses presents a corresponding inverted V-shaped pattern.6Fig. 3 Time-varying cross-coefficient representing the directed spillover effect from sector i (the ith column of the figure) to sector j (the jth row of the figure) To further analyze the spillover effect among sectors, we study the time-varying cross-coefficients of the TVP-VAR model. Figure 3 gives the time-varying cross-coefficient (the value of Bt(ji) where i≠j), which represents directed spillover effect from sector i (the ith column of Fig. 3) to sector j (the jth row of Fig. 3). The sign of Bt(ji) indicates whether the spillover effect is positive or negative. We first analyze the interconnectedness between banks and other sectors. We find that the outgoing connections from banks to both brokers and real estates are first rising and then falling, while the connections to insurers are becoming smaller. Meanwhile, the incoming connections from other sectors to banks show some degree of symmetry with the outgoing case, but not quite. For example, the spillover impact of banks on brokers is obviously greater than that of brokers on banks, although their trends are consistent. It is worth noting that the spillover values are negative from insurers to banks, so the dependency trend is likewise decline as that from banks to insurers. For brokers, the evolutionary patterns of outgoing and incoming connections are quite similar with those for banks. For insurers, the spillover effects are asymmetric, that is, the connection value from insurers to other sectors is negative, while the connection from other sectors to insurers is positive. This may be due to the unique industry attributes of insurers. Compared with other financial institutions, insurers essentially short risks, just like option sellers. Therefore, as counterparties of other investors, insurers often have negative spillovers with other financial sectors. The (absolute) outgoing connections to banks show a downward trend, which is consistent with incoming connections from banks. This may reflect the competitive relationship between insurance and banking. However, after 2017, the (absolute) outgoing connections from insurers to both brokers and real estates show an increasing trend. It reflects that the house price control policies intensify the connections among different sectors. This is consistent with the empirical results for the developed countries (Jourde 2022). For real estates, the outgoing connections to brokers show a upward trend, but a downward trend to insurers. The outgoing connections to banks are more complicated; namely, the curve shows an upward trend first and then decline between 2014 and 2016 and finally rise again. The real estate is a typical capital-intensive industry. Its development depends on credit support from banks. When housing prices rise too quickly as in 2014–2016 and exceed the purchasing power of residents, housing supply exceeds demand. As a result, a large amount of real estate inventory is stored. The intensified backlog funds of real estates will cause an increase in the non-performing loan ratio of commercial banks, which has a negative impact on the development of commercial banks. Therefore, the performances of real estates and banking industry may emerge some separation and cause the connection decline in that period. After 2017, the house price control policies intensify the spillovers from real estates to banks. Interconnectedness at financial institution level In this section, we analyze the interconnectedness among individual financial institutions. We select 20 systemically important financial institutions in China, including all 4 G-SIBs and 1 G-SII released by the Financial Stability Board (FSB) and other 15 head institutions.7 The sample consists of 7 banks, 5 brokers, 3 insurers, and 5 real estates, as the following: Bank of China, China Construction Bank, Industrial and Commercial Bank of China, Agricultural Bank of China, China Minsheng Bank, China Merchants Bank, China CITIC Bank, CITIC Securities, Haitong Securities, GF Securities, Huatai Securities, China Merchants Securities, Ping An Insurance, China Pacific Insurance, China Life Insurance, Vanke, Poly Group, Gemdale Group, Shimao Group, and GM Real Estate. The study conducts a pairwise analysis similar to that of Billio et al. (2012) and Geraci and Gnabo (2018). That is, we estimate a bivariate TVP-VAR for each pair of financial institutions. Financial institution centrality In order to compare the importance of each institution in the dynamic financial network, we introduce two measures: incoming centrality (ICC) and outgoing centrality (OGC), according to the definition of weighted degree in complex network. The absolute value of Bt(ij) (Bt(ji)) is viewed as the in-degree (out-degree) weight for financial institute i. Thus, ICC measures the average extent influenced by other financial institutions and OGC measures the average propagation to other financial institutions. These two measures are given as follows:6 ICCi,t=1N-1∑j≠ij→ti×Bt(ij)OGCi,t=1N-1∑j≠ii→tj×Bt(ji) where i,j∈{1,2,⋯,20} and N = 20 and Bt(ij) represents the cross coefficient from j to i estimated by the bivariate TVP-VAR model.Fig. 4 Incoming centrality for the Chinese systemically important financial institutions Fig. 5 Outgoing centrality for the Chinese systemically important financial institutions Figures 4 and 5 represent the time-varying ICC and OGC for 20 systemically important financial institutions, respectively. Figure 4 shows that Ping An Insurance is the most connected financial institution with the highest level of ICC at the beginning of sample period. Ping An Insurance is the unique Chinese insurance company listed in the global systemically important insurers (G-SIIs). However, the ICC level of Ping An Insurance decreases year by year from 0.085 to 0.058, showing that spillover from other institutions to Ping An Insurance is declining, and Ping An Insurance is becoming more robust. At the end of sample period, the ICC level of China Merchants Securities exceeds that of Ping An Insurance. For real estates, we find that almost all real estate companies present V-shaped ICC, as the overall density in Fig. 1. On the whole, the trends of ICC level for different financial institutions are diverse. Figure 5 shows that OGC trends are more homogeneous among financial institutions compared with the ICC trends. Specifically, all 5 brokers experience a decline in OGC during this sample period. All insurers and real estates show a pattern of first falling and then rising. However, the trends of OGC for banks are not consistent, that is, increasing for Bank of China, Agricultural Bank of China, China CITIC Bank, and China Minsheng Banking, but decreasing for China Merchants Bank. Stability of centrality rankings According to the ICC and OGC, we can rank the 20 systemically important financial institutions at every time point. However, if the rankings of institutions are highly volatile, it is difficult for regulators to supervise these institutions and adjust supervision policies in time. Therefore, the stability of ranking is important to implement an effective supervision. In order to assess the stability of interconnectedness ranking, we refer to two stability indicators proposed by Geraci and Gnabo (2018). The quadratic stability indicators are defined as:7 SIQIN=1T-1∑t=2T∑i=1NZi,tIN-Zi,t-1IN2NSIQOUT=1T-1∑t=2T∑i=1NZi,tOUT-Zi,t-1OUT2N where Zi,tIN represents the ranking of institution i at time t according to the ICC and Zi,tOUT represents the ranking of institution i at time t according to the OGC. Similarly, the absolute stability indicators are defined as:8 SIAIN=∑t=2T∑i=1NZi,tIN-Zi,t-1INN(T-1)SIAOUT=∑t=2T∑i=1NZi,tOUT-Zi,t-1OUTN(T-1) In order to compare the ranking stability of ICC and OGC estimated by the TVP-VAR model, we also compute these stability indicators for several additional rankings, such as MES (Marginal Expected Shortfall), leverage ratio, and SES (Systemic Expected Shortfall) proposed by Acharya et al. (2017). The MES refers to the expected equity loss of each institution in the financial crisis, defined as the average return of its equity when the market returns appear in the 5% worst days:9 MESi=1number of days∑{t: system is in its5%tail}Ri,t The quasi-leverage ratio is defined as:10 LVGi=quasi-marketvalueofassetsmarketvalueofequity=bookassets-bookequity+marketequitymarketequity SES measures the conditional capital shortfall of a financial institution when a systemic crisis has already happened. It is based on a structural assumption that requires observing a realization of systemic crisis. For instance, the 2008 financial crisis is a typical systemic crash. We can use the returns of each firm from July 2008 to June 2009 to estimate MES and use the balance sheet data at June 2009 to estimate the LVG. Correspondingly, we use the cumulative equity return from July 2009 to June 2010 as the proxy of SES for each firm. Then, we run a cross-sectional regression for SES on MES and LVG:11 SESi,t=a+bMESi,t-1+cLVGi,t-1+ϵi,t After having the triplet (a, b, c), the predicted SES that firm i poses at a future time t+1 is calculated as:12 SES^i,t+1=b^b^+c^MESi,t+c^b^+c^LVGi,t Bisias et al. (2012) give a detailed review on systemic risk measures, including MES, SES, and Leverage described above. We notice that an alternative measure, called SRISK (Brownlees and Engle 2017), estimates the conditional capital shortfall the same as SES, but using a different method.Table 1 Absolute and quadratic stability indicators for centrality rankings. Panel A reports the results computed by the TVP-VAR model. Panel B reports the results computed by SES, MES and leverage ratio. “% Invariance” means the proportion that the ranking of financial institutions remains unchanged in adjacent time periods Panel A. 2010–2021 Stability indicators % invariance SIQIN SIQOUT SIAIN SIAOUT IN OUT TVP-VAR 2.51 2.37 0.99 0.93 84.58 86.04 Panel B. 2010–2021 SIQ SIA % invariance SES 4.46 3.10 25.65 MES 7.96 6.47 5.65 Leverage 3.38 1.99 47.61 Table 1 reports the absolute and quadratic stability indicators for centrality rankings computed by the TVP-VAR model (Panel A), as well as by SES, MES and leverage ratio (Panel B). The columns headed “% Invariance” denote the proportion that the ranking of financial institutions remains unchanged in adjacent time periods. From Panel A, the quadratic stability value for the ICC and OGC measured by TVP-VAR is 2.51 and 2.37, and the absolute stability value is 0.99 and 0.93, respectively. We can see that the ranking of risk propagator is more stable than that of risk receiver. The average percentages of institutions that maintain the same ranking in adjacent time periods are 84.58% and 86.04%. We can conclude that the ranking of most institutions in the adjacent time periods remains unchanged. From Panel B, the quadratic and absolute stability indicators measured by MES are the highest, while the lowest calculated by leverage. The leverage is computed based on the book value of assets reported in the firm’s balance sheet, which is updated less frequently. Therefore, the ranking measured by leverage is more stable. In contrast, MES is calculated by the firm and the market’s stock return and therefore has the lowest stability. SES is the weighted average of MES and leverage ratio, so its stability lies in between. The average percentages of institutions maintaining the same position in adjacent time periods are 25.65%, 5.65%, and 47.61%, respectively. Only a few proportion of institutions remain unchanged under MES method, while more than half of institutions under SES and leverage ratio change the rank. Compared with Panel A and Panel B, we show that the TVP-VAR model provides much more stable institution ranking than other methods.Table 2 Top 5 financial institutions in terms of incoming and outgoing centrality at 3 time points Ranking Incoming centrality Outgoing centrality End of Jun, 2012 1 Ping An Insurance Ping An Insurance 2 China Merchants Securities China Vanke 3 CITIC Securities Haitong Securities 4 China Life Insurance Bank of China 5 China Pacific Insurance CITIC Securities End of Jun, 2018 1 China Merchants Securities Ping An Insurance 2 Ping An Insurance Bank of China 3 China Life Insurance Haitong Securities 4 CITIC Securities China Vanke 5 GM Real Estate CITIC Securities End of Jun, 2021 1 China Merchants Securities Ping An Insurance 2 Ping An Insurance Bank of China 3 China Life Insurance China Vanke 4 CITIC Securities Haitong Securities 5 GM Real Estate CITIC Securities To illustrate the ranking stability by the TVP-VAR model intuitively, Table 2 lists the top 5 institutions, in terms of incoming and outgoing centrality at 3 time points, i.e., the end of Jun 2012, the end of Jun 2018, and the end of Jun 2021. We can find that the rankings are fairly stable at different time points for both ICC and OGC. For ICC, Ping An Insurance and China Merchants Securities are always the top 2. CITIC Securities and China Life Insurance always list in top 5, although their rankings changes slightly. Similarly, for OGC, Ping An Insurance, Bank of China, Haitong Securities, China Vanke, and CITIC Securities never rank out of top 5. Moreover, Ping An Insurance always ranks the top 1. Interestingly, the top 5 include head institutions of all 4 financial sectors, indicating that any sector can contribute to systemic risk significantly. Alternative measure In order to demonstrate the robustness of the interconnectedness measure, we estimate the total interconnectedness using the variance decomposition method proposed by Diebold and Yılmaz (2014). We select 2 different rolling window widths to estimate, i.e., 50 weeks and 130 weeks (see Fig. 6). We also see that before the implementation of the house price control policies, the interconnectedness continued to decline, while after that, the interconnectedness turned to increase, especially for the longer rolling window case (130 weeks). This evolutionary pattern is similar with that in Fig. 1, but with higher volatility.Fig. 6 Total interconnectedness estimated by the variance decomposition method proposed by Diebold and Yılmaz (2014). The rolling estimation window widths are 50 weeks (left) and 130 weeks (right), respectively. The predictive horizon for the underlying variance decomposition is 5 weeks Conclusion In this paper, we apply a time-varying parameter vector autoregression model to analyze the spillover effects in the Chinese financial system. We compute the time-varying sectoral density and institution centrality, respectively. To do this, we collect the weekly closing prices of four sectoral indexes and 20 systemically important financial institutions from 2010 to 2021. Since we allow stochastic volatility in the model, we take a Bayesian approach using the Markov Chain Monte Carlo sampling algorithm for an efficient estimation. We calculate an overall network density for the Chinese financial system, which reflects the average strength of dependence among four sectors at every time point. We find that the trend of overall density is highly related with the house price control policies since 2017. Before 2017, with the prosperity of the real estate market, the interconnectedness of the Chinese financial system continued to decline. However, after the house price control policies, with the slowdown of house price growth and the downturn of the real estate market, the interconnectedness turned to increase. This implies that the 2017 house price control policies have significantly increased the risk of China’s financial system. For different sectors, the trends and the magnitudes of the spillover effects are diverse, and any sector can contribute to systemic risk in a dynamic way. At the financial institution level, we estimate two importance measures proposed by Geraci and Gnabo (2018): incoming centrality and outgoing centrality. These centrality measures are time varying and asymmetric between incoming and outgoing connections. According to these measures, we rank the 20 systemically important financial institutions at every time point. As the unique Chinese insurance company among G-SIIs, Ping An insurance is still the most important risk taker (top 2) and the most important risk spreader, although its connection strength is declining. We also confirm that the rankings are fairly stable at different time points for the Chinese financial system, as in Geraci and Gnabo (2018) for the American financial system. The stable institution ranking provides less noisy information for regulators to formulate a policy and intervene in the market effectively. Acknowledgements We are grateful to the editor Bertrand Candelon and an anonymous referee for their constructive comments and helpful suggestions. We thank participants at the 5th International Workshop on “Financial Market and Nonlinear Dynamics” organized in Paris in June 2021. We acknowledge financial support from the National Natural Science Foundation of China (U1811462 and 71971081) and the Fundamental Research Funds for the Central Universities. Declarations Competing interests The authors have no competing interests to declare that are relevant to the content of this article. 1 We focus on the spillover effect among financial institutions, which is directional and intertemporal dependence. Therefore, we use “interconnectedness” instead of “connectedness”. 2 http://www.pbc.gov.cn/en/3688247/3688978/3709143/4509019/index.html. 3 https://www.fsb.org/wp-content/uploads/P231121.pdf. 4 https://www.fsb.org/wp-content/uploads/2016-list-of-global-systemically-important-insurers-G-SIIs.pdf. 5 We start the analysis in 2010 because some systemically important financial institutions were listed in 2010. For example, the Agricultural Bank of China was listed on the Shanghai Stock Exchange on July 15, 2010. 6 Note that we can’t simply say that the interconnectedness measures are procyclical or countercyclical. Before 2017, the house price index surged, while the overall interconnectedness showed a downward trend. After 2017, although both the house price index and the interconnectedness have been rising, the growth of house price index slowed down significantly. 7 We select stocks based on 2 criteria. First, its listing date is no later than that of the Agricultural Bank of China. Second, in addition to 4 G-SIBs and 1 G-SII, we select other institutions based on its industry reputation. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Acharya V Pedersen L Philippon T Richardson M Measuring systemic risk Rev Financ Stud 2017 30 1 4 27 10.1093/rfs/hhw088 Adams Z Füss R Gropp R Spillover effects among financial institutions: a state-dependent sensitivity Value-at-Risk approach J Financ Quant Anal 2014 49 3 575 598 10.1017/S0022109014000325 Adams Z Füss R Schindler F The sources of risk spillovers among U.S. REITs: financial characteristics and regional proximity Real Estate Econ 2015 43 1 67 100 10.1111/1540-6229.12060 Adrian T Brunnermeier MK CoVaR Am Econ Rev 2016 106 7 1705 1741 10.1257/aer.20120555 Arnold B Borio C Ellis L Moshirian F Systemic risk, macroprudential policy frameworks, monitoring financial systems and the evolution of capital adequacy J Bank Financ 2012 36 12 3125 3132 10.1016/j.jbankfin.2012.07.023 Barigozzi M Brownlees C NETS: network estimation for time series J Appl Economet 2019 34 3 347 364 10.1002/jae.2676 Barigozzi M Hallin M A network analysis of the volatility of high dimensional financial series J R Stat Soc C 2017 66 3 581 605 10.1111/rssc.12177 Benoit S Colliard J Hurlin C Perignon C Where the risks lie: a survey on systemic risk Rev Finance 2017 21 1 109 152 10.1093/rof/rfw026 Bernal O Gnabo J Guilmin G Assessing the contribution of banks, insurance and other financial services to systemic risk J Bank Financ 2014 47 270 287 10.1016/j.jbankfin.2014.05.030 Bierth C Irresberger F Weiss G Systemic risk of insurers around the globe J Bank Financ 2015 55 232 245 10.1016/j.jbankfin.2015.02.014 Billio M Getmansky M Lo A Pelizzon L Econometric measures of connectedness and systemic risk in the finance and insurance sectors J Financ Econ 2012 104 3 535 559 10.1016/j.jfineco.2011.12.010 Bisias D Flood M Lo A Valavanis S A survey of systemic risk analytics Annu Rev Financ Econ 2012 4 255 296 10.1146/annurev-financial-110311-101754 Brownlees C Engle R SRISK: a conditional capital shortfall measure of systemic risk Rev Financ Stud 2017 30 1 48 79 10.1093/rfs/hhw060 Chan JCC Eisenstat E Strachan RW Reducing the state space dimension in a large TVP-VAR J Econom 2020 218 1 105 118 10.1016/j.jeconom.2019.11.006 Cogley T Sargent T Drifts and volatilities: monetary policies and outcomes in the post WWII US Rev Econ Dyn 2005 8 2 262 302 10.1016/j.red.2004.10.009 Diebold F Yilmaz K Measuring financial asset return and volatility spillovers, with application to global equity markets Econ J 2009 119 534 158 171 10.1111/j.1468-0297.2008.02208.x Diebold F Yılmaz K On the network topology of variance decompositions: measuring the connectedness of financial firms J Econom 2014 182 119 134 10.1016/j.jeconom.2014.04.012 Drehmann M Tarashev N Measuring the systemic importance of interconnected banks J Financ Intermed 2013 22 4 586 607 10.1016/j.jfi.2013.08.001 Geraci MV Gnabo J-Y Measuring interconnectedness between financial institutions with Bayesian time-varying vector autoregressions J Financ Quant Anal 2018 53 3 1371 1390 10.1017/S0022109018000108 Giglio S Kelly B Pruitt S Systemic risk and the macroeconomy: an empirical evaluation J Financ Econ 2016 119 3 457 471 10.1016/j.jfineco.2016.01.010 Giudici P Sarlin P Spelta A The interconnected nature of financial systems: direct and common exposures J Bank Financ 2020 112 105149 10.1016/j.jbankfin.2017.05.010 Gong X-L Liu X-H Xiong X Zhang W Financial systemic risk measurement based on causal network connectedness analysis Int Rev Econ Financ 2019 64 290 307 10.1016/j.iref.2019.07.004 Hautsch N Schaumburg J Schienle M Financial network systemic risk contributions Rev Financ 2015 19 2 685 738 10.1093/rof/rfu010 Hong Y Liu Y Wang S Granger causality in risk and detection of extreme risk spillover between financial markets J Econom 2009 150 2 271 287 10.1016/j.jeconom.2008.12.013 Jourde T The rising interconnectedness of the insurance sector J Risk Insur 2022 89 2 397 425 10.1111/jori.12373 Kupiec P Guntay L Testing for systemic risk using stock returns J Financ Serv Res 2016 49 2–3 203 227 10.1007/s10693-016-0254-1 Marco D Primiceri G Time varying structural vector autoregressions and monetary policy: a corrigendum Rev Econ Stud 2015 82 4 1342 1345 10.1093/restud/rdv024 Nakajima J Time-varying parameter VAR model with stochastic volatility: an overview of methodology and empirical applications Monet Econ Stud 2011 29 107 142 Nakajima J Omori Y Stochastic volatility model with leverage and asymmetrically heavy-tailed error using GH skew Student’s t-distribution Comput Stat Data Anal 2012 56 11 3690 3704 10.1016/j.csda.2010.07.012 Nijskens R Wagner W Credit risk transfer activities and systemic risk: how banks became less risky individually but posed greater risks to the financial system at the same time J Bank Financ 2011 35 6 1391 1398 10.1016/j.jbankfin.2010.10.001 Nucera F Schwaab B Koopman S Lucas A The information in systemic risk rankings J Empir Financ 2016 38 461 475 10.1016/j.jempfin.2016.01.002 Pelger M Understanding systematic risk: a high-frequency approach J Financ 2020 75 4 2179 2220 10.1111/jofi.12898 Primiceri G Time varying structural vector autoregressions and monetary policy Rev Econ Stud 2005 72 3 821 852 10.1111/j.1467-937X.2005.00353.x Rampini A Viswanathan S Vuillemey G Risk management in financial institutions J Financ 2020 75 2 591 637 10.1111/jofi.12868 Wang G-J, Jiang Z-Q, Lin M, Xie C, Stanley HE (2018a) Interconnectedness and systemic risk of China’s financial institutions. Emerg Mark Rev 35:1–18. 10.1016/j.ememar.2017.12.001 Wang G-J, Xie C, Zhao L, Jiang Z-Q (2018b) Volatility connectedness in the Chinese banking system: do state-owned commercial banks contribute more? J Int Financ Mark Inst Money 57:205–230. 10.1016/j.inthn.2018.07.008 Wang G-J Chen Y-Y Si H-B Xie C Chevallier J Multilayer information spillover networks analysis of China’s financial institutions based on variance decompositions Int Rev Econ Financ 2021 73 325 347 10.1016/j.iref.2021.01.005 White H Kim T Manganelli S VAR for VaR: measuring tail dependence using multivariate regression quantiles J Econom 2015 187 1 169 188 10.1016/j.jeconom.2015.02.004 Yang J Zhou Y Credit risk spillovers among financial institutions around the global credit crisis: firm-level evidence Manag Sci 2013 59 10 2343 2359 10.1287/mnsc.2013.1706	0	PMC9734361	NO-CC CODE	2022-12-14 23:28:28	no	Empir Econ. 2022 Dec 6;:1-18	utf-8	Empir Econ	2,022	10.1007/s00181-022-02338-x	oa_other
==== Front Cent Eur J Oper Res Cent Eur J Oper Res Central European Journal of Operations Research 1435-246X 1613-9178 Springer Berlin Heidelberg Berlin/Heidelberg 831 10.1007/s10100-022-00831-3 Article A deterministic model for the inventory policy of countries for procurement of vaccines http://orcid.org/0000-0002-6494-6343 Pınarbaşı Aysun [email protected] https://orcid.org/0000-0002-1349-1035 Vizvári Béla [email protected] grid.461270.6 0000 0004 0595 6570 Department of Industrial Engineering, Eastern Mediterranean University, Mersin 10, IE-C105 Famagusta, North Cyprus Turkey 9 12 2022 113 17 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. The countries are the units that procure the vaccines during the COVID-19 pandemic. The delivered quantities are huge. The countries must bear the inventory holding cost according to the variation of stock quantities. This cost depends on the speed of the vaccination in the country. This speed is time-dependent. The vaccinated portion of the population can be approximated by the cumulative distribution function of the Cauchy distribution. A model is provided for determining the minimal-cost inventory policy and its optimality conditions are provided. The model is solved for 20 countries for different numbers of procurements. The results reveal the individual behavior of each country. Keywords COVID-19 Vaccination Inventory policy Invendory holding cost Bounded total demand Cauchy distribution Sigmoid function ==== Body pmcIntroduction Yellow fever, caused by Flavivirus, a filterable agent, was reported as the first virus infected human beings. It caused pandemic in 1901. Afterward, the 1918–1920 pandemic, Spanish flu (H1N1), is regarded as the greatest medical disease of the time, affecting millions of people worldwide. Furthermore, in 1997, the first human cases of a new and extremely dangerous avian influenza virus—the H5N1 strain—were discovered in Hong Kong. The most recent serious case is the outbreak of the COVID-19 pandemic (Kinlaw et al. 2009; Pradhan et al. 2020). The outbreak of the 2019 novel coronavirus disease (COVID-19) has brought challenges in many areas of the world (Dwyer et al. 2020; Gabster et al. 2020; Mamun and Ullah 2020; Wollina 2020; Zhou et al. 2020). (World Health Organization, no date), it is a pandemic viral disease that caused more than 620,000,000 people infected and more than 6,540,000 deaths worldwide until 12 October 2022. Vaccination has began at the end of 2020 with the full approval of Pfizer by the Food and Drug Administration (FDA) of the USA. In order to accomplish vaccination; vaccine supplies, people to implement them and people to be vaccinated are required (U.S. Food and Drug Administration, no date; Mills and Salisbury 2021). Other methods of prevention including wearing a mask, washing hands, and obeying the social distance rule, reduce the risk of transmission in the community as well (Güner, Hasanoğlu and Aktaş 2020). However, it has been discovered that the vaccine must have an efficacy of at least 70% to prevent an epidemic, and this emphasizes the impact of the vaccine on epidemics (Bartsch et al. 2020). In inventory problems, stockouts and overstock can be handled to meet the demand of dynamic customers (Shekhawat et al. 2016). Both parameters are costly situations for companies and their risks of them are reducible. Improvements in ordering policy cope with stockouts (Antic, Djordjevic Milutinovic and Lisec 2022). From a supply chain management perspective, perishable products such as vaccines are difficult to transport and store and need accurate inventory control models. In other words, inventory optimization strategies need to connect with the right balance to customer demand to avoid inventory holding (O’Neill and Sanni 2018). This is especially true in highly uncertain and dynamic markets, such as those created by the pandemic period. It is of critical importance today due to the large size demand for vaccination required for social needs in the recent Covid-19 epidemic, and because it is subject to public procurement with urgent timing (Patriarca et al. 2020). We assume that the number of vaccines is enough to cover a significant part of the population. Decision-makers are taking a significant role in vaccine allocation in order to reduce problems caused by the disease. The effectiveness of distribution of a limited vaccine inventory in a certain population can be measured by certain outcomes. After a proper distribution model, it is expected to reduce the total number of cases and the total number of losses of life. However, delivering the vaccine to people does not only include the activity of transportation in order to dispatch a certain amount of vaccine and also includes storing in stock points. In this case, some inventory policies should apply which provide a yield of minimizing the cost of inventory holding. It is necessary to provide vaccines uninterruptedly and to create a vaccination policy for this purpose. Properties of perishable products should also take into consideration. On the other hand, decisions depend on the willingness of the population for being vaccinated in the society, because it directly affects the number of requirements of the countries. The paper suggests a new policy for the procurement of vaccines on the level of individual countries. The policy enables the governments to act in a rational and economic way such that still every citizen who volunteers for obtaining the vaccine can get it. Here, we provide an inventory policy for the pandemic period for the countries. This paper presents a deterministic model for vaccines with a demand rate variable over time for the countries. It is aimed to provide an analytical model to deal with minimization of holding cost and develop inventory policies regarding this aim to be used for a variety of perishable products such as vaccines. In Sect. 2, the saturation process is introduced. In Sect. 3, an approximation of the vaccination curve of the countries has been discussed. In Sect. 4, a deterministic model for inventory policy has been developed. Section 5 discusses the numerical results obtained for 20 countries. The paper is finished with conclusions. Saturation processes The saturation process is best illustrated by the life cycle of a product. The life cycle of a product begins by entering the market. Only a few people know the new product, so consumption is growing slowly. As time goes on, more and more people get to know the product. A nearly constant consumption is formed. Later, other, alternative and substitute products will appear on the market. Therefore, the consumption of the product under investigation slows down and eventually stops. The first finding that is immediately apparent is that it is not possible to an infinite amount of the product. The market is saturating and even the curve of total consumption does not cross a certain threshold. If we look at the curve that shows the volume of all products sold from the product release to a given moment, it will be like an elongated letter S. This particular curve is constantly increasing. If the events occur as described above, the curve will have at least one inflection point. For very famous products, such as the release of a new volume from a well-known book series, or the start of selling a new mobile phone from a popular brand, when interested customers line up for either the book or the phone, the event curve may be different. Namely, consumption is very high in the first period and then decreases later. Thus, the curve of total sales increases steeply first and then decreases. Then the curve has no inflection point. In that case, the curve itself is like a concave function. The vaccination is very similar to the life cycle of a product. It is not possible to vaccinate more people than the total population. The actual value is even less as some people cannot be vaccinated because of medical reasons, and others resist or are unavailable. Thus, the vaccine is a product with a bounded total demand. The mathematical definition of a saturation function is as follows: Definition A function fx is called saturation function if fx is a real-valued function on the [-∞,∞] interval such that (i) it is monotone increasing, (ii) limx→-∞fx=0 , (iii) there is a positive number L such that limx→+∞fx=L . Obviously, it can be seen that the saturation function is very similar to the cumulative distribution function of probabilistic distributions. The only difference is that the limit at +∞ is not necessarily 1 but it can be another positive number. Figure 1 shows a saturation function which is a cumulative distribution function at the same time. It is the cumulative distribution function of the Cauchy distribution.Fig. 1 The function arctanxπ+0.5 In mathematics, a function is called the sigmoid function which is very similar to the saturation function. The sigmoid function must have exactly one inflection point. There is no such a claim in the case of saturation function. On the other hand, the sigmoid function can converge at -∞ to any value. There are many functions which are both sigmoid and saturation. The word saturation expresses better the role of the function in the applications. Approximation of the vaccination curve of the countries The total percentage of vaccinated people in a country divided by 100 is a number between 0 and 1 (Our World in Data, no date). This total number is an increasing function of time. Thus, it can be approximated by the cumulative distribution functions of a probability distribution. There are plenty of known functions which are the cumulative distribution function of frequently used probability distributions. The function of Fig. 1 is based on the inverse of the tangent function and its formula is.arctanxπ+0.5. It is both a sigmoid and saturation function. The function is not suitable for regression in this pure form. Three parameters must be introduced. One is the position of the inflection point and is denoted by b. The next one determines its steepness at the inflection point. It is a. The last one is the height of the shift vertically such that at the beginning the value of the function is 0. The general formula of the function is1 arctanax-bπ+c. Many saturation processes have the feature that the saturating property can be uncovered earlier than the process reached the inflection point. It is the case with vaccination as well. It is illustrated in the example of Denmark. The time series of the country consists of 319 days. Denmark did not reach yet the 80 percent vaccination until the end of the period. The inflection point is about the 160th day. Regression is made after 32, 64, 96, and 300 days. Figure 2 shows the final result of the regression after 300 days. Notice that the two curves, i.e. the original time series and the regression function are very close to one another. The first three curves are compared with the last one in Fig. 3. Table 1 summarizes the basic data. The saturation is not recognized yet after 32 days. The regression function in Fig. 3a is almost linear. The reason is that the process is at the beginning. The achieved vaccination is less than 3 percent. The saturation is uncovered after 64 days. The steepness of the curve is somewhat overestimated as Fig. 3 (b) shows. However, the regression function after 96 days is very close and the inflection point is still not achieved, i.e. the process is still in its first half in Fig. 3c.Fig. 2 The final result of the regression after 300 days Fig. 3 The results of regression for after a 32 days; b 64 days; c 96 days Table 1 Time length of the regression period and vaccination percentage Day Achieved percentage of vaccination 32 2.95 64 6.94 96 12.48 300 76.86 The saturation is still not uncovered as can be seen in Fig. 3a. After 64 days. The saturation is uncovered. However, the steepness is overestimated in Fig. 3b. After 96 days, the two regression curves are close to one another in Fig. 3c. It turned out that function (1) works for every country well. High correlations are achieved. Table 2 summarizes the results for a selected set of countries.Table 2 The parameters of the countries Country Parameters Correlation Available percentage a b c Belgium 0.019440264 142.3695511 0.372352365 0.998409235 78 Canada 0.023610645 141.5326286 0.367132913 0.998222367 79 Cyprus 0.045090581 20.01509258 0.224373146 0.994908854 65 Czecia 0.011369405 132.4762014 0.264628608 0.993386019 62 Denmark 0.020214063 159.1129644 0.407710152 0.998316487 81 Finland 0.021794171 94.2677651 0.367909632 0.999739483 77 France 0.016971981 149.3836653 0.381809048 0.998825287 77 Hungary 0.018408277 73.06371651 0.252787089 0.995603327 64 Iceland 0.063514378 54.17233745 0.434359993 0.99793347 84 Ireland 0.018075924 150.5716412 0.389356595 0.998955126 78 Italy 0.018014458 151.3257418 0.377070492 0.999624445 77 Malta 0.022652701 95.41681316 0.39760535 0.999183087 82 Mexico 0.010377526 188.0011619 0.32312548 0.998082526 56 Peru 0.012379217 200 0.349426928 0.995727225 58 Portugal 0.025237491 162.0226288 0.461236454 0.99488844 83 South Korea 0.021931008 161.1951162 0.437882706 0.991209179 79 Spain 0.028466217 107.8294504 0.417967064 0.998390744 82 Turkey 0.012053373 140.355595 0.368700663 0.993605162 68 UK 0.015992445 55.39616283 0.303422349 0.995669144 72 USA 0.013378891 91.25265859 0.249549082 0.99807531 65 A deterministic model for inventory policy This section elaborates on a model for minimizing the inventory holding cost of vaccines. It is based on the results of the previous section. It was shown that the vaccination process of the countries can be approximated within a short time from the beginning of the vaccination with high accuracy. The model is based on a sequence of assumptions as follows:It is assumed that the vaccination process is deterministic and the total percentage of the vaccinated population has an a priori course in time. This course is described by a saturation function. The number of procuring of the vaccine is given. Each procuring is made by issuing an order. The lead time is 0. i.e. the ordered quantity arrives immediately. The first order is issued at the beginning of the process when still nobody is vaccinated. Any further order is issued when the stock level becomes 0. The process is finished when a certain percentage is achieved. This vaccination level can be different in different countries as the willingness for vaccination is depending on the country. The inventory holding process is always ahead of the vaccination, i.e. there is no negative stock level. The total vaccinated percentage of the population must be compared to the total arrived vaccine quantity. The latter one is an increasing step function. The first one is a saturation function obtained by regression of the previous section. The two can be seen in Fig. 4.Fig. 4 The solution for Denmark is if the number of procurements is 3 The total inventory holding cost is proportional to the area between the two functions as this is the quantity in the stock. It is what should be minimized primarily. However, the area under the saturation function is fixed. See a more detailed mathematical explanation below. Thus, the area between the two functions depends only on the step function. Consequently, it is enough to minimize the area under the step function. To formalize the mathematical model, some notations are used as follows:f(t) The saturation function, i.e. the function of the total demand n The number of orders ti The time of order i(i=1,2,⋯,n+1) T The time when the target percentage of the vaccination is achieved The time unit that can be used in real practice is a day. A dummy order is added to the set of orders. It is when the process is finished at the target percentage, i.e. a day T. Its ordered quantity is 0. The stocking process consists of a first-order at time 0, and there are n-1 further reorders. A reorder is issued exactly at the moment when the stock becomes 0. The ordered quantity arrives immediately as the lead time is 0. In other words, the ordered quantity is enough exactly until the next reorder point. Thus, the total ordered quantity until the time ti is f(ti+1), as it must be exactly enough until ti+1. Hence, the ordered quantity at ti is f(ti+1)-f(ti). The latter quantity is the demand in the time interval ti,ti+1. The inventory holding cost in the time interval ti,ti+1 is proportional to the area between the step function and the saturation function. This area isti+1-tifti+1-∫titi+1ftdt. Thus, the total inventory holding cost is proportional to∑i=1nti+1-tifti+1-∫titi+1ftdt=∑i=1nti+1-tifti+1-∑i=1n∫titi+1ftdt=∑i=1nti+1-tifti+1-∫0Tftdt. The last term is independent of the selection of the reorder points. Thus, it is enough to minimize the last but one term. Hence, the optimization problem determining the best inventory policy is as follows:2 min∑i=1nti+1-tifti+1 3 t1=0 4 ti+1≥tii=1,…,n 5 tn+1=T. The results obtained from the model are in Table 3. Also, population information of the countries is provided in same table (Worldometer, no date).Table 3 Direct results obtained from the model Countries The number of procurements Target percentage Population 3 4 5 Belgium 118,98 113,38 109,91 70 11,589,623 Canada 119,33 114,21 111,02 70 37,742,154 Cyprus 24,43 22,46 21,31 60 1,207,359 Czechia 119,12 111,61 107,10 60 10,708,981 Denmark 158,97 150,83 145,81 80 5,792,202 Finland 81,15 76,68 73,95 70 5,540,720 France 121,81 115,92 112,28 70 65,273,511 Hungary 70,34 65,59 62,76 60 9,660,351 Iceland 50,05 47,76 46,33 80 341,243 Ireland 120,41 114,86 111,42 70 4,937,786 Italy 125,03 119,15 115,51 70 60,461,826 Malta 105,01 98,13 93,99 80 441,543 Mexico 99,22 94,67 91,86 50 128,932,753 Peru 101,73 97,61 95,04 50 32,971,854 Portugal 137,94 132,63 129,28 80 10,196,709 South Korea 114,42 110,18 107,51 70 51,269,185 Spain 105,09 99,73 96,42 80 46,754,778 Evaluation of the numerical results The optimality condition of problem (2)–(5) can be obtained by the Karush–Kuhn–Tucker Theorem (May 2020) of optimization theory. However, the same condition can be obtained in a simpler way as well. It is provided in the Appendix. On the other hand, the problem has a very easy structure. Excel solver got the optimal solutions in all cases within 1 min. Thus, the optimality condition was not used directly in the research. The optimal value given in Table 2 has the unitproportionofthepopulation×vaccine×day. The result can be transferred vaccine×days if it is multiplied by the population. Still, it is necessary to subtract the area under the approximation curve again multiplied by the population. To obtain the latter quantity, the function given by formula (1) must be integrated between 0 and the day when the target percentage of the vaccination is reached according to the approximation function. Notice, that the stocking policy is determined far before the vaccination process is finished. Thus, the only approximation can be used in the calculations. The requested formula of∫0Tarctanat-bπ+cdt is based on the well-known equation as follows:∫arctantdt=tarctant-ln1+t22. Hence6 ∫0Tarctanat-bπ+cdt=t-barctanat-bπ-ln(1+at-b)22πa+ct0T=T-barctanaT-bπ-ln(1+aT-b)22πa+cT+barctan-abπ+ln1+a2b22πa. If the value of formula (6) is subtracted from the optimal value given in Table 2 and the result is multiplied by the population, then the inventory holding cost is obtained in vaccine×days. These results are in Table 4. It is converted to money if this quantity is multiplied by the inventory holding unit cost, i.e. the cost of storing one unit of vaccine for one day.Table 4 Numerical results for 20 countries Counties a b c Final Day (T) Formula (F) 3 4 5 Belgium 0.019440264 142.3695511 0.372352365 228 64.29026361 90.19075904 83.57677798 79.64537806 Canada 0.023610645 141.5326286 0.367132913 215 53.24148951 78.03480276 71.64457211 67.86551724 Cyprus 0.045090581 20.01509258 0.224373146 74 27.49832022 33.94322493 32.43122393 31.4920501 Czecia 0.011369405 132.4762014 0.264628608 287 83.11237951 112.3800927 105.1931561 100.8328068 Denmark 0.020214063 159.1129644 0.407710152 300 115.0464986 149.7767118 141.273454 136.0812595 Finland 0.021794171 94.2677651 0.367909632 173 58.29396609 77.20242481 72.47309479 69.63043015 France 0.016971981 149.3836653 0.381809048 241 71.59856407 98.5424194 91.66398699 87.57603406 Hungary 0.018408277 73.06371651 0.252787089 178 55.34248674 72.99695498 68.71574096 66.10102004 Iceland 0.063514378 54.17233745 0.434359993 90 31.90457574 42.43743208 39.78585016 38.19445836 Ireland 0.018075924 150.5716412 0.389356595 233 66.45330793 92.56298544 85.84476148 81.86999212 Italy 0.018014458 151.3257418 0.377070492 241 68.64570778 95.90475771 88.93389157 84.79419076 Malta 0.022652701 95.41681316 0.39760535 235 110.4281795 135.8404753 129.8105321 126.0704568 Mexico 0.010377526 188.0011619 0.32312548 258 49.26915341 66.27082275 60.45730972 57.06268812 Peru 0.012379217 200 0.349426928 242 37.56293105 59.13462631 53.34917208 50.00984351 Portugal 0.025237491 162.0226288 0.461236454 234 72.65974997 99.74732113 92.69207673 88.54613668 South Korea 0.021931008 161.1951162 0.437882706 211 52.2574544 74.99464779 68.90816708 65.39065074 Spain 0.028466217 107.8294504 0.417967064 199 76.64587247 99.72076437 94.03076002 90.57292356 Turkey 0.012053373 140.355595 0.368700663 215 60.35264529 80.66341144 75.43952095 72.35184566 UK 0.015992445 55.39616283 0.303422349 241 117.0078174 137.442241 132.7286793 129.7746195 USA 0.013378891 91.25265859 0.249549082 239 77.73800687 100.7775641 95.22356232 91.8213829 Counties 3-F 4-F 5-F Population (P) 3-FP 4-FP 5-FP Belgium 25.90049544 19.28651438 15.35511445 11,589,623 300,176,977.6 223,523,430.6 177,959,987.6 Canada 24.79331325 18.4030826 14.62402773 37,742,154 935,753,046.7 694,571,977.5 551,942,306.8 Cyprus 6.444904705 4.932903707 3.993729877 1,207,359 7,781,313.699 5,955,785.687 4,821,865.711 Czecia 29.26771315 22.08077662 17.72042726 10,708,981 313,427,384 236,462,617.3 189,767,718.8 Denmark 34.73021323 26.22695542 21.03476097 5,792,202 201,164,410.6 151,911,823.6 121,837,584.6 Finland 18.90845872 14.1791287 11.33646406 5,540,720 104,766,475.4 78,562,581.95 62,812,173.15 France 26.94385533 20.06542292 15.97746999 65,273,511 1,758,720,037 1,309,740,604 1,042,905,563 Hungary 17.65446823 13.37325421 10.7585333 9,660,351 170,548,359.9 129,190,329.7 103,931,207.9 Iceland 10.53285634 7.881274419 6.289882625 341,243 3,594,263.497 2,689,429.727 2,146,378.417 Ireland 26.10967751 19.39145355 15.41668419 4,937,786 128,924,000.1 95,750,847.85 76,124,287.36 Italy 27.25904993 20.28818379 16.14848298 60,461,826 1,648,131,934 1,226,660,638 976,366,768.1 Malta 25.41229585 19.38235261 15.64227735 441,543 11,220,621.35 8,558,142.119 6,906,738.067 Mexico 17.00166933 11.1881563 7.793534703 128,932,753 2,192,072,033 1,442,519,793 1,004,841,885 Peru 21.57169526 15.78624103 12.44691246 32,971,854 711,258,786.8 520,501,634.3 410,397,780.4 Portugal 27.08757117 20.03232676 15.88638671 10,196,709 276,204,080.7 204,263,806.6 161,988,862.3 South Korea 22.73719339 16.65071268 13.13319634 51,269,185 1,165,717,374 853,668,468.9 673,328,272.9 Spain 23.0748919 17.38488756 13.92705109 46,754,778 1,078,861,448 812,826,558.4 651,156,182.1 Turkey 20.31076614 15.08687566 11.99920037 84,339,067 1,712,991,067 1,272,413,017 1,012,001,364 UK 20.43442357 15.72086188 12.76680214 67,886,011 1,387,211,503 1,067,226,602 866,687,270.5 USA 23.03955727 17.48555545 14.08337603 331,002,651 7,626,154,535 5,787,765,209 4,661,634,801 The final inventory cost can be obtained if the order costs are added. In the case of the order cost, only the transportation cost can be a non-negotiable amount of money. The decision-maker can solve the model easily for different numbers of procurements. The best solution can be selected after taking into account the order cost. Conclusions The countries bear the costs of vaccination in case of a pandemic. The paper intends to reduce public expenses. The main contributions of the paper are as follows:The increase in the proportion of the vaccinated population is analyzed in the COVID-19 pandemic. It is shown on selected 20 countries that the countries have individual behavior. However, the cumulative distribution function of the Cauchy distribution is a suitable tool to model the increase of the aforementioned proportion in the case of each country. A simple model is elaborated to minimize the inventory holding cost if the number of procurements is fixed. The model can be solved for a country within one minute by an excel solver. The numerical results are presented for the 20 countries. Three results are provided for each country such that the number of procurements is 3, 4, and 5, respectively. Appendix: The optimality condition of (2)–(5) The Karush–Kuhn–Tucker optimality conditions are first order conditions. The easy way to obtain the same conditions in this particular case is as follows: Assume that the times of orders i-1 and i+1 are fixed. What is the optimal time of order i? The area under the step functio in the interval [ti-1,ti+1] isgti=ti-ti-1fti+ti+1-tifti+1. Notice that the only unknown quantity in the formula is ti. The derivative of gti must be 0 at the optimal position of ti. Hence, g′ti=fti+ti-ti-1fti-fti+1=0. Thus, the optimality condition is that (*) holds for i=1,2,⋯,n. Acknowledgements The authors are grateful to János Tóth, Gergely Kovács, and Zoltán Lakner for their valuable comments. Declarations Conflict of interest We confirm that the manuscript is the authors’ original work and the manuscript has not received prior publication and is not under consideration for publication elsewhere. All authors have contributed to this paper, reviewed and approved the current form of the manuscript to be submitted. We confirm that all authors of the manuscript have no conflict of interest to declare. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Antic S Djordjevic Milutinovic L Lisec A Dynamic discrete inventory control model with deterministic and stochastic demand in pharmaceutical distribution Appl Sci 2022 12 1536 10.3390/app12031536 Bartsch SM Vaccine efficacy needed for a COVID-19 coronavirus vaccine to prevent or stop an epidemic as the sole intervention Am J Prev Med 2020 59 4 493 503 10.1016/j.amepre.2020.06.011 32778354 Dwyer MJ Physical activity: Benefits and challenges during the COVID-19 pandemic Scand J Med Sci Sports 2020 30 7 1291 1294 10.1111/sms.13710 32542719 Gabster BP Challenges for the female academic during the COVID-19 pandemic Lancet 2020 395 10242 1968 1970 10.1016/S0140-6736(20)31412-4 Güner R Hasanoğlu İ Aktaş F COVID-19: Prevention and control measures in community Turkish J Med Sci 2020 50 SI-1 571 577 10.3906/sag-2004-146 Kinlaw K Barrett DH Levine RJ Ethical guidelines in pandemic influenza: recommendations of the ethics subcommittee of the advisory committee of the director, Centers for Disease Control and Prevention Disaster Med Public Health Prep 2009 3 S2 S185 S192 10.1097/DMP.0b013e3181ac194f 19675459 Mamun MA Ullah I COVID-19 suicides in Pakistan, dying off not COVID-19 fear but poverty?—the forthcoming economic challenges for a developing country Brain Behav Immun 2020 87 163 166 10.1016/j.bbi.2020.05.028 32407859 May R (2020) A simple proof of the Karush-Kuhn-Tucker theorem with finite number of equality and inequality constraints. Available at: http://arxiv.org/abs/2007.12483 Mills MC Salisbury D The challenges of distributing COVID-19 vaccinations EClinicalMedicine 2021 31 100674 10.1016/j.eclinm.2020.100674 33319186 O’Neill B Sanni S Profit optimisation for deterministic inventory systems with linear cost Comput Ind Eng 2018 122 303 317 10.1016/j.cie.2018.05.032 Our World in Data (no date) Share of people who received at least one dose of COVID-19 vaccine. Available at: https://ourworldindata.org/grapher/share-people-vaccinated-covid. Accessed: 11 December 2021 Patriarca R EOQ inventory model for perishable products under uncertainty Prod Eng Res Devel 2020 14 5–6 601 612 10.1007/s11740-020-00986-5 Pradhan D A review of current interventions for COVID-19 prevention Arch Med Res 2020 51 5 363 374 10.1016/j.arcmed.2020.04.020 32409144 Shekhawat S Integration of deterministic and probabilistic inventory methods to optimize the balance between overstock and stockout IOP Conf Series Mater Sci Eng 2016 10.1088/1757-899X/722/1/012060 U.S. Food and Drug Administration (no date) FDA Approves First COVID-19 Vaccine\|FDA. Available at: https://www.fda.gov/news-events/press-announcements/fda-approves-first-covid-19-vaccine. Accessed: 12 December 2021) Wollina U Challenges of COVID-19 pandemic for dermatology Dermatol Therapy 2020 10.1111/dth.13430 World Health Organization (no date) WHO Coronavirus (COVID-19) Dashboard\|WHO Coronavirus (COVID-19) Dashboard With Vaccination Data. Available at: https://covid19.who.int/ (Accessed: 12 December 2021) Worldometer (no date) Population by Country (2022)—Worldometer. Available at: https://www.worldometers.info/world-population/population-by-country/. Accessed: 11 January 2022) Zhou C COVID-19: challenges to GIS with big data Geogr Sustain 2020 1 1 77 87 10.1016/j.geosus.2020.03.005	0	PMC9734365	NO-CC CODE	2022-12-14 23:28:28	no	Cent Eur J Oper Res. 2022 Dec 9;:1-13	utf-8	Cent Eur J Oper Res	2,022	10.1007/s10100-022-00831-3	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 36469286 1437 10.1007/s40615-022-01437-w Article Black Americans Receiving the COVID-19 Vaccine and Effective Strategies to Overcome Barriers: An Integrative Literature Review Roat Chad http://orcid.org/0000-0002-5282-3362 Webber-Ritchey Kashica J. http://orcid.org/0000-0003-4387-5221 Spurlark Roxanne S. [email protected] Lee Young-Me grid.254920.8 0000 0001 0707 2013 School of Nursing, DePaul University, Chicago, IL USA 5 12 2022 111 6 6 2022 20 10 2022 21 10 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. Background Black Americans have a greater likelihood of serious morbidity or mortality from contracting the coronavirus and represent the lowest percentage of vaccinated individuals by race. This integrative literature review aims to identify the major barriers to Black Americans receiving the COVID-19 vaccine and proposed solutions to improve vaccination rates among this population. Method Databases CINAHL and LitCovid from the National Library of Medicine were utilized to find the articles included in this review. Results A total of seven articles were identified indicating five barriers preventing Black Americans from being vaccinated against COVID-19 that included (1) mistrust of the medical establishment, (2) uncertainty in vaccine safety, (3) limited access to healthcare, (4) inequitable access to resources, and (5) lower health literacy. The studies also indicated five strategies to increase the desire of Black Americans to be vaccinated including (1) utilizing trusted community leaders, (2) acknowledgment of the history of discrimination and trauma, (3) building more representative clinical trial cohorts, (4) continual investment into community-based organizations, and (5) mobile vaccine clinics. Conclusion The medical establishment in the USA has significant work to do to gain the trust of Black Americans. Many of the strategies to increase vaccine uptake among Black Americans have yet to be implemented which limits the conclusions that can be drawn from them. A future study should examine the outcomes of these proposed solutions to see if they do indeed work as intended and increase vaccination rates among this population. Keywords COVID-19 Vaccine hesitancy Vaccination barriers Black Americans Medical mistrust ==== Body pmcBackground The coronavirus (COVID-19), first found in Wuhan, China, in late 2019, has spread rapidly and widely across the globe and led to significant mortality and morbidity worldwide [1, 2]. In early 2021, COVID-19 vaccines were developed, and studies report that vaccination is highly effective in reducing COVID-19 infections, hospitalizations, and even deaths [3, 4]. Despite this fact, Black Americans in the United States (US), who have experienced disproportionately higher rates of COVID-19-related illness and death since the pandemic began, have consistently been vaccinated at lower rates. Funk and Tyson reported data from different time periods ranging from before COVD-19 vaccine availability through the first 3 months of 2021, when vaccines first became available [5]. In a 2020 vaccination uptake study, 42% of Back Americans were planning to receive the COVID-19 vaccine if available [5]. Although this increased to 61% of Black Americans for those that received the vaccination and planning to receive the vaccination combined in 2021, Black Americans still possess the lowest rates for vaccination in comparison to all other racial/ethnic groups, Asian (91%), Hispanic (70%), and White (69%) [5]. As of July 6, 2022, these rates of vaccination have shown to be similar to the Centers for Disease Control and Prevention (CDC) [6] reports of adults who have received at least one dose of the COVID-19 vaccine, with 59% for Black adults when compared to 87% for Asian adults, 67% for Hispanic adults, and 65% for White adults [7]. As the COVID-19 is an ongoing pandemic and different variants like Omicron are emerging, booster doses of COVID-19 vaccine are now recommended for all adults to shore up protection against more transmissible variants. It is primarily in response to concerns about possible waning immunity, the transmission of breakthrough infections, and the emergence of new viral variants with increased transmissibility [8–10]. Current studies have reported that people who had booster shots showed a significant decline in protection against COVID-19 infections a few months after the initial vaccination [9, 11, 12]. Those studies reported the similar results that vaccines’ effectiveness against infections declined over time and increased the risk for COVID-19-related hospitalizations and severe illness. As a result of this, booster shots were recommended and emerged as the vital option to boost immunity and improve protection against COVID-19 [13]. However, the disparities existed in COVID-19 vaccination among Black Americans returned with additional recommendations for booster shots [8]. Many of the same issues that initially led to slower vaccination among Black adults exist, contributing to the consistently lagged booster rates, including high levels of vaccine hesitancy, concerns about safety, and deep-seated distrust of the medical system [8, 9]. Data from the CDC [14] shows that the rate ratio of Black Americans contracting COVID-19 is almost identical when compared to that of White Americans, but the death toll for Blacks is about double. Race and ethnicity are connected to other health determinants such as socioeconomic status and access to healthcare [14]. Egede and Walker [15] postulate that structural racism is a determinant of population health. They define structural racism as the methods a society uses to promote prejudice by way of mutually reinforcing inequitable systems [15]. Structural racism in the US molds the dispersal of social determinants of health and social risk factors, creating disparities in vaccine hesitancy due to a lack of trust in terms of adverse effects of the COVID-19 vaccine and medical system through historical and contemporary trauma [15]. A study reported that vaccine hesitancy was higher among African Americans than any other race/ethnic groups that contribute to lower vaccination uptake [16]. Research Problem Black Americans have a 2.8 times higher rate ratio than their White counterparts in needing to be hospitalized after contracting the COVID-19 [17]. Black Americans in the US work in the service industry or other blue-collar occupations in greater prevalence than Whites [18]. Many of these jobs are unable to be done remotely and became part of the essential work during the pandemic putting a myriad of Black Americans at an increased risk of contracting the virus. According to a survey conducted by the CDC, only 59% of Black adults in the US reported they had been vaccinated; this is much lower than the national average of 75% [6]. However, there is limited synthesis of published data that examine the barriers that this minority population faces to receiving the COVID-19 vaccination and successful strategies to overcome these hurdles to improve COVID-19 vaccine uptake. Research Purpose The purpose of this literature review was to identify barriers Black Americans face to receiving the COVID-19 vaccination, and to identify strategies that have been successfully implemented to increase the willingness of Black Americans to be vaccinated. Identifying these barriers and strategies enables the government to develop programming to target these specific obstacles to yield a higher percentage of COVID-19 vaccinations in the US. Research Question The following research questions are addressed in this literature review:What are the barriers that prevent Black Americans from receiving the COVID-19 vaccine? What strategies are currently being used to increase the desire of Black Americans to be vaccinated and what more can be done to continue to improve this percentage? Methods Research Design An integrative literature review design was employed to examine the barriers Black Americans face in receiving the COVID-19 vaccine and proposed solutions. Integrative literature reviews identify gaps in current publications while illuminating the need for further research [19]. An integrative literature review is appropriate for this study because there is no single answer to the research questions posed, eliminating the possibility of a systematic review. Of the studies identified for this review, none utilized methods similar enough to be reviewed by meta-analysis. Integrative literature reviews examine what methods have been utilized successfully in hopes of contributing to the body of knowledge and to nursing practice [19]. This integrative review followed Whittemore and Knafl’s [20] framework that includes five stages: problem identification (identification of the problem that the review will be addressing), literature search (comprehensive search of relevant literature using well-defined literature search strategies), data evaluation (evaluation of the quality of primary sources), data analysis (thorough interpretation of primary sources and synthesis of literature consisting of data reduction, data display, data comparison, conclusion drawing, and verification), and presentation (conclusion about results in table or diagrammatic form). Problem Identification The problem identified for this study is that Black Americans face barriers to getting vaccinated for COVID-19. After deciding upon this topic, a search of the existing literature was performed to accumulate information on the related concepts. All the primary sources were evaluated for overall quality and assessed to determine their relevance to the study. Tapering the sources found to only those pertinent to the study allows for a more thorough analysis of the data. The most important citations were interpreted and presented to draw conclusions across multiple studies [19]. Literature Search Strategies and Limitations We performed searches in the following databases—the Cumulative Index to Nursing and Allied Health Literature (CINAHL) and LitCovid from the National Library of Medicine—to identify research studies up to August 12, 2022. The database of CINAHL provides access to nursing literature, so this database was chosen because of researchers’ nursing background and the topic being highly relevant to nursing professionals. LitCovid (National Library of Medicine) was chosen as database because it covers up-to-date scientific information about COVID-19 and provides access to relevant articles in PubMed. The following key search terms were used to guide the searches: vaccine/vaccination, Black/African American, hesitant, barrier, overcoming, strategies/solutions, and COVID-19/corona. The following key term combinations were used to address the two research questions: “Black* OR African American* AND Covid* OR corona AND vaccin* AND barrier OR hesita* AND strategy OR solution.” The search across the two databases using Boolean combinations of the keywords “Black* OR African American* AND Covid* OR corona AND vaccin* AND barrier OR hesita* AND strategy OR solution” resulted in 243 research articles. The first author reviewed the 243 publication titles to determine their eligibility. Inclusion criteria consisted of articles relating to barriers that Black Americans face in getting the COVID-19 vaccine. Due to the recent nature of this disease, all studies included in the review were published within the last 2 years. Articles not available through the university’s library subscription and without open access were excluded from this review. Due to the topic being on Blacks or African Americans being the sole focus, any studies conducted in another country were excluded from this review along with duplicate studies found. Of the 243 publications, 56 articles did not meet the inclusion criteria. The titles and abstracts of the remaining 74 articles were reviewed. The articles selected included the keywords previously listed. The titles and abstracts of the sources were evaluated to sift out those with applicable information for this review and any that lacked pertinent information were excluded. Therefore, 14 articles were included in the final review. Figure 1 displays the process of article selection for this integrative literature review.Fig. 1 Analysis breakdown of research articles Data Evaluation When initial searches were conducted, a total of 238 articles pertaining to Black Americans and coronavirus vaccines were found. To reduce the number of articles to only those relevant to the study, inclusion and exclusion criteria were applied. The titles and abstracts of the articles were analyzed for relevance. This led to a selection of 14 articles to be included in this integrative literature review. Data Analysis and Synthesis The purpose of this integrative literature review was fulfilled through examination of these 14 articles. Data obtained was organized into a table that lists the purpose, results/variables of interest, limitations, barriers, and solutions. From the articles selected, data was compiled into Table 1 below. A total of two research questions were posed for this integrative literature review. Research question one examined the barriers Black Americans face to receiving a COVID-19 vaccine. Data from the selected articles relating to healthcare disparities and other hurdles African Americans face in recruitment for vaccine trials and receiving vaccines were studied. After reviewing the articles, overarching barriers that Black Americans faced were identified. Research question two examined what can be done and is currently happening to recruit a higher percentage of Black Americans to receive a vaccination against COVID-19. Through the review of the articles, overarching effective strategies to increase COVID-19 vaccination were identified.Table 1 Data collection table Source (author(s)/year) Purpose Results/variable of interest Limitations Barriers Solutions Ojikutu et al. (2021) Investigate ways in which trust can be built between the historically mistreated Black community and medical professionals Community investment is the preferred method of reaching minority communities. This is presented in contrast to simply community outreach Community outreach is temporary Deep-rooted mistrust of medical industry. Suboptimal enrollment of representative diversity in clinical trials. Pandemic amplified existing health inequalities True investment in communities of color. Community engagement projects that are ever evolving to address the issues/needs of the people Brandt et al. (2021) Looking into possible barriers and promoters of COVID-19 vaccination in youth populations Primary reason for rejecting vaccination is lack of trust. Disparities in healthcare need to be addressed to engage historically marginalized communities in vaccine dissemination Only 911 participants, diversity percentages of which did not match up with current population trends Concerns about vaccine side effects, safety, and effectiveness Black Americans have a general lack of trust in medical system Materials specifically designed to meet the concerns of unique populations should be created and distributed. Engagement of historically marginalized community members at every level of planning and dissemination of information Privor-Dumm & King (2020) Address vaccine hesitancy by engaging trusted community messengers—pastors Past negative experiences with healthcare and unequal treatment have led to mistrust in medicine. Acknowledging the history of mistreatment and providing evidence-based responses to inquiries can help address anxieties Even trusted community messengers like pastors are unlikely to change the minds of individuals who are extremely untrusting Distrust in the vaccine and medical institution due to history of mistreatment. Unequal access to care, health disparities. Feelings of being treated inequitably by providers Vaccine acceptance information needs to be catered specifically for the communities it is being implemented in. Pastors and other respected leaders in the community can work with the government to build trust and share information. Acknowledging past and present mistreatment Woko et al. (2020) [16] Investigate potential contributors to the disparity of Black Americans willing to be vaccinated against COVID-19 COVID-19 vaccine–related behavioral beliefs and trust in four COVID-19 information sources—mainstream media, social media, former President Trump, and public health officials and agencies This study only investigated two of the potential contributors to this disparity History of discrimination and unequal treatment by the medical establishment has bred mistrust. Black Americans are less likely to be insured. Unfavorable beliefs about vaccine Mass communication efforts utilizing trusted authorities to distribute accurate information. Future studies to understand the different forces driving resistance Abdul-Mutakabbir et al. (2021) Propose solutions to overcoming barriers to Black Americans receiving the COVID-19 vaccine A three-tiered approach consisting of engaging Black faith leaders, healthcare professionals delivering accurate educational information, and developing a mobile vaccination effort This study has just one example of the mobile vaccination effort, although a positive one History of distrust in, and inadequate access to, healthcare. Limited access to a computer, internet, and transportation Using faith leaders from the Black community to deliver accurate information about vaccine. Mobile vaccination effort to reach underserved communities. A faith summit about vaccines. Equitable allocation of vaccines Ferdinand, K. C. (2021, April) Discussing the need for cultural humility when educating Black Americans on vaccine information Overcoming the mistrust Black Americans hold against the medical institution No data available on the efficacy of the solutions posed Deeply rooted mistrust in healthcare. Social determinants of health (limited finances, healthy food, education, healthcare coverage, job flexibility) Public health messaging. Recognition of injustices suffered by Black Americans in the healthcare system. Use of trusted messengers to disseminate vaccine information Ferdinand, K. C., Nedunchezhian, S., & Reddy, T. K. (2020, December 6) Compare and contrast vaccine hesitancy between influenza and COVID-19 vaccines Barriers to Black Americans receiving the influenza vaccine is used to discuss the similar barriers to COVID-19 vaccination This study was published prior to the release of the vaccine to the public, so some things have changed since it was authored Years of bias and mistrust in orthodox medicine. Safety concerns and environmental barriers to vaccine access Educational campaigns and policy initiatives—creating and funding a COVID-19 vaccine risk communication and community engagement program. Governmental agencies need to work with trusted members of the community to develop programs to get accurate information out. Eliminating disparities to access the vaccine Guitierrez et al. (2022) To identify predictors for low likelihood of COVID-19 vaccination among women in the US and determine whether reasons for low intention were modified by race, ethnicity, or other characteristics to better understand the factors that shape attitudes toward the COVID-19 vaccine and help inform multilevel interventions Self-reported low vaccination likelihood and concerns influencing vaccination decision Data was collected in January 2021, so there is possibility that vaccine sentiments have changed Lack of trust and vaccine-related concerns There is a need to target messaging to specific populations, including pregnant and breastfeeding women Kicorian and Turner (2021) To explore COVID-19 vaccine hesitancy and examine factors that may help remedy such hesitancy among Black and Hispanic Americans COVID-19 vaccine hesitancy Cros-sectional design only captures a single point in time when COVID-19 vaccines became available Mistrust Black medical professionals discussing the vaccine Osake et al., (2022) To explore barriers and facilitators of COVID19 vaccine acceptance among Black and Hispanic New Yorkers Facilitators of vaccine acceptance Underrepresentation of younger age groups (ages 18–21 years) and limited generalizability due to study conducted in a New York metropolitan area Mistrust, lack of adequate information, misinformation about the vaccine The key drivers of vaccine acceptance are provision of reliable vaccine-related information, use of social networks, seeing people like themselves receive the vaccination, and trusted doctors Momplasir et al. (2021) To assess their attitudes, beliefs, and norms around a COVID-19 vaccine among Blacks Drivers of vaccine hesitancy Focus groups were held Focus groups took place between July and August 2020 which sentiments toward COVID-19 vaccine may have changed Mistrust in the medical establishment, concerns with the accelerated timeline for vaccine development, limited data on short- and long-term side effects, and the political environment promoting racial injustice Recommendation from a trusted healthcare provider Budhwani et al. (2021) To ascertain sentiments toward COVID-19 vaccination among rural African American or Black (AAB) adolescents Sentiments toward COVID-19 vaccine Findings not generalizable to all rural AAB adolescents Fear of side effects, misinformation, and institutional distrust A socioecological approach is needed to deliver vaccine-promoting messaging at multiple levels. Public health professionals and clinical providers should tailor messaging to rural AAB adolescents Stoler et al., (2021) To test the hypothesis that Black race would interact with medical trust to undermine COVID-19 vaccine willingness Vaccine hesitancy Findings may not be generalizable to Black population Medical mistrust Build medical trust by addressing structural racism Stoler et al. (2022) To assess a diverse set of correlates of COVID-19 vaccine hesitancy identified in previous studies using US survey data collected in July–August 2021 Vaccine hesitancy While findings highlight wide sources of vaccine hesitancy, they may not be generalizable to Black population Mistrust, belief in misinformation about the COVID vaccines, and political identity Leadership should develop policy focused on mitigating misinformation about the COVID-19 vaccines and further assessment of assessing other predictors of belief in COVID-19 misinformation and vaccine hesitancy Results All 14 studies revealed barriers that Black Americans face to receiving the COVID-19, vaccine hesitancy, and strategies used to increase vaccination. Studies identified four barriers to receiving the COVID-19 vaccine that included (1) mistrust of the medical establishment, (2) limited access to healthcare, (3) inequitable access to resources, and (4) lower health literacy. In addition, studies identified the uncertainty in vaccine safety and five strategies to increase COVID-19 vaccination in Black Americans. Barriers Mistrust in the Medical Establishment An overall mistrust in the medical establishment was identified as a major barrier in each of the articles chosen for this review [21–27]. This absence of trust from Black Americans comes from a long history of abuse and exploitation by medical entities. Many of the articles chosen point to the tragedy of the Tuskegee syphilis experiment, which has become the hallmark example of the mistreatment Black persons have experienced in healthcare [23, 24, 26, 27]. The water crisis in Flint, Michigan, as well as the controversial usage of Henrietta Lacks’ cancer cells have been cited as further examples of the establishment’s abuse of Black Americans [28]. With the COVID-19 vaccine in particular, political influences have tainted the public’s perception of the pandemic and vaccination [28]. Limited Access to Healthcare In general, Black Americans tend to have insufficient access to healthcare and lower rates of having health insurance [28]. Healthcare is not a guaranteed right in America and as a result millions of citizens are uninsured or underinsured. This leads to many not having a regular primary care doctor that they know and trust to ask their opinion [22]. When one is uninsured or underinsured, seeing a doctor is often viewed as a luxury that is avoided due to a fear of unexpected and unaffordable bills. Inequitable Access to Resources Black Americans are disproportionally affected by poverty and inequitable access to resources in America. Limited access to a computer, a connection to the internet, and reliable transportation are all financial-based barriers that Black Americans face [21]. The structural racism still prevalent throughout this country has led to significant shortages of resources for numerous communities of color [25]. Pharmacies are one of the locations that vaccinations are distributed, but if there is not a pharmacy in an economically disadvantaged neighborhood or town, it makes access more difficult. Lower Health Literacy Kutner et al. [17] did a study showing that 58% of Black Americans have a basic or below grasp on health literacy whereas in non-Hispanic Whites, this percentage was only 28%. Average health literacy was shown to increase with each higher level of education obtained starting with a high school diploma or general educational development (GED) and elevating from there [17]. It was also found that adults of all races living below the poverty line in America had lower health literacy than those above it [17]. For Black Americans, issues with health literacy can begin even prior to adulthood [29]. A study of Black American teenagers found that 65% had low health literacy [29]. A robust understanding of health literacy is essential to make informed decisions about one’s healthcare, including whether to get vaccinated. Vaccine Hesitancy Uncertainty in Vaccine Safety Black Americans revealed not trusting the safety and efficacy of the vaccine based on the lack of comprehensive diversity represented in the clinical trials [22, 24–27]. By October of 2020, only 3% of COVID-19 vaccine trial volunteers were composed of Black Americans [24]. During the third phase of vaccine trials, it was brought to light that participant demographics were not being transparently disclose when they were Black Americans and made up a smaller percentage of participants than their respective percentage of the country’s population [25]. Among three cross-sectional studies, lack of trust in the vaccine and access to comprehensive and correct information about the vaccine served as a barrier in Black Americans while Budhawi et al. [25] and Kricorian and Turner [30] identified concerns related to adverse events [10–16]. Four qualitative studies revealed mistrust as a barrier to receiving COVID-19 vaccination. Fear and lack of information emerged as primary themes influencing vaccine hesitancy in a qualitative study [31] while a cross-sectional study identified beliefs that the vaccine is dangerous, more harmful than getting COVID-19, causes COVID-19, and not worth the risk as contributors to COVID-19 vaccine hesitancy among Black study participants. Another qualitative study among Black-American adolescents identified fear of side effects and misinformation in relation to vaccine hesitancy. Stoler and colleagues [32] found misinformation about the COVID-19 vaccines positively correlated with vaccine hesitancy and structural racism as likely an attribute of vaccine hesitancy to further support their previous findings revealing higher levels of vaccine hesitancy in Black Americans [33]. Black Americans believe that clinical trials reward pharmaceutical companies who, in turn, do not reciprocate benefits back into the community [26]. There is also a mindset that not enough clinical trials have been done to prove the safety and efficacy of the vaccines [26]. Strategies to Increase COVID-19 Vaccination of Black Americans Although strategies and solutions were not the sole focus nor tested, 14 articles proposed some strategies and solutions to increase COVID-19 vaccination in Black Americans. The five strategies proposed to increase the desire of Black Americans to be vaccinated included (1) utilizing trusted community leaders, (2) acknowledgement of the history of discrimination and trauma, (3) building more representative clinical trial cohorts, (4) continual investment into community-based organizations, and (5) mobile vaccine clinics. Disseminating Information Through Trusted Community Leaders One strategy to address vaccine hesitancy in the Black American population is utilizing trusted community leaders to be messengers on behalf of healthcare and government entities [21–24, 26, 27]. These people are respected community members that can be faith leaders, Black medical professionals, or community organizers. Having a trusted messenger disseminating accurate data may be more well received than information from an etic or outsider’s perspective [21, 26]. Specifically, policy focused on mitigating misinformation is needed [32] to aid trusted community leaders in providing consistent, comprehensive messages [31]. Building trusted patient–provider relationships are necessary to educate and recommend the COVID-19 vaccine especially among patients experiencing vaccine hesitancy [34]. Policies and multi-modal interventions should promote community engagement to decrease vaccine hesitancy [31, 35]. Given that religion is a large facet of the Black-American experience, partnering with pastors and churches can be an excellent way of reaching an audience that may otherwise hesitate to trust government or healthcare institutions [26]. These trusted messengers should be known to have a history of kindness, compassion, and collaboration with minority communities [23] and utilize culturally relevant vaccine communications [30]. Acknowledgement of the History of Mistreatment and Trauma Black Americans deserve recognition of the long chronicle of discrimination and abuse by the medical establishment. By acknowledging past wrongs, medical institutions can begin mending fragmented relationships with communities and individuals of color. Doctors and other healthcare professionals have historically perpetuated racist ideologies that substantiated the subjugation of and prejudice against Black Americans [23].This wretched history cannot be forgotten and must be addressed for Black communities to be able to harbor trust in the healthcare system. In fact, difficult community dialogues are crucial to engaging Black Americans and addressing vaccine hesitancy [33]. More trustworthy healthcare systems that address vaccine hesitancy and improve access to the COVID-19 vaccine through the lens of equity based on the communities they serve are needed [36]. Making Clinical Trials More Inclusive This solution is predicated on the first two strategies being successful. If Black Americans cannot trust the medical community or the vaccine safety, then increasing the number of people of color in clinical trials will be difficult, if not impossible. In October of 2020, Black Americans made up only 3% of those willing to partake in a COVID-19 vaccine clinical trial [23]. Other factors influence this discrepancy as well such as jobs that do not offer paid time off and clinical trial reimbursement not being competitive versus compensation made by working. Vaccine manufacturers also need to build their trial cohorts with complete transparency so that any racial and ethnic discrepancies can be noted and addressed [23]. Investment in Communities This pandemic offers a chance for real contribution to communities of color beyond just fleeting outreach programs. True investment in these groups of marginalized individuals can help to diminish distrust and ultimately improve vaccination rates [23, 25]. Providing capital to communities can work to generate financial worth, while also providing support for societal needs and problems [25, 39]. The purpose of this funding is to nurture new relationships between Black Americans and institutions of power. The historical lack of investment into community organizations led by Black Americans is a result of structural racism [39]. Financing stakeholder organizations in the present can cause more collaborations on experimentation and intervention subsequently [25]. Establishing these relationships can help to build the trust necessary to change Black American’s perception of the COVID-19 vaccine [25]. Mobile Vaccine Clinics One proposed mechanism to counter the discrepancies in local availability of vaccine distribution sites is to offer mobile clinics. Only one of the reviewed articles makes mention of this intervention but shows promising data for its efficacy. In one study [21], the mobile vaccination clinic is just one part of a three-pronged approach to address barriers to vaccination. First, they partnered with Black congregation leaders, followed by disseminating knowledge provided by Black medical experts [21]. The final step was to institute a vaccine clinic in a church parking lot of a predominantly Black community [21]. Bringing vaccine distribution directly into Black communities demolishes barriers such as the need to register on the internet to schedule a vaccination appointment or to have to travel to a vaccine clinic [21]. Discussion The purpose of this integrative literature review was to identify barriers Black Americans face to receiving the COVID-19 vaccination, and to identify strategies that have been successfully implemented to increase the willingness of Black Americans to be vaccinated. Consistent with literature, this current review indicates the barriers to vaccination in Black communities. The long history of discrimination and unequal treatment Black Americans have faced makes it most evident that this population has deep-seated suspicion and hesitation regarding the healthcare system [21–27]. Every article selected for this review touches on the mistrust of the medical establishment as a barrier to Black Americans getting vaccinated against COVID-19. This is clearly one of the largest hurdles that must be overcome to increase willingness to be vaccinated in this community. Decades of inequality and injustices against Black Americans cannot be erased or forgotten overnight. To heal this community and build a bridge of trust between Black Americans and the medical establishment, future work and resources must be put into motion. Building trust is the paramount solution to existing barriers to vaccination, but it cannot be done without first addressing the major health disparities Black Americans face [24, 26]. Massive investments into communities of color are needed to increase the availability and affordability of healthcare [25, 39]. Medical care is a human right and the for-profit healthcare system in America is unethical and fosters the health disparities that run rampant here. Perhaps if substantial and sustained funding was being used toward building up Black American communities, it would prove easier to recruit representative numbers of people of color to participate in clinical trials [25]. Increasing diversity in the patients represented in future clinical trials can help to alleviate some of the concerns regarding safety [23, 24]. Recognizing that vaccine hesitancy is deeply rooted in mistrust is by far the most fundamental yet the most glaring point of reference in the beginning to understand the resolution to the problem itself. Mistrust in healthcare systems, government agencies, timelines for research development, social media content, and safety and efficacy of the vaccine itself continues to be among the top discussions for those who choose not to vaccinate. Healthcare providers will best serve the ability to build trust through the ability to appeal to the community through the trusted voices of healthcare professionals from the communities they have served, offering culturally competent education and the ability to test and vaccinate. Other opportunities include promoting digital platforms with information appealing to the Black-American population. The use of the digital platform provides social advocacy and empowerment for those aged 18–24 and with perceptions of low risk of contracting COVID-19. Other studies have also shown that the use of mobile devices to support the dissemination of information that is both timely and trusted, is highly beneficial. There is no single reason to explain the lesser percentage of Black Americans receiving vaccination for the COVID-19. However, vaccine hesitancy is less likely the sole result [40, 41]. Similarly, there is not just one solution that will eliminate the barriers being faced. Those looking to have a firm grasp on problems identified with vaccine should take into consideration that residency and political affiliation could also be factorial considerations. Conversations in consideration of increasing vaccination status should include participants from leaders in regions identified as having lower vaccination rates and their trust in the strategies ability to be safe. Engaging trusted community leaders to disseminate important information to the population has proven to be beneficial [21–24, 26, 27, 41]. Communities of color are more likely to respond positively to guidance when it is coming from an inherently trusted source [21–24, 26, 27]. The US medical establishment needs a multifaceted approach to combat health disparities and build trust in healthcare and COVID-19 vaccinations [21, 23, 24]. Future exploration of how to ameliorate the barriers to Black Americans receiving the COVID-19 vaccine is needed. Limitations This literature review is limited by the small number of research articles that were included. This is reflective of the relative lack of data and articles that had been written on this topic when the research was performed. Many of the articles chosen for this review propose solutions that are logically sound but have either not been implemented or have not been followed up with post-implementation. This limits the conclusions that can be drawn from the strategies suggested. Implications The COVID-19 pandemic has dramatically impacted the healthcare system in the US and the nurses on the frontlines during this pandemic. Data is clear that symptoms of COVID-19 are much more manageable and less frequently require hospitalization in those that are vaccinated. Therefore, increasing vaccination rates among all Americans should be a priority of healthcare professionals everywhere. Specifically, research supports the significant role that Black healthcare professionals can play in reaching the Black community through the provision of vaccine education [21]. More doses of the vaccine being distributed will equal less strain on an already crumbling medical establishment. It is important that nurses be armed with the latest data and best ways to communicate it to combat vaccine hesitancy and misinformation with their patients. Implications for Research More research may be necessary to isolate the specific concerns of unique communities to produce educational material that is catered to these populations. Research should also be performed to determine the best means of disseminating this information to these groups. Now that booster vaccine doses are being recommended and distributed, future research should expand to include barriers Black Americans face to receiving these vaccinations as well. Researchers must prioritize building trust [28]. Conclusion The COVID-19 will likely not be the last novel virus spreading around the globe, so it is important to learn from what went wrong during this pandemic to not let history repeat itself. The COVID-19 pandemic is not over yet, and the solutions studied in this review can still be relevant and helpful if they are instigated properly. The science is clear that vaccination is the best tool available to help mitigate the disasters caused by this disease. It is imperative that the US starts doing the long and hard work of building trust in healthcare in Black Americans. Previous wrongs cannot be undone, but the medical establishment should begin by acknowledging the atrocities of the past so that people of color have a reason to give their trust. Author Contribution All authors contributed to the conceptual development of the manuscript and needed edits. All authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. All authors agree to be accountable for all aspects of the work and their appropriateness in integrity and accuracy. 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 References 1. Phua J Weng L Ling L Egi M Lim CM Divatia JV Shrestha B Arabi Y Ng J Gomersall C Nishmura M M Koh Y Du B Intensive care management of coronavirus disease 2019 (COVID-19): challenges and recommendations Lancet Res Med 2020 8 5 506 517 10.1016/S2213-2600(20)30161-2 2. World Health Organization. World Health Organization. 2022. Available at: https://covid19.who.int/data/. Accessed 5 May 2022. 3. Pawlowski C, et al. FDA-authorized covid-19 vaccines are effective per real-world evidence synthesized across a multi-state health system. 2021. 10.1101/2021.02.15.21251623. 4. Dagan N Barda N Kepten E Miron O Perchik S Katz MA Hernán MA Lipsitch M Reis B Balicer RD BNT162b2_MRNA COVID-19 vaccine in a nationwide mass vaccination setting N Engl J Med 2021 384 1412 1423 10.1056/NEJMoa2101765 33626250 5. Funk C, Tyson A. Growing share of americans say they plan to get a COVID-19 vaccine – or already have, center for disease control and prevention. 2021. Available at: https://www.pewresearch.org/science/2021/03/05/growing-share-of-americans-say-they-plan-to-get-a-covid-19-vaccine-or-already-have/. 6. CDC Covid Data tracker. Centers for Disease Control and Prevention. Centers for Disease Control and Prevention. 2020. Available at: https://covid.cdc.gov/covid-data-tracker/. Accessed 5 May 2022. 7. KFF. Available at: https://www.kff.org/. Accessed 5 May 2022. 8. Division N. Joint statement from HHS Public Health and medical experts on COVID-19 booster shots, HHS.gov. 2022. Available at: https://www.hhs.gov/about/news/2021/08/18/joint-statement-hhs-public-health-and-medical-experts-covid-19-booster-shots.html. Accessed 15 Aug 2022. 9. Vegel L Duong D What is the evidence for COVID-19 booster shots? Can Med Assoc J 2021 193 35 E1400 E1401 10.1503/cmaj.1095959 34493570 10. Silva-Cayetano A, et al. A booster dose enhances immunogenicity of the COVID-19 vaccine candidate Chadox1 nCoV-19 in aged mice. Med. 2021;2(3). 10.1016/j.medj.2020.12.006. 11. Chenchula S Karunakaran P Sharma S Chavan M Current evidence on efficacy of COVID-19 booster dose vaccination against the Omicron variant: a systemic review J Med Virol 2022 94 7 2969 2976 10.1002/jmv.27697 35246846 12. Nanduri S, et al. Effectiveness of pfizer-biontech and moderna vaccines in preventing SARS-COV-2 infection among nursing home residents before and during widespread circulation of the SARS-COV-2 B.1.617.2 (delta) variant - national healthcare safety network, march 1–august 1, 2021, centers for disease control and prevention. Ctr Dis Control Prev. 2021. Available at: https://www.cdc.gov/mmwr/volumes/70/wr/mm7034e3.htm. Accessed 11 Aug 2022. 13. Frisco ML, Hook JV, Thomas KJA. Racial/ethnic and nativity disparities in U.S. covid-19 vaccination hesitancy during vaccine rollout and factors that explain them, Penn State. Elsevier Limited. 2022. Available at: https://pennstate.pure.elsevier.com/en/publications/racialethnic-and-nativity-disparities-in-us-covid-19-vaccination. Accessed 11 Aug 2022. 14. Risk for COVID-19 infection, hospitalization, and death by Race/Ethnicity. Centers for Disease Control and Prevention. 2021. Available at: https://www.cdc.gov/coronavirus/2019-ncov/covid-data/investigations-discovery/hospitalization-death-by-race-ethnicity.html. Accessed 20 May 2022. 15. Egede LE, Walker RJ. Structural Racism, Social Risk Factors, and Covid-19 — A Dangerous Convergence for Black Americans. New England Journal of Medicine. 2020;383(12):e77. Available at: 10.1056/nejmp2023616. 16. Khubchandani J Sarman S Price J Wiblishauser M Sharma M Wenn F COVID-19 Vaccination hesitancy in the United States: a rapid national assessment J Community Health 2021 46 270 277 10.1007/s10900-020-00958-x 33389421 17. Kutner M, Greenberg E, Paulsen C. The health literacy of america's adults: results from the 2003 national assessment of adult literacy. National Center for Education Statistics. 2006. Available at: https://nces.ed.gov/pubs2006/2006483.pdf. 18. Employment \| Healthy People. Office of Disease Prevention and Health Promotion. 2020. Available at: https://www.healthypeople.gov/2020/topics-objectives/topic/social-determinants-health/interventions-resources/employment#30. 19. Russell C An overview of the integrative research review Prog Transplant 2005 15 1 813 10.1177/152692480501500102 20. Whittemore R Knafl K The integrative review: updated methodology J Adv Nurs 2005 52 5 546 553 10.1111/j.1365-2648.2005.03621.x 16268861 21. Abdul-Mutakabbir J Casey S Jews V King A Simmons K Hogue M Belliard J Peverini R Veltman J A three-tiered approach to address barriers to COVID-19 vaccine delivery in the Black community Lancet Glob Health 2021 9 6 e749 e750 10.1016/S2214-109X(21)00099-1 33713634 22. Brandt E Rosenberg J Waselewski M Amaro X Wasag J Chang T National study of youth opinions on vaccination for COVID-19 in the U.S J Adol Health 2021 68 5 869 872 10.1016/j.jadohealth.2021.02.013 23. Ferdinand K Overcoming barriers to COVID-19 vaccination in African Americans: the need for cultural humility Am J Public Health 2021 111 4 586 588 10.2105/AJPH.2020.306135 33689446 24. Ferdinand K Nedunchezhian S Reddy T The COVID-19 and influenza “twindemic”: barriers to influenza vaccination and potential acceptance of SARS-CoV2 vaccination in African Americans J Natl Med Assoc 2020 112 6 681 687 33276969 25 Ojikutu B Stephenson K Mayer K Emmons K Building Trust in COVID-19 Vaccines and beyond through authentic community investment American J Public Health 2021 111 3 366 368.34 10.2105/AJPH.2020.306087 26. Privor-Dumm L King T Community-based strategies to engage pastors can help address vaccine hesitancy and health disparities in black communities J Health Commun 2020 25 10 827 830 10.1080/10810730.2021.1873463 33719889 27. Woko C Siegel L Hornik R An investigation of low COVID-19 vaccination intentions among Black Americans: the role of behavioral beliefs and trust in COVID-19 information sources J Health Commun 2020 25 10 819 826 10.1080/10810730.2020.1864521 33719874 28. Webber-Ritchey K, Lane-Cordova K. Health disparities and COVID-19 pandemic: increasing clinical research participation among african americans. Journal of Health Disparities and Practice. 2022;14(2):53–63. Available at: https://digitalscholarship.unlv.edu/jhdrp/vol14/iss2/5. 29. Manganello J Sojka C An exploratory study of health literacy and African American adolescents Compre Child Adol Nursing 2016 39 3 221 239 10.1080/24694193.2016.1196264 30. Kricorian K Turner K COVID-19 Vaccine acceptance and beliefs among Black and Hispanic Americans PLoS ONE 2021 16 8 e0256122 10.1371/journal.pone.0256122 34428216 31. Osakwe ZT, et al. Facilitators of COVID-19 vaccine acceptance among black and hispanic individuals in New York: a qualitative study. Am J Infect Control. 2022;50(3):268–272. 10.1016/j.ajic.2021.11.004. 32. Stoler J, et al. Sociopolitical and psychological correlates of COVID-19 vaccine hesitancy in the United States during summer 2021. Soc Sci Med. 2022;306:115112. 10.1016/j.socscimed.2022.115112. 33. Stoler J, et al. The limits of medical trust in mitigating COVID-19 vaccine hesitancy among black americans. J Gen Intern Med. 2021;36(11):3629–3631. Available at: 10.1007/s11606-021-06743-3. 34. Momplaisir F Haynes N Nkwihoreze H Nelson M Werner RM Jemmott J Understanding drivers of coronavirus disease 2019 vaccine hesitancy among blacks Clin Infect Dis 2021 73 10 1784 1789 10.1093/cid/ciab102 33560346 35. Gutierrez S, et al. Predictors of COVID-19 vaccination likelihood among reproductive-aged women in the United States. Public Health Reports. 2022;137(3):588–596. Available at: 10.1177/00333549221081123. 36. Balasuriya L Santilli A Morone J Ainooson J Roy B Njoku A Mendiola-Iparraguirre A O'Connor Duffany K Macklin B Higginbottom J Fernández-Ayala C Vicente G Venkatesh A COVID-19 vaccine acceptance and access among Black and Latinx communities JAMA Netw Open 2021 4 10 e2128575 10.1001/jamanetworkopen.2021.28575.PMID:34643719;PMCID:PMC8515205 34643719 37. Purnell M, et al. Exploring covid-19 vaccine hesitancy at a rural historically black college and university. J Am Pharm Assoc. 2022;62(1):340–344. Available at: 10.1016/j.japh.2021.09.008. 38. Budhwani H, et al. Covid-19 vaccine sentiments among african american or black adolescents in rural Alabama. J Adolesc Health. 2021;69(6):1041–1043. Available at: 10.1016/j.jadohealth.2021.09.010. 39 Kline N Quiroga M Organizing for Black lives and funding COVID-19 relief: community responses to systemic racism and Imagining Public Health 4.0 American J Public Health 2021 111 S3 S201 S203 10.2105/AJPH.2021.306408 40. Padamsee TJ Bond RM Dixon GN Hovick SR Na K Nisbet EC Wegener DT Garrett RK Changes in COVID-19 vaccine hesitancy among Black and White individuals in the US JAMA Netw Open 2022 5 1 e2144470 10.1001/jamanetworkopen.2021.44470 35061038 41. Sharma M Batra K Batra R A theory-based analysis of COVID-19 vaccine hesitancy among African Americans in the United States: a recent evidence Healthcare (Basel) 2021 9 10 1273 10.3390/healthcare9101273 34682953	36469286	PMC9734369	NO-CC CODE	2022-12-14 23:28:28	no	J Racial Ethn Health Disparities. 2022 Dec 5;:1-11	utf-8	J Racial Ethn Health Disparities	2,022	10.1007/s40615-022-01437-w	oa_other
==== Front Mach Vis Appl Mach Vis Appl Machine Vision and Applications 0932-8092 1432-1769 Springer Berlin Heidelberg Berlin/Heidelberg 1356 10.1007/s00138-022-01356-0 Original Paper Real-time pedestrian pose estimation, tracking and localization for social distancing http://orcid.org/0000-0002-1164-1976 Abdulrahman Bilal [email protected] 1Bilal Abdulrahman received his master’s in Data Science from the CUNY Graduate Center, in 2022. He is, currently pursuing a PhD in computer science at the CUNY Graduate Center. He is a member of the City College Visual Computing Laboratory since 2020. His research interests include computer vision, and machine learning, currently focusing on human interaction and human body reconstruction. http://orcid.org/0000-0002-9990-1137 Zhu Zhigang [email protected] 2Zhigang Zhu is current Herbert G. Kayser Chair Professor of Computer Science, at The City College of New York (CCNY) and The CUNY Graduate Center, where he directs the City College Visual Computing Laboratory (CcvcL). Dr. Zhu is an Associate Editor of the Machine Vision Applications Journal, Springer (2006 - now), and was Technical Editor, IEEE/ASME Transactions on Mechatronics (09/2010 - 09/2014). His research interests include 3D computer vision, multimodal sensing, human-computer interaction, virtual/augmented reality, and various applications in assistive technology, robotics, surveillance and transportation. He has published over 200 peer -reviewed technical papers in the related fields. His research has been supported by AFOSR, AFRL, ARO, DARPA, DHS, NSF, ODNI as well as industry. Dr. Zhu is a senior member of the IEEE, and a senior member of the ACM. 1 grid.253482.a 0000 0001 0170 7903 The Graduate Center, The City University of New York, New York, NY, 10016 USA 2 grid.212340.6 0000000122985718 The City College and The Graduate Center, The City University of New York, New York, NY, 10031 USA 5 12 2022 2023 34 1 87 7 2021 28 5 2022 3 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. The corona virus pandemic has introduced limitations which were previously not a cause for concern. Chief among them are wearing face masks in public and constraints on the physical distance between people as an effective measure to reduce the virus spread. Visual surveillance systems, which are common in urban environments and initially commissioned for security surveillance, can be re-purposed to help limit the spread of COVID-19 and prevent future pandemics. In this work, we propose a novel integration technique for real-time pose estimation and multiple human tracking in a pedestrian setting, primarily for social distancing, using CCTV camera footage. Our technique promises a sizeable increase in processing speed and improved detection in very low-resolution scenarios. Using existing surveillance systems, pedestrian pose estimation, tracking and localization for social distancing (PETL4SD) is proposed for measuring social distancing, which combines the output of multiple neural networks aided with fundamental 2D/3D vision techniques. We leverage state-of-the-art object and pose estimation algorithms, combining their strengths, for increase in speed and improvement in detections. These detections are then tracked using a bespoke version of the FASTMOT algorithm. Temporal and analogous estimation techniques are used to deal with occlusions when estimating posture. Projective geometry along with the aforementioned posture tracking is then used to localize the pedestrians. Inter-personal distances are calculated and locally inspected to detect possible violations of the social distancing rules. Furthermore, a “smart violations detector” is employed which estimates if people are together based on their current actions and eliminates false social distancing violations within groups. Finally, distances are intuitively visualized with the right perspective. All implementation is in real time and is performed on Python. Experimental results are provided to validate our proposed method quantitatively and qualitatively on public domain datasets using only a single CCTV camera feed as input. Our results show our technique to outperform the baseline in speed and accuracy in low-resolution scenarios. The code of this work will be made publicly available on GitHub at https://github.com/bilalze/PETL4SD. Keywords Multiple object tracking Human pose estimation Pedestrian localization http://dx.doi.org/10.13039/100000001 national science foundation 1827505 Zhu Zhigang http://dx.doi.org/10.13039/100011038 office of the director of national intelligence HHM402-19-1- 0003 HHM402-18-1-0007 Zhu Zhigang issue-copyright-statement© Springer-Verlag GmbH Germany, part of Springer Nature 2023 ==== Body pmcIntroduction Fig. 1 Workflow of the PETL4SD system pipeline: pedestrian pose estimation, tracking and localization for social distancing It has been over two years since the initial cases of the COVID-19 virus surfaced [33]. The world is now still amidst a global pandemic. Even though multiple vaccines have been introduced around the world and are being swiftly implemented, new variants which are deadlier and spread faster keep appearing [11]. The pandemic has introduced limitations which were previously not a cause for concern. Chief among them are wearing face masks in public and social distancing measures. In public health, social distancing is a set of non-pharmaceutical interventions or measures intended to prevent the spread of a contagious disease by maintaining a physical distance between people and reducing the number of times people come into close contact [17]. The CDC (Center for Disease Control and Prevention) in the USA recommends 6 feet as a safe distance between individuals. Visual surveillance systems, which are common in urban environments, aim at providing safety in everyday life. High-quality surveillance cameras are already present in most urban streets and departmental stores. Although they were initially commissioned for security surveillance, they can be re-purposed to help limit the spread of COVID-19 and prevent future pandemics (Fig. 1). In this work we propose a novel and systematic approach to social distancing measurement in a pedestrian setting, called PETL4SD, which stands for Pose Estimation, Tracking and Localization for Social Distancing. This technique functions as it states in its name. Pose estimates are tracked, which are then used for localization, and thus, inter-personal distances are calculated and social distancing violations are estimated. That being said, this is a high-level overview and there are many nuances that add to the novelty of our technique. These will be explained in the coming sections. Furthermore, although we employ PETL4SD for social distancing estimation in this work, its modular nature allows it to be generalized for a bevy of applications, such as detecting and tracking human behavior, action estimation, analyzing pattern of movement (for example in clothing shops), training pose estimator with labels, etc. During initialization of the PETL4SD pipeline, we compute homography or a projective transformation between the ground plane in the real world and ground plane as it is represented in the image plane of the camera. Homography is required to be computed only once and can be done before hand. We will use the estimated homograph along with the point of intersection of pedestrians with the ground plane (i.e., their feet) to localize them. After initial setup, our technique utilizes a pipeline of tasks that are performed on each frame of the video feed. These tasks are as follows: (1) pedestrian detection; (2) pedestrian pose estimation; (3) pedestrian feature extraction; (4) pedestrian track/tracks creation/update; (5) pedestrian localization; (6) social distancing calculation; and (7) visualization of the tracks and social distance violations. In the first step, a neural network object detector, in this case YOLOv5 [19], is used for the detection of pedestrians. In the second step, the detected bounding boxes of pedestrians are resized and then fed to the OpenPifPaf network [20] for posture estimation. The detections from the object detector are also used to extract features in step three using the OSNET REID network [37] for use in tracking and re-identification. All models used in this paper are pretrained models; for training parameters and training methodology please refer to the respective publications. Tracking process begins in the fourth step, and the tracking method used is similar to FASTMOT [36] with modifications made for this implementation. This technique uses a custom version of the DeepSORT algorithm [34] complimented with optical flow estimation. In the fifth step, we update the tracks or create new ones by matching the detected pedestrian bounding boxes, the optical flow points and the REID template with the previous frame. Then we employ our temporal and analogous techniques for pose completion to estimate feet of pedestrians if not detected due to occlusion. Homography is then used to project the feet coordinates from the image onto real-world coordinates and, thus, localize the pedestrians. In the sixth step, the ground points history of the pedestrians is used to estimate whether there has been a social distancing violation. Finally, the seventh step is when the distances between pedestrians are visualized with the right perspective and occlusion relations using a simple Augmented Reality (AR) composition of the real scene background, the social distance marks on the ground and the images of the detected pedestrians, as if the marks are actually on the ground. Along with other improvements, the main contribution of our work is: We exploit the standard surveillance setting, where people appear up-right and this heuristic allows us to implement a bottom-up approach to pose estimation as essentially a top-down approach. Using prepossessing tricks we are able to get a sizeable increase in speed and improvement in pose estimation, making the system function in real time on less capable hardware. We utilize posture tracking history along with localized mean distributions of human body dimensions to estimate the position of feet even if they are not detected due to occlusions. Even if the pose detector is unable to detect any posture, the posture and bounding box history are used to get a reasonable estimate of the pedestrians feet position and thus improve localization accuracy. We created a “smart violations detector” which uses the track and posture history to infer the pedestrians current action and uses this knowledge to group people as “together.” This is vital for eliminating false social distance violations between pedestrians in a group. Furthermore, this approach only requires 4 points of correspondence on the ground visible in the video feed for estimating the planar homography, which can easily be sourced from online maps (if the view is outdoors) or floor plans (if the view is indoors). If correspondence cannot be established, a camera model with intrinsic and extrinsic parameters will also suffice. Therefore, after the initial setup, this approach will work requiring only the video feed. All tests and implementations are done on Python. All models are run using the PyTorch library [24], and most of the computer vision tasks are performed using the OpenCV library [8]. The paper is organized as follows: Sect. 2 discusses some related work. Section 3 describes the mapping from 2D images to the ground plane for a foundation of social distancing measures in the metric space. Section 4 discusses our choice of deep learning model for real-time human bounding box detection. Section 5 presents our integrated solution for human pose estimation with the YOLO object detector and the bottom-up pose estimator—OpenPifPaf. Section 6 describes the combination of DeepSORT with optical flow estimation for human pose tracking. Section 7 presents our temporal and analogous techniques for pedestrian localization and distance calculation. Sections 8 and 9 describe two final steps—social distancing measurement and visualization. Some implementation details and experimental results (including a video demo) are presented in Sect. 10. Finally Sect. 11 concludes the work with discussions of some future directions. Related work Due to the importance of social distancing measures, a number of works have appeared since the beginning of the pandemic. These works tackle problems such as human detection, human tracking, human localization and combination of such techniques for social distancing monitoring. One such work [28] focuses on detecting small and close by humans from CCTV footage for the purpose of social distancing monitoring. They propose a SSD architecture for this task. Although presenting novel techniques, the work does not highlight its performance when compared to existing techniques and no posture estimation is performed for more accurate tracking and localization estimation is trivial. Another work [2] utilizes transfer learning on YOLOv3 pretrained models to better detect human bounding boxes and then uses tracking to keep track of the violations. This work again lacks localization accuracy and only bounding boxes are tracked leading to vague estimations of the humans actual feet position. Another technique [3] proposes a wearable smart tag to detect and alert for social distancing violations. This is helpful for health authorities to implant on a known COVID patient. However, it can be cause for a privacy concern and large-scale adoption does not seem feasible. A social distancing monitoring approach [27] that utilizes YOLOv3 and DeepSORT on CCTV footage was proposed to detect and track pedestrians followed by calculating a violation index for non-social distancing behaviors. The way this approach estimates the position of pedestrians is unclear as there is no implementation discussion on this part and the work mostly focuses on the DeepSORT algorithm [34]. Several prototypes utilizing machine learning and AI along with cameras and sensors have been developed as commercial solutions for social distancing monitoring and enforcing. Landing AI [21] has proposed a social distancing detector using a surveillance camera to highlight social distancing violations. The work in [26] details a systems that was deployed in a manufacturing plant to monitor worker activity. This type of implementation requires knowledge of a lot more camera parameters and complex transformations compared to our work. In addition to surveillance cameras, LiDAR-based [16] and stereo camera-based [31] systems are also commercially deployed. These approaches require special hardware made for the specific applications. A recent work [35] proposes a non-intrusive warning system with softer omnidirectional audio-visual cues. It evaluates critical social density and modulates inflow into a region-of-interest. Another work [12] defines social distancing monitoring as a visual social distancing problem. This work introduces a skeleton for a detection-based approach to inter-personal distance measuring. It also discusses the effect of social context on people’s social distancing and raises privacy concerns. As a continuation of this work, [1] uses skeletons detection for estimating the homography by assuming parameters and estimates scales using average body part dimensions. This work does not address occlusions or any sort of tracking. It also does not address false social violations caused by people who are together such as family or friends. Another technique [7] localizes human posture estimations using only the intrinsic camera parameters and a machine learning model. This work seems to be useful when the camera is moving and not stationary, i.e., the extrinsic parameters cannot be determined. Planar homography estimation Homography is a special case of projective geometry. It enables the mapping of points in spaces with different dimensionality. A point x′ observed in a view (such as an image) can be mapped into its corresponding point x in another perspective or coordinate system (such as a ground plane) (Fig. 2).Fig. 2 A visual representation of homography and the corresponding points We can utilize this property to localize the pedestrians in view and calculate their locations on the ground with respect to each other. For this purpose, we need to find the homography matrix between ground plane in real-world coordinates and ground plane as it is represented in the image plane. This can be achieved by finding at least four corresponding points between the two planes and using them to compute the matrix elements. The only restriction is that the four points must be in a “general position,” which means that no three points are collinear. One thing to note here is the computation of the homography matrix H in this way does not require knowledge of any of the camera’s parameters or the pose of the plane [18]. That being said, if correspondence cannot be established, but the intrinsic and extrinsic cameras parameters are known, the corresponding ground points can be computed by using a camera model [25]. There are also techniques available for automatic homography estimation using scene geometry [1] or machine learning [7]. These methods do not require any correspondence or extrinsic parameters. However, their reliability when it comes to accuracy is debatable. Therefore, it is recommended these techniques can be employed only as a last resort. Pedestrian detection with YOLOv5 The primary objective in this step is to detect bounding boxes of pedestrians in the image frame. We chose YOLOv5 [19] for this task. Yolo is a fast and accurate neural network for object detection [29]. The Yolo architecture splits the input image into grids and for each grid generates bounding boxes and class probabilities for those bounding boxes. This type of model has several advantages over traditional classifier-based systems. It looks at the whole image at test time, so its predictions are informed by global context in the image. It also makes predictions with a single-network evaluation unlike systems like R-CNN which require thousands of passes for a single image. This makes it extremely fast, more than 1000x faster than R-CNN and 100x faster than Fast R-CNN [29]. Yolo has had many improvements done to it over the years as newer versions come out. The YOLOv5 repository is a natural extension of the YOLOv3 PyTorch repository by Glenn Jocher. YOLOv5 is implemented natively in PyTorch, whereas all prior models in the YOLO family leverage Darknet. The principal motivation behind the choice of YOLOv5 over v4 or v3 is speed. Since this implementation requires real-time inference, YOLOv5 has an advantage over its competition. YOLOv5 also provides us with flexibility and ease of use. We can pick and choose between speed and accuracy by leveraging its different network models (Fig. 3). “YOLOv5s” is the best when it comes to speed of inference, whereas ’YOLOV5m’ is the optimal middle ground between speed and accuracy. Each frame is forwarded to the Yolo neural net as input and it outputs predictions with confidence levels as to what objects are present in the frame. All predictions below a threshold confidence level τ (0.5 in our case) are rejected. The locations of these predictions are returned as a center point (cx′,cy′) and height (hb) and width (wb) of an enclosing bounding box. Finally, the detections go through a process of non-maximal suppression to eliminate duplicates; Fig. 4 shows the final results of a frame.For our purposes the pre-trained models provided by the authors are adequate and perform well without any extra training.Fig. 3 Graph showcasing performance of different YOLOv5 models taken from Yolov5 GitHub repository Fig. 4 Bounding boxes drawn on an original frame from the result of the YOLOv5 neural net Fig. 5 Two examples of how our system is able to detect very low-resolution skeletons where solo OpenPifPaf fails. In both (a) and (b) results with solo OpenPifPaf are shown on the left and results with our system on the right Human pose estimation using OpenPifPaf with detection For the estimation of poses we use a OpenPifPaf model pretrained on the COCO dataset. OpenPifPaf is a bottom-up method for multiperson 2D human pose estimation [20]. OpenPifPaf’s architecture is based on a fully convolutional singleshot design. The architecture has two head networks. For each body part or joint, one head network predicts the confidence score, the precise location and the size of this joint, namely part intensity field (PIF). The other head network called the part association field (PAF) predicts associations between parts to form full human poses. OpenPifPaf’s strength lies in its exceptional performance at low resolution and in crowded, and occluded scenes thanks to composite field PAF encoding fine-grained information and Laplace loss for regressions. OpenPifPaf specifically addresses challenges of limited resolution and high-density crowds where pedestrians occlude each other. Although OpenPifPaf is meant as a bottom-up approach to pose estimation (estimating joints without a person detector), we do not use it standalone. We combine it with the YOLOv5 object detector to effectively implement it as a top-down method (i.e., using a person detector first and estimating joints within the detected bounding boxes).The main motivation behind this change is the methodology of training the OpenPifPaf model: using square image crops to create uniform batches and its scale-dependent adaptive regression loss [20]. Essentially, providing the model with images similar to training environment should, in theory, improve performance. This is because OpenPifPaf—despite its superior performance on low resolution, crowded, and cluttered scenes and scenes with occlusion—struggles when the resolution of the pedestrian is very poor (far away from the camera), which goes below 40 pixels height in our experiments. Fortunately, YOLOv5 has far better low-resolution performance and is adequately able to identify these pedestrians (Fig. 5). This combination thus results in greater speed and greater estimation of poses without further training in a pedestrian setting. Through our testing, we discovered that resizing the bounding boxes of the pedestrians into an aspect ratio of 1:1 or 1:2 improved detection of joints by the OpenPifPaf network on these very low resolutions. Table 1 Comparison between OpenPifPaf solo and our method on a system with a GTX 1050tiM and core i7-8750H on the PETS 2009 s2, L1 View 001 [13] Methods Skeletons Bounding box fps Total (ground truth) 5020 5020 – OpenPifPaf solo 4498 4498 4.6 Our method (with YOLOv5s) 4504 4556 22.3 Our method (with YOLOv5m) 4563 4679 15.6 The number of skeletons estimated and average FPS is recorded and displayed The bold specifies better detection of skeletons and bounding boxes from frames by our method when compared with OpenPifPaf. While the last column specifies best FPS These discoveries led to the following implementation: Firstly, the pedestrian bounding boxes are detected using an object detector (YOLOv5). Second, as we are assuming a pedestrian setting, most people are assumed to have an upright posture, i.e., standing, walking or running (although other poses perform adequately in our testing as well). With this heuristic in mind, all detected bounding boxes are cropped out from the image and then resized to a pixel size of 50×100 as shown in Fig. 6. We then perform batch inference, using the OpenPifPaf network, on all these cropped detections as all are the same resolution. As a by-product of this batch inference the speed is dramatically increased, which is another major goal of this work. Thus, not only does this method improve detection of very low-resolution pedestrians, it also speeds up inference more than 5 times when measuring fps (frame per second). To demonstrate this, we run OpenPifPaf solo, and combinations of Yolov5s and OpenPifPaf, and Yolov5m and OpenPifPaf, on the PETS dataset [13]. The pretrained backbone used for all testing and implementation is shufflenet. The number of skeletons detected and average FPS is recorded on the PETS 2009 s2,L1 View 001 and is displayed in Table 1. To evaluate the performance on this dataset, the ground truth is calculated by the manually counting the number of individuals visible in each frame. Our method results in more detections and up to 5x higher fps on lower end graphics hardware. Examples from this test can also be seen in Fig. 5 where the pose of women in the background is not detected when only OpenPifPaf is used. They are, however, detected when the combination of the two is used. This can also be used for training the stand-alone OpenPifPaf network, further improving its accuracy as manually labeling poses can be very expensive. As the skeletons were counted manually frame by frame and there are no ground truth annotations, thus, accuracy of the labels cannot be judged by this test. We further evaluate accuracy in validation listed below.Table 2 Validation results on 100 pedestrian setting images from the CrowdPose dataset [22], after revising the validation set to include missed labels Method Average precision (AP) Average recall (AR) FPS IoU=0.50:0.95 IoU=0.75 IoU=0.50 IoU=0.50:0.95 IoU=0.75 IoU=0.50 OpenPifPaf (solo) 0.421 0.451 0.625 0.573 0.613 0.722 2.85 Our method (with YOLOv5x) 0.451 0.470 0.652 0.644 0.676 0.797 2.87 Our method (with YOLOv5l) 0.441 0.467 0.633 0.638 0.677 0.789 3.45 Our method (with YOLOv5m) 0.458 0.491 0.667 0.640 0.680 0.792 4 Our method (with YOLOv5s) 0.455 0.493 0.650 0.623 0.669 0.762 5 DEKR 0.515 0.585 0.652 0.593 0.644 0.680 1.66 Crops are resized to 100×200 The bold specifies the best performance in the column Fig. 6 All detected bounding boxes cropped out (top) and then resized to 50×100 (bottom) Fig. 7 Two examples (shown in (a) and (b)) of how our system is able to detect very low-resolution skeletons on the CrowdPose dataset [22]. In (b), an overexposed crop of the pedestrian in the dark tunnel is also shown, which is very hard to see in the regular image Fig. 8 A comparison from the CrowdPose dataset showcasing superior performance of our method on far away pedestrians: a shows output from our method (YOLOv5 bounding box and OpenPifPaf pose estimation); b shows output from DEKR [14] Our system is also validated on 100 images picked randomly from the CrowdPose dataset [22] with the only limitation being the image should be somewhat similar to a pedestrian setting, i.e., people are standing, walking or running. For consistency with OpenPifPaf, evaluation is done using the COCO API. As CrowdPose annotations use different keypoints for the head and neck when compared to COCO, to keep the testing consistent, these are not used when validating. The results are shown in Table 2. As ground truth labels in the original dataset do not include labels for pedestrians that are farther away from the camera, these annotations are added by us to the ground truth using COCO annotator [9]. These updated annotations will be released along side the code of this work. For completeness, we also compare with a SOTA technique DEKR [14]. We use DEKR with HRNet-W32 backbone pretrained on the COCO dataset by its authors. The average precision of our technique is higher when compared with OpenPifPaf solo, whereas average recall is higher across the board. The higher recall results from our model being able to estimate poses for pedestrians very far away from the camera. These are not even predicted by DEKR despite slower inference speed. Figure 8 shows an example. Average precision performance with a 50 percent overlap is superior for our model. Lower performance in the 75 to 95 percent overlap precision could be attributed to resize losses. Some examples are shown in Fig. 7. An unwanted side effect of our method is when bounding boxes overlap there can be duplicate pose estimations. However, this can easily be remedied, with minimal hit to performance, by comparing the mean of Euclidean distances between corresponding joints of the estimations in these overlapping bounding boxes. If Ani is a set joint coordinates estimated to be corresponding to a single pedestrian, such that:1 Ani={(xi,1,yi,1),(xi,2,yi,2),...(xi,17,yi,17)} then two annotation sets Anj and Ank from overlapping bounding boxes can be compared as follows:2 X=1n∑i=1n(xj,i-xk,i)2+(yj,i-yk,i)2 If X is less than a threshold τ (=5 in our case) , the duplicate with fewer keypoints is discarded. We also implement non-maximal suppression from [10] as an alternative to remove duplicates. Users have the option to choose between them. The resize resolution of 50×100 can also be increased, maintaining aspect ratio, if higher accuracy is preferred over speed. Tracking using DeepSORT with optical flow For tracking we use a system similar to FASTMOT [36], but with modifications made to the object detector, feature extractor, tracking system and how the optical flow is Incorporated. Also changes were made to make the system compatible with Pytorch library. This system has MOTA scores close to state-of-the-art trackers (Table 3) while still being able to run in real time. Multi-Object Tracking Accuracy (MOTA) is the standard metric used to evaluate multi-object tracker systems [6]. In this section we will focus on how the tracking system works, as at the time of writing there is no paper published to support the work in [36] and also go in to the detail of the modifications we made to improve the system for our implementation. The system core function in [36] is similar to the DeepSORT algorithm in [34]. Optical flow is added in order to improve tracking accuracy, camera motion detection and increase fps performance by skipping running the detector on intermediary frames using optical flow prediction to fill in the gaps. We replace the object detector with YOLOv5 which results in a drastic speed improvement. We also abandon the camera motion detection as CCTV camera is assumed to be stationary, and therefore we avoid this unnecessary overhead.Fig. 9 Tracking using our system described in Sect. 6: a paths drawn corresponding to the tracks. b A top-down view As stated earlier the tracking system functions similarly to the DeepSORT algorithm. The bounding boxes from the object detector are used to initiate tracks and extract features. One key difference from DeepSORT is that the feature extractor neural net is replaced by OSNET REID [37]. Dubbed the omni-scale network (OSNET) by its authors, its architecture consists of a residual block composed of multiple convolutional streams, each detecting features at a certain scale. This type of design allows it to capture features of both homogeneous and heterogeneous scales. A unified aggregation gate is employed to then dynamically fuse multi-scale features with input-dependent channel-wise weights. By using point-wise and depth-wise convolutions, the model is able to efficiently learn spatial correlations and avoid over-fitting. In addition, OSNET is extremely lightweight. This allows inference in real time, and despite its small model size, it achieves state-of-the-art performance outperforming most large-sized models. In our implementation the model generates a 512 feature set from a bounding box which is bespoke to each person and can be used for tracking along with re-identification after a track is lost or goes out of frame.Table 3 FASTMOT on MOT20 train set MOTA IDF1 HOTA MOTP MT MLs 66.8% 56.4% 45.0% 79.3% 912 274 Taken from [36] With each successive re-identification of a track, the feature set is updated fractionally by the newly extracted feature set by a factor θ. If fold are the old features and fext are the newly extracted feature from the current frame, then the new feature set for the track fnew would be:3 fnew=fold∗(1-θ)+fext∗θ Now we discuss how optical flow is incorporated into the algorithm. At initialization, image frame is converted into gray scale. Then corners inside each bounding box are detected using OpenCV’s implementation of the Shi-Tomasi corner detection method [30] to find the strongest corner points. These points are then passed through an ellipsoid filter. This filters out all keypoints not inside an ellipsoid drawn with the bounding box center as its center, and height and width of the bounding box as its major and minor axes. This is done to remove points at the edges of the bounding box. The remaining points are to be used as keypoints for optical-flow calculation. This step is only performed when none or too few previously predicted keypoints are present in the current bounding box. Otherwise the predicted keypoints are recycled. After keypoint detection, their motion in the current frame is predicted using OpenCV’s implementation of the Lucas-Kanade Optical Flow [23], trying to find their matches in the current frame. All points without matches or those that exceed max error threshold (100 in our case) are dropped. A partial affine transformation matrix between the previous and current matched points is computed. This matrix is used to transform the current bounding box to the optical flow prediction of its position in the next frame. This prediction is used to update the Kalman filter in the DeepSORT algorithm. Thus the Kalman filter receives updates twice per frame, one from the optical flow prediction (giving higher uncertainty to the optical flow prediction) and the other when the track is updated with the final match. The bounding-box matching is done identically to the DeepSORT algorithm. First, the Mahalanobis distance between the Kalman filter predictions and detected bounding boxes and Euclidean distance between the feature embeddings are calculated and then fused together into a single cost using a weighting metric. The distances are also checked to see whether they are within the gating region for both metrics. A linear assignment algorithm is used to match detected bounding boxes to the tracks. Then those unmatched are matched using IOU cost, i.e., how much the bounding boxes overlap and if it is over a certain threshold (over 40 percent overlap in our case), giving priority to those that have previously been matched. Finally the tracks left unmatched are analyzed for re-identification with previously lost tracks, whether the extracted feature embeddings distance is below a certain threshold (0.6 in our case). This works as people exiting and then reappearing in the frame or coming out of occlusion will almost always not match with any other track. An example of our tracking system can be seen in Fig. 9. The visualization technique used in this figure will be described in Sect. 9. Pedestrian localization and distance calculation As previously discussed in Sect. 3, we need to find the piercing point of the pedestrians with the ground plane in order to localize them on the ground using homography. The skeletons of each pedestrian detected inside their respective bounding boxes are tracked using our tracking method described in Sect. 6. This allows us to not only infer the piercing point of each pedestrian, on the image, using their feet but also if the feet are occluded, we can infer the piercing point using the dimensions and location of other estimated body parts or using previous skeleton estimations of the track (Fig. 10). Note that if only the bounding boxes were detected, the system would be unable to deal with occlusions, or rather it would be unaware if an occlusion has even occurred. Since it would have no information as to how much of the body is captured inside the bounding box. This is the main advantage of our method over other more trivial implementations [12, 27].Fig. 10 Left shows the previous frame where the complete leg is detected. Middle shows the current frame where the lower leg is not detected by OpenPifPaf. Right shows the lower leg estimation in the current frame using our pedestrian localization technique Fig. 11 In the case that only torso is detected (for the first person on the right of the image who is occluded by the person behind), both legs are reconstructed from previous skeletons of the track In the optimal scenario where there are no occlusions, the piercing point is estimated as the center point between the two feet of a pedestrian. As all body parts detected by the pose estimator are labelled as such, if both or either one of the feet is occluded, but the thigh is visible and detected, we assume the rest of the leg to be a natural extension and thus extract the dimension of the lower leg from the previous skeleton estimations of the track and add it to the end of the thigh using the thigh’s slope as the slope for the lower leg. If there is no previous skeleton, we assume symmetry and extract the dimension of the other lower leg and append it to the end of the thigh, thus completing the skeleton. We want to note here that the machine learning model OpenPifPaf’s output has labels on which body part is detected. Therefore, even if the upper body is not detected, the thigh (i.e., the upper part of the leg) can be identified by the network. If m is the slope of the thigh and (xt,yt) are the coordinates for the knee(thigh endpoint) and d is the extracted dimension of the lower leg, then we can estimate the coordinate for the corresponding foot F as follows:4 F=xt+d.11+m2,yt+d.m.11+m2 If the complete leg is occluded, i.e., even the thigh is not detected, we look for the lower end of the torso or corresponding side of the torso as shown in Fig. 11. If it is detected, we can still infer the leg positions in a similar fashion. Even if none of the above-mentioned resources are available, we can still use estimations based on the average human proportion distributions in the camera feed to infer the piercing point. As the distribution is localized to each camera, the effect of the camera angle skewing proportions is minimized. Thus a piercing point can be established to a reasonable accuracy even if only the head of the pedestrian is free of occlusions. If the skeleton cannot be estimated from the image but the bounding box is detected, we estimate the piercing point as the mid-point of the bottom side of the bounding box as a fallback. We also look at the track history to estimate average size of pedestrian to prevent abrupt changes to bounding box size and improve piercing point accuracy. Finally, the piercing point can be transformed into real-world coordinates (px,py) using the homography matrix H calculated in Sect. 3. The knowledge of the coordinates of all pedestrians in the image with respect to the ground plane in the real world reduces the problem of calculating distance between them to Euclidean distance between two points in 2D plane. Regardless of the coordinate system selected on the ground plane, we can find the distance between any two pedestrians by calculating the Euclidean distance [32] between their points of intersections on the ground plane as5 dij=(xi-xj)2+(yi-yj)2 where dij is the distance between any two pedestrians having piercing points (xi, yi) and (xj,yj) on the ground. Smart violations detector When calculating whether there has been a social distancing violation, we also want to take into account detecting whether two or more pedestrians are together. For example, family members walking together would not be considered a social distancing violation. For this reason, we developed a “smart” social distancing measure technique for estimating if pedestrians are together. Walking or standing. We first detect and segregate pedestrians into two categories: walking or standing. If the pedestrians are deemed walking, then the Fréchet distance between their tracks for a predetermined amount of time is computed. The Fréchet distance is used to measure similarity between curves taking into account the location and ordering of the points along the curves [4]. The ordering part is useful as we measure max distance between pedestrians relative to time; what position they both were at that specific time or on that specific frame. If it is less than a certain threshold, then these two pedestrians are deemed together. We use 1.5 meters as the threshold. Using this technique we are able to detect even if multiple people are together and assign them to the same group (family or friends) (Fig. 12). These groups are assumed together only within the group and any violation with a pedestrian or group outside the current group is detected as such. For weeding out crowds, any group of people over 6 is automatically removed of its assignment as a group and considered violating social distancing. This number can be programmatically altered if desired.The system is limited in the fact that it lacks semantic understanding. When people are walking together due to crowded areas or narrow walkways, in such cases only a semantic understanding of the scene can help evaluate if two people are together. As no dataset is currently available for this task, training a model is not possible. Some fail-safes are put in place to somewhat remedy this including the crowd limit stated above or the tweaking the together-distance threshold. Fig. 12 People walking and having Fréchet distance of their paths less than a threshold are considered together and have blue circles around them Fig. 13 People standing and facing each other are considered together and have blue circles around them Fig. 14 The vector between the legs of each person is obtained (v1, v2) and the relation between the these vectors is used to figure out if they are standing in front or behind each other Fig. 15 People standing in front of each other but not facing each other. The angle between v1 and v2 is close to 90 degrees People are estimated as standing if they do not move for a predetermined amount of time. When standing two people are considered to be together only if they are facing each other. As it makes sense for family or friends standing together to interact with each other. Facing each other. Detecting if people are facing each other without 3D skeletons can be a daunting task. Hence, we use heuristics to determine whether two people are facing each other. First, a vector v1 is drawn from the left to the right foot coordinates localized on the ground plane, of one of the pedestrians. The slope of v1 is used to draw a rectangle R1 where the v1 is the line passing through the center of the rectangle as shown in Fig. 14). The width of R1 is equal to the width of v1 and length is set to 3m (i.e., 1.5m each side). We then check if one of the feet of the other pedestrian lies inside this rectangle or the vector v2 between the feet (left to right) of the other pedestrian intersects this rectangle and angle θ1 between v1 and v2 is not close to 90 degrees. The reason for checking the angle is to see if the person is standing facing toward or away and not at a 90 degree angle as shown in Fig. 15. If this check is validated, we can move to the next step. Now we have determined that the other person is standing in front or behind of the first pedestrian. When pedestrians are facing each other, they can be viewed in frame either as one pedestrian facing the camera and the other facing away or both can be seen from a side view. An example can be seen in Fig. 13. In the first case, we check for both or either eyes detected for one person and only shoulders and/or ear detected for the other to validate if they are facing each other. To demonstrate this in Algorithm 1, we use Booleans el1, er1, el2 and er2 which correspond to the left and right eyes of the first pedestrian and left and right eyes of the second pedestrian, respectively. These Booleans are true when the corresponding eyes are visible in frame. We also employ two other Booleans ch1 and ch2 to check if either of the shoulders or ears are visible, one Boolean for each pedestrian, respectively, to tackle occlusions. If both eyes for both pedestrians are detected or vice versa that means they are facing the same direction and thus not facing each other.Fig. 16 Limited detection if people are together when complete occlusion occurs Fig. 17 People seen talking with each other and holding hands are detected as together by the smart violations detector For the second scenario, we measure the distance between the eyes/ear/nose of one pedestrian to the eye/ear/nose of the other pedestrian. If looking at each other the distance from the eyes of one pedestrian to the eyes of the other pedestrian should be smaller than to the ears of the other pedestrian and vice versa if facing opposite or the same direction. A final test for robustness is put in which is optional and can be omitted: we measure the angle θ2 of the vector between the ears and eyes for both pedestrians. If both are facing the same direction, it will be less than 90 degrees. We do not actually need to know which scenario the pedestrians are standing in as the checks are sequenced to work for both scenarios. The standing methodology is limited in cases of severe occlusions for example, when one of the pedestrians is completely occluded by the other in the frame as shown in Fig. 16 . Evaluation of togetherness. The social distancing together groups are reevaluated every second. For evaluation of the system on a real-world scenario we employ Oxford Town Centre dataset [5]. This dataset is a CCTV video of pedestrians in a busy downtown area in Oxford. We calculate homography without camera parameters by employing Google Earth [15] and using bench locations visible in the camera and on Google Earth to collect corresponding points. We were able to obtain under 30 cm of error when localizing pedestrians on the walkway. As no ground truth is available for people being together we only look at people holding hands or interacting with each other (mostly talking to each other) to establish ground truth, Fig. 17. The smart violations detector is able to categorize these pedestrians together 81% of the time (17 out of 21), as shown in Table 4. Only 2 times the detector categorized people together which were not picked as convincingly together in the ground truth. The smart violations detector misses cases where the object detector fails to consistently detect a pedestrian due to constant occlusions or the pedestrians are not in frame long enough for the necessary amount of localization data required. We also loop the dataset and have the system running for 2 hours on our machine to test robustness in a streaming scenario. The footage is resized from 1920× 1080 to 1280× 720, and a frame skip is employed processing 5 out of the 25 frames a second in order to maintain real-time operation on our machine. A two second buffer is provided. The results are shown in Table 5. The system is able to maintain consistent frame rate, and detections remain consistent with frames on subsequent loops.Table 4 Total together detections from the Oxford Town Centre dataset [5] that are interacting with each other Ground truth (GT) Total detected In GT Not in GT 21 19 17 2 Table 5 CPU, GPU,RAM usage and fps reported on the Oxford Town Centre dataset [5] Component Average Max CPU utilization 44% 72% GPU utilization 18.48% (747MB) 18.48% (747MB) RAM utilization 2GB 2GB FPS 5 7 Pedestrians visible in frame 15 24 The specifications of the machine are GPU:GTX 1050tiM and CPU:core i7-8750H Social distancing visualization The distances need to be represented in an intuitive way that can be easily interpreted and is not overwhelming to the viewer. Initially, the raw representation was adding green distance lines between pedestrians if the distance was less than 5 meters (16.4 feet) and displaying the distance on top of the lines. These lines would turn red if the distance went below 2 meters or roughly 6 feet (Fig. 18a). With blue lines representing pedestrians that were together, this representation becomes overwhelming and confusing if there are many pedestrians in view (Fig. 18b).Fig. 18 Social distancing initial representation with bounding boxes and lines: a Bounding boxes for pedestrians and lines between them in green ( greater than 6 feet) and red ( less than 6 feet). b Dense crowds may overwhelm the viewer in this kind of view For an image-based visualization, the best representation we could come up with was green circles of a 3-foot radius around each pedestrian. When any two circles intersect it can be interpreted as the two pedestrians are closer than 6 feet and the circles turn red alerting the viewer. Similarly blue circles to represent people that were together. Now the challenge we faced implementing this was drawing circles in real time that would correctly reflect the distance in the real world on the image. As drawing circles of fixed size would not show the distance accurately. Also, to make it work in real time transforming each point of a circle before drawing it on the image was not computationally viable. Therefore, what we ended up creating was a separate blank image which represented a scaled top-down view of the real-world ground plane. This blank image would act as a buffer between the ground plane and image. The reasons for scaling the blank image were twofold: so that all of the ground plane coordinates visible in the camera feed were represented as positive pixel values on the blank image and the blank image size would not be excessive. The circles and the moving average of the tracks (to smooth them out) were then drawn on this blank image (Fig. 19) and transformed using homography and overlaid on to the original image using the following relations. A point Pi on the image of the ground plane could be represented as:6 Pi=Hip.Pp where Pp is the point on the ground plane in real-world coordinates and Hip is the projective transformation or homography between the two planes. The same point on the blank image Pb can be represented as7 Pb=Hbp.Pp where Pb is the point on the blank image and Hbp is the projective transformation between real-world ground plane and blank image (Fig. 20). Now Eq. (7) can be written as:8 Pp=Hbp-1.Pb Putting value of Pp in Eq. (6) from Eq. (8) we get:9 Pi=Hip.Hbp-1.Pb Thus, by using this relation each point in the blank image is mapped onto the camera frame. And since the blank image is just a scaled representation of the real-world coordinates, this resulted in circles which were correctly scaled (Fig. 21a). The blank image in Fig. 21b is also used in the result to represent the top-down view of the ground plane. Implementation details and results This work is implemented on the PETS 2009 s2, L1 View 001. We use a calibrated camera with calibration parameters already provided in the data set to find corresponding points for the purpose of computing homography. That being said, it is not necessary for the camera to be calibrated for this work to be implemented [18]. To compute planar homography only 4 corresponding points between 2 planes are required as already explained in Sect. 3. The application is run on a Intel core i7-8750H CPU and a Nvidia GTX 1050tiM GPU and 16GB RAM. The CCTV video feed has 7 fps; our system runs at 9fps on this hardware. This results in smooth real-time operation. A video demo of the results of our work can be seen on YouTube at the following link: https://youtu.be/OUjvAlYy_vs. The videos show a top-down view and a regular view with overlay of the system’s results as shown in Fig. 21. The blank image of the top-down view has the satellite image of the scene added to it courtesy of Google Earth [15] for visual appeal (Fig. 21b, Table 6).Fig. 19 Scaled blank image top-down view Fig. 20 The homography between two images induced by a world plane (the concatenation of two homography matrices) Fig. 21 Social distancing final representation. a Social distancing enhanced visualization with circles around pedestrians. The image is a real-time composition of background, circles and pedestrian images, showing the right perspectives, sizes and occlusions. b Top down view with the social distancing circles added to the corresponding satellite image Table 6 Breakdown of the total time taken by each module when processing a frame Module Percentage of total time Object detector (yolov5) 14 % Pose estimation (OpenPifPaf) 55 % Tracking+optical flow (FastMOT) 28 % Social distancing (smart violations detector) <1 % Visualization 2 % Conclusion and discussion We have presented a novel system called PETL4SD for tracking posture skeletons in real time in a pedestrian setting using CCTV cameras. We then employ this technique to detect and visualize social distancing violations. The pedestrian detection was performed using YOLO neural net. The posture estimation was done using OpenPifPaf. We validated our pose estimation on the CrowdPose dataset and the PETS 2009 dataset. Our system demonstrated superior pose estimation performance with no additional training. Tracking was performed using a modified implementation of [36]. The localization, relative distance and realistic circular overlays are achieved using planar homography and image composition (background, circles, pedestrians). All this sophistication leads to better social distance tracking and monitoring when compared with more trivial systems, with no extra infrastructure and minimal requirements for initial setup. We also introduce a novel technique of detecting whether pedestrians are together and therefore negating false violations. We also evaluate this technique on a real-world CCTV dataset: Oxford Town Centre dataset. Great cares have been taken to optimize the time and detection performance of all the components; therefore, this work in its current form can be implemented for real-world use. Nevertheless, the performance and tracking can be further enhanced by adding multiple camera angles to deal with occlusions. Using panoptic segmentation of the ground plane and estimates of average human dimensions and average walking speed we can develop a system to estimate homography from a video feed, thus eliminating the need for manual correspondence. With ground positions known we can estimate the homography for multiple imaginary planes with regular intervals across the video feed using the human body dimension estimates and, therefore, with some heuristics, be able to recreate a 3d scene from the CCTV camera feed. We can also employ the posture tracking data to train an action recognition neural net for other applications, for example, to infer fall detection, hostile behavior, etc. Acknowledgements The work is supported in part by NSF via the Partnerships for Innovation Program (Award #1827505) and the CISE-MSI Program (Award #1737533), AFOSR Dynamic Data Driven Applications Systems (Award #FA9550-21-1-0082) and ODNI via the Intelligence Community Center for Academic Excellence (IC CAE) at Rutgers University (Awards #HHM402-19-1-0003 and #HHM402-18-1-0007). Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Aghaei, M., Bustreo, M., Wang, Y., Bailo, G., Morerio, P., Bue, A.D.: Single image human proxemics estimation for visual social distancing. arXiv:1905.00953 (2020) 2. Ahmed, I., Ahmad, M., Rodrigues, J.J., Jeon, G., Din, S.: A deep learning-based social distance monitoring framework for covid-19. Sustain. Cities Soc. 65, 102571 (2021). 10.1016/j.scs.2020.102571, https://www.sciencedirect.com/science/article/pii/S2210670720307897 3. Alhmiedat T Aborokbah M Social distance monitoring approach using wearable smart tags Electronics 2021 10.3390/electronics10192435 4. Alt H Godau M Computing the fréchet distance between two polygonal curves Int. J. Comput. Geom. Appl. 1995 05 01–02 75 91 10.1142/s0218195995000064 5. Benfold, B., Reid, I.: Stable multi-target tracking in real-time surveillance video. In: CVPR 2011, pp. 3457–3464 (2011). 10.1109/CVPR.2011.5995667 6. Bernardin, K., Elbs, A., Stiefelhagen, R.: Multiple object tracking performance metrics and evaluation in a smart room environment. In: Sixth IEEE International Workshop on Visual Surveillance, in conjunction with ECCV, vol. 90. Citeseer (2006) 7. Bertoni, L., Kreiss, S., Alahi, A.: Monoloco: Monocular 3d pedestrian localization and uncertainty estimation. CoRR (2019). arXiv:1906.06059 8. Bradski, G.: The OpenCV Library. Dr. Dobb’s J. Softw. Tools (2000) 9. Brooks, J.: COCO Annotator. https://github.com/jsbroks/coco-annotator/ (2019) 10. Cao, Z., Hidalgo, G., Simon, T., Wei, S., Sheikh, Y.: Openpose: Realtime multi-person 2d pose estimation using part affinity fields. CoRR (2018). arXiv:1812.08008 11. CDC: COVID-19 and Your Health (2020). https://www.cdc.gov/coronavirus/2019-ncov/transmission/variant.html 12. Cristani, M., Del Bue, A., Murino, V., Setti, F., Vinciarelli, A.: The visual social distancing problem. In: IEEE Access (2020). 10.1109/ACCESS.2020.3008370 13. Ferryman, J., Shahrokni, A.: Pets2009: Dataset and challenge. In: 2009 Twelfth IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, pp. 1–6 (2009). 10.1109/PETS-WINTER.2009.5399556 14. Geng, Z., Sun, K., Xiao, B., Zhang, Z., Wang, J.: Bottom-up human pose estimation via disentangled keypoint regression. CoRR (2021). arXiv:2104.02300 15. Google earth. https://earth.google.com/ 16. Hall, J.: Social distance monitoring (2020). https://levelfivesupplies.com/social-distance-monitoring/ 17. Harris, M., Ghebreyesus, T., Tu, R.M., Kerkhove, V., Maria, D., Imogen, C.: Corinne: Covid-19 - world health organization (2020). https://www.who.int/docs/default-source/coronaviruse/transcripts/who-audio-emergencies-coronavirus-press-conference-full-20mar2020.pdf?sfvrsn=1eafbff_0 18. Hartley R Zisserman A Multiple View Geometry in Computer Vision 2003 2 Cambridge Cambridge University Press 19. Jocher, G., Stoken, A., Borovec, J.: ultralytics/yolov5: v3.1 (2020). 10.5281/zenodo.4154370 20. Kreiss, S., Bertoni, L., Alahi, A.: Pifpaf: Composite fields for human pose estimation. CoRR (2019). arXiv:1903.06593 21. Landing AI: Landing AI creates an AI tool to help customers monitor social distancing in the workplace (2020). https://landing.ai/landing-ai-creates-an-ai-tool-to-help-customers-monitor-social-distancing-in-the-workplace 22. Li, J., Wang, C., Zhu, H., Mao, Y., Fang, H.S., Lu, C.: Crowdpose: Efficient crowded scenes pose estimation and a new benchmark (2018). 10.48550/ARXIV.1812.00324 23. Lucas, B.D., Kanade, T.: An iterative image registration technique with an application to stereo vision. In: Proceedings of the 7th International Joint Conference on Artificial Intelligence - Volume 2, IJCAI’81, p. 674-679. Morgan Kaufmann Publishers Inc., San Francisco, CA, USA (1981) 24. Paszke, A., Gross, S., Chintala, S., Chanan, G., Yang, E., DeVito, Z., Lin, Z., Desmaison, A., Antiga, L., Lerer, A.: Automatic differentiation in pytorch. In: NIPS 2017 Workshop on Autodiff (2017). https://openreview.net/forum?id=BJJsrmfCZ 25. Sinha, P.K.: Camera Calibration. SPIE. 10.1117/3.858360.ch8 26. Khandelwal, P., Khandelwal, A., Agarwal, S., Thomas, D., Xavier, N., Raghuraman, A.: Using computer vision to enhance safety of workforce in manufacturing in a post covid world. arXiv:2005.05287 (2020) 27. Punn, N.S., Sonbhadra, S.K., Agarwal, S.: Monitoring covid-19 social distancing with person detection and tracking via fine-tuned yolo v3 and deepsort techniques. arXiv:2005.01385 (2020) 28. Qin, J., Xu, N.: Reaserch and implementation of social distancing monitoring technology based on ssd. Procedia Comput. Sci. 183, 768–775 (2021). Proceedings of the 10th International Conference of Information and Communication Technology 10.1016/j.procs.2021.02.127https://www.sciencedirect.com/science/article/pii/S1877050921006037 29. Redmon, J., Divvala, S.K., Girshick, R.B., Farhadi, A.: You only look once: Unified, real-time object detection. CoRR (2015). arxiv:1506.02640 30. Shi, J., Tomasi: Good features to track. In: 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, pp. 593–600 (1994). 10.1109/CVPR.1994.323794 31. STEREOLABS: Using 3D cameras to monitor social distancing (2020). https://www.stereolabs.com/blog/using-3d-cameras-to-monitor-social-distancing/ 32. Szabo F The Linear Algebra Survival Guide: Illustrated with Mathematica 2015 Cambridge Academic Press Inc 33. WHO: Listings of WHO’s response to COVID-19 (2020). https://www.who.int/news/item/29-06-2020-covidtimeline 34. Wojke, N., Bewley, A., Paulus, D.: Simple online and realtime tracking with a deep association metric. In: 2017 IEEE International Conference on Image Processing (ICIP), pp. 3645–3649. IEEE (2017). 10.1109/ICIP.2017.8296962 35. Yang, D., Yurtsever, E., Renganathan, V., Redmill, K.A., Ozguner, U.: A vision-based social distancing and critical density detection system for covid-19. arXiv:2007.03578 (2020) 36. Yang, Y.: FastMOT: High-Performance Multiple Object Tracking Based on YOLO, Deep SORT, and Optical Flow (2020). 10.5281/zenodo.4294717 37. Zhou, K., Yang, Y., Cavallaro, A., Xiang, T.: Omni-scale feature learning for person re-identification. arXiv:1905.00953 (2019)	0	PMC9734371	NO-CC CODE	2022-12-14 23:28:28	no	Mach Vis Appl. 2023 Dec 5; 34(1):8	utf-8	Mach Vis Appl	2,022	10.1007/s00138-022-01356-0	oa_other
==== Front World J Pediatr World J Pediatr World Journal of Pediatrics 1708-8569 1867-0687 Springer Nature Singapore Singapore 36480134 662 10.1007/s12519-022-00662-x Systematic Review Pediatric endocrinopathies related to COVID-19: an update Memar Elmira Haji Esmaeli 1 Mohsenipour Reihaneh 2 Sadrosadat Seyedeh Taravat 34 http://orcid.org/0000-0001-9691-7558 Rostami Parastoo [email protected] 5 1 grid.411705.6 0000 0001 0166 0922 Pediatric Center of Excellence, Tehran University of Medical Sciences, Tehran, Iran 2 grid.411705.6 0000 0001 0166 0922 Growth and Development Research Center, Children’s Medical Center, Tehran University of Medical Sciences, Tehran, Iran 3 grid.510410.1 0000 0004 8010 4431 Universal Scientific Education and Research Network (USERN), Tehran, Iran 4 grid.411705.6 0000 0001 0166 0922 Children’s Medical Center, Pediatric Center of Excellence, Tehran University of Medical Sciences, Tehran, Iran 5 grid.411705.6 0000 0001 0166 0922 Growth and Development Research Center, Department of Endocrinology and Metabolism, Pediatric Center of Excellence, Children’s Medical Center, Tehran University of Medical Sciences, Tehran, Iran 8 12 2022 112 1 6 2022 20 11 2022 © Children's Hospital, Zhejiang University School of 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. Background Coronavirus disease 2019 (COVID-19) is a disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the seventh coronavirus to be linked to human disease. The SARS-CoV-2 virus may have several pathophysiologic interactions with endocrine systems, resulting in disruptions in glucose metabolism, hypothalamus and pituitary function, adrenal function, and mineral metabolism. An increasing amount of evidence demonstrates both the influence of underlying endocrine abnormalities on the outcome of COVID-19 and the effect of the SARS-CoV-2 virus on endocrine systems. However, a systematic examination of the link to pediatric endocrine diseases has been missing. Data sources The purpose of this review is to discuss the impact of SARS-CoV-2 infection on endocrine systems and to summarize the available knowledge on COVID-19 consequences in children with underlying endocrine abnormalities. For this purpose, a literature search was conducted in EMBASE, and data that were discussed about the effects of COVID-19 on endocrine systems were used in the current study. Results Treatment suggestions were provided for endocrinopathies associated with SARS-CoV-2 infection. Conclusions With the global outbreak of COVID-19, it is critical for pediatric endocrinologists to understand how SARS-CoV-2 interacts with the endocrine system and the therapeutic concerns for children with underlying problems who develop COVID-19. While children and adults share certain risk factors for SARS-CoV-2 infection sequelae, it is becoming obvious that pediatric responses are different and that adult study results cannot be generalized. While pediatric research gives some insight, it also shows the need for more study in this area. Supplementary Information The online version contains supplementary material available at 10.1007/s12519-022-00662-x. Keywords Coronavirus disease 2019 (COVID-19) Endocrine Infection Pediatrics ==== Body pmcIntroduction Coronavirus disease 19 (COVID-19) is caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the seventh coronavirus to be linked to human disease. The major receptor for SARS-CoV-2 infection is believed to be angiotensin-converting enzyme 2 (ACE2) [1]. The binding of SARS-CoV-2 to ACE2 initiates a cascade that activates the nuclear factor-κB pathway, resulting in very high levels of proinflammatory cytokines and chemokines, which contribute to the development of acute respiratory distress syndrome (ARDS) in severe COVID-19 [2, 3]. The lethality of ARDS and non-pulmonary consequences in COVID-19 are considered to be caused by a cytokine storm, in which immune and non-immune cells produce massive quantities of proinflammatory cytokines, causing damage both inside and outside the respiratory system [4]. A limited but rising body of evidence suggests that underlying endocrine abnormalities may influence the outcome of COVID-19 and that the SARS-CoV-2 may have an effect on endocrine systems. At first, it was assumed that infection with SARS-CoV-2 would cause similar but less severe symptoms and complications in children and adolescents [5]. Multisystem inflammatory syndrome in children (MIS-C) and atypical endocrine responses have been observed in children infected with SARS-CoV-2 [6]. Recent understanding concerning the impact of SARS-CoV-2 infection on endocrine systems is summarized here, and current material on COVID-19 pandemic complications in endocrine diseases is reviewed. SARS-CoV-2 and endocrine disorders have been studied extensively in adults, but we have attempted to include all relevant pediatric data from PubMed as of March 2022. If available, we included data from adults, and we made a notation in the text when articles spoke of adult vs. pediatric research teams. SARS-CoV-2 infection in children has been shown to have a wide range of responses in the pediatric population [7]. Since children in the COVID-19 period need specific therapy, pediatric endocrinologists must comprehend the disease's symptoms and the limits of current knowledge. For this purpose, a literature search was conducted in EMBASE, and data that were discussed about the effects of COVID-19 on endocrine systems were used in the current study (Supplementary Table 1). There are few data about the effects of COVID-19 on pediatric endocrine systems, so we tried to discuss related data in adults. SARS-CoV-2 infection and subsequent endocrine dysfunction The SARS-CoV-2 virus has several pathophysiologic linkages to the endocrine system and hence has the potential to disrupt pituitary, adrenal, and thyroid function, as well as glucose and mineral metabolism. Existing data are mostly favorable in terms of COVID-19-related endocrine problems in children (Table 1).Table 1 The summary of the endocrinopathies in children and adults and their management Diseases Suggested mechanisms Managements Obesity Low physical activity times Overfeeding Provide age-appropriate nutrition; increase physical activity Adrenal insufficiency COVID-19 inhibits the adrenal stress response Presence of a hypercoagulable condition Established recommendations for stress dosage of the corticosteroid Diabetes mellitus Insulin resistance due to high levels of cytokines during COVID-19 T1DM: parenteral insulin T2DM: (1) outpatients: glucagon-like peptide 1 analogs and dipeptidyl peptidase 4 inhibitors; (2) critical patients: insulin, metformin, TZD, SGLT2 inhibitors, and TZD, SGLT2 inhibitor Thyroid diseases (subacute thyroiditis, hypothyroidism, hyperthyroidism) Direct injury virus infection of the thyroidal cells and indirect destruction by cytokine storm Nonsteroidal anti-inflammatory drugs, steroids, and nonspecific beta-blockers are suggested Levothyroxine Methimazole Hypopituitarism Hematogenous spread of the virus and infect the pituitary Hydrocortisone Water and electrolyte Metabolic bone disease Vitamin D insufficiency may due to restrict outside time Vitamin D supplementation Home delivery of burosumab injections for patients with hypophosphatemic rickets during the COVID-19 pandemic COVID-19 coronavirus disease 2019, TZD thiazolidinedione, SGLT2 sodium-glucose cotransporter receptor-2, T1DM type 1 diabetes mellitus, T2DM type 2 diabetes mellitus The similarities between COVID-19, SARS, and Middle Eastern respiratory syndrome show that the virus may enter the central nervous system through the olfactory bulb, including the hypothalamus [8]. Adult observational studies have indicated that COVID-19 disrupts posterior pituitary function and causes an immediate onset of syndrome of inappropriate antidiuretic hormone [9–11]. SARS survivors have been shown to have hypothalamic/pituitary dysfunction [12]. However, the only evidence of pituitary involvement in COVID-19 is a magnetic resonance imaging finding of pituitary stalk involvement in two adult patients; there have been no reports of pituitary hormone deficiency in adults or children with COVID-19 thus far [13]. There is evidence that people with COVID-19 may be at risk for adrenal and thyroid problems. In one study, it was shown that acutely unwell people with COVID-19 illness had greater cortisol levels than those without COVID-19, but there was a negative link between the degree of cortisol response and survival rate in those who were COVID-19 positive [14]. Computed tomography (acute adrenal infarction) and post-mortem investigations in individuals with severe COVID-19 and SARS-CoV-2 infection have also shown adrenal involvement [15, 16]. Adults with COVID-19 have been diagnosed with both thyrotoxicosis [as a result of elevated interleukin (IL)-6 levels] and hypothyroidism [17–19]. Although ACE2 is extensively expressed in the thyroid and to a lesser amount in adrenal tissue, and therefore children may potentially be at risk, thyroid and adrenal pathology in children with COVID-19 and MIS-C have not been documented [20]. Due to the high expression of ACE2 in pancreatic islet cells, SARS-CoV-2 infection may have a diabetogenic impact irrespective of the stress response associated with severe sickness [21]. Adults, but not children, with COVID-19 have been reported to develop diabetes mellitus for the first time [22, 23]. Contrary to expectations, the most critically ill individuals with COVID-19 have low lipid levels. These individuals had very low levels of total cholesterol, low density lipoprotein cholesterol, and high density lipoprotein cholesterol, indicating a strong inflammatory (cytokine) impact. Lipid levels rise concurrently with a decrease in inflammatory markers in recovering intensive care unit (ICU) patients. While the long-term repercussions of this occurrence are unknown, decreased lipid levels in conjunction with higher inflammatory markers do seem to be associated with a poor result [24, 25]. COVID-19 does not seem to have a direct effect on the parathyroid glands or on mineral ion balance at the moment. However, research suggests that persons with severe COVID-19 may have a decrease in blood calcium levels [26, 27]. There are few occurrences of hypocalcemia in pediatric patients with MIS-C; however, there are no systemic findings in children at the moment, and probable reasons remain unknown [28, 29]. Complications and management of COVID-19 in children with pre-existing endocrine disorders The complications of SARS-CoV-2 infection have been predominantly documented in adults with underlying endocrine problems. While adult results should not be generalized to children, the studies clearly indicate potential dangers for the pediatric group. The purpose of this section is to summarize what is currently known about COVID-19 in patients with underlying endocrine problems and how it may affect pediatric patients. There is currently no evidence that central hormone deficiency increases the risk of contracting SARS-CoV-2. Children and adolescents with multiple pituitary hormone deficiency, on the other hand, have distinct treatment challenges due to the complexity of their medical condition. Infants and toddlers with diabetes insipidus who have respiratory issues as a result of COVID-19 exposure have a significantly increased risk of acquiring abnormal blood salt levels [30]. Due to variables including lower fluid intake, higher irreversible losses, and the difficulty for adipsic individuals with diabetes insipidus to tolerate oral desmopressin, patients with diabetes insipidus are at an increased risk of severe hypernatremia, which may be exacerbated by venous thrombosis in the acute illness [31–34]. Patients with adrenal insufficiency (including those on corticosteroid replacement therapy) are at greater risk of respiratory and adrenal issues if they get COVID-19, a virus that has not been extensively studied in adults or children. Currently, there is no evidence that children and adolescents with underlying thyroid abnormalities are at a greater risk of contracting SARS-CoV-2 infection or that their clinical course is changed. It is necessary to bear in mind, however, that patients with Graves' disease who receive antithyroid drug (ATD) treatment are at an increased risk of agranulocytosis and secondary infections [35]. This is critical since one study found that more than half of COVID-19 non-survivors had a subsequent infection [36]. In adults with COVID-19, underlying thyroid illness, especially hypothyroidism, does seem to be a risk factor for a more severe disease course [37–39]. Adults with diabetes mellitus, obesity, or hypertension are at increased risk of COVID-19 infection and have a greater incidence of complications and mortality [39–43]. The T1D Exchange provided data on 64 persons with type 1 diabetes (T1D); 33 had COVID-19-positive symptoms but were not tested or had COVID-19-negative symptoms; 65.5% of participants were between the ages of 19 and 20 [44]. COVID-19-positive individuals had a higher mean glycosylated hemoglobin A1c (HbA1c) (8.5% vs. 8%), were more likely to present with diabetic ketoacidosis (DKA) (45.5% vs. 13.3%) and needed a greater level of care than COVID-like individuals. In England, a population-based study found that patients with an HbA1c of 86 mmol/mol (10.0%) or greater had an increased risk of COVID-19-related mortality [hazard ratio = 2.23, 95% confidence interval (CI) = 1.50–3.30; P < 0.0001 in T1D] [45]. Children with T1D and COVID-19 were shown to have higher HbA1c, higher hospitalization rates, non-Hispanic Black ethnicity, and public insurance, according to T1D Exchange data in the pediatric population (unpublished data). During the coronavirus pandemic, children newly diagnosed with T1D are more likely to develop DKA and have more severe DKA [46]. However, evidence suggests that children with T1D and COVID-19 have clinical outcomes that are equivalent to those of children who do not have diabetes [47]. Children with diabetes have encountered significant obstacles as a result of the COVID-19 epidemic, most notably due to extensive school and childcare facility closures. In a study from Greece, 34 children with T1D who used insulin pumps and continuous glucose monitoring had no significant increase in time in range during lockdown but did have increased blood glucose variability compared to the prelockdown period [48]. Additionally, the children in this research had significant alterations in their food routines during lockdown. Restrictions imposed in response to the COVID-19 pandemic have led to reduced physical activity and dietary modifications, as well as changed diabetes management behaviors, all of which may raise the risk of poor nutrition, excessive weight gain, and increased diabetes management stress [47]. Numerous investigations have shown no difference in obesity rates between children with moderate COVID-19 and those with severe COVID-19 [49–51]. While obesity is not more prevalent in pediatric patients hospitalized for COVID-19 than in the general pediatric population, the severity of COVID-19 illness may be connected with obesity, as it is in adults. Obesity was found to be a risk factor for mechanical ventilation in one investigation of 50 pediatric patients hospitalized with COVID-19 [52]. Additionally, a recent multicenter investigation of COVID-19 infection in 281 hospitalized children under the age of 22 years showed obesity (odds ratio = 3.39, 95% CI = 1.26–9.10; P = 0.02) and hypoxia on admission as the only two underlying risk factors for severe respiratory illness [53]. In adults, obesity-related adverse outcomes may be mediated by pre-existing cardiovascular and renal illness, as well as hypertension [54]. Children who have metabolic bone disease or skeletal dysplasia that leads to respiratory insufficiency as a consequence of an irregular chest wall structure may be more susceptible to COVID-19 complications [55]. Immunity and autophagy are both influenced by vitamin D; however, it is not known whether vitamin D deficiency enhances the chance of COVID-19 infection or its repercussions. Data from the UK Biobank did not show a correlation between SARS-CoV-2 infection and decreased blood 25-OH-vitamin D concentrations after controlling for confounding factors, despite the findings of certain observational studies of adults [56]. Obesity Obesity is taken into consideration during the outbreak of COVID-19 for two reasons. First, according to reports from the World Health Organization, during an outbreak of COVID-19, the prevalence of childhood and adolescent obesity due to low physical activity times and overfeeding has increased [57]. A large pediatric primary care study showed a significant increase in obesity rates among children ages 2 through 17 since the onset of the COVID-19 pandemic [58], and obesity-related lifestyle behaviors have also changed [59]. Second, obesity as a comorbidity likely increases susceptibility and severe COVID-19 infection in pediatrics and adolescents [60]. During the COVID-19 outbreak in Canada, obesity was introduced as the third factor among children with severe infection after malignancies and immunosuppression [51]. Obesity may be a risk factor regardless of age, gender, and higher body mass index (BMI) related to the risk of requiring invasive mechanical ventilation [61]. Furthermore, adult patients with obesity are more likely to have some symptoms, including fever, dyspnea, and caught [62]. Because of the lack of enough studies, the impact of obesity on COVID-19 in the pediatric population has not been well explained. Poor immune response, cardiopulmonary disease, and chronic inflammation that link obesity to COVID-19 and have been proven in adult patients have also been shown in children [63]. Obesity is associated with overexpression of angiotensin 2 [64]. Obesity limits respiratory muscle movement and worsens general conditions in pediatric patients [65]. Asthma and obstructive sleep apnea are linked to obesity, which is related to a higher risk of pulmonary infections [66]. Cardiac anatomy changes, higher blood pressure and medications [67], intima layer artery thickening [68], endothelial dysfunction, and damage to the endothelium due to leptin perivascular adipose tissue have been explained in obese children. In obese individuals, chronic inflammation impairs the regulation of anticoagulant factors that contribute to venous thromboembolism during COVID-19 [69]. Angiotensin converting enzyme converted angiotensin 2 to angiotensin type 1–7. Unlike angiotensin 2, which has inflammatory effects, angiotensin 1–7 acts as an anti-inflammatory [70]. Therefore, it is assumed that this imbalance may cause immune response dysregulation [71]. In patients with obesity, adipocytokines, such as leptin, may change the function and number of immune cells, leading to an increase in the number of activated macrophages (M1) and effector T helper 1 (Th1) and Th7 cytotoxic T cells and then a decrease in the number of regulatory T cells and M2 macrophages. Macrophages derived from visceral adipose tissue secrete large amounts of inflammatory cytokines, including nitric oxide IL-12, IL-1b, IL-6, and tumor necrosis factor-α (TNF-α) [72]. In individuals with obesity, SARS-CoV-2 enters fat cells by binding to ACE2, which could lead to increased viral load and prolonged viral spread due to their altered immune responses and cytokine response in adipose tissue [73]. There is evidence of endothelial dysfunction in obesity [74] and kidney disease [75]. Obesity is usually associated with comorbidities, such as hypertension, hyperinsulinism, type 2 diabetes (T2D), and thrombogenic events, which decrease the body’s ability to overcome COVID-19 [76]. Fortunately, these comorbidities are not common in children. It seems that obese individuals spread more enormous amounts of the virus through inhalation and can infect other people [77]. Since the effectiveness of the influenza vaccine due to alterations in the immune system decreased in obese people, this phenomenon may also occur in the vaccination against SARS-CoV-2, which leads to decreased immunization against COVID-19 in obese people patients [78]. A large retrospective cohort study from the Society of Critical Care Medicine Viral Respiratory Illness Universal Study registry assessed all children hospitalized with COVID-19 from March 2020 to January 2021; 31.5% of patients had obesity. In adult patients, obesity was associated with more severe COVID-19 diseases and more extended hospitalization, but unlike adults, the risk of mortality in obese children was not higher than that in non-obese patients. They found no relation between obesity and a higher risk of MIS-C because inflammatory and proinflammatory alterations due to obesity may have less impact on the immune dysregulation response in MIS-C. Furthermore, they have shown a negative correlation between age and high BMI with illness severity from COVID-19 in pediatric patients [60]. Management Due to social isolation, lack of exercise, and food insecurity in socioeconomically challenged homes, the pandemic-related closure of schools, camps, and extracurricular sports and activities has dramatically affected children’s and adolescents' health [79]. In one of our centers, we have seen a large rise in the number of children under 19 who arrive with new-onset diabetes. The majority of the increase is attributed to the rise in T2D. It is critical to promote good eating habits and provide age-appropriate nutrition and physical activity guidelines. During the COVID-19 pandemic, racial/ethnic and socioeconomic differences exacerbated health inequities, notably concerning weight control. Healthcare practitioners should continue to educate patients about the importance of eating healthily and exercising regularly. We advocate for the use of telemedicine treatments to supplement pediatric weight control. The likelihood of developing premature atherosclerotic cardiovascular disease (ASCVD) in children who have undergone MIS-C is unknown; however, individuals with a history of Kawasaki disease and persistent aneurysmal dilatation are regarded to be at a high risk of developing ASCVD. Coronary artery aneurysms have been detected in 6%–24% of patients with MIS-C, and Kawasaki disease characteristics have been observed in up to 40% of patients [79–81]. Recent evidence suggests that pediatric children with MIS-C who receive intravenous immunoglobulin or an IL-6 antagonist, such as tocilizumab, recover without complications [82]. Given the uncertain risk of complications, children who have recovered from MIS-C should receive a longer-term cardiac follow-up. Adrenal insufficiency ACE2 is expressed in small amounts in adrenal tissue, but COVID-19 can inhibit the adrenal stress response through different mechanisms. It has been shown that the virus contains amino acids such as adrenocorticotropin hormone. Antibodies against these amino acids and adrenocorticotropin hormone function cause a limited adrenal stress response and adrenal insufficiency [83]. Another mechanism may be due to the hypercoagulable condition during the infection leading to acute adrenal infarction [84]. On the other hand, patients with previous adrenal insufficiency are at a higher risk of infection because of reduced cortisol secretion, especially during the acute phase of adrenal insufficiency. Adrenocortical necrosis and hemorrhage during severe COVID-19 infection may cause adrenal insufficiency. Medications may cause adrenal insufficiency during COVID-19 infection. Martino et al. [85] showed that the risk of developing COVID-19 in patients with adrenal insufficiency is lower than that in the normal population, which may be due to more attention and care. On the other hand, patients taking supraphysiological doses of glucocorticoids may be more prone to COVID-19 [86]. Although adrenal involvement has been shown in adults with COVID-19, significant adrenal disease in children with COVID-19 has not been reported [15]. There are limited data available in pediatric patients with adrenal insufficiency and COVID-19. Nevertheless, Raisingani [87] showed that the mortality rate and severe disease increased in children with adrenal insufficiency and COVID-19 compared with children with COVID-19 and no adrenal insufficiency. Management Individuals who are steroid-dependent or fear that they may have adrenal suppression should exercise vigilance to prevent infection with SARS-CoV-2. Patients with symptomatic COVID-19 should be handled according to established recommendations for stress dosage [88, 89]. Diabetes mellitus and impaired glucose metabolism Is diabetes mellitus a risk factor for increasing the severity of COVID-19? Diabetes mellitus, prediabetes mellitus, and hyperglycemia are independent risk factors associated with moderate to severe COVID-19, increased morbidity and mortality, and a significant increase in inflammatory cytokines compared with non-diabetic patients [45, 90, 91], but diabetes is not related to an increased risk of SARS-CoV-2 [92]. Immunological dysregulation associated with hyperglycemia, endothelial damage, and increased levels of cytokines and oxidative stress related to coronavirus infection led to multiple organ dysfunction and thromboembolic risk [93]. One cohort study revealed that patients with an HbA1c of 10.0% or higher had increased COVID-19-related mortality [45]. Furthermore, antiviral drugs and systemic corticosteroids used for COVID-19 may exacerbate hyperglycemia and the severity of the disease in diabetic patients [94]. In human studies, high glucose levels lead to increased SARS duplication, the production of mitochondrial reactive oxygen, and the activation of hepatocyte nuclear factor-1α, and glycolysis maintains this duplication [95]. In patients with T2D, neutrophil, and complement dysfunction, high proinflammatory cytokines, such as TNF-α, IL-6, and IL-1β, and decreased viral clearance can exacerbate the clinical course of COVID-19 [96]. During the COVID-19 outbreak, children with new-onset-dependent diabetes may develop DKA but have more severe clinical manifestations. Their prognosis has not been worse than that of patients without diabetes [46, 47]. On the other hand, one study among adult patients reported a mortality rate of approximately fifty percent among patients with DKA in COVID-19. Although DKA may occur in T2D, a combined DKA/hyperglycemic hyperosmolar state should be considered and has higher mortality than DKA alone [93, 97]. Is COVID-19 a risk factor for developing new-onset diabetes mellitus? High levels of IL-6, IL-1β, TNF-α, monocyte chemoattractant protein-1, and inducible protein-10 during COVID-19 can cause insulin resistance. As mentioned above, obesity associated with T2D exacerbates the cytokine reaction that contributes to lowering insulin sensitivity and worsening hyperglycemia in patients with diabetes and prediabetes. COVID-19, as a viral infection, can trigger the presentation of diabetes in genetically predisposed individuals [98]. The expression of ACE2 in β-cells as a receptor for COVID-19 may cause pancreatic damage and result in pancreatic endocrine and exocrine dysfunction [99]. In one study, among the patients hospitalized for COVID-19, 2.8% had newly diagnosed diabetes, and a systematic review showed 10% newly diagnosed diabetes mellitus among the patients with COVID-19 [93, 100]. Management During COVID-19 infection, regular consultation and monitoring of blood glucose levels (via finger sticks or continuous glucose monitors) are necessary. Insulin doses should be adjusted according to self-monitoring blood glucose. Furthermore, patients should check ketones in addition to blood glucose monitoring, and if ketones are present, increase correction doses of insulin. Adequate hydration of the patient effectively controls blood glucose levels [101]. Regular physical activity and correcting eating habits are helpful. In patients with T2D acute hyperglycemia due to exacerbation of inflammation, insulin resistance must be controlled rapidly and effectively. Some medications are used for glycemic control in patients with T2D, but the choice of medication during COVID-19 infection should be based on the patient’s condition. Glucagon-like peptide 1 analogs and dipeptidyl peptidase 4 inhibitors in mild to moderate symptoms have glucose-lowering efficacy and inflammatory action in outpatient and hospitalized patients. Insulin infusion in critically ill diabetic patients has been suggested for adequate glycemic control and D-dimer and IL-6 levels [94]. Metformin, as a lowering glucose agent, had anti-inflammatory action related to significantly lower mortality compared to patients not receiving metformin [102]. During severe COVID-19 infection (ICU-admitted patients), metformin, thiazolidinedione (TZD), sodium-glucose cotransporter receptor-2 (SGLT2) inhibitors, and TZD, SGLT2 inhibitors were used in hospitalized patients, but moderate disease has not been recommended to use. [94]. Due to increased insulin resistance and ketone generation, viral infections might be more challenging to control in patients with diabetes [103, 104]. Given the positive correlation between a higher HbA1c level and the prevalence of DKA in children with T1D, a large proportion of the pediatric T1D population may be at increased risk of developing DKA in the presence of COVID-19 infection [105, 106]. To reduce the risk of DKA, practitioners should familiarize themselves with sick day guidelines and be available for guidance during times of illness. Patients should be advised to monitor their blood glucose levels more often during COVID-19 illness, either with continuous glucose monitors or with finger sticks [101]. Insulin doses may need to be adjusted more often, and additional fast-acting insulin correction boluses may be required to avoid severe hyperglycemia or ketoacidosis [107]. Patients should monitor ketones regardless of blood sugar levels and increase correction dosage and fluid intake if ketones are detected. Encouraging remote blood glucose monitoring with continuous glucose monitoring is a great tool that should be made available to all families. Additional guidelines for children with T2D include those discussed in the obesity section below. All children must engage in regular physical exercise and practice good dietary practices during this epidemic. Telemedicine is a helpful method for maintaining regular contact with the diabetes team and counseling and managing patients with T1D and T2D during COVID-19 infection. However, there are limited data, and further research is needed. A review among patients with T2D and T1D showed that telemedicine could reduce HbA1c by approximately − 0.01% to − 1.13% and − 0.12% to − 0.86%, respectively [108]. Thyroid disease Thyroid cells express transmembrane protease serine 2 and ACE2 as receptors for COVID-19 even more than the lungs, explaining distinct thyroid manifestations during infection [20]. Direct injury virus infection of the thyroidal cells and indirect destruction by cytokine storms during COVID-19 may lead to thyroid gland damage that presents to subacute thyroiditis with three-phase thyrotoxicosis, hypothyroidism, and finally euthyroidism [109]. Thyroid hormones act as modulators of the immune system [110]. Normal values of T4 and T3 can exacerbate the production and release of inflammatory and proinflammatory cytokines, which develop into a “cytokine storm”. T4 activates human platelets, which can contribute to clotting as a complication of virus infections [111]. There is no evidence that thyroid disease predisposes individuals to COVID-19 infection, but uncontrolled hyperthyroidism is a risk factor for complications. Patients with thyroid disease should continue their medication as before [84]. Three thyroid disorders, hypothyroidism, subacute thyroiditis, thyrotoxicosis, and non-thyroidal illness, have been described during the COVID-19 pandemic. The severity of COVID-19 is an essential determinant of the type of thyroid disorder [112]. One study reported a prevalence of 5.2% hypothyroidism in hospitalized patients with COVID-19, predominantly subclinical hypothyroidism, and another showed 20.2% of patients with thyrotoxicosis [18]. Fibrillation in a patient without a history of previous cardiovascular disease may indicate a thyrotoxicosis state [113]. Furthermore, lower thyroid stimulating hormone (TSH) and free T3 values correlated with the severity of COVID-19 [114]. Management All collected data about thyroid damage and COVID-19 have recommended regular monitoring of thyroid function during acute infection and treatment. There are no data on the treatment of thyrotoxicosis and sick euthyroid syndrome in hospitalized patients with COVID-19 [112]. Since the clinical course of thyroiditis is benign, a watch-and-wait approach is sufficient, but in severely symptomatic patients, non-steroidal anti-inflammatory drugs, steroids, and non-specific beta-blockers are suggested [115, 116]. If hypothyroidism and TSH are > 10 mIU/mL, levothyroxine therapy should be started [84]. Patients with primary hypothyroidism or hyperthyroidism should be treated as usual. ATDs such as methimazole should be warned about the risk of agranulocytosis and consequent bacterial infection. Because the symptoms of ATDs (fever, sore throat, and cough) are similar to COVID-19, a fast medical check is advised if symptoms emerge. Similar to other viruses, COVID-19 may cause thyroid storms in people with uncontrolled hyperthyroidism. Thyroid cancer patients who underwent surgery with replacement levothyroxine had no increased risk of infection. Immunosuppression may increase the infection risk in a portion of chemo patients. The American Thyroid Association and the American Association of Endocrine Surgeons say that due to the tumor's slow growth, most thyroid cancer operations may be safely deferred [117]. Hypopituitarism ACE2, as a COVID-19 receptor on central nervous tissues, may play a role in COVID-19 neural invasion [118]. In addition, the hematogenous spread of the virus can infect the pituitary [119]. Children with hypopituitarism are not at increased risk for COVID-19, and they are not at increased risk of developing severe COVID-19 disease following infection. Management During the COVID-19 outbreak, regular monitoring is necessary for children with hypopituitarism to take their medication correctly. They have been taught that increased dose of hydrocortisone according to the severity of their disease. Some reports in adult patients with growth hormone deficiency suggest stopping growth hormone during hospitalization with COVID-19, but there are no data among pediatric patients [120, 121]. In diabetes insipidus, the water and electrolyte balance should be considered. Children with numerous pituitary hormone deficits may be at a greater risk of COVID-19 complications and death, especially if central adrenal insufficiency is not treated appropriately. In the absence of published data or standards, we propose that children's pituitary hormone deficits be managed according to known protocols. While current guidelines for adults with growth hormone insufficiency suggest discontinuing growth hormone therapy during hospitalization with COVID-19, there is a dearth of research on the impact of growth hormone treatment in children with COVID-19 [120]. Metabolic bone disease Vitamin D supplementation has not been shown to protect against COVID-19 or its consequences. Randomized experiments are now being conducted to ascertain whether vitamin D supplementation may help prevent or lessen the severity of COVID-19. Prolonged home quarantine to prevent the spread of COVID-19 may restrict outside time, increasing the risk of vitamin D insufficiency and related consequences such as rickets, osteomalacia, and symptomatic hypocalcemia [122, 123]. Clinicians should continue to adhere to existing vitamin D supplementation guidelines for individuals at risk of insufficiency. The Food and Drug Administration has authorized home delivery of burosumab injections for patients with hypophosphatemic rickets during the COVID-19 pandemic. The American Society of Bone and Mineral Research, the American Association of Clinical Endocrinology, the Endocrine Society, the European Calcified Tissue Society, and the National Osteoporosis Foundation have issued a joint statement outlining guidelines for managing osteoporosis in adults during the COVID-19 pandemic, some of which may need to be modified for pediatric patients. Essential measures to preserve bone health should be advocated in children with poor bone density, including proper calcium and vitamin D consumption, as well as sex steroid replacement where necessary. Weight-bearing exercise is critical for bone health optimization and should not be overlooked; physical therapy services are readily accessible through telemedicine and can be used to lead treatment in the home when necessary. Advanced therapy should be maintained when practicable, especially in children at high risk of fragility fractures. Due to the long-acting nature of intravenous bisphosphonates, it is typically regarded as safe to postpone therapy for at least 6–9 months in adults [124]. In youngsters, however, continued new bone formation and the emergence of stress risers during intermittent bisphosphonate medication may raise the risk of fracture if treatment is significantly delayed [125]. Consider switching patients from pamidronate to zoledronic acid, which is injected over a shorter time period and needs fewer infusions. Clinicians may elect to forego pre-treatment laboratory testing for repeat infusions if the patient has no history of hypocalcemia with previous infusions, is consuming an adequate amount of calcium and vitamin D through diet or supplementation, has no renal disease, and has a stable overall health status. In most patients, DXA scans and other imaging may be postponed if the findings of the dual-energy X-ray absorptiometry (DXA) scan do not need a change in care. Precocious puberty There are some reports that represent precocious puberty in girls during COVID-19 [126–128]. According to the first report, an increase in diagnoses and a quicker puberty progression were observed. It was hypothesized that these trends were related to an increase in electronic device use, an increase in BMI, and possibly an increase in eating. The second study confirmed the rise in precocious puberty diagnoses as well as the net prevalence of females. In a study conducted by Turriziani Colonna et al., children who presented with suspected central precocious puberty (CCP) during the COVID-19 outbreak were assessed. It was observed that among all patients who were referred for CCP, 11 (39.3%) new CCP were diagnosed during COVID-19, and 22/45 (48.9%) of the patients showed an increased pubertal progression rate, with more children in the T3 and T4-5 phases [128]. In another study, it was mentioned that of 124 patients with idiopathic precocious puberty, 46.8% were diagnosed during one year of this pandemic, and the rest were diagnosed during three years before the COVID-19 pandemic [129]. Additionally, other studies confirmed that COVID-19 is related to precocious puberty in children [130–133]. Conclusions SARS-CoV-2 has the capacity to affect the majority of endocrine systems mechanistically. While little is known regarding the association between COVID-19 and endocrine problems in children, published data in adults have indicated pathology associated with diabetes mellitus, obesity, pituitary, adrenal, and thyroid illness. The following aspects of COVID-19 and endocrine diseases in children and adolescents are highlighted by current data: (1) antithyroid medication and continuous corticosteroid therapy are likely to enhance the risk of SARS-CoV-2 infection; patients should be counseled about this risk and advised to take further measures to avoid infection; (2) MIS-C seems to increase the incidence of hypocalcemia in children, and those who develop coronary artery aneurysms may be at increased risk of developing ASCVD later in life; (3) children with T1D and a higher HbA1c are more likely than children with better diabetes management to be hospitalized with COVID-19. The COVID-19 pandemic has produced a variety of difficulties for pediatric diabetes treatment, including school closures, interrupted schedules, and stress associated with diabetes control during lockdown times; (4) obesity does not seem to enhance the incidence of infection with SARS-CoV-2 in children, but it may be a risk factor for COVID-19-related problems in this group; and (5) COVID-19 treatment in children with diabetes insipidus needs special attention to fluid and salt balance. It may necessitate a modification in the method of desmopressin administration if the intranasal version is used. The COVID-19 pandemic will aggravate obesity, metabolic syndrome, and its related comorbidities if enforced shelter-in-home laws and physical distance limitations are not taken into account, which we believe is very important. There is a direct link between sedentary activity and increased risks of developing T2D and precocious puberty over time [126–136]. We must continue to give the necessary assistance to our patients to mitigate the pandemic's health repercussions for the future generation. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file 1 (PDF 406 KB) Acknowledgements We wish to thank our colleagues in Endocrinology Ward of Children’s Medical Center for their cooperation and helps. Author contributions MEHE and SST wrote the first draft of manuscript. MR developed the concept, gathered some data and published article. RP developed the concept, gathered some data and published article, and wrote the first draft of manuscript. All authors read and approved the final version of manuscript. Funding The authors received no specific funding for this work. Data availability Data of this manuscript are available in the archive of Children’s Medical Center Hospital and will be presented in case of request. Declarations Ethical approval This research was done under permission of Ethics Committee of Tehran University of Medical Sciences. Conflict of interest The authors declare that they 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. Mason RJ Pathogenesis of COVID-19 from a cell biology perspective Eur Respir J 2020 55 2000607 10.1183/13993003.00607-2020 32269085 2. Huang C Wang Y Li X Ren L Zhao J Hu Y Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China Lancet 2020 395 497 506 10.1016/S0140-6736(20)30183-5 31986264 3. Rothan HA Byrareddy SN The epidemiology and pathogenesis of coronavirus disease (COVID-19) outbreak J Autoimmun 2020 109 102433 10.1016/j.jaut.2020.102433 32113704 4. Hirano T Murakami M COVID-19: a new virus, but a familiar receptor and cytokine release syndrome Immunity 2020 52 731 733 10.1016/j.immuni.2020.04.003 32325025 5. Zimmermann P Curtis N COVID-19 in children, pregnancy and neonates: a review of epidemiologic and clinical features Pediatr Infect Dis J 2020 39 469 477 10.1097/INF.0000000000002700 32398569 6. Diorio C Henrickson SE Vella LA McNerney KO Chase J Burudpakdee C Multisystem inflammatory syndrome in children and COVID-19 are distinct presentations of SARS–CoV-2 J Clin Invest 2020 130 5967 5975 10.1172/JCI140970 32730233 7. Naja M Wedderburn L Ciurtin C COVID-19 infection in children and adolescents Br J Hosp Med (Lond) 2020 81 1 10 10.12968/hmed.2020.0321 8. Steardo L Steardo L Jr Zorec R Verkhratsky A Neuroinfection may contribute to pathophysiology and clinical manifestations of COVID-19 Acta Physiol (Oxf) 2020 229 e13473 10.1111/apha.13473 32223077 9. Ata F Almasri H Sajid J Yousaf Z COVID-19 presenting with diarrhoea and hyponatraemia BMJ Case Rep 2020 13 e235456 10.1136/bcr-2020-235456 10. Habib MB Sardar S Sajid J Acute symptomatic hyponatremia in setting of SIADH as an isolated presentation of COVID-19 IDCases 2020 21 e00859 10.1016/j.idcr.2020.e00859 32523871 11. Wu Y Xu X Chen Z Duan J Hashimoto K Yang L Nervous system involvement after infection with COVID-19 and other coronaviruses Brain Behav Immun 2020 87 18 22 10.1016/j.bbi.2020.03.031 32240762 12. Leow MKS Kwek DSK Ng AWK Ong KC Kaw GJL Lee LSU Hypocortisolism in survivors of severe acute respiratory syndrome (SARS) Clin Endocrinol (Oxf) 2005 63 197 202 10.1111/j.1365-2265.2005.02325.x 16060914 13. Pascual-Goñi E Fortea J Martínez-Domeño A Rabella N Tecame M Gómez-Oliva C COVID-19-associated ophthalmoparesis and hypothalamic involvement Neurol Neuroimmunol Neuroinflamm 2020 7 e823 10.1212/NXI.0000000000000823 32587102 14. Tan T Khoo B Mills EG Phylactou M Patel B Eng PC Association between high serum total cortisol concentrations and mortality from COVID-19 Lancet Diabetes Endocrinol 2020 8 659 660 10.1016/S2213-8587(20)30216-3 32563278 15. Leyendecker P Ritter S Riou M Wackenthaler A Meziani F Roy C Acute adrenal infarction as an incidental CT finding and a potential prognosis factor in severe SARS-CoV-2 infection: a retrospective cohort analysis on 219 patients Eur Radiol 2021 31 895 900 10.1007/s00330-020-07226-5 32852586 16. Santana MF Borba MGS Baía-da-Silva DC Val F Alexandre MAA Brito-Sousa JD Case report: adrenal pathology findings in severe COVID-19: an autopsy study Am J Trop Med Hyg 2020 103 1604 1607 10.4269/ajtmh.20-0787 32876012 17. Ippolito S Dentali F Tanda M SARS-CoV-2: a potential trigger for subacute thyroiditis? Insights from a case report J Endocrinol Invest 2020 43 1171 1172 10.1007/s40618-020-01312-7 32488726 18. Lania A Sandri MT Cellini M Mirani M Lavezzi E Mazziotti G Thyrotoxicosis in patients with COVID-19: the THYRCOV study Eur J Endocrinol 2020 183 381 387 10.1530/EJE-20-0335 32698147 19. Muller I Cannavaro D Dazzi D Covelli D Mantovani G Muscatello A SARS-CoV-2-related atypical thyroiditis Lancet Diabetes Endocrinol 2020 8 739 741 10.1016/S2213-8587(20)30266-7 32738929 20. Li MY Li L Zhang Y Wang XS 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 21. 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 22. Chee YJ Ng SJH Yeoh E Diabetic ketoacidosis precipitated by Covid-19 in a patient with newly diagnosed diabetes mellitus Diabetes Res Clin Pract 2020 164 108166 10.1016/j.diabres.2020.108166 32339533 23. Hollstein T Schulte DM Schulz J Glück A Ziegler AG Bonifacio E Autoantibody-negative insulin-dependent diabetes mellitus after SARS-CoV-2 infection: a case report Nat Metab 2020 2 1021 1024 10.1038/s42255-020-00281-8 32879473 24. Wei X Zeng W Su J Wan H Yu X Cao X Hypolipidemia is associated with the severity of COVID-19 J Clin Lipidol 2020 14 297 304 10.1016/j.jacl.2020.04.008 32430154 25. Wang G Zhang Q Zhao X Dong H Wu C Wu F Low high-density lipoprotein level is correlated with the severity of COVID-19 patients: an observational study Lipids Health Dis 2020 19 204 10.1186/s12944-020-01382-9 32892746 26. Lippi G South AM Henry BM Electrolyte imbalances in patients with severe coronavirus disease 2019 (COVID-19) Ann Clin Biochem 2020 57 262 265 10.1177/0004563220922255 32266828 27. Qian GQ Yang NB Ding F Ma AHY Wang ZY Shen YF Epidemiologic and clinical characteristics of 91 hospitalized patients with COVID-19 in Zhejiang, China: a retrospective, multi-centre case series QJM 2020 113 474 481 10.1093/qjmed/hcaa089 32181807 28. Almoosa ZA Al Ameer HH AlKadhem SM Busaleh F AlMuhanna FA Kattih O Multisystem inflammatory syndrome in children, the real disease of COVID-19 in pediatrics-a multicenter case series from Al-Ahsa Saudi Arabia Cureus 2020 12 e11064 33240687 29. Samuel S Friedman RA Sharma C Ganigara M Mitchell E Schleien C Incidence of arrhythmias and electrocardiographic abnormalities in symptomatic pediatric patients with PCR-positive SARS-CoV-2 infection, including drug-induced changes in the corrected QT interval Heart Rhythm 2020 17 1960 1966 10.1016/j.hrthm.2020.06.033 32621881 30. Christ-Crain M Hoorn EJ Sherlock M Thompson CJ Wass JAH Endocrinology in the time of COVID-19: management of diabetes insipidus and hyponatraemia Eur J Endocrinol 2020 183 G9 15 10.1530/EJE-20-0338 32380474 31. Behan L Sherlock M Moyles P Renshaw O Thompson C Orr C Abnormal plasma sodium concentrations in patients treated with desmopressin for cranial diabetes insipidus: results of a long-term retrospective study Eur J Endocrinol 2015 172 243 250 10.1530/EJE-14-0719 25430399 32. Crowley RK Sherlock M Agha A Smith D Thompson CJ Clinical insights into adipsic diabetes insipidus: a large case series Clin Endocrinol (Oxf) 2007 66 475 482 17371462 33. Cuesta M Hannon MJ Thompson CJ Adipsic diabetes insipidus in adult patients Pituitary 2017 20 372 380 10.1007/s11102-016-0784-4 28074401 34. Darmon M Timsit JF Francais A Nguile-Makao M Adrie C Cohen Y Association between hypernatraemia acquired in the ICU and mortality: a cohort study Nephrol Dial Transplant 2010 25 2510 2515 10.1093/ndt/gfq067 20167570 35. Ross DS Burch HB Cooper DS Greenlee MC Laurberg P Maia AL 2016 American Thyroid Association guidelines for diagnosis and management of hyperthyroidism and other causes of thyrotoxicosis Thyroid 2016 26 1343 1421 10.1089/thy.2016.0229 27521067 36. Zhou F Yu T Du R Fan G Liu Y Liu Z Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study Lancet 2020 395 1054 1062 10.1016/S0140-6736(20)30566-3 32171076 37. Boelaert K Visser WE Taylor PN Moran C Léger J Persani L Endocrinology in the time of COVID-19: management of hyperthyroidism and hypothyroidism Eur J Endocrinol 2020 183 G33 G39 10.1530/EJE-20-0445 32438340 38. Dworakowska D Grossman AB Thyroid disease in the time of COVID-19 Endocrine 2020 68 471 474 10.1007/s12020-020-02364-8 32507963 39. Zhang JJ Dong X Cao YY Yuan YD Yang YB Yan YQ Clinical characteristics of 140 patients infected with SARS-CoV-2 in Wuhan, China Allergy 2020 75 1730 1741 10.1111/all.14238 32077115 40. Guan WJ Ni ZY Hu Y Liang WH Ou CQ He JX Clinical characteristics of coronavirus disease 2019 in China N Engl J Med 2020 382 1708 1720 10.1056/NEJMoa2002032 32109013 41. 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 32203977 42. Yang J Zheng Y Gou X Pu K Chen Z Guo Q Prevalence of comorbidities and its effects in patients infected with SARS-CoV-2: a systematic review and meta-analysis Int J Infect Dis 2020 94 91 95 10.1016/j.ijid.2020.03.017 32173574 43. Zheng YY Ma YT Zhang JY Xie X COVID-19 and the cardiovascular system Nat Rev Cardiol 2020 17 259 260 10.1038/s41569-020-0360-5 32139904 44. Ebekozien OA Noor N Gallagher MP Alonso GT Type 1 diabetes and COVID-19: preliminary findings from a multicenter surveillance study in the US Diabetes Care 2020 43 e83 e85 10.2337/dc20-1088 32503837 45. Holman N Knighton P Kar P O'Keefe J Curley M Weaver A Risk factors for COVID-19-related mortality in people with type 1 and type 2 diabetes in England: a population-based cohort study Lancet Diabetes Endocrinol 2020 8 823 833 10.1016/S2213-8587(20)30271-0 32798471 46. Ho J Rosolowsky E Pacaud D Huang C Lemay JA Brockman N Diabetic ketoacidosis at type 1 diabetes diagnosis in children during the COVID-19 pandemic Pediatr Diabetes 2021 22 552 557 10.1111/pedi.13205 33745226 47. DiMeglio LA Albanese-O’Neill A Muñoz CE Maahs DM COVID-19 and children with diabetes—updates, unknowns, and next steps: first, do no extrapolation Diabetes Care 2020 43 2631 2634 10.2337/dci20-0044 32887703 48. Christoforidis A Kavoura E Nemtsa A Pappa K Dimitriadou M Coronavirus lockdown effect on type 1 diabetes management οn children wearing insulin pump equipped with continuous glucose monitoring system Diabetes Res Clin Pract 2020 166 108307 10.1016/j.diabres.2020.108307 32650036 49. Belhadjer Z Méot M Bajolle F Khraiche D Legendre A Abakka S Acute heart failure in multisystem inflammatory syndrome in children in the context of global SARS-CoV-2 pandemic Circulation 2020 142 429 436 10.1161/CIRCULATIONAHA.120.048360 32418446 50. DeBiasi RL Song X Delaney M Bell M Smith K Pershad J Severe coronavirus disease-2019 in children and young adults in the Washington, DC, metropolitan region J Pediatr 2020 223 199 203.e1 10.1016/j.jpeds.2020.05.007 32405091 51. Shekerdemian LS Mahmood NR Wolfe KK Riggs BJ Ross CE McKiernan CA Characteristics and outcomes of children with coronavirus disease 2019 (COVID-19) infection admitted to US and Canadian pediatric intensive care units JAMA Pediatr 2020 174 868 873 10.1001/jamapediatrics.2020.1948 32392288 52. Zachariah P Johnson CL Halabi KC Ahn D Sen AI Fischer A Epidemiology, clinical features, and disease severity in patients with coronavirus disease 2019 (COVID-19) in a children’s hospital in New York City New York JAMA Pediatr 2020 174 e202430 10.1001/jamapediatrics.2020.2430 32492092 53. Fernandes DM Oliveira CR Guerguis S Eisenberg R Choi J Kim M Severe acute respiratory syndrome coronavirus 2 clinical syndromes and predictors of disease severity in hospitalized children and youth J Pediatr 2021 230 23 31.e10 10.1016/j.jpeds.2020.11.016 33197493 54. Petrilli CM Jones SA Yang J Rajagopalan H O’Donnell L Chernyak Y Factors associated with hospital admission and critical illness among 5279 people with coronavirus disease 2019 in New York City: prospective cohort study BMJ 2020 369 m1966 10.1136/bmj.m1966 32444366 55. Brizola E Adami G Baroncelli GI Bedeschi MF Berardi P Boero S Providing high-quality care remotely to patients with rare bone diseases during COVID-19 pandemic Orphanet J Rare Dis 2020 15 228 10.1186/s13023-020-01513-6 32867855 56. Hernández JL Nan D Fernandez-Ayala M García-Unzueta M Hernández-Hernández MA López-Hoyos M Vitamin D status in hospitalized patients with SARS-CoV-2 infection J Clin Endocrinol Metab 2021 106 e1343 e1353 10.1210/clinem/dgaa733 57. World Health Organization. WHO Director-General’s opening remarks at the media briefing on COVID-19, 11 March, 2020. https://www.who.int/director-general/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19—11-march-2020. Accessed 11 Mar 2020. 58. Jenssen BP Kelly MK Powell M Bouchelle Z Mayne SL Fiks AG COVID-19 and changes in child obesity Pediatrics 2021 147 e2021050123 10.1542/peds.2021-050123 33653879 59. Yang D Luo C Feng X Qi W Qu S Zhou Y Changes in obesity and lifestyle behaviours during the COVID-19 pandemic in Chinese adolescents: a longitudinal analysis from 2019 to 2020 Pediatr Obes 2022 17 e12874 10.1111/ijpo.12874 34937132 60. Tripathi S Christison AL Levy E McGravery J Tekin A Bolliger D The impact of obesity on disease severity and outcomes among hospitalized children with COVID-19 Hosp Pediatr 2021 11 e297 316 10.1542/hpeds.2021-006087 34168067 61. Simonnet A Chetboun M Poissy J Raverdy V Noulette J Duhamel A High prevalence of obesity in severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) requiring invasive mechanical ventilation Obesity (Silver Spring) 2020 28 1195 1199 10.1002/oby.22831 32271993 62. Hajifathalian K Kumar S Newberry C Shah S Fortune B Krisko T Obesity is associated with worse outcomes in COVID-19: analysis of early data from New York City Obesity (Silver Spring) 2020 28 1606 1612 10.1002/oby.22923 32470210 63. Nogueira-de-Almeida CA We need to look at the comorbidities of obesity during childhood and adolescence Future 2017 30 431 437 64. Saiki A Ohira M Endo K Koide N Oyama T Murano T Circulating angiotensin II is associated with body fat accumulation and insulin resistance in obese subjects with type 2 diabetes mellitus Metabolism 2009 58 708 713 10.1016/j.metabol.2009.01.013 19375596 65. Rychter AM Zawada A Ratajczak AE Dobrowolska A Krela-Kaźmierczak I Should patients with obesity be more afraid of COVID-19? Obes Rev 2020 21 e13083 10.1111/obr.13083 32583537 66. Samuels JD Obesity phenotype is a predictor of COVID-19 disease susceptibility Obesity (Silver Spring) 2020 10.1002/oby.22866 67. Caixe SH Benedeti ACGS Garcia J de Paula MW Mauad Filho F Del Ciampo LA Evaluation of echocardiography as a marker of cardiovascular risk in obese children and adolescents Int J Clin Pediatr 2014 3 72 78 10.14740/ijcp164w 68. Garcia J Benedeti ACGS Caixe SH Mauad F Nogueira-de-Almeida CA Ultrasonographic evaluation of the common carotid intima-media complex in healthy and overweight/obese children J Vasc Bras 2019 18 e20190003 10.1590/1677-5449.190003 31645861 69. Korakas E Ikonomidis I Kousathana F Balampanis K Kountouri A Raptis A Obesity and COVID-19: immune and metabolic derangement as a possible link to adverse clinical outcomes Am J Physiol Endocrinol Metab 2020 319 E105 E109 10.1152/ajpendo.00198.2020 32459524 70. Souza LL Costa-Neto CM Angiotensin-(1–7) decreases LPS-induced inflammatory response in macrophages J Cell Physiol 2012 227 2117 2122 10.1002/jcp.22940 21769868 71. Radzikowska U Ding M Tan G Zhakparov D Peng Y Wawrzyniak P Distribution of ACE2, CD147, CD26, and other SARS-CoV-2 associated molecules in tissues and immune cells in health and in asthma, COPD, obesity, hypertension, and COVID-19 risk factors Allergy 2020 75 2829 2845 10.1111/all.14429 32496587 72. Magrone T Jirillo E Childhood obesity: immune response and nutritional approaches Front Immunol 2015 6 76 10.3389/fimmu.2015.00076 25759691 73. Ryan PM Caplice NM Is adipose tissue a reservoir for viral spread, immune activation, and cytokine amplification in coronavirus disease 2019? Obesity (Silver Spring) 2020 28 1191 1194 10.1002/oby.22843 32314868 74. Elagizi A Kachur S Lavie CJ Carbone S Pandey A Ortega FB An overview and update on obesity and the obesity paradox in cardiovascular diseases Prog Cardiovasc Dis 2018 61 142 150 10.1016/j.pcad.2018.07.003 29981771 75. Lakkis JI Weir MR Obesity and kidney disease Prog Cardiovasc Dis 2018 61 157 167 10.1016/j.pcad.2018.07.005 29981350 76. Sattar N McInnes IB McMurray JJV Obesity is a risk factor for severe COVID-19 infection: multiple potential mechanisms Circulation 2020 142 4 6 10.1161/CIRCULATIONAHA.120.047659 32320270 77. Yan J Grantham M Pantelic J De Mesquita PJB Albert B Liu F Infectious virus in exhaled breath of symptomatic seasonal influenza cases from a college community Proc Natl Acad Sci U S A 2018 115 1081 1086 10.1073/pnas.1716561115 29348203 78. Karlsson EA Hertz T Johnson C Mehle A Krammer F Schultz-Cherry S Obesity outweighs protection conferred by adjuvanted influenza vaccination MBio 2016 7 e01144 e1216 10.1128/mBio.01144-16 27486196 79. Kinsey EW Hammer J Dupuis R Feuerstein-Simon R Cannuscio CC Planning for food access during emergencies: missed meals in Philadelphia Am J Public Health 2019 109 781 783 10.2105/AJPH.2019.304996 30896998 80. Feldstein LR Rose EB Horwitz SM Collins JP Newhams MM Son MBF Multisystem inflammatory syndrome in US children and adolescents N Engl J Med 2020 383 334 346 10.1056/NEJMoa2021680 32598831 81. Sperotto F Friedman KG Son MBF VanderPluym CJ Newburger JW Dionne A Cardiac manifestations in SARS-CoV-2-associated multisystem inflammatory syndrome in children: a comprehensive review and proposed clinical approach Eur J Pediatr 2021 180 307 322 10.1007/s00431-020-03766-6 32803422 82. Gruber CN Patel RS Trachtman R Lepow L Amanat F Krammer F Mapping systemic inflammation and antibody responses in multisystem inflammatory syndrome in children (MIS-C) Cell 2020 183 982 95.e14 10.1016/j.cell.2020.09.034 32991843 83. Wheatland R Molecular mimicry of ACTH in SARS–implications for corticosteroid treatment and prophylaxis Med Hypotheses 2004 63 855 862 10.1016/j.mehy.2004.04.009 15488660 84. Mung SM Jude EB Interplay between endocrinology, metabolism and COVID-19 infection Clin Med (Lond) 2021 21 e499 504 10.7861/clinmed.2021-0200 34507934 85. Martino M Aboud N Cola MF Giancola G Ciarloni A Salvio G Impact of COVID-19 pandemic on psychophysical stress in patients with adrenal insufficiency: the CORTI-COVID study J Endocrinol Invest 2021 44 1075 1084 10.1007/s40618-020-01422-2 32946078 86. Kaiser UB Mirmira RG Stewart PM Our response to COVID-19 as endocrinologists and diabetologists J Clin Endocrinol Metab 2020 10.1210/clinem/dgaa148 87. Raisingani MG Risk of complications in children with adrenal insufficiency and Covid-19 J Endocr Soc 2021 5 Suppl 1 A721 10.1210/jendso/bvab048.1467 88. American Association of Clinical Endocrinologists. AACE position statement: coronavirus (COVID-19) and people with thyroid disease. 2020. https://www.aace.com/recent-news-and-updates/aace-position-statement-coronavirus-covid-19-and-people-thyroid-updated. Accessed 20 Mar 2020. 89. Arlt W Baldeweg SE Pearce SH Simpson HL Endocrinology in the time of COVID-19: management of adrenal insufficiency Eur J Endocrinol 2020 183 G25 32 10.1530/EJE-20-0361 32379699 90. Smith SM Boppana A Traupman JA Unson E Maddock DA Chao K Impaired glucose metabolism in patients with diabetes, prediabetes, and obesity is associated with severe COVID-19 J Med Virol 2021 93 409 415 10.1002/jmv.26227 32589756 91. Wang X Liu Z Li J Zhang J Tian S Lu S Impacts of type 2 diabetes on disease severity, therapeutic effect, and mortality of patients with COVID-19 J Clin Endocrinol Metab 2020 10.1210/clinem/dgaa535 92. Fadini GP Morieri ML Longato E Avogaro A Prevalence and impact of diabetes among people infected with SARS-CoV-2 J Endocrinol Invest 2020 43 867 869 10.1007/s40618-020-01236-2 32222956 93. Pal R Banerjee M Yadav U Bhattacharjee S Clinical profile and outcomes in COVID-19 patients with diabetic ketoacidosis: a systematic review of literature Diabetes Metab Syndr 2020 14 1563 1569 10.1016/j.dsx.2020.08.015 32853901 94. Lim S Bae JH Kwon HS Nauck MA COVID-19 and diabetes mellitus: from pathophysiology to clinical management Nat Rev Endocrinol 2021 17 11 30 10.1038/s41574-020-00435-4 33188364 95. Codo AC Davanzo GG de Brito ML de Souza GF Muraro SP Virgilio-da-Silva JV Elevated glucose levels favor SARS-CoV-2 infection and monocyte response through a HIF-1α/glycolysis-dependent axis Cell Metab 2020 32 437 46.e5 10.1016/j.cmet.2020.07.007 32697943 96. Chatterjee S Ghosh R Biswas P Dubey S Guria RT Sharma CB COVID-19: the endocrine opportunity in a pandemic Minerva Endocrinol 2020 45 204 227 10.23736/S0391-1977.20.03216-2 32548995 97. Armeni E Aziz U Qamar S Nasir S Nethaji C Negus R Protracted ketonaemia in hyperglycaemic emergencies in COVID-19: a retrospective case series Lancet Diabetes Endocrinol 2020 8 660 663 10.1016/S2213-8587(20)30221-7 32621809 98. Lönnrot M Lynch KF Elding Larsson H Lernmark Å Rewers MJ Törn C Respiratory infections are temporally associated with initiation of type 1 diabetes autoimmunity: the TEDDY study Diabetologia 2017 60 1931 1940 10.1007/s00125-017-4365-5 28770319 99. Liu F Long X Zhang B Zhang W Chen X Zhang Z ACE2 expression in pancreas may cause pancreatic damage after SARS-CoV-2 infection Clin Gastroenterol Hepatol 2020 18 2128 30.e2 10.1016/j.cgh.2020.04.040 32334082 100. Cariou B Pichelin M Goronflot T Gonfroy C Marre M Raffaitin-Cardin C Phenotypic characteristics and prognosis of newly diagnosed diabetes in hospitalized patients with COVID-19: results from the CORONADO study Diabetes Res Clin Pract 2021 175 108695 10.1016/j.diabres.2021.108695 33577905 101. Hill MA Mantzoros C Sowers JR Commentary: COVID-19 in patients with diabetes Metabolism 2020 107 154217 10.1016/j.metabol.2020.154217 32220611 102. Luo P Qiu L Liu Y Liu XL Zheng JL Xue HY Metformin treatment was associated with decreased mortality in COVID-19 patients with diabetes in a retrospective analysis Am J Trop Med Hyg 2020 103 69 72 10.4269/ajtmh.20-0375 32446312 103. She J Liu L Liu W COVID-19 epidemic: disease characteristics in children J Med Virol 2020 92 747 754 10.1002/jmv.25807 32232980 104. Dong Y Mo X Hu Y Qi X Jiang F Jiang Z Epidemiology of COVID-19 among children in China Pediatrics 2020 145 e20200702 10.1542/peds.2020-0702 32179660 105. Cengiz E Xing D Wong JC Wolfsdorf JI Haymond MW Rewers A Severe hypoglycemia and diabetic ketoacidosis among youth with type 1 diabetes in the T1D Exchange clinic registry Pediatr Diabetes 2013 14 447 454 10.1111/pedi.12030 23469984 106. Kahkoska AR Shay CM Crandell J Dabelea D Imperatore G Lawrence JM Association of race and ethnicity with glycemic control and hemoglobin A1c levels in youth with type 1 diabetes JAMA Netw Open 2018 1 e181851 10.1001/jamanetworkopen.2018.1851 30370425 107. Gupta R Ghosh A Singh AK Misra A Clinical considerations for patients with diabetes in times of COVID-19 epidemic Diabetes Metab Syndr 2020 14 211 212 10.1016/j.dsx.2020.03.002 32172175 108. Ghosh A Gupta R Misra A Telemedicine for diabetes care in India during COVID19 pandemic and national lockdown period: guidelines for physicians Diabetes Metab Syndr 2020 14 273 276 10.1016/j.dsx.2020.04.001 32283497 109. Lui DTW Lee CH Chow WS Lee ACH Tam AR Fong CHY Thyroid dysfunction in relation to immune profile, disease status, and outcome in 191 patients with COVID-19 J Clin Endocrinol Metab 2021 106 e926 e935 10.1210/clinem/dgaa813 33141191 110. De Vito P Incerpi S Pedersen JZ Luly P Davis FB Davis PJ Thyroid hormones as modulators of immune activities at the cellular level Thyroid 2011 21 879 890 10.1089/thy.2010.0429 21745103 111. Davis PJ Glinsky GV Lin HY Mousa SA Actions of thyroid hormone analogues on chemokines J Immunol Res 2016 2016 3147671 10.1155/2016/3147671 27493972 112. Scappaticcio L Pitoia F Esposito K Piccardo A Trimboli P Impact of COVID-19 on the thyroid gland: an update Rev Endocr Metab Disord 2021 22 803 815 10.1007/s11154-020-09615-z 33241508 113. Dhakal BP Sweitzer NK Indik JH Acharya D William P SARS-CoV-2 infection and cardiovascular disease: COVID-19 heart Heart Lung Circ 2020 29 973 987 10.1016/j.hlc.2020.05.101 32601020 114. 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 115. Chakraborty U Ghosh S Chandra A Ray AK Subacute thyroiditis as a presenting manifestation of COVID-19: a report of an exceedingly rare clinical entity BMJ Case Rep 2020 13 e239953 10.1136/bcr-2020-239953 116. Brancatella A Ricci D Viola N Sgrò D Santini F Latrofa F Subacute thyroiditis after Sars-COV-2 infection J Clin Endocrinol Metab 2020 10.1210/clinem/dgaa276 117. Surgical Affairs Committee. Comment on thyroid surgery during the Covid‐19 pandemic. Updated 7/13/2020. https://www.thyroid.org/covid‐19/comment‐on‐thyroid‐surgery‐during‐pandemic/. Accessed 13 July 2020. 118. Mao L Jin H Wang M Hu Y Chen S He Q Neurologic manifestations of hospitalized patients with coronavirus disease 2019 in Wuhan, China JAMA Neurol 2020 77 683 690 10.1001/jamaneurol.2020.1127 32275288 119. Mussa BM Srivastava A Verberne AJM COVID-19 and neurological impairment: hypothalamic circuits and beyond Viruses 2021 13 498 10.3390/v13030498 33802995 120. Society for Endocrinology. COVID‐19 resources for managing endocrine conditions. https://www.endocrinology.org/clinical‐practice/covid‐19‐resources‐for‐managing‐endocrine‐conditions/. Accessed 30 Aug 2020. 121. European Centre for Disease Prevention and Control. COVID-19 in children and the role of school settings in COVID-19 transmission. https://www.ecdc.europa.eu/en/publicationsdata/children-and-school-settings-covid-19-transmission. Accessed 31 Aug 2020. 122. Carpenter TO Shaw NJ Portale AA Ward LM Abrams SA Pettifor JM Rickets Nat Rev Dis Primers 2017 3 17101 10.1038/nrdp.2017.101 29265106 123. Absoud M Cummins C Lim MJ Wassmer E Shaw N Prevalence and predictors of vitamin D insufficiency in children: a Great Britain population based study PLoS ONE 2011 6 e22179 10.1371/journal.pone.0022179 21799790 124. Reid IR Horne AM Mihov B Stewart A Garratt E Wong S Fracture prevention with zoledronate in older women with osteopenia N Engl J Med 2018 379 2407 2416 10.1056/NEJMoa1808082 30575489 125. Biggin A Briody JN Ormshaw E Wong KKY Bennetts BH Munns CF Fracture during intravenous bisphosphonate treatment in a child with osteogenesis imperfecta: an argument for a more frequent, low-dose treatment regimen Horm Res Paediatr 2014 81 204 210 10.1159/000355111 24356182 126. Stagi S De Masi S Bencini E Losi S Paci S Parpagnoli M Increased incidence of precocious and accelerated puberty in females during and after the Italian lockdown for the coronavirus 2019 (COVID-19) pandemic Ital J Pediatr 2020 46 165 175 10.1186/s13052-020-00931-3 33148304 127. Verzani M Bizzarri C Chioma L Bottaro G Pedicelli S Cappa M Impact of COVID-19 pandemic lockdown on puberty: experience of an Italian tertiary center Ital J Pediatr 2021 47 52 10.1186/s13052-021-01015-6 33673836 128. Turriziani Colonna A Curatola A Sodero G Lazzareschi I Cammisa I Cipolla C Central precocious puberty in children after COVID-19 outbreak: a single-center retrospective study Minerva Pediatr (Torino) 2022 10.23736/S2724-5276.22.06827-6 129. Acar S Özkan B Increased frequency of idiopathic central precocious puberty in girls during the COVID-19 pandemic: preliminary results of a tertiary center study J Pediatr Endocrinol Metab 2021 35 249 251 10.1515/jpem-2021-0565 34881532 130. Mondkar SA Oza C Khadilkar V Shah N Gondhalekar K Kajale N Impact of COVID-19 lockdown on idiopathic central precocious puberty–experience from an Indian centre J Pediatr Endocrinol Metab 2022 35 895 900 10.1515/jpem-2022-0157 35658967 131. Oliveira Neto CP Azulay RSS Almeida AGFP Tavares MDGR Vaz LHG Leal IRL Differences in puberty of girls before and during the COVID-19 Pandemic Int J Environ Res Public Health 2022 19 4733 10.3390/ijerph19084733 35457600 132. Chioma L Bizzarri C Verzani M Fava D Salerno M Capalbo D Sedentary lifestyle and precocious puberty in girls during the COVID-19 pandemic: an Italian experience Endocr Connect 2022 11 e210650 10.1530/EC-21-0650 35029543 133. Umano GR Maddaluno I Riccio S Lanzaro F Antignani R Giuliano M Central precocious puberty during COVID-19 pandemic and sleep disturbance: an exploratory study Ital J Pediatr 2022 48 60 10.1186/s13052-022-01256-z 35461296 134. Hamburg NM McMackin CJ Huang AL Shenouda SM Widlansky ME Schulz E Physical inactivity rapidly induces insulin resistance and microvascular dysfunction in healthy volunteers Arterioscler Thromb Vasc Biol 2007 27 2650 2656 10.1161/ATVBAHA.107.153288 17932315 135. Proper KI Singh AS van Mechelen W Chinapaw MJ Sedentary behaviors and health outcomes among adults: a systematic review of prospective studies Am J Prev Med 2011 40 174 182 10.1016/j.amepre.2010.10.015 21238866 136. Tsai AC Lee SH Determinants of new-onset diabetes in older adults–results of a national cohort study Clin Nutr 2015 34 937 942 10.1016/j.clnu.2014.09.021 25453397	36480134	PMC9734372	NO-CC CODE	2022-12-14 23:28:28	no	World J Pediatr. 2022 Dec 8;:1-12	utf-8	World J Pediatr	2,022	10.1007/s12519-022-00662-x	oa_other
==== Front Adv Fiber Mater Adv Fiber Mater Advanced Fiber Materials 2524-7921 2524-793X Springer Nature Singapore Singapore 237 10.1007/s42765-022-00237-5 Review A Review on Electrospinning as Versatile Supports for Diverse Nanofibers and Their Applications in Environmental Sensing Song Jialing 13Jialing Song is now an enrolled postgraduate student in the College of Environmental Science and Engineering, Donghua University. She has joined the group of professor Manhong Huang since 2017. In 2021, she joined Prof. Li's research group at the Department of Chemistry, National University of Singapore (NUS) through China Scholarship Council. Her main research direction is the nanomaterials, environmental analysis and sensing, and membrane separation technique. Lin Xuanhao 3Xuanhao Lin is research fellow of analytical science at National University of Singapore. He is a graduate of National University of Singapore and Shandong University. Prior to joining National University of Singapore, Dr. Lin worked for Hewlett-Package on thermal inkjet printing technology and Institute of Materials Research & Engineering on nanotechnology. His current research interests include chemical sensors, biosensors, and quantum dot bioconjugates for imaging, labeling and sensing, as well as photocatalytic remediation of wastewater. Ee Liang Ying 3Liang Ying Ee received his accelerated BSc (Hons.) in Chemistry from the National University of Singapore (NUS) in 2017. He has gained significant industrial work and research experience from Hyflux Pte. Ltd. and Agency for Science, Technology and Research (A*STAR), respectively. In August 2018, he was awarded the NUS Research Scholarship administered by the Singapore Ministry of Education to undertake the PhD program under Prof. Li at NUS. His research area involves the extraction of nanocellulose from agricultural waste and 3D printing of nanocellulosic desalination membrane. Li Sam Fong Yau 35Sam Fong Yau Li is working as a Full Professor at the Department of Chemistry, National University of Singapore (NUS). He received both his BSc (Hons.) and Phd in Chemistry from the Imperial College and DSc from the University of London. Prof. Li has particular expertise and research interest in nanomaterials, 3D printing, environmental analysis and sensing, environmental remediation technologies, capillary electrophoresis, and metabolomics. To date, he has authored/co-authored over 400 papers in peer-reviewed journals (h-index = 56), with numerous papers on the preparation, analysis, and environmental toxicity of nanomaterials that are closely relevant to the reviewed field. Huang Manhong [email protected] 124Manhong Huang is working as a Full Professor in the College of Environmental Science and Engineering, Donghua University. She received her PhD in the College of Environmental Science and Engineering, Tongji University, China. In 2011 and 2013, she went to the Department of Environment of Tsinghua University and Geogia Tech West Institute for exchange study, respectively. Prof. Huang has particular expertise and research interest in nanomaterials, environmental analysis and sensing, MFC, and membrane technology. To date, she has authored/co-authored over 100 papers in peer-reviewed journals and many researches have been transformed and applied in the field of practical environment. 1 grid.255169.c 0000 0000 9141 4786 College of Environmental Science and Engineering, Key Laboratory of Science and Technology of Eco-Textile, Ministry of Education, Donghua University, Shanghai, 201620 People’s Republic of China 2 grid.24516.34 0000000123704535 Shanghai Institute of Pollution Control and Ecological Security, Shanghai, 200092 People’s Republic of China 3 grid.4280.e 0000 0001 2180 6431 Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore, 117543 Singapore 4 grid.255169.c 0000 0000 9141 4786 State Key Laboratory for Modification of Chemical Fibers and Polymer Materials, Donghua University, Shanghai, 201620 People’s Republic of China 5 grid.4280.e 0000 0001 2180 6431 National University of Singapore Environmental Research Institute, T Lab Bldg, 5A Engineering Drive 1, Singapore, 117411 Singapore 5 12 2022 132 19 8 2022 13 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. Rapid industrialization is accompanied by the deterioration of the natural environment. The deepening crisis associated with the ecological environment has garnered widespread attention toward strengthening environmental monitoring and protection. Environmental sensors are one of the key technologies for environmental monitoring, ultimately enabling environmental protection. In recent decades, micro/nanomaterials have been widely studied and applied in environmental sensing owing to their unique dimensional properties. Electrospinning has been developed and adopted as a facile, quick, and effective technology to produce continuous micro- and nanofiber materials. The technology has advanced rapidly and become one of the hotspots in the field of nanomaterials research. Environmental sensors made from electrospun nanofibers possess many advantages, such as having a porous structure and high specific surface area, which effectively improve their performance in environmental sensing. Furthermore, by introducing functional nanomaterials (carbon nanotubes, metal oxides, conjugated polymers, etc.) into electrospun fibers, synergistic effects between different materials can be utilized to improve the catalytic activity and sensitivity of the sensors. In this review, we aimed to outline the progress of research over the past decade on electrospinning nanofibers with different morphologies and functional characteristics in environmental sensors. Graphical Abstract Keywords Electrospinning Nanofibers Environmental sensing Functional nanomaterials the National Key Research and Development Project2019YFC0408304 Huang Manhong the Fundamental Research Funds for the Central UniversitiesNo. 2232022G-04 Huang Manhong the Fundamental Research Funds for the Central Universities and Graduate Student Innovation Fund of Donghua UniversityCUSF-DH-D-2021037 BCZD2022005 Song Jialing China Scholarship Council202006630085 Song Jialing ==== Body pmcIntroduction Rapid global industrialization and urbanization not only have resulted in significant economic benefits, but also led to resource depletion and grave environmental pollution problems, including air, soil, water, and energy (light, sound, heat, and radioactivity) pollution [1]. The large amounts of flammable, toxic, and harmful gases emitted from intensive industrial production and urbanization exacerbate air pollution. Motor vehicle emissions, comprising a large amount of greenhouse gases, toxic nitrogen oxide, sulfur oxide, and other gases, are one of the leading reasons for air pollution [2, 3]. Many indoor decoration materials are not environmentally-friendly and produce large amounts of carcinogenic formaldehyde, toluene, and other volatile solvent vapors. These gaseous pollutants not only cause global warming, haze, acid rain, and other environmental pollution problems, but also acutely jeopardize human health [3]. In the aquatic environment, pollution is aggravated by a large amount of improperly discharged wastewater, including wastewater contaminated by antibiotics from pharmaceutical companies and livestock farms, heavy metals from industrial manufacturing processes, and pesticides from agriculture and horticulture. Numerous harmful pollutants in the aquatic environment can easily enter the human body through several pathways, especially through the food chain where they can cause different food-borne diseases and even the formation of "super bacteria" [4–6]. Agricultural pollutants, and their by-products produced in soil, can cause serious damage to human health after entering the food chain. Among the many food safety problems caused by soil pollution, contamination by heavy metals, such as Pb, As, Cr, and Hg, is the most serious. Therefore, the development of sensor technologies with superior performance metrics, such as high sensitivity, rapid response, and good selectivity and stability, is crucial; however, it remains a major challenge. Recently, there has been increasing interest in the use of chemical sensors for the analysis of environmental contaminants. Owing to the benefits of easy sample preparation, miniaturization, portability, and the low cost of producing the new chemical sensors, these sensors can complement or even replace classical analytical instruments in real-time on-site or online detection [7–10]. The type and structure of the sensing material are pivotal to the performance of the chemical sensing technology. Previously, a strategy has been proposed to improve the performance of a sensor by enhancing the transfer capability between the sensing material and the target analytes by adjusting the specific surface area of the sensing material [11]. In this context, nanomaterials play an important role in the field of sensor technology owing to their unique dimensional properties. Furthermore, sensors fabricated using diverse nanomaterials with tunable sizes and surface functional groups have been reported to remarkably increase sensitivity and expand the range of target contaminants. Nanomaterials are widely recognized for their excellent prospects in the development of advanced sensing technologies [11]. By adapting and optimizing electrospinning, nanofibers with unique structures, morphologies, and functions can be produced. Therefore, electrospinning technology has been recognized as a highly promising method in the design and development of nanomaterial-based ultra-sensitive sensor systems [12–14]. During electrospinning, a polymer solution or melt is charged and deformed by a high-voltage electrostatic field, forming a pendant cone-shaped droplet at the end of a nozzle [13, 15]. When the charge repulsion on the surface of the droplet exceeds its surface tension, tiny jets of the polymer fluid, referred to as "jets", are ejected at a high speed from the surface of the droplet towards the plate electrode at a relatively short distance to form the polymer fibers [16, 17]. When the diameters are scaled down from micron to sub-micron or even nanometer, polymer fibers exhibit several unique properties. Conventional preparation methods for nanofibers include stretching, template synthesis, phase separation, and self-assembly [5, 12, 18, 19]. The diameters of fibers prepared using conventional approaches are in hundreds of microns, whereas those obtained by electrospinning are 2-3 orders of magnitude smaller. The specific surface area of conventional fibers is normally 0.4 m2 g−1, while that of electrospun fibers is typically 40 m2 g−1 [18, 19]. Electrospinning substantially increases the surface area-to-volume ratio of the nanomaterials (up to 1,000-fold higher for ultrafine fibers) [13]. Moreover, the mechanical strength, hydrophilicity/hydrophobicity, electrical conductivity, and flexibility of nanomaterials can be tailored as required via polymer selection, concentration modulation, optimization of electrospinning parameters, and chemical modification through functionalization [13, 20, 21]. There are two main approaches for the preparation of sensors by electrospinning. The first approach produces nanofibers with direct sensing functions by electrospinning functional polymers, such as polyacrylic acid and polyacrylonitrile (PAN) and utilizing these nanofibers directly as the sensing elements of the sensors [13, 20]. The second approach uses electrospun nanofibers as stencils and deposits responsive sensing materials with surface functionalization on the fiber surface to create micro- and nanostructures with selective sensing properties [22]. Nanofibers prepared via electrospinning generally exhibit a large specific surface area with most atoms located on the surface or within interfacial regions, exhibiting high chemical reactivity owing to their unique and complex surface structure [20]. One-dimensional nanomaterials prepared via electrospinning are not limited to nanometer size and can have a high aspect ratio for the fast transfer of electrons [13, 14]. Therefore, the introduction of nanomaterials with a high specific surface area during the design of sensing materials favors both sensitivity enhancement and response time reduction. Compared with the hydrothermal and templating methods, the electrospinning technology is more flexible for nanomaterial preparation. It can regulate the morphology, structure, composition, and function of the nanofibers in a diverse and multifaceted manner via polymer concentration, co-blending of different polymers, and chemical cross-linking between polymers [14, 20]. The electrospinning technology enables the continuous preparation of nanomaterials with different structures and morphologies, including nanofibers, nanowires (nanorods), nanoribbons, hollow nanofibers, core-shell structures, hollow nanotubes, nanodendrites, etc. [22–30]. In addition, nanomaterials prepared via electrospinning exhibit high axial strength and can achieve continuous electron transfer, which facilitates charge transfer along the long-axis direction and enables higher sensitivity for sensors fabricated in this way [13, 15, 31]. Electrospinning is facile, and the derived materials are highly reproducible and scalable for industrial production. Owing to these characteristics, the electrospun nanofibers provide greater selectivity towards environmental sensing. In this review, we sought to provide an overview of the applications of electrospinning in the field of environmental sensing over the past decade. This review further discusses progress from multiple perspectives on environmental sensing by comparing and summarizing the morphology, function, and composition of nanomaterials prepared using the electrospinning technology. Electrospinning and Nanofibers Background Electrospinning is a technology in which polymer solutions are sprayed and stretched under the action of static electricity to obtain nanoscale fibers. The electrospinning equipment is mainly composed of three parts: a high-voltage power supply, micro-injection pump and nozzle, and fiber-receiving substrate. A high-voltage direct current (several thousands to tens of thousands volts) is applied to the polymer solution. When the electric field force is sufficiently large, the charged polymer droplets overcome the surface tension to form a jet stream that evaporates or solidifies with the solvent during the jetting process [13, 32]. Finally, it falls on the collection substrate to form a fiber mat that is similar to a non-woven fabric. The diameters of the nanofibers formed range between tens of nanometers (nm) and micrometers (μm). Generally, the fiber membranes collected on the collector substrate are composed of randomly oriented and entangled fibers. With a specially designed collector substrate, fibers with high strength and toughness can be obtained in highly aligned patterns. Special collectors (Fig. 1), such as high-speed rotating drum collectors, parallel collectors, and tip collectors, are suitable for obtaining aligned fibers [16, 33, 34].Fig. 1 Special collectors for obtaining aligned fibers: a high-speed rotating drum collectors; reproduced with permission from ref. [16], Copyright 2018, Elsevier. b Tip collectors; reproduced with permission from ref. [33], Copyright 2003, American Chemical Society. c Parallel plates collectors; reproduced with permission from ref. [34], Copyright 2010, Elsevier Traditional electrospinning equipment commonly adopts a single capillary nozzle and is generally employed to prepare solid and smooth nanofibers. However, only nanofibers composed of a single material can be obtained from the traditional equipment, and hence accompanied with several shortcomings that include insufficient surface specificity and poor mechanical properties. It is difficult to prepare composite materials with multiple functional structures. Although these materials can be adopted for a wide range of applications [12, 35]. To overcome these shortcomings, researchers have improved electrospinning via three strategies: (1) surface modification of nanofibrous membranes, (2) blending multiple functional materials, and (3) coaxial electrospinning technique. These strategies proved convenience in preparing continuous core-shell nanofibers or hollow-structured nanofibers in a single step. In the coaxial electrospinning process, mixing does not occur before solidification due to the short confluence time and low diffusion coefficient of the two spinning solutions at the capillary outlet. When a high-voltage electric field is applied to the liquids in the inner and outer orifices of the capillary, the charges of the solution in the inner orifice gradually migrate to the solution surface in the outer orifice [32]. Increase in voltage, and electric field, results in a gradual increase in the amount of charge on the solution surface in the outer orifice [12, 36]. When the charge reaches a certain limit, its repulsive force causes the outer solution to form a composite Taylor cone at the electrospinning nozzle followed by a coaxial composite structure with a core-shell double layer. A hollow nanofiber is formed when the core (inner) layer is removed through heating or dissolution in suitable solvent system. Key Regulating Factors in the Electrospinning Process Characteristics of the Electrospinning Solution Polymer: Electrospinning has been recognized as a low-cost, convenient, and environmentally friendly method that is widely adopted to produce nanofibers with various morphologies and functions [12]. Water-soluble polymers, such as poly (vinylpyrrolidone), polyethylene oxide, poly (vinyl acetate), and polyvinyl alcohol (PVA), and solvent-soluble polymers, such as polyimide, PAN, polyvinylidene fluoride (PVDF), poly (methyl methacrylate), and polystyrene (PS) are the most common polymers used in electrospun nanofibers. Notably, the addition of multifunctional materials, such as metal oxides, inorganic non-metals, metal–organic frameworks, and covalent organic frameworks, to polymers allows the preparation of composite nanofibers with different morphologies, structures, and functionalities under controlled conditions [37–39]. Conductivity: The conductivity of the electrospinning solution is another key parameter that affects the morphology of electrospun fibers and contributes to jet formation. The addition of salts, ionic polymers, or conductive polymers to the electrospinning solution increases the solution conductivity and supports the formation of high-quality fibers with fewer defects and smaller diameters [40]. Surface Tension: In electrospinning, a trickle is ejected when the charged solution overcomes its surface tension. Generally, high surface tension in a liquid reduces its surface area, maintains its spherical shape, and is the predominant contributing factor towards the formation of liquid beads [40]. Therefore, the surface tension of the electrospinning solution is minimized to enlarge its surface area as much as possible to prevent the formation of liquid beads and reduce the fiber diameter. Routine approaches to reduce the surface tension of the solution include (1) the use of a solvent with low surface tension, such as ethanol or acetone, and (2) the addition of surfactants into the electrospinning solution. Concentration and Viscosity of the Electrospinning Solution: Low polymer concentration typically results in low viscosity that causes the jet to be unstable and discontinuous. Such a low concentration also results in the formation of beaded nanofibers in which this phenomenon is known as the "flying filaments". Viscosity increases with increasing polymer concentration [41]. However, the flow of the solution is hindered when the polymer concentration exceeds a certain limit, causing nozzle blockage or what some authors refer to as "non-spinning" [13]. Equipment Parameters Electric Field Intensity: Low voltage applied cannot overcome the surface tension of the electrospinning solution due to a low electric field, which hinders the stretching and splitting of the solution, resulting in larger nanofiber diameters [35]. As the electric field strength increases, higher surface charge density and electrostatic repulsion accumulate on the jet of the polymeric electrospinning solution. Simultaneously, a higher electric field strength accelerates the jet to a higher velocity. These factors contribute to the formation of fibers with greater tensile stress and finer textures. However, a high electric field strength over a certain limit induces a very fast jet flow and hampers its stretching and splitting, which consequently results in larger fiber diameters, worse uniformity, and beaded fibers [42–44]. Needle Pitch: The distance between the electrospinning needle and collector affects the electric field strength and volatilization of the electrospinning solution, thereby altering the diameter of the nanofibers. When the voltage is kept constant, the electric field strength is inversely proportional to the distance. In a typical electrospinning setup, the distance ranges from 10 to 15 cm, which usually allows sufficient evaporation of the solvent from the jet before being deposited as a dried fiber bundle. If the distance is too short to allow sufficient evaporation of the solvent of the jet, unfavorable fused fibers may be formed [44, 45]. Flow Rate: The electrospinning rate is controlled by the flow rate of the syringe. The electrospinning rate not only governs the production of high-quality nanofibers, but more importantly influences the jet stability and fiber diameter, which increases with decreasing electrospinning speed [32, 46]. Environmental Factors Environmental temperature and humidity in the electrospinning process have a significant impact on the evaporation rate of jet solvents and the morphology and diameter of the produced fibers. Temperature is known to affect solvent evaporation, Brownian motion, and the diffusion of ions, atoms, molecules, and particles. The increased temperature of the electrospinning solution reduces its viscosity and elevates its solvent evaporation and Brownian motion, ultimately altering the fiber diameter [47]. Humidity also changes the volatilization rate of solvents in the spinning solution, affecting the void distribution on the surface of the electrospun nanofibers. Several studies have revealed a relatively smoother surface of electrospun fibers produced with less than 25% humidity and a rougher surface with more pores on the fiber surface as humidity exceeds 30%. Notably, very low humidity speeds up the solvent evaporation of a jet and increases its viscosity, both of which may cause nozzle clogging at the capillary tip. In contrast, high humidity aggravates the formation of dangerous conducting air and slows solvent volatilization [38, 48]. Diversity of the Fiber Structure The structure of electrospun fibers is one of the factors governing their performance. Fibers fabricated using multiple electrospinning techniques produce different types of structures and morphologies than those using a single type of electrospinning. Fibers with unique structures, morphologies, and functions can be produced by selecting and optimizing the suitable type of electrospinning technique. Individual Single Types of Electrospinning Beaded Fiber: Beaded fibers are formed by unstable fluid jets under an external electric field. Owing to their unique structure, they can be applied in different applications. Moreover, it has become possible to carry out the corresponding theoretical modelling on the formation of beaded fibers with the rapid advancement in electrospinning technology in the recent years [30, 49, 50]. Core-Shell/Hollow Nanofiber: Electrospun core-shell/hollow nanofibers have been widely assessed and applied by many researchers in sensing systems. These nanofibers can be produced by coaxial or emulsion electrospinning. In the former process, the outer and inner electrospinning solutions are delivered to two capillaries with different inner diameters. Thereafter, they combine to form a coaxial Taylor cone followed by coaxial fluid jets in a high-voltage electrostatic field, of which the jets solidify into core-shell nanofibers with different interior and exterior contents. If the core layer is removed via heating or dissolution, hollow nanofibers are formed with a hollow inner lumen surrounded by an intact outer layer. Prior studies revealed that polymeric electrolytes based on ionic liquids do not only have high conductivity, but also excellent mechanical properties [51–53]. Ionic gels with high capacitance can be used as thin films to prepare capacitive sensors [54]. However, the surface of a hydrogel (ionic gel) is prone to dehydration. Therefore, a protective mechanism must be developed to prevent water from evaporating from its surface [55]. In addition, surface area affects the capacitance of an ionic hydrogel matrix film and limits the sensitivity and stability of its capacitive sensor device. Lin et al. [29] overcame the above-mentioned problems using the coaxial electrospinning method. An ionic gel was employed as the core with the copolymer PVDF as the outer shell to produce a core-shell nanofiber mat. The advantages of a co-polymer shell include its larger surface area and sealed protective layer that effectively prevents dehydration of the core-shell ionic gel. Liu et al. [56] developed a chemical sensor based on the core-shell structured semiconductor nanofibers through rapid single-nozzle electrospinning process and spontaneous phase separation. The rich active sites of the core-shell structured nanofibers effectively promote interactions between charge carriers in their conductive channels and chemical analytes, and change the output current and charge carrier mobility that lead to a high sensitivity in the resulting sensor. The shell thickness in a core-shell structure can be adjusted by changing the ratio of the outer and inner electrospinning solutions in the mixed solution. A nanofiber-based sensor with an appropriate core-shell ratio exhibits high sensitivity toward low concentrations of ammonia (the limit of detection was as low as 50 ppb). Hollow nanofibers with porous surfaces and tubular structures also exhibit outstanding properties. Electric current can easily flow around their disturbance zone and target contaminants can freely diffuse in and out of them, thereby shortening the response and recovery time [57]. Park et al. [28] took advantage of the Kirkendall effect to fabricate porous hollow nanofibers of SnO2-CuO nanocomposite by single-needle electrospinning. The Kirkendall effect arises from the supersaturation of lattice vacancies caused by the different diffusivities of distinct atoms. A nanotube structure is formed after calcination, which is attributed to the Kirkendall effect [58]. A fast response was achieved with the prepared SnO2-CuO hollow nanofiber-felt for the detection of H2S gas with a rapid detection time of 5.27 s and a low concentration range of 1–40 ppm. Porous Fiber: Emerging porous materials generally have channels, pores, and gaps on their surfaces or interiors. These materials also usually have the characteristics of low density, high specific surface area, high porosity, and high adsorption capacity and thus widely adopted in sensors, filtration, catalysis, and other applications [59–61] reported that the calcination of electrospun PVA composite nanofibers containing silica sol nanoparticles at 450 °C produced porous inorganic nanofibers with high specific surface areas. Liu et al. [62] prepared porous alumina nanofibers with hollow structures by sintering aluminum-nitrate/PAN nanofibers through high-temperature calcination. Ma et al. [63] prepared the porous PAN nanofibers by washing electrospun PAN/NaHCO3 composite nanofibers with a 10% hydrochloric acid solution to dissolve and remove NaHCO3, and the escaping CO2 gas "blown" nanopores in the PAN nanofibers. The electrospun porous nanofibers also had discontinuous pore structures in the interior or open pore structures on the surface. The porous structure of nanofiber materials provides superior properties, including large specific surface areas and reduced thermal conductivity, which further broadens their application range. Others: Electrospinning can be employed to generate fibers of various shapes, such as beaded, core-shell, ribbon, porous, spiral, square, and honeycomb (Fig. 2). Zhang et al. [22] prepared membranes with electrospun ribbon nanofibers and suggested that the formation of a fiber-ribbon shape could be attributed to the positive charge of the polymer precursor. According to the study, a jet carrying negative charge on its surface attracts the positively charged polymer precursor to its surface during the electrospinning process, leaving its center solvent-rich and forming a tube-shaped structure with a thin outer layer. When the residual solvent in the tube center evaporated, a partially collapsed hollow tube was formed with wrinkles on the surface. When the hollow tube collapsed completely, a ribbon-like structure was formed. Shin et al. [26] reported a facile process for producing electrospun spiral nanofibers with a single polymer and demonstrated the transformation of nanofibers from spiral to linearly oriented with an adjustment in electric field strength.Fig. 2 Various morphologies of electrospun fibers: a beaded fibers; reproduced with permission from ref. [30], Copyright 2015, Wiley. b Core-shell fibers; reproduced with permission from ref. [29], Copyright 2018, Elsevier. c Hollow fibers; reproduced with permission from ref. [28], Copyright 2020, Elsevier. d Porous fibers; reproduced with permission from ref. [27], Copyright 2008, American Chemical Society. e Ribbon fibers; reproduced with permission from ref. [22], Copyright 2019, Wiley. f Spiral fibers; reproduced with permission from ref. [26], Copyright 2006, American Institute of Physics. g Random fibers and h aligned fibers; reproduced with permission from ref. [25], Copyright 2016, American Chemical Society. i Square-patterned fibers; reproduced with permission from ref. [24], Copyright 2021, Wiley. j Honeycomb-patterned fibers; reproduced with permission from ref. [23], Copyright 2021, Elsevier Aggregate Electrospinning Structure Figure 2g–j shows the great diversity in the structures and morphologies of electrospun fiber membranes. The structural diversity mainly depends on the collector type, which is one of the most important parts of the electrospinning equipment and determines the distribution and macrostructure of the electrospun fibers. Other aspects of the processing equipment, such as the electric field strength, presence of a magnetic field, and composition of the electrospinning solution, can be optimized to obtain the desired fiber mat structures. The collected electrospun fibers or fiber mats are generally divided into three groups: random fibers, arranged fibers, and pattern fibers. Random Electrospinning: Random fibers are the most common form of electrospun fiber collection. Randomly oriented fibers can be obtained simply using a traditional electrospinning collector. Charged polymer jets naturally travel along spiral trajectories during the electrospinning process and finally solidify into randomly oriented continuous fine fibers on a ground plate substrate. Oriented Electrospinning: Regularly oriented electrospun nanofibers can be prepared using a special collector and electrodes. Park et al. [25] prepared a hybrid scaffold with a dual configuration of aligned and random electrospun fibers. Wang et al. [62] prepared ordered parallel arrays of PVA nanofibers using similar hybrid scaffolds and successfully applied them to the detection of low concentrations of NO2. Patterned Electrospinning: Patterned electrospun nanofibers can be produced using appropriate solutions and a special collecting device. Liang et al. [24] prepared an oriented square microfiber mat using an electrostatic template for nanofibers deposition on a pattern collector made of regularly distributed protrusions. Gangolphe et al. [23] developed honeycomb electrospun scaffolds and successfully applied them to prepare honeycomb fibrous membranes with appropriate macropore size and fiber arrangement. These researchers also assessed the anisotropic and mechanical properties, and the degradation mechanism of honeycomb membranes. Application of Electrospinning in Environmental Sensing Electrospun sensing and interface materials have gradually been adopted in the construction of various sensors, mainly using the following characteristics: (1) nanofibrous membrane materials used as sensors with a range of specific surface areas that can be obtained by adjusting the fiber geometry and pore size; (2) the specificity and diversity of fiber materials offer different characteristics and a wide range of applications; (3) the strong molecular recognition ability of electrospun sensors eliminates tedious sample pre-treatment, and their reusability leads to simple and rapid detection processes; (4) the sensor interface has high sensitivity and quick response to a small quantity of samples; (5) electrospun sensors have good repeatability, stability, and reusability; and (6) their detection cost is low. More specifically, the high surface area of electrospun nanofibers offers more active adsorption sites for target analytes to be adsorbed on the sensing interfaces. In addition, the adsorption and desorption rates of electrospun nanofibers are significantly accelerated for rapid sensing and improved sensitivity [64]. According to prior studies, the enhanced sensing rate can be attributed to the high surface-area-to-volume ratio [65]. The Debye lengths of the nanofibers, which refers to the distance over which significant charge separation can occur, are equivalent to their diameters. Charge accumulation/consumption areas, where the adsorption of analyte molecules occurs, extend from their surface into their interior and seriously affect current flow. Compared with two-dimensional thin films, the current in 1D nanofibers can still flow easily around the disturbed area [28, 57]. Therefore, their response is fast and the recovery time is short when the target molecules quickly diffuse into the sensing material. In addition to polymers, several other functional materials such as metals and organic/inorganic materials can be included into the electrospinning solutions to prepare multifunctional electrospun nanofibers in sensors. Gas Sensor With accelerated industrialization, the global air pollution problem has become increasingly serious. Air pollutants contain different toxic gases, such as CO, SO2, NO2, and H2S, and volatile organic compounds (VOCs), such as benzene, toluene, and formaldehyde. These hazards can cause respiratory system damage, physiological dysfunction, digestive system disorders, nervous system abnormalities, mental decline, carcinogenesis, and disability. High concentrations of air pollutants can cause acute poisoning or even diseases [66–68]. According to the "2019 Global State of the Air" report recently released by the US Institute of Health Effects (based on 2017 data), the number of people who died of stroke, heart disease, lung cancer, diabetes, and chronic lung disease worldwide due to air pollution reached nearly 5 million in 2017. Notably, severe air pollution can even cause abnormalities in the components of pulmonary surfactants or even lung damage, making humans more susceptible to diseases, including COVID-19 [69]. Air pollution has become an urgent global problem, and authorities in countries and organizations have specified short-term exposure limits (STEL) for various toxic gases and VOCs [70, 71]. Environmental protection first requires the establishment of an environmental supervision mechanism; gas sensors, as one of the essential sensors for environmental supervision, help the set-up of an environmental Internet of Things [72]. Air quality affects our daily lives, and effective actions must be taken to ensure good air quality. Continuous monitoring of pollutant gases in the air through gas sensors can effectively identify the potential risks of air pollution to human health. Over the past 5 years, many gas sensors with excellent sensitivity, stability, reversibility, response time, and reproducibility have been reported. These gas sensors include resistive gas sensors, quartz crystal microbalance (QCM) gas sensors, and optical sensors with electrospun nanostructured layers composed of polymers and functional TiO2, SnO2, ZnO, In2O3, Fe2O3, WO3, or other metal oxides [73–75]. In the following sections, we aimed to outline the recent progress and development in this emerging field and highlight the sensing layer microstructures and gas-sensitive properties of electrospun gas sensors. Resistive Gas Sensor Compared with traditional gas detection equipment, portable and inexpensive gas sensors have outstanding advantages in terms of shorter detection times, without sacrificing detection accuracy. Resistive gas sensors are one of the most common types of gas detection sensors [76–79]. A resistive sensor is based on changes in the electrical resistance or conductivity caused by the adsorption of gas molecules. Therefore, the response rate depends on the interaction speed and strength between the target gas molecules and the sensor material surface, and the specific surface area available for adsorption and desorption [80]. In an oxygen-rich environment, oxygen molecules adsorbed onto sensor surfaces extract electrons from the conduction band and trap them as negatively charged ions on the surface, forming different surface energy levels. Accordingly, an electron depletion layer and a hole accumulation layer are formed on the sensor surface for the n- and p-type materials, respectively. When switched to a target gas-rich environment, the target gas molecules are adsorbed onto the sensor surface and exchange electrons with it, causing a conductivity change in the sensing material. The resistance/conductivity of gas sensors depends on the charge transfer mechanism between the adsorbed gas molecules and gas-sensitive adsorbents, as well as their surface reactions. When an n-type sensing material is exposed to oxidizing gas, its resistance increases, whereas exposure to reducing gas decreases its resistance [81, 82]. Figure 3 shows the principle of the gas-sensing mechanism of a resistive gas sensor based on the n-type and p-type sensing materials [83].Fig. 3 Schematic illustration of basic sensing principles for a n- and b p-type MOS gas sensors; reproduced with permission from ref. [84], Copyright 2014, Elsevier According to the configuration, resistance gas sensors can be further divided into ceramic tubular, flat, and flexible sensors. The structure and physical diagrams of various resistance gas sensors are shown in Fig. 4. For a long time, the ceramic tubular sensor was the main type owing to its simple preparation process and low cost [85, 86]. As shown in Fig. 4a, the ceramic tube is composed of four parts: ceramic tube, Ni-Cr heater, Au electrode, and Pt wire. The heater is placed in a ceramic tube to maintain the working temperature. Two rings of gold electrodes are placed on the outer wall of the ceramic tube, on top of which is the deposited sensing material for electrical signal production.Fig. 4 Schematic and optical images of various resistive gas sensor configurations: a tubular; reproduced with permission from ref. [87], Copyright 2020, Elsevier. b Plate-like; reproduced with permission from ref. [88], Copyright 2020, Elsevier. c Wearable gas sensor; reproduced with permission [89], Copyright 2017, Tsinghua University Press When the sensing interface of the gas sensor is porous, the target gas molecules can penetrate the entire sensing layer, thereby improving sensitivity. In addition, a large specific surface area facilitates the interaction between the gas molecules and the sensing material. Commonly used chemical solution deposition technology and ceramic slurry sintering technology for preparing gas-sensitive materials can hardly meet the requirements of porosity and large specific surface area at the same time. However, it is difficult to uniformly coat gas-sensitive materials onto the ceramic tube surface using traditional techniques. Hence, they easily aggregate, their pore sizes are not uniform, and gas molecules at their sensing interface cannot be well diffused. In contrast, sensing layers prepared via electrospinning can fulfill the requirements of both high porosity and large specific surface area, and nanofibers have controllable fiber and pore sizes. For different gases, sensor layers with suitable pore sizes can be developed to effectively overcome the gas diffusion limitation, enabling unimpeded penetration of the sensor layer by gas molecules. Han et al. [90] used atomic layer deposition to deposit n-type ZnO on the surface of CuO-polymer hybrid nanofibers. Subsequently, the polymer molecules were removed via calcination to form hollow p-CuO/n-ZnO nanofibers (Fig. 5a, b). The optimal responses of the gas sensor based on the p-CuO/n-ZnO heterostructure were approximately 6- and 45-fold higher than those of pure ZnO and CuO, respectively. Similarly, Zhang et al. [91] reported the use of electrospinning to prepare nanofibers with hierarchical porous heterostructures (p-type Co3O4 and n-type In2O3). Based on their findings, Co2+/Co3+ with high catalytic activity diffused in In2O3 with uniform distribution (Fig. 5c). Their results revealed that the gas sensor had a remarkably high response to VOCs, including acetone and formaldehyde. In addition to the heterogeneous structure, the size of the sensing material has a significant impact on its sensitivity. Li et al. [92] successfully prepared electrospun ZnO/ZnFe2O4/Au nano-mixtures with controllable particle sizes using a combination of electrospinning and atomic layer deposition techniques (Fig. 5d). Their results showed that the response of the prepared ZnO/ZnFe2O4/Au nanocomposite to acetone was 3- and 5.5-fold higher than that of the ZnO/ZnFe2O4 composites and ZnO, respectively. More interestingly, Zhang et al. [93] synthesized Cr2O3-TiO2 core-shell fibers with different shell thicknesses through a simple coaxial electrospinning technology and discovered that the shell thickness affected the gas sensor performance. As the TiO2 shell thickness increased, the response characteristics of the prepared core-shell fiber gas sensor to acetone showed a transition from the p-type to n-type. In addition, the formation of a Cr2O3-TiO2 heterojunction in the gas sensor enabled a quicker response to the target gas molecules and a shorter recovery time.Fig. 5 a Schematic illustration of 1D p-CuO/n-ZnO hollow nanofibers synthesis using a three-step method and b XRD, SEM, and TEM images of 0.3CuO/ZnO200, 0.3CuO/ZnO400 and 0.3CuO/ZnO600 HNFs; reproduced with permission from ref. [90], Copyright 2019, Elsevier. c SEM and XRD pattern of In2O3-Co3O4 ribbon; reproduced with permission from ref. [91], Copyright 2020, Elsevier. d SEM and TEM images of pure ZnO hollow nanomeshes and ZnO/ZnFe2O4 composites; reproduced with permission from ref. [92], Copyright 2019, Elsevier In flat-type sensors, sensing materials are often deposited on the insulating substrate surface of an interdigital electrode (IDE) as shown in Fig. 4b. IDE sensors are favored in the fields of biomedicine and the environment owing to advantages such as miniaturization, low-cost, and large-scale production [94, 95]. Bulemo et al. [96] suggested that gas analytes diffusing into semiconductor metal oxide-based sensing layers require gas-sensing materials to be thin and porous, and exhibit high performance. In their study, hollow SiO2-SnO2 core-shell microstrips were prepared by etching away the SiO2 core with NaOH solution (pH 12) from electrospun calcined highly porous SiO2-SnO2 core-shell microstrips (Fig. 6a, b). After sensitizing with platinum nanoparticles (Pt NPs), the prepared hollow nanofibers were used to detect acetone, exhibiting a highly stable response (Ra/Rg = 93.70 ± 0.89) and significant selectivity to as low as 2 ppm acetone in air. Electrospinning has also been employed as an innovative functionalization method for manufacturing polymer nanofibers with a diameter range of 50–500 nm and a nano-network on the movable plate of a capacitive micromachined ultrasonic transducer (CMUT). Zhao et al. [97] reported a CMUT-based resonant biochemical sensor that was prepared to detect SO2. The resonance frequency shift of this sensor rapidly changed as the SO2 concentration changed (Fig. 6c, d). Bai et al. [98] synthesized a layered heterostructure composed of thin MoS2 nanosheets vertically grown on SnO2 nanotubes through simple electrospinning and hydrothermal strategies (Fig. 6e). This heterojunction expands the surface area of MoS2@SnO2, thereby providing more adsorption sites for gas adsorption. The heterostructure formed by the nanofibers facilitated electron communication between MoS2 and SnO2, which promoted charge generation and enhanced the charge separation efficiency (Fig. 6f). The prepared sensing interface MoS2@SnO2 responded to 34.67–100 ppm NO2 at room temperature in a time-period as less as 2.2 s.Fig. 6 Schematic illustrations of a morphological evolution about Pt NPs loaded as spun composite nanostructure and b morphology of Pt_SiO2@SnO2 belts obtained upon calcination and corresponding Pt_SnO2 hollow belts obtained upon etching; reproduced with permission from ref. [96]; Copyright 2021, Elsevier. c Dynamic gas testing system and d the sensing mechanism of a CMUT-based resonant biochemical sensor; reproduced with permission from ref. [97], Copyright 2019, IEEE Industrial Electronics Society. e SEM images of pure SnO2, MoS2@SnO2-1, MoS2@SnO2-2, and MoS2@SnO2-3 and f energy band structure of a nanocomposite of SnO2 and MoS2 in air and in NO2 atmosphere; reproduced with permission from ref. [98], Copyright 2021, Elsevier In recent years, flexible sensors have attracted considerable attention. Flexible gas sensors are key components of portable electronic devices, which have been widely used for on-site environmental and health monitoring owing to their small size, light weight, and strong flexibility [99–101]. Nair et al. [102] fabricated a CNFs@Ni-Pt nanohybrid on a flexible polyester substrate by electrospinning and chemical reduction and tested it with an H2 sensor (Fig. 7a, b). The integration of bimetallic Ni and Pt nanocatalysts on CNF provides more active sites for hydrogen sorption and improves the hydrogen sensing performance of the sensor. Based on this sensor, the research team achieved a wide range of concentration (0.01–4%) of hydrogen detection. Furthermore, after several bending cycles, a good response to H2 (50% strength compared with that before bending) was still exhibited owing to the high aspect ratio of the carbon nanofibers. Khalifa and Anandhan [103] developed a gas sensor based on electrospun blend nanocomposite film (EBNC), which was shown in Fig. 7c, d. The EBNC sensor showed good response (approximately 92% saturation with 108 ppm analyte) and high selectivity for NO2 gas. Furthermore, this sensor was stable for more than 30 days and 100 operation cycles, demonstrating excellent durability. Notably, even after repeated bending cycles, the performance of the sensor did not decrease. This result demonstrates that the electrospun nanofibers were flexible, and their sensing performance was not affected by the bending stress.Fig. 7 a Fabrication process adopted for a CNF@Ni-Pt-based flexible hydrogen sensor and b photograph of the flexible CNFs@Ni-Pt (3:1)-based sensor; reproduced with permission from ref. [102], Copyright 2021, American Chemical Society. c Schematic of a homemade gas-sensing set-up used for evaluating gas sensors and d two gas sensors: one nanocomposite films based and the other EBNC based; reproduced with permission from ref. [103], Copyright 2021, IOP Publishing Ltd. Optical Gas Sensor Although optical sensing technology appeared relatively late, its development is the fastest in gas-sensing technology. Commonly used types in the industry include ultraviolet analyzers, infrared gas analyzers, light scattering analyzers, photoelectric colorimetric analyzers, chemiluminescence analyzers, etc. Fluorescent probes have been widely explored because of their convenient non-invasive operations [104]. The fluorescence of traditional organic fluorophores is often affected by aggregation-caused quenching (ACQ), hence, the stability of sensors made of them is also seriously affected. Fluorescent agents that exhibit aggregation-induced emission (AIE) characteristics have attracted much attention because they can overcome the ACQ problem [105–107]. In addition, volatile acidic and alkaline gases such as hydrochloric acid and ammonia are common hazardous components in flue gases emitted from industrial production processes. They are usually toxic, harmful, and corrosive, and are extremely detrimental to human health. It was demonstrated that organosilicon precursors with AIE characteristics covalently attached to periodic mesoporous organosilicon (PMO) frameworks had high fluorescence efficiency [108, 109]. Electrospun three-dimensional ordered porous materials provide abundant active sites for gas adsorption/desorption, allowing them to be used for the quick and sensitive monitoring of pollutant gases. Gao et al. [110] prepared a flexible film of dispersed periodic mesoporous organosilicas (PMOs) nanospheres in a mixed fiber matrix (PMO-CFs) by electrospinning and assembly technologies and successfully applied it as an efficient fluorescent probe for the detection of ammonia and hydrochloric acid vapor by naked eyes, which was shown in Fig. 8a–c. In addition, the PMO-CF sensor was easy to be regenerated and very stable to light over a long period of time (fluorescence still above 94% after 10 cycles of regeneration).Fig. 8 a Schematic illustration of the construction of PMOs and PMO-CF films for sensor applications, b fluorescence sensing mechanism of PMOs and c optical properties of the PMO-CF film; reproduced with permission from ref. [110], Copyright 2021, American Chemical Society. d Chemical structure and photographs of contact angle measurement on the electrospun fibrous film and e the CCD images of microfiber films under photoluminescence excitation at different air pressures; reproduced with permission from ref. [111]; Copyright 2019, Elsevier Polymers that have been utilized in optical oxygen sensors thus far include polystyrene, hydrogel, silicone rubber, various copolymers, and fluorinated polymers [112–114]. The inclusion of fluorine can increase the rate of diffusion of oxygen molecules in fluorinated copolymers [115], which promotes the sensor to have higher sensitivity owing to the high quenching efficiency [114]. Polymer microfibers, notably fluorocopolymer fibers, are promising functional materials for optoelectronic devices, sensing, and energy storage due to their different surface properties [116, 117]. However, the existing fluorine-containing copolymers broadly face notorious drawbacks of low fluorine content, poor fiber morphology and non-uniformity, hampering their application in sensor devices [118, 119]. For this reason, Mao et al. [111] originally proposed a thin porous microfiber membrane of platinum porphyrin grafted poly(isobutyl methacrylate-co-dodecafluoromethacrylate) copolymer with high fluorine content and applied it to an optical oxygen sensor (Fig. 8d, e). Unlike many solid sensor membranes, porous sensor membranes can promote the permeation of oxygen molecules, and the large surface area of fluorine in the fiber can offer more sites for analyte interactions and signal transduction [120, 121]. Therefore, the sensitivity of the fabricated optical gas sensor with a fluorine-containing microfiber membrane was greatly enhanced to be 584% higher than that of the solid sensor membrane. Electrospun liquid crystal fiber mats have also been established to be a promising medium for gas sensing [122, 123]. Liquid crystals can operate at room temperature and do not require any input of energy as they are powered only by thermal energy and provide a strong optical response. They can conveniently detect gases without intricate spectroscopy equipment. Even low-concentration gases will strongly affect the self-assembly of liquid crystals [124, 125]. Reyes et al. [126] demonstrated the application of a polyvinylpyrrolidone-5CB fiber mat to qualitatively detect the presence of toluene gas by optical measurement. Wang et al. [127] suggested that the electrospun fiber mat made from polylactic acid (PLA) and amyl-cyanobiphenyl (5CB) could quantitatively detect toluene and acetone vapor. Agra-Kooijman et al. [128] reported a resistive liquid crystal core polymer fiber mat (LCC PFM) sensor, in which LC/polymer fibers were obtained after electrospinning on the IDE substrate. The results showed that they responded well to low concentrations of acetone at room temperature. These studies have revealed that chemicals can pass through the polymeric shell of the fiber and be absorbed by the liquid crystal in the core. Upon absorption, the optical properties of the fiber mat were altered by phase transition, which could be adopted for the sensitive and selective detection of volatile organic compounds. Lead(II) acetate ((Pb(Ac)2) reacts with hydrogen sulfide to form a brown lead sulfide precipitate. Up to now, Pb(Ac)2 has been used as an indicator in the form of test papers with a detection limit of 5 ppm to detect H2S gas leaks. The Cha research group [129] successfully fabricated porous nanofibers with high surface area and thermal stability using Pb(Ac)2, overcoming the limitations of traditional Pb(Ac)2-based fibers H2S sensors (Fig. 9). The nanostructure could not only prevent the aggregation of particles but also provided a variety of reaction sites. This sensor material detected H2S as low as 400 ppb at 90% relative humidity.Fig. 9 SEM micrographs of a Pb(Ac)2 particles and b Pb(Ac)2@NFs; schematic representation of c the gas-sensing test and d simulated halitosis breath test equipment; e schematic illustrations of Pb(Ac)2@NFs; f images of Pb(Ac)2@NFs after exposure to various H2S concentrations; reproduced with permission [129]; Copyright 2018, American Chemical Society QCM Gas Sensor Quartz crystal microbalances (QCM) have attracted attention in gas detection owing to their high accuracy, low power consumption, and high stability. The detection mechanism is based on the load mass changes after the sensor layer absorbs gas molecules, which translates to changes in the resonant frequency of the quartz crystal [130–132]. The gas detection capability is highly dependent on the appropriate choice of nanomaterials, including metal oxides, organic polymers, and carbonaceous materials that make up the QCM sensor layer. For instance, a tungsten oxide film can detect NO2, while graphene oxide can detect formaldehyde (HCHO) [133–135]. Despite the plethora of nanomaterials to select from, they lack porous structure and sufficient adsorption sites to absorb a considerable amount of target gas molecules. To further improve the ability of QCM sensor materials to absorb more gas molecules and their sensitivity, electrospinning technology has been introduced to produce highly porous sensor materials. Kang et al. [136] used electrospinning and oxidative polymerization techniques to successfully prepare SnO2 nanofibers (NFs)/PDA composites that were quantitatively deposited onto the QCM electrode surface. The SnO2 nanofibers prepared in polymeric composites had rough morphology, which promoted the adsorption of gas molecules. The agglomerated PDA microspheres adhered to the surface of SnO2 NFs and revealed a compact composite structure with a larger specific surface area exposing more active sites. The gas sensor prepared in this way showed a high sensitivity for formaldehyde sensing. Gas adsorption capacity is generally related to the surface functional groups and defect sites of the sensor materials. In addition to the porous nanofibers fabricated by electrospinning technology, the active metal oxides on the fiber surface were also crucial for gas adsorption, wherein the oxygen atom and imine functional group hydrogen-bonded with formaldehyde molecules [137, 138]. Moreover, the amine functional group in PDA nanospheres had undergone condensation with the aldehyde groups in formaldehyde to some extent [139, 140], thereby greatly improving the sensitivity. The fabricated nanofibrous QCM sensor achieved a detection limit of 500 ppb formaldehyde. Diltemiz’s research team [141] employed similar principles to prepare CuO-ZnO nanofibers for detecting formaldehyde as low as 41 ppb and the optimal detection frequency reached 8 Hz ppm−1. Zhang et al. [142] made use of centrifugal electrospinning equipment to fabricate aligned nanofibers to selectively adsorb CO molecules against control nitrogen gas. Based on the Sauerbrey equation, it was found that an exceptionally high amount of CO (72.36 ng) was adsorbed on the nanofibers with a maximum frequency change of − 54 Hz for 50 ppm of CO feed, which was attributed to the large nanofiber surface area. Metal Ion Sensor Rapid industrialization is accompanied by several grim environmental problems, most notably water pollution [143]. Heavy metal pollution is a prevalent environmental issue that plays a detrimental role to aquatic life and human health [144, 145]. Therefore, society has paid close attention to the excessive discharge of heavy metal effluents because of their universality, trace-level toxicity, risk of carcinogenesis, and multiple organ failure. Table 1 lists the sources and effects of some highly toxic heavy metal ions that are the cause of many heavy metal-related diseases. In summary, in water pollution control, methods that can accurately monitor the heavy metals content in real time are of great significance for protecting the environment and human health [146].Table 1 The sources and side effects of some highly toxic heavy metal and emission limit values for heavy metal ions under the Chinese, WHO, and BIS regulations Heavy metal Source Side effects Emission limit (mg L−1) GB 5749-2005a WHOb BISc Arsenic (As) [147, 148] Mining, metallurgy, decolorizing agent, pesticide, chemical fertilizer, etc Highly toxic Carcinogenicity 0.01 0.05 0.05 Cadmium (Cd) [149, 150] Electroplating, mining, smelting, fuel, battery Hypertension Cardiovascular and cerebrovascular disease Kidney failure 0.005 0.005 0.01 Chromium (hexavalent) (Cr(VI)) [151, 152] Cosmetic raw materials, leather preparation, electroplating, industrial pigments, rubber Cancer Liver damage Stomach ulcer Muscle spasm 0.05 0.05 0.05 Lead (Pb) [153] Paint, coating, battery, smelting, electroplating, cosmetics, etc Highly toxic Nervous System damage Intellectual damage 0.01 0.05 0.05 Mercury (Hg) [154, 155] Instrument factory, salt electrolysis, smelting, cosmetics, dental materials, etc Liver damage Nervous system damage Vision damage 0.001 0.001 0.001 Copper (Cu) [156, 157] Smelting, metal processing, machinery manufacturing, rubber, etc Gastrointestinal mucosa ulcer Hemolysis Liver necrosis Kidney damage 1 1.3 1.3 Zinc (Zn) [158] Zinc mining, smelting, electroplating, machinery manufacturing, paper industry, etc Diarrhea Nausea Vomiting 1 5 5 aStandards for drinking water quality (GB 5749-2005): Legal health standards for chemicals related to human health in drinking water issued by the Ministry of Health of China bWHO: World Health Organization recommended permissible limits for heavy metal ions in water cBIS: Permissible limits for heavy metal ions as per Bureau of Indian Standards Electrochemical Sensors Electrochemical techniques have been extensively studied for their application in the detection and speciation of heavy metals. These electrochemical analyses normally use cyclic voltammetry (CV), differential pulse voltammetry (DPV), electrochemical impedance spectroscopy (EIS), chronoamperometry (i–t), linear sweep voltammetry (LSV), and stripping voltammetry (SWSV) for that purpose. The measurements from the electrochemical analysis are typically registered as electrical signals such as resistance and current. Therefore, the electrochemical analysis equipment is relatively simple to analyze, easy to automate, and performs continuous analysis. Chemically modified electrodes refer to the molecular design and modification of the electrode surface by chemical means to produce the electrode with specific electrochemical properties [159]. By endowing materials with excellent properties (catalytic effect, photoelectric effect, etc.) on the electrode surface, the conductivity of the electrode can be greatly improved, accelerating the electron transfer efficiency, while increasing the electrode response, thereby improving their selectivity and sensitivity as potential sensors [159, 160]. Common materials used in electrospinning are conductive polymers, carbonaceous, and metal nanoparticles [160]. Among the polymeric nanofibers, polyaniline (PANi) turned out to be the most commonly investigated conductive polymer owing to its one-dimensional structure that ensures high conductivity and surface area, which plays an extremely important role in the preparation of the electrode. Promphet et al. [161] uncovered the development of graphene/polyaniline/polystyrene (GO/PANi/PS) nanoporous fiber-modified screen-printed carbon electrode (SPCE), which was successfully solved by square wave anodic stripping voltammetry (SWASV) for the simultaneous determination of lead (Pb2+) and cadmium (Cd2+) in the range between 10 and 500 μg L−1, with detection limits of 3.30 μg L−1 (Pb2+) and 4.43 μg L−1 (Cd2+). Similarly, Huang et al. [162] used phytic acid-doped polyaniline nanofiber-based nanocomposites to modify glassy carbon electrodes for Cd2+ and Pb2+ detection with EIS and differential pulse anodic stripping voltammetry (DPASV) proved that the synergistic effect of polyaniline nanofibers and phytic acid had enhanced the charge transfer efficiency of metal ions. Due to the synergistic effect of PANi, the detection limit of Cd2+ in the range of 0.0560 μg L−1 (S/N = 3) was 0.02 μg L−1, and that of Pb2+ in the range of 0.160 μg L−1 was 0.05 μg L−1. Besides PANi, carbon nanofibers (CNFs) have attracted attention because of their high mechanical strength, large specific surface area, excellent electrical conductivity, and strong corrosion resistance. The PAN-based CNFs prepared by Zhao et al. [163] using electrospinning technology was eight times more sensitive to trace amounts of Pb2+ with a detection limit of 0.9 nM under anodic dissolution voltammetry analysis. Furthermore, Gao et al. [164] prepared nitrogen- and sulfur-codoped PAN-based CNFs by the pyrolysis of trithiocyanuric acid and silica nanospheres, which showed the advantages of exhibiting large surface area (109 m2 g−1), porous structure and high proportion of heteroatoms (19 at.% of N and 0.75 at.% S). The electrochemical sensor fabricated using the high-performance porous CNFs was highly responsive to trace Cd2+ in the range of 2.0500 μg L−1 using DPASV. Tang et al. [165] formed PANi nanosheet arrays on electrospun Fe-CNFs substrates and deposited AuNPs uniformly on the nanosheet surfaces with reducing characteristics, which was shown in Fig. 10a, b. The presence of Fe in CNF accelerated the growth of PANi nanosheets and improved their adsorption of As3+, responding to a wide linear range (5–400 ppb) and low detection limit of 0.5 ppb (S/N ≥ 3).Fig. 10 a The morphology of Au–PANi–Fe–CNF and b the current response of As(III) in different concentrations; reproduced with permission from ref. [165], Copyright 2020, Elsevier. c The mechanism of colorimetric membrane; reproduced with permission from ref. [170], Copyright 2016, Elsevier. d Optical images of the sensor immersed in Au/Ag NPs colloid solution at different period; reproduced with permission from ref. [171], Copyright 2018, Springer Vienna. e Ion detection mechanism and f color-differentiation map about Fe3+ and Cu2+; reproduced with permission from ref. [172], Copyright 2019, American Chemical Society Metal oxides and metal nanoparticles are also widely applied in electrochemical sensors due to their significant electron transfer kinetic ability, larger specific surface area, and availability of adsorption sites [166]. ZnO is an excellent sensor material due to its high adsorption capacity, biocompatibility, and high chemical stability. The one-dimensional nanostructure can facilitate the diffusion of ions from the electrolyte to the surface of the sensor material more rapidly. Therefore, the production of nanofibers with highly uniform one-dimensional nanostructures by electrospinning is a feasible approach. Oliveira et al. [167] reported the synthesis of ZnO nanofibers/l-cysteine (ZnO/l-cys) nanocomposite electrode for electrochemical sensing of Pb2+. Under the SWASV analysis, highly sensitive quantification of Pb2+ (LOD = 0.397 μg L−1) was achieved in the linear range of 10,140 μg L−1. Girija et al. [168] corroborated that cobalt and zinc ions form a stable tetrahedral structure with the zeolite imidazole ester framework under electrostatic action. On that basis, the cobalt–zinc–zeolite imidazole framework (Co/Zn-ZIF) nanofibers were successfully synthesized to detect heavy metal ions using CV analysis, uncovering the catalysis effect of the electrode to facilitate electron transfer between the electrode surface and metal ion. The Co/Zn-ZIF nanofibers effectively detected Cd2+ with low interference in the range from 100 nM to 1 mM and a detection limit of 27.27 nM. Teodoro et al. [169] used polyamide (PA6), cellulose nanowhiskers (CNW), and silver nanoparticles (AgNPs) to fabricate electrospun nanofiber electronic tongues for the quality assessment of water samples. It was found that the nanofiber mat was rich with pores and channels, which provided a multidisciplinary interaction between the binding site and analytes. With high aspect ratio, high biodegradability, and low density, CNW was uniformly blended in a co-system with metal nanoparticles for increased conductivity, enabling the prepared electronic tongue sensor to effectively discriminate different heavy metal ions with detection limit for Pb2+ in aqueous solution only 10 nM. Colorimetric and Fluorescence Sensor The binding constant of the analyte of interest and the nature of sensor molecules are the main factors affecting the sensitivity, specificity, reusability, and stability of the sensor. For the colorimetric sensor interface, the signaling molecule would need to induce responsive color change [173, 174]. Electrospinning has the advantages of fabricating a three-dimensional structure with a large specific surface area and excellent biocompatibility, which can provide more binding sites in the sensor molecules, and effectively improve the sensitivity and response rate of the sensor [174, 175]. Moreover, the low cost, simple operation and facile surface functionalization of electrospinning make it widely employed in many fields [176]. In recent years, colorimetric sensors constructed using electrospinning as the sensor interface have been used to detect metal ions with careful design of organic chemistry. Thence, with the development of nanotechnology, colorimetric methods based on surface plasmon resonance (SPR) of metal nanomaterials have progressively been introduced. The core of the color sensing system is represented by the specific recognition molecules that can generate and transmit color signals, mainly using organic dyes or transition metal complexes. The recognition molecule and the analyte selectively interact by non-covalent bonding or covalent bonding, which induces the charge transfer or spatial structural change of the recognition molecule leading to the macroscopic manifestation in color change. Rhodamine B (RhB), widely used for industrial coloring and as a fluorescent probe, is revealed to be a reliable color signal transmitter because of its long absorption and emission wavelengths, high photostability, high absorption coefficient, and quantum efficiency. Parsaee et al. [177] prepared Au NPs using Gracilaria under ultrasonication, and used electrospinning technology to modify silica gel membranes in the presence of RhB. The nanofibers immobilized with gold nanoparticles and RhB were used to detect Hg2+ in real water samples using colorimetry and fluorescence at detection limits of 2.21 nM and 1.10 nM, respectively. Moreover, the sensor can be reused multiple times with high integrity upon regeneration under oxidation by air. Similarly, Li et al. [170] used copolymers of rhodamine and quinoline propylene-based monomers to produce electrospun NFMs that acted as solid-state sensors and exhibited a high degree of Fe3+ selectivity, which was shown in Fig. 10c. In addition, the NFM realized the obvious color change from colorless to pink within a minute and its LOD was 1.19 μM. Poly(aspartic acid) (PASP) is a synthetic poly(amino acid) with several unique properties, including strong metal ion binding ability, biocompatibility, biodegradability, water solubility, and low toxicity. Zhang et al. [172] developed a reuseable PASP-based colorimetric sensor with nanofiber structure for Cu2+ and Fe3+ detection, with a large amount of Cu2+ or Fe3+ accumulating on the PASP-based NFM after filtration, which was shown in Fig. 10e, f. Upon exposure to an aqueous Cu2+ solution, the color of the sensor changed from white to blue and the detection limit was reported to be 0.3 mg L−1. In contrast, the membrane changed color from white to yellow, with a detection limit of 0.1 mg L−1 for Fe3+ aqueous solution. Noble metal nanoparticles have a small size effect, surface effect, and quantum tunneling effect, and exhibit unique physicochemical properties. SPR can be described by the reaction of a large number of freely conducting electrons in precious metals to external incident electromagnetic waves. When the electronic oscillation frequency is equal to the incident light frequency, surface plasmon oscillation occurs, enabling the noble metal nanoparticles to produce strong absorption and scattering spectra in the visible light range. The spectral peak position is highly dependent on the size and distribution of the nanoparticles, and any changes in the external environment. When the radius of the nanoparticles is larger than 3.5 nm, the clustering of nanoparticles causes surface plasmon coupling, and a strong color change is observed in the visible light range. The Au/Ag NPs were fixed on the aminated PAN nanofiber membrans (NFMs) to obtain a test strip with a porous structure of a large surface area (38.6 m2 g−1). The color of the NFM measured at a wavelength of 420 nm, underwent a redshift when exposed to Cu2+ and the color changed from yellow to pink to colorless. This effect was due to the leaching of Au/Ag NPs from NFM in the presence of ammonium chloride, thiosulfate, and Cu2+ and the formation of soluble thiosulfate complexes of Ag+, Au3+, and Cu2+. On that basis, Abedalwafa et al. [171] achieved colorimetric detection of Cu2+ in drinking water samples by developing electrospun nanofiber membranes with Au/Ag core-shell nanoparticles, which was shown in Fig. 10d. Under optimized conditions, this method has the advantages of low detection limit (50 nM at S/N = 3), fast determination time (3 min), good specificity, and excellent reversibility. Spectrometric Sensor Surface-enhanced Raman Scattering (SERS) technology overcomes the inherent shortcomings of traditional Raman spectroscopy with weak signals and can increase the Raman intensity by several orders of magnitude. One of the main research directions in the SERS research area is the development of SERS substrates. The fabrication of SERS substrates usually relies on the surface plasmonic properties of the noble metal nanoparticles [178]. The nanoparticles loaded onto the electrospun polymer fiber mat, provides a larger surface area that can effectively enhance the detection signal and is considered to be an effective SERS substrate [179]. Xu et al. [180] fabricated a 3D SERS substrate composed of electrospun polycaprolactone (PCL) fibers and silver-coated gold nanorods (Ag/AuNRs) to detect traces of heavy metals in the environment. The successful fabrication and fixation of Ag/AuNRs on PCL fibers benefited from electrostatic interaction and excellent charge transfer between the gold core and the silver layer in the bimetallic structure. Following that, the detection of trace concentrations of organic arsenic, arsenic acid and roxarsone was achieved. Zhang et al. [181] used electrospinning technology to produce for the first time polyacrylonitrile (PAN)/noble metal/SiO2 nanofiber mats with plasma-enhanced fluorescence activity. These nanofiber mats selectively increased the fluorescence intensity of conjugated polyelectrolytes (CPE), making the polymer/noble metal nanofibers a promising substrate with improved sensitivity for metal ion detection. Antibiotic Sensor Since penicillin was discovered by Fleming in 1929 and used clinically by Florey and Chain, more than a hundred types of antibiotics have been developed and played a huge role in treating infectious diseases [182]. Antibiotics have an inhibitory effect on bacteria. Therefore, they are widely used to treat or prevent diseases in humans and animals [183]. Moreover, antibiotics are often added to animal feeds to promote animal growth and harvest. The types and properties of common antibiotics are listed in Table 2. Unfortunately, the problem of antibiotic contamination has become increasingly serious due to their mass production and abuse [184–186].Table 2 Types and properties of common antibiotics Types of antibiotics Chemical formula Applications Side effects Tetracyclines [200, 201] Tetracycline C22H24N2O8 Mycoplasma Gastrointestinal discomfort Liver toxicity Oxytetracycline C22H24O9N2 Chlamydia infection Doxycycline C22H24N2O8 Lyme disease Aminoglycosides [202] Gentamicin C60H123N15O21 Gram-negative bacterial infections, such as Escherichia coli, Klebsiella, Pseudomonas aeruginosa Hearing Damage Dizziness Nephrotoxicity Kanamycin C18H36N4O11 Streptomycin C21H39N7O12 Tobramycin C18H37N5O9 Macrolides [203] Erythromycin C37H67NO13 Treatment of streptococcal infections, syphilis, respiratory tract infections, and mycoplasma infections Nausea Vomiting Diarrhea Clarithromycin C38H69NO13 Dirithromycin C42H78N2O14 Roxithromycin C41H76N2O15 Azithromycin C38H72N2O12 β-Lactam [204, 205] Pannixilin C16H19N3O5S Broad-spectrum antibacterial, streptococcal infection, syphilis Gastrointestinal upset Diarrhea Severe allergies Kidney toxicity Ampicillin C16H19N3O4S penicillin C16H18N2O4S Azlocillin C20H23N5O6S Quinolones [206, 207] Ciprofloxacin C17H18FN3O3 Urinary tract infections, bacterial prostatitis, bacterial diarrhea, gonorrhea Nausea Tendon degeneration Levofloxacin C18H20FN3O4 Norfloxacin C16H18FN3O3 Ofloxacin C18H20FN3O4 Sulfonamides [208] Trimethoprim C14H18N4O3 Urinary tract infection Nausea Vomiting Diarrhea Allergic rash Urinary stones Kidney failure Sulfamethizole C9H10N4O2S2 Sulfamethoxazole C10H11N3O3S Antibiotics can enter the water environment in many ways and most commonly through urinary or fecal excretion [187]. The antibiotics discharged from the human body typically flow through the sewers and end up in the sewage treatment plant [186, 188]. However, the current sewage waste treatment technology could not completely remove antibiotics, causing residual antibiotics to be discharged from the sewage treatment plant and causing pollution of the aquatic environment and sources of drinking water [189–191]. Similarly, veterinary antibiotics cannot be completely metabolized and degraded in animals, and can enter the soil in the form of organic fertilizers and migrate to the ground- and surface water through runoff, infiltration, and leaching [192]. In this regard, antibiotics released into water bodies can affect the composition and activity of microorganisms, and detrimentally alter their ecological structure [193]. To make things worse, the challenge of bacterial resistance has been looming [191, 194], and the mixed consumption of antibiotics can cause the same kind of bacteria to develop resistance to multiple antibiotics, thus creating "Superbugs". The detection techniques of antibiotics in trade effluents conventionally include capillary electrophoresis, spectrophotometry, SPR, gas chromatography and liquid chromatography, hyphenated techniques, and immunoassays [195–199]. The accuracy of the detection from these methods is very credible, with outstanding sensitivity and stability, but they place high demands on the instrument user and where sample preparation and operation procedures are often complicated and take a long time [191]. Therefore, on-site and real-time monitoring of antibiotics in the aquatic environment requires easy-to-operate and faster detection methods without compromising on their detection performance. At present, there is not much research on using electrospinning technology to detect antibiotics, and the detection methods involved are still mainly focused on electrochemical detection and optical detection. The electrochemical detection method mainly includes two forms of representation: voltammetry and impedance. Non-specific bare electrodes typically limit the performance of the electrodes, modification strategies must be introduced to improve their performance. On the one hand, modifying the electrode surface by physical or chemical means can improve the hydrophilicity of the electrode–solution interface and thereby increase the concentration of antibiotics in the sensing area [209, 210]. Functional materials such as nanomaterials, organic materials, enzymes, or antibodies are assembled on the electrode surface through modification methods such as self-assembly, coating, electrodeposition, and electropolymerization. As a result, the modified electrode displays a special surface effect, size effect, quantum tunneling effect, and catalytic activity, which can enhance the conductivity of the electrode, accelerate the electron transfer in the interface, and reduce the redox overpotential of antibiotics [211, 212]. The effective amalgamation of the two above-mentioned aspects promotes a synergistic effect and significantly improves the sensitivity of antibiotic detection [182]. The three-dimensional porous network of CNFs network with extensive ion transport channels and a large aspect ratio facilitates the electron transport rate and effectively adjusts the charge diffusion length [213]. The composite structure formed of CNFs and other functional materials can significantly improve the surface chemical properties and conductivity, thereby changing its electron-donating ability and further improving the selectivity and stability of the sensor [214, 215]. Rare earth metal orthovanadate, among reported electrocatalysts, has become the most important functional material because of its tunable bandgap, abundant oxygen vacancies, non-cytotoxicity, excellent stability, and electrical conductivity. Baby et al. [195] presented a samarium vanadate/carbon nanofiber (SmV/CNF) composite material for the quantification of sulfadiazine (SFZ) over a wide linear range of 0.009–445 µM. The synergistic effect created by the composite structure of SmV and CNF accelerated the charge transfer while creating more active sites for detection. The prepared sensor had proven significant electrocatalytic activity, a low detection limit of 0.0013 µM, and high sensitivity of 4.03 µA µM−1 cm2. In addition, Thangavelu Kokulnathan et al. [214] used zirconia/carbon nanofibers (ZrO2/CNF) to fabricate a modified glassy carbon electrode (GCE), which was successfully applied for selective detection of chloramphenicol (CPL). DPV was used to measure the CPL reduction, which translates to quantification over a linear range from 0.005 to 903.76 μM with LOD of 0.0018 μM. Scagion et al. [216] assembled a new type of nanomaterial based on electrospun PA6/PANi nanofibers on a gold interdigital microelectrode and used it as the electronic impedance tongue sensing device to selectively detect tetracycline at a concentration as low as 1 ppb. Selective sensing has always been the focus and bottleneck in antibiotic sensor development since their chemical structures can be very similar [217]. Common antigen-antibodies have strong specific binding ability despite the complicated antibody synthesis steps and strict storage conditions that have always limited their application in antibiotic sensors [218]. Easily synthesized and functionalized aptamers have been recognized as viable alternative sensing probes with high chemical stability [219, 220]. Song et al. [196] prepared a novel electrochemical aptamer sensor based on iron-based MOF that detected tetracycline in the range from 0.1 to 105 nM. CNFs based on the metal–organic framework NH2-MIL-101(Fe) could not only provide more active sites to support the aptamer, the presence of amino groups (–NH2) also promoted electron transfer and proton coupling [221]. Furthermore, NH2-MIL-101(Fe) strongly interacted with the functional groups of the negatively charged nucleic acid sequence, which allowed it to capture a considerable amount of target pollutants, thereby improving the sensitivity of the electrochemical sensor [222, 223]. Similarly, Vafaye et al. [224] electrodeposited gold nanoparticles on the surface of a carbon nanofiber mat to fix the aptamer, realizing a highly sensitive detection of penicillin (1–400 ng mL−1). Among the optical sensors, colorimetry is the most intuitive detection method and, Fe3+ is commonly utilized in this instance to drive color change. The chemical structure of tetracycline is highly prone to complex with the Fe atoms as it contains multiple O- and N-containing functional groups [225, 226]. The oxygen atoms of the carboxylate in the alginate skeleton preferentially chelate to Fe3+ that eventually bind to the target antibiotic molecules. Based on this principle, Yan et al. [227] functionalized alginate directly on the surface of PAN nanofibers for Fe3+ fixation and detected 5 μg kg−1 of tetracycline (TC) within 10 min with a color change that could be observed easily with the naked eyes, which was shown in Fig. 11a, b.Fig. 11 a Three stages scheme of the colorimetric strips for TC determination, b the visual colorimetric response and the UV-Vis spectra of the strips against different concentrations of TC; reproduced with permission from ref. [227], Copyright 2018, Royal Society of Chemistry. c Illustration of the design, fabrication, and detection mechanism of the MA@GNP-immobilized PA6 NFM colorimetric strips for metronidazole and d the linear relationship between the color difference values; reproduced with permission from ref. [229], Copyright 2019, Royal Society of Chemistry Due to its highly effective antibacterial properties, metronidazole (MTZ) is widely used as a food additive in animal husbandry to promote the growth of animals and plants. However, MTZ can cause serious diseases through genotoxic, carcinogenic, and mutagenic effects [228]. Therefore, appropriate detection methods are required to prevent excessive MTZ residues from entering o the aquatic environment, which would adversely affect human health. Under neutral pH, the amino groups in melamine (MA) form a covalent bond with the Au nanoparticles with increased stability. When MTZ binds with the composite material (MA@GNP) via hydrogen bonding, the color of MA@GNP changes from pink to purple due to the aggregation of gold nanoparticles. Mohammed et al. [229] reproduced this phenomenon by assembling melamine-functionalized Au nanoparticles on polyamide NFMs. The resulting colorimetric strips, which shown in Fig. 11c, d showed good sensitivity for MTZ with low LOD (2 nM at S/N = 3) and a fast response time (2.5 min). Apart from colorimetry, SERS is also a powerful spectroscopy technique that can be used for label-free detection of (bio)chemical substances. The principle behind this technology is the electromagnetic amplification by noble metal nanostructures as mentioned previously. Kang et al. [230] reported a SERS sensor based on assembled Au@Ag core-shell nanoparticles on an electrospun nanofiber matrix that was used for the selective detection of methamphetamine at logarithmic concentration from 10–1 to 104 ppb, with the detection limit reaching an impressive 7.2 ppt. Tang et al. [231] further found that amorphous silica nanofibers uniformly loaded with Ag nanoparticles fabricated using electrospinning technology showed strong SERS enhancement effects and achieved high efficiency for trace antibiotics detection. Pesticide Sensor There have been reports of pesticide pollution of the aquatic environment around the world [232, 233]. According to reports, many famous rivers such as the Yangtze River, Songhua River, Tingjiang River, and Heilongjiang River in China have been polluted by pesticides. More than 130 pesticides or their degraded products have also been identified in groundwater in the United States. Pesticide residues such as atrazine, acetochlor, and dimehypo were found in groundwater in Jiangsu, Jiangxi, and Hebei, China. In general, the higher the water solubility of pesticides, the more residues can be found in the water and studies have shown that surface waters can easily promote the accumulation of pesticides or chemical substances [234]. Additionally, the pollutants in the surface water are easily displaced to the groundwater through the hydrological cycle or soil infiltration. Therefore, pollution of the aquatic environment caused by pesticides is a very urgent global issue to be addressed. Today, the detection of pesticides in the aquatic environment relies mainly on manual on-site sampling and then routine analytical techniques such as gas chromatography and high-performance liquid chromatography [235–237]. It is undeniable that chromatography is highly accurate and a mature detection technology, but the underlying factors such as expensive instrumentation, cumbersome sample preparation process, and high operational requirements make pesticide detection challenging. Over the past 5 years, a variety of rapid detection methods have been developed, including electrochemical sensors, colorimetric sensors, and detection cards [232, 238]. These new detection technologies help environmental inspectors to localize the pollutants in the aquatic environment in time, and effectively avoid the large-scale release and spread of pollutants. Feng et al. [238] combined electrospinning and hydrophilic modification to develop an emerging type of nano-/microstructure detection card based on PCL fiber mats to detect indole acetate and acetylcholinesterase (AChE). Pretreating the fiber mat with ethanol promoted hydrolysis of the PCL fiber mat, which improved the surface wettability, and the minimum detectable concentrations of carbofuran, malathion, and trichlorfon were reduced by fivefold, twofold, and 1.5-fold, respectively (Fig. 12a, b). Using a similar approach, Moghazy et al. [239] electrospun chitosan/PVA nanofibers to fabricate a mutant acetylcholinesterase-based biosensor with LOD determined to be 0.2 nM for phosphorus and thus far below the maximum residue limit (MRL) of 164 nM, set by international regulations. The low LOD could be achieved owing to the unique spatial structure, high porosity, and large specific surface area of the electrospun fibers. Shao et al. [240] used electrospinning technology to manufacture flexible and hydrophobic Ag/SB nanofiber films as SERS substrates on a large scale. Using R6G as a probe molecule, it was observed that the Ag/SB-SERS method had excellent sensitivity and stable signal reproducibility towards triazophos with LOD of 2.5 × 10–8 M. More importantly, the method did not require complicated sample preparation.Fig. 12 a The preparation process and principle of pesticide determination of the nano-/microstructured detection card and b storage stability of the detection card below 4 °C and room temperature (RT); reproduced with permission from ref. [238], Copyright 2021, MDPI. c Schematic representation of the proposed biosensor platform and d the stability, selectivity, and interference suppression of the immobilized electrode; reproduced with permission from ref. [247], Copyright 2019, Elsevier. e Schematic illustration of the synthesis process of PANi-array@CNF, f FE-SEM micrographs of PANi-array@CNF prepared at different feeding speeds, and g DPVs of the PANi-array@CNF/GCE with different concentrations of o-NP, m-NP and simultaneous addition of different concentrations of o-NP and p-NP; reproduced with permission from ref. [232], Copyright 2020, Elsevier Atrazine is a type 3A carcinogen and an insecticide in the chlorotriazine family, and is known to seriously affect the human endocrine system if consumed. The high toxicity of atrazine extends its effect on animals and plants. Supraja et al. [232] electrospun SnO nanofibers to perform label-free, and ultra-sensitive electrochemical detection of atrazine within the wide dynamic detection range of μM, achieving a detection limit for atrazine of 0.9 zM and sensitivity of 4.11 μA μM−1 cm−2 (Fig. 12c, d). Nitrophenols, including three isomers o-nitrophenol (o-NP), m-nitrophenol (m-NP), and p-nitrophenol (p-NP), are all important raw materials for the production of pesticides and are suspected carcinogens [241]. They have high chemical stability and will not be easily degraded by microorganisms, which seriously upset the ecosystem balance [242], hence are listed by the United States Environmental Protection Agency (USEPA) as toxic pollutants and hazardous wastes of top priority. According to the European Communities Commission regulations, the maximum allowable amount of nitrophenol in drinking water is 5.0 μM [243–245]. Unfortunately, o-NP and p-NP as hydrolysates of pesticides (parathion and methyl parathion, etc.) widely contribute to water and food contamination and are commonly used as markers to diagnose pesticide contamination [246]. Zhu's research group [247] prepared CNFs with a high surface area–volume ratio and aspect ratio by the carbonization of electrospun PAN fibers. By controlling the polymerization time and ammonium persulfate (APS) feed rate, a one-dimensional nanoconical PANi array with favorable uniformity was successfully assembled on the electrospun CNFs (PANi-array@CNF) at low temperatures yielding abundant mesoporous structures necessary for electrochemical reactions (Fig. 12e–g). Due to the large specific surface area and highly ordered nanostructure, the CNF electrode displayed excellent activity and response to the electrochemical reduction of nitrophenol, enabling ultra-sensitive and selective detection of p-nitrophenol with a LOD as low as 1.5 nM. Moreover, the fabricated sensor could quantify and differentiate o-NP and p-NP simultaneously. Antohe et al. [248] reported a PANi/platinum coating for the highly sensitive detection of 4-nitrophenol (4-NP) contaminant with a layered fiber surface plasmon resonance sensor, which revealed excellent detection performance in the low picomolar range (LOD = 0.34 pM). Others The composition of pollutants in the aquatic environment is complex. Besides chemical pollutants, biological contamination such as bacteria, viruses, and parasites are of high prevalence and relevance today. Schistosomiasis is one of the deadliest diseases caused by parasites from water pollution and their transmission can be much faster than chemical contamination to cause an outbreak of diseases. The Coronavirus Disease 2019 (COVID-19) pandemic has taken millions of lives worldwide and there is a need to rapidly detect the virus. Currently, real-time (quantitative) reverse transcription polymerase chain reaction (RT-PCR) is considered the gold standard for COVID-19 diagnosis. However, RT-PCR-based tests are complex, expensive, time-consuming, and require sample pre-treatment by trained personnel. Over the past 2 years, researchers around the world have developed several solutions to quickly diagnose COVID-19. However, in the early stages of infection, there are many false-positive misdiagnoses with these developed sensors. To address the current challenges, Jadhav et al. [249] proposed a diagnostic scheme based on SERS combined with microfluidics integrating microchannels. The channels consisted of Au/Ag-coated carbon nanotubes and an electrospun micro/nanofiltration membrane reported to have the potential to successfully capture viruses from various biological fluids/secretions. In other instances of monitoring biological contaminants, Xu et al. [250] reported a novel type of aptamer@AuNPs@UiO-66-NH2 nanofiber sensor for specific detection of microcystin (MC-LR). High loading of MOFs and aptamers on nanofiber fibers was achieved and successfully applied to accurately identify MC-LR by solid-phase microextraction (SPME) combined with LC–MS. Highly specific recognition of MC-LR was achieved with an extremely low LOD (0.004 ng mL−1) and good precision (CV% < 11.0%). Yang et al. [251] used the electrospun nanofiber mat prepared by the in situ assembly of tunable Ag nanoparticles on TiO2 nanofibers to detect bacteria. The Ag NPs were uniformly deposited on the nanofibers on the surface of TiO2 so that the composite nanofiber mat had excellent SERS properties. The minimum detection limit of the sensor was around 10–9 mol L−1. More importantly, the nanofiber mat fabricated could be used as a SERS substrate to detect E. coli, S. aureus and other biological macromolecules without the need for using aptamers. Niri et al. [252] developed a CNF-based electrochemical biosensor for the detection of Hepatitis B virus in a linear range of 10–12–10–6 M, with a detection limit of 1.58 × 10–12 M. Arshad et al. [253] developed an impedance sensor based on molecularly imprinted polymer (MIP) and electrospun PS nanofibers for early detection of dengue infection, successfully attaining a linear response ranging from 1 to 200 ng mL−1, and the LOD as low as 0.3 ng mL−1. Conclusion and Future Trends In the present comprehensive review, electrospinning technology has been uncovered to be one of the most effective methods that can be used to achieve mass production of nanofibers for different applications. Electrospun nanofiber membrane is empowered with the advantages of displaying three-dimensional structure, high porosity, large specific surface area, and controllable structure. It is an ideal nanomaterial for fabricating high-performance sensor elements. Compared to methods such as hydrothermal and templating, electrospinning allows the blending and in situ polymerization or cross-linking, of a variety of functional materials to realize the preparation of nanomaterials with unique structures and morphologies including nanofibers, nanowires (nanorods), nanoribbons, hollow nanofibers, core-shell structures, hollow nanotubes, and nanodendrites. Moreover, these electrospun nanomaterials have high axial strength and can facilitate continuous electron transfer along the long-axis direction, so that they can achieve higher sensitivity in the field of environmental sensing. Finally, the electrospinning technology is easy to mass-produce and commercialize, laying the foundation for the market application of nanofiber-based sensors in the near future. Although the prospects are optimistic, there are many challenges ahead in bringing sensors based on electrospinning technology into the market and realizing industrial-scale production. These include: (1) improving the dispersion of nanomaterials on the surface or inside nanofibers so that the functionalized materials are evenly distributed; (2) the fabrication of electrospun nanofibers is highly susceptible to environmental factors, and the process of nanofibers needs to be improved. Parameters to ensure the stability and repeatability of nanofibers produced under large-scale manufacturing conditions; (3) the integration of electrospinning nanofibers with chips, wearable devices, and other sensor platforms; (4) at this stage, the laboratory-based environmental sensors are only used for detection in a relatively clean gas and water environment, and the actual environment contains much more complex interferences, which poses requirements for the selectivity and anti-interference of the sensor; and (5) environmental sensors are mainly divided into real-time detection and long-term monitoring. Long-term monitoring in particular poses a challenge to the stability of the sensor. In conclusion, nanofibers produced by electrospinning technology offer many effective binding sites for analytes due to their high surface area and porosity, which greatly improves the sensitivity and response time of the sensor. Despite the evident challenges, the authors have great confidence that the future commercialization of electrospinning technology for environmental monitoring can be expected with the continuous research efforts. Acknowledgements This work is supported by the National Key Research and Development Project (2019YFC0408304), the Fundamental Research Funds for the Central Universities and Graduate Student Innovation Fund of Donghua University (no. 2232022G-04, BCZD2022005, and CUSF-DH-D-2021037). The support provided by China Scholarship Council (CSC) (no. 202006630085) during a visit to the National University of Singapore is also acknowledged. Declarations Conflict of Interest On behalf of all authors, the corresponding author states that there is no conflict of interest. ==== Refs References 1. Liang LW Wang ZB Li JX The effect of urbanization on environmental pollution in rapidly developing urban agglomerations J Clean Prod 2019 237 117649 10.1016/j.jclepro.2019.117649 2. Chowdhary P Raj A Bharagava RN Environmental pollution and health hazards from distillery wastewater and treatment approaches to combat the environmental threats: a review Chemosphere 2018 194 229 10.1016/j.chemosphere.2017.11.163 29207355 3. Manisalidis I, Stavropoulou E, Stavropoulos A, Bezirtzoglou E. Environmental and health impacts of air pollution: a review. Frontiers Public Health 2020;8:14. 4. Novo A Andre S Viana P Nunes OC Manaia CM Antibiotic resistance, antimicrobial residues and bacterial community composition in urban wastewater Water Res 1875 2013 47 5. Cheng DL Ngo HH Guo WS Chang SW Nguyen DD Liu YW Wei Q Wei D A critical review on antibiotics and hormones in swine wastewater: water pollution problems and control approaches J Hazard Mater 2020 387 121682 10.1016/j.jhazmat.2019.121682 31744723 6. Cacace D Fatta-Kassinos D Manaia CM Cytryn E Kreuzinger N Rizzo L Karaolia P Schwartz T Alexander J Merlin C Garelick H Schmitt H de Vries D Schwermer CU Meric S Ozkal CB Pons MN Kneis D Berendonk TU Antibiotic resistance genes in treated wastewater and in the receiving water bodies: a pan-European survey of urban settings Water Res 2019 162 320 10.1016/j.watres.2019.06.039 31288142 7. Sutka A Gross KA Spinel ferrite oxide semiconductor gas sensors Sens Actuators B Chem 2016 222 95 10.1016/j.snb.2015.08.027 8. Fang X Zong BY Mao S Metal-organic framework-based sensors for environmental contaminant sensing Nano Micro Lett 2018 10 92 110 10.1007/s40820-018-0218-0 9. Li M Gou HL Al-Ogaidi I Wu NQ Nanostructured sensors for detection of heavy metals: a review Acs Sustain Chem Eng 2013 1 713 10.1021/sc400019a 10. Qi YY Chen YT Xiu FR Hou JX An aptamer-based colorimetric sensing of acetamiprid in environmental samples: convenience, sensitivity and practicability Sens Actuators B Chem 2020 304 127359 10.1016/j.snb.2019.127359 11. Zhu L Wang JN Liu JW Nasir MS Zhu JW Li SS Liang JD Yan W Smart formaldehyde detection enabled by metal organic framework-derived doped electrospun hollow nanofibers Sens Actuators B Chem 2021 326 128819 10.1016/j.snb.2020.128819 12. 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 13. Wang XX Yu GF Zhang J Yu M Ramakrishna S Long YZ Conductive polymer ultrafine fibers via electrospinning: preparation, physical properties and applications Prog Mater Sci 2021 115 100704 10.1016/j.pmatsci.2020.100704 14. Chen SX Li RQ Li XR Xie JW Electrospinning: an enabling nanotechnology platform for drug delivery and regenerative medicine Adv Drug Deliv Rev 2018 132 188 10.1016/j.addr.2018.05.001 29729295 15. Ghosal K Agatemor C Spitalsky Z Thomas S Kny E Electrospinning tissue engineering and wound dressing scaffolds from polymer-titanium dioxide nanocomposites Chem Eng J 2019 358 1262 10.1016/j.cej.2018.10.117 16. Liu Q Zhu JH Zhang LW Qiu YJ Recent advances in energy materials by electrospinning Renew Sustain Energy Rev 1825 2018 81 17. Sun B Long YZ Zhang HD Li MM Duvail JL Jiang XY Yin HL Advances in three-dimensional nanofibrous macrostructures via electrospinning Prog Polym Sci 2014 39 862 10.1016/j.progpolymsci.2013.06.002 18. Greiner A Wendorff JH Electrospinning: a fascinating method for the preparation of ultrathin fibres Angew Chem Int Ed 2007 46 5670 10.1002/anie.200604646 19. Li D Xia YN Electrospinning of nanofibers: reinventing the wheel? Adv Mater 2004 16 1151 10.1002/adma.200400719 20. 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 21. Ahmed FE Lalia BS Hashaikeh R A review on electrospinning for membrane fabrication: challenges and applications Desalination 2015 356 15 10.1016/j.desal.2014.09.033 22. Zhang H Liu YN Cui K Zhao QL Huang J Mao SH Jiang T Ma Z Electrospun ribbon-like microfiber films of a novel guanidine-based ABA triblock copolymer: fabrication, antibacterial activity, and cytotoxicity Macromol Chem Phys 2019 220 1900138 10.1002/macp.201900138 23. Gangolphe L Leon-Valdivieso CY Nottelet B Dejean S Bethry A Pinese C Bossard F Garric X Electrospun microstructured PLA-based scaffolds featuring relevant anisotropic, mechanical and degradation characteristics for soft tissue engineering Mater Sci Eng C Mater Biol Appl 2021 129 112339 10.1016/j.msec.2021.112339 34579931 24. Liang M Hebraud A Sutter C Bardin A Lobry E Schlatter G Measurement and modeling of the nanofiber surface potential during electrospinning on a patterned collector: toward directed 3D microstructuration Adv Mater Interfaces 2021 8 2101302 10.1002/admi.202101302 25. Park SH Kim MS Lee B Park JH Lee HJ Lee NK Jeon NL Suh KY Creation of a hybrid scaffold with dual configuration of aligned and random electrospun fibers ACS Appl Mater Interfaces 2016 8 2826 10.1021/acsami.5b11529 26756644 26. Shin MK Kim SI Kim SJ Controlled assembly of polymer nanofibers: from helical springs to fully extended Appl Phys Lett 2006 88 223109 10.1063/1.2208689 27. Kongkhlang T Kotaki M Kousaka Y Umemura T Nakaya D Chirachanchai S Electrospun polyoxymethylene: spinning conditions and its consequent nanoporous nanofiber Macromolecules 2008 41 4746 10.1021/ma800731r 28. Park KR Cho HB Lee J Song Y Kim WB Choa YH Design of highly porous SnO2–CuO nanotubes for enhancing H2S gas sensor performance Sens Actuators B Chem 2020 302 127179 10.1016/j.snb.2019.127179 29. Lin MF Xiong JQ Wang JX Parida K Lee PS Core-shell nanofiber mats for tactile pressure sensor and nanogenerator applications Nano Energy 2018 44 248 10.1016/j.nanoen.2017.12.004 30. Wang LG Topham PD Mykhaylyk OO Yu H Ryan AJ Fairclough JPA Bras W Self-assembly-driven electrospinning: the transition from fibers to intact beaded morphologies Macromol Rapid Commun 2015 36 1437 10.1002/marc.201500149 26036661 31. Zhang CL Yu SH Nanoparticles meet electrospinning: recent advances and future prospects Chem Soc Rev 2014 43 4423 10.1039/c3cs60426h 24695773 32. Mercante LA Scagion VP Migliorini FL Mattoso LHC Correa DS Electrospinning-based (bio)sensors for food and agricultural applications: a review Trac Trends Anal Chem 2017 91 91 10.1016/j.trac.2017.04.004 33. Li D Wang YL Xia YN Electrospinning of polymeric and ceramic nanofibers as uniaxially aligned arrays Nano Lett 2003 3 1167 10.1021/nl0344256 34. Baji A Mai YW Wong SC Abtahi M Chen P Electrospinning of polymer nanofibers: effects on oriented morphology, structures and tensile properties Compos Sci Technol 2010 70 703 10.1016/j.compscitech.2010.01.010 35. Agarwal S Greiner A Wendorff JH Functional materials by electrospinning of polymers Prog Polym Sci 2013 38 963 10.1016/j.progpolymsci.2013.02.001 36. Ding YC Hou HQ Zhao Y Zhu ZT Fong H Electrospun polyimide nanofibers and their applications Prog Polym Sci 2016 61 67 10.1016/j.progpolymsci.2016.06.006 37. Thakkar S Misra M Electrospun polymeric nanofibers: new horizons in drug delivery Eur J Pharm Sci 2017 107 148 10.1016/j.ejps.2017.07.001 28690099 38. Bhardwaj N Kundu SC Electrospinning: a fascinating fiber fabrication technique Biotechnol Adv 2010 28 325 10.1016/j.biotechadv.2010.01.004 20100560 39. Deitzel JM Kleinmeyer J Harris D Tan NCB The effect of processing variables on the morphology of electrospun nanofibers and textiles Polymer 2001 42 261 10.1016/S0032-3861(00)00250-0 40. Cheng J Jun Y Qin JH Lee SH Electrospinning versus microfluidic spinning of functional fibers for biomedical applications Biomaterials 2017 114 121 10.1016/j.biomaterials.2016.10.040 27880892 41. Rutledge GC Fridrikh SV Formation of fibers by electrospinning Adv Drug Deliv Rev 2007 59 1384 10.1016/j.addr.2007.04.020 17889398 42. Frenot A Chronakis IS Polymer nanofibers assembled by electrospinning Curr Opin Colloid Interface Sci 2003 8 64 10.1016/S1359-0294(03)00004-9 43. Greiner A, Wendorff JH. Functional Self-Assembled Nanofibers by Electrospinning. In: Shimizu T, editor. Self-assembled nanomaterials I: nanofibers. Advances in Polymer Science, 2008. p. 107. 44. Wang XM Sun FZ Yin GC Wang YT Liu B Dong MD Tactile-sensing based on flexible PVDF nanofibers via electrospinning: a review Sensors 2018 18 330 10.3390/s18020330 29364175 45. Luo CJ Stoyanov SD Stride E Pelan E Edirisinghe M Electrospinning versus fibre production methods: from specifics to technological convergence Chem Soc Rev 2012 41 4708 10.1039/c2cs35083a 22618026 46. Chen K Chou WM Liu LC Cui YH Xue P Jia MY Electrochemical sensors fabricated by electrospinning technology: an overview Sensors 2019 19 3676 10.3390/s19173676 31450877 47. Reneker DH Chun I Nanometre diameter fibres of polymer, produced by electrospinning Nanotechnology 1996 7 216 10.1088/0957-4484/7/3/009 48. Mit-uppatham C Nithitanakul M Supaphol P Ultratine electrospun polyamide-6 fibers: effect of solution conditions on morphology and average fiber diameter Macromol Chem Phys 2004 205 2327 10.1002/macp.200400225 49. Saad EM El Gohary NA El-Shenawy BM Handoussa H Klingner A Elwi M Hamed Y Khalil ISM El Nashar RM Mizaikoff B Fabrication of magnetic molecularly imprinted beaded fibers for rosmarinic acid Nanomaterials 2020 10 1478 10.3390/nano10081478 32731560 50. Vicente A Rivero PJ Palacio JF Rodriguez R The role of the fiber/bead hierarchical microstructure on the properties of PVDF coatings deposited by electrospinning Polymers 2021 13 464 10.3390/polym13030464 33535449 51. Lee KH Zhang SP Lodge TP Frisbie CD Electrical impedance of spin-coatable ion gel films J Phys Chem B 2011 115 3315 10.1021/jp110166u 21410209 52. Schwartz G Tee BCK Mei JG Appleton AL Kim DH Wang HL Bao ZN Flexible polymer transistors with high pressure sensitivity for application in electronic skin and health monitoring Nat Commun 2013 4 1859 10.1038/ncomms2832 23673644 53. Fujimoto T Awaga K Electric-double-layer field-effect transistors with ionic liquids Phys Chem Chem Phys 2013 15 8983 10.1039/c3cp50755f 23665738 54. Nie BQ Li RY Cao J Brandt JD Pan TR Flexible transparent iontronic film for interfacial capacitive pressure sensing Adv Mater 2015 27 6055 10.1002/adma.201502556 26333011 55. Yuk H Zhang T Parada GA Liu XY Zhao XH Skin-inspired hydrogel-elastomer hybrids with robust interfaces and functional microstructures Nat Commun 2016 7 12028 10.1038/ncomms12028 27345380 56. Liu DP Shi QQ Jin S Shao YL Huang J Self-assembled core-shell structured organic nanofibers fabricated by single-nozzle electrospinning for highly sensitive ammonia sensors Infomat 2019 1 525 10.1002/inf2.12036 57. Chartuprayoon N Zhang ML Bosze W Choa YH Myung NV One-dimensional nanostructures based bio-detection Biosens Bioelectron 2015 63 432 10.1016/j.bios.2014.07.043 25128623 58. Fan HJ Gosele U Zacharias M Formation of nanotubes and hollow nanoparticles based on Kirkendall and diffusion processes: a review Small 2007 3 1660 10.1002/smll.200700382 17890644 59. Liu RR Hou LL Yue GC Li HK Zhang JS Liu J Miao BB Wang N Bai J Cui ZM Liu TX Zhao Y Progress of fabrication and applications of electrospun hierarchically porous nanofibers Adv Fiber Mater 2022 4 604 10.1007/s42765-022-00132-z 60. Xu F Yang J Dong RZ Jiang HX Wang CH Liu WL Jiang ZX Zhang XQ Zhu GD Wave-shaped piezoelectric nanofiber membrane nanogenerator for acoustic detection and recognition Adv Fiber Mater 2021 3 368 10.1007/s42765-021-00095-7 61. Kang SX Zhao K Yu DG Zheng XL Huang CX Advances in biosensing and environmental monitoring based on electrospun nanofibers Adv Fiber Mater 2022 4 404 10.1007/s42765-021-00129-0 62. Liu PC Zhu YZ Ma JH Yang SG Gong JH Xu J Preparation of continuous porous alumina nanofibers with hollow structure by single capillary electrospinning Colloids Surf Physicochem Eng Asp 2013 436 489 10.1016/j.colsurfa.2013.07.023 63. Ma GP Yang DZ Nie J Preparation of porous ultrafine polyacrylonitrile (PAN) fibers by electrospinning Polym Adv Technol 2009 20 147 10.1002/pat.1180 64. Scampicchio M Arecchi A Lawrence NS Mannino S Nylon nanofibrous membrane for mediated glucose biosensing Sens Actuators B Chem 2010 145 394 10.1016/j.snb.2009.12.042 65. Wanekaya AK Chen W Myung NV Mulchandani A Nanowire-based electrochemical biosensors Electroanalysis 2006 18 533 10.1002/elan.200503449 66. Miller MR Shaw CA Langrish JP From particles to patients: oxidative stress and the cardiovascular effects of air pollution Future Cardiol 2012 8 577 10.2217/fca.12.43 22871197 67. Glencross DA Ho TR Camilla N Hawrylowicz CM Pfeffer PE Air pollution and its effects on the immune system Free Radic Biol Med 2020 151 56 10.1016/j.freeradbiomed.2020.01.179 32007522 68. Mahajan S Jagtap S Metal-oxide semiconductors for carbon monoxide (CO) gas sensing: a review Appl Mater Today 2019 18 100483 10.1016/j.apmt.2019.100483 69. Zhang ZB Xue T Jin XY Effects of meteorological conditions and air pollution on COVID-19 transmission: evidence from 219 Chinese cities Sci Total Environ 2020 741 140244 10.1016/j.scitotenv.2020.140244 32592975 70. Mirzaei A Leonardi SG Neri G Detection of hazardous volatile organic compounds (VOCs) by metal oxide nanostructures-based gas sensors: a review Ceram Int 2016 42 15119 10.1016/j.ceramint.2016.06.145 71. Kumar P Deep A Kim KH Brown RJC Coordination polymers: opportunities and challenges for monitoring volatile organic compounds Prog Polym Sci 2015 45 102 10.1016/j.progpolymsci.2015.01.002 72. Heft-Neal S Burney J Bendavid E Burke M Robust relationship between air quality and infant mortality in Africa Nature 2018 559 254 10.1038/s41586-018-0263-3 29950722 73. Nikfarjam A Hosseini S Salehifar N Fabrication of a highly sensitive single aligned TiO2 and gold nanoparticle embedded TiO2 nano-fiber gas sensor ACS Appl Mater Interfaces 2017 9 15662 10.1021/acsami.6b15554 28422485 74. Su HY Li H Lin H Shi XW Du YM Luo Y Deng HB Highly sensitive formaldehyde sensors based on CuO/ZnO composite nanofibrous mats using porous cellulose acetate fibers as templates Int J Biol Macromol 2022 206 653 10.1016/j.ijbiomac.2022.02.167 35240217 75. Wang N Wang XF Jia YT Li XQ Yu JY Ding B Electrospun nanofibrous chitosan membranes modified with polyethyleneimine for formaldehyde detection Carbohyd Polym 2014 108 192 10.1016/j.carbpol.2014.02.088 76. Zou YD Zhou XR Zhu YH Cheng XW Zhao DY Deng YH sp2-Hybridized carbon-containing block copolymer templated synthesis of mesoporous semiconducting metal oxides with excellent gas sensing property Acc Chem Res 2019 52 714 10.1021/acs.accounts.8b00598 30829473 77. Ma JH Ren Y Zhou XR Liu LL Zhu YH Cheng XW Xu PC Li XX Deng YH Zhao DY Pt nanoparticles sensitized ordered mesoporous WO3 semiconductor: gas sensing performance and mechanism study Adv Funct Mater 2018 28 1705268 10.1002/adfm.201705268 78. Yang YR Gao W Wearable and flexible electronics for continuous molecular monitoring Chem Soc Rev 2019 48 1465 10.1039/C7CS00730B 29611861 79. Yao MS Lv XJ Fu ZH Li WH Deng WH Wu GD Xu G Layer-by-layer assembled conductive metal-organic framework nanofilms for room-temperature chemiresistive sensing Angew Chem Int Ed 2017 56 16510 10.1002/anie.201709558 80. Nasri A Petrissans M Fierro V Celzard A Gas sensing based on organic composite materials: review of sensor types, progresses and challenges Mater Sci Semicond Process 2021 128 105744 10.1016/j.mssp.2021.105744 81. Srinivasan P Ezhilan M Kulandaisamy AJ Babu KJ Rayappan JBB Room temperature chemiresistive gas sensors: challenges and strategies—a mini review J Mater Sci Mater Electron 2019 30 15825 10.1007/s10854-019-02025-1 82. Majhi SM Mirzaei A Kim HW Kim SS Kim TW Recent advances in energy-saving chemiresistive gas sensors: a review Nano Energy 2021 79 105369 10.1016/j.nanoen.2020.105369 32959010 83. UlAbideen Z Kim JH Lee JH Kim JY Mirzaei A Kim HW Kim SS Electrospun metal oxide composite nanofibers gas sensors: a review J Korean Ceram Soc 2017 54 366 10.4191/kcers.2017.54.5.12 84. Kim HJ Lee JH Highly sensitive and selective gas sensors using p-type oxide semiconductors: overview Sens Actuators B Chem 2014 192 607 10.1016/j.snb.2013.11.005 85. Xu HY Ju DX Chen ZR Han R Zhai T Yu HQ Liu CY Wu XW Wang JQ Cao BQ A novel hetero-structure sensor based on Au/Mg-doped TiO2/SnO2 nanosheets directly grown on Al2O3 ceramic tubes Sens Actuators B Chem 2018 273 328 10.1016/j.snb.2018.06.055 86. Wang C, Wang YL, Cheng PF, Xu LP, Dang F, Wang TL, Lei ZH. In-situ generated TiO2/alpha-Fe2O3 heterojunction arrays for batch manufacturing of conductometric acetone gas sensors. Sensors and Actuators B-Chemical. 2021, 340. 87. Liang QH Zou XX Chen H Fan MH Li GD High-performance formaldehyde sensing realized by alkaline-earth metals doped In2O3 nanotubes with optimized surface properties Sens Actuators B Chem 2020 304 127241 10.1016/j.snb.2019.127241 88. Kou XY Meng FQ Chen K Wang TS Sun P Liu FM Yan X Sun YF Liu FM Shimanoe K Lu GY High-performance acetone gas sensor based on Ru-doped SnO2 nanofibers Sens Actuators B Chem 2020 320 128292 10.1016/j.snb.2020.128292 89. Trung TQ Duy LT Ramasundaram S Lee NE Transparent, stretchable, and rapid-response humidity sensor for body-attachable wearable electronics Nano Res 2021 2017 10 90. Han CH Li XW Shao CL Li XH Ma JG Zhang XT Liu YC Composition-controllable p-CuO/n-ZnO hollow nanofibers for high-performance H2S detection Sens Actuators B Chem 2019 285 495 10.1016/j.snb.2019.01.077 91. Zhang R Cao S Zhou TT Fei T Wang R Zhang T Rational design and tunable synthesis of Co3O4 nanoparticle-incorporating into In2O3 one-dimensional ribbon as effective sensing material for gas detection Sens Actuators B Chem 2020 310 127695 10.1016/j.snb.2020.127695 92. Li XW Han CH Lu DX Shao CL Li XH Liu YC Highly electron-depleted ZnO/ZnFe2O4/Au hollow meshes as an advanced material for gas sensing application Sens Actuators B Chem 2019 297 126769 10.1016/j.snb.2019.126769 93. Zhang M Sui N Wang R Zhang T The effect of shell thickness on gas sensing properties of core-shell fibers Sens Actuators B Chem 2021 332 129456 10.1016/j.snb.2021.129456 94. Filippidou MK Chatzichristidi M Chatzandroulis S A fabrication process of flexible IDE capacitive chemical sensors using a two step lift-off method based on PVA patterning Sens Actuators B Chem 2019 284 7 10.1016/j.snb.2018.12.095 95. Rivadeneyra A Fernandez-Salmeron J Banqueri J Lopez-Villanueva JA Capitan-Vallvey LF Palma AJ A novel electrode structure compared with interdigitated electrodes as capacitive sensor Sens Actuators B Chem 2014 204 552 10.1016/j.snb.2014.08.010 96. Bulemo PM Kim DH Kim ID Controlled synthesis of electrospun hollow Pt-loaded SnO2 microbelts for acetone sensing Sens Actuators B Chem 2021 344 130208 10.1016/j.snb.2021.130208 97. Zhao LB Zhao YH Xia Y Li ZK Li J Zhang JW Wang JH Zhou XY Li YJ Zhao YL Jiang ZD A novel CMUT-based resonant biochemical sensor using electrospinning technology IEEE Trans Ind Electron 2019 66 7356 10.1109/TIE.2018.2878123 98. Bai X Lv H Liu Z Chen JK Wang J Sun BH Zhang Y Wang RH Shi KY Thin-layered MoS2 nanoflakes vertically grown on SnO2 nanotubes as highly effective room-temperature NO2 gas sensor J Hazard Mater 2021 416 125830 10.1016/j.jhazmat.2021.125830 33865111 99. Zhao DW Zhu Y Cheng WK Chen WS Wu YQ Yu HP Cellulose-based flexible functional materials for emerging intelligent electronics Adv Mater 2021 33 2000619 10.1002/adma.202000619 100. Yang LY Xu XR Liu MD Chen C Cui J Chen X Wu K Sun DP Wearable and flexible bacterial cellulose/polyaniline ammonia sensor based on a synergistic doping strategy Sens Actuators B Chem 2021 334 129647 10.1016/j.snb.2021.129647 101. Zhu ZY Liu CC Jiang FX Liu J Liu GQ Ma XM Liu PP Huang R Xu JK Wang L Flexible fiber-shaped hydrogen gas sensor via coupling palladium with conductive polymer gel fiber J Hazard Mater 2021 411 125008 10.1016/j.jhazmat.2020.125008 33445047 102. Nair KG Vishnuraj R Pullithadathil B Highly sensitive, flexible H2 gas sensors based on less platinum bimetallic Ni-Pt nanocatalyst-functionalized carbon nanofibers ACS Appl Electron Mater 2021 3 1621 10.1021/acsaelm.0c01103 103. Khalifa M Anandhan S Highly sensitive and wearable NO2 gas sensor based on PVDF nanofabric containing embedded polyaniline/g-C3N4 nanosheet composites Nanotechnology 2021 32 485504 10.1088/1361-6528/ac1f54 104. Guo CX Sedgwick AC Hirao T Sessler JL Supramolecular fluorescent sensors: an historical overview and update Coord Chem Rev 2021 427 213560 10.1016/j.ccr.2020.213560 34108734 105. Wu Z Liu MM Liu ZC Tian Y Real-time imaging and simultaneous quantification of mitochondrial H2O2 and ATP in neurons with a single two-photon fluorescence-lifetime-based probe J Am Chem Soc 2020 142 7532 10.1021/jacs.0c00771 32233469 106. Tavakoli J Pye S Reza A Xie N Qin J Raston CL Tang BZ Tang YH Tuning aggregation-induced emission nanoparticle properties under thin film formation Mater Chem Front 2020 4 537 10.1039/C9QM00585D 107. Feng HT Yuan YX Xiong JB Zheng YS Tang BZ Macrocycles and cages based on tetraphenylethylene with aggregation-induced emission effect Chem Soc Rev 2018 47 7452 10.1039/C8CS00444G 30177975 108. Li DD Yu JH AIEgens-functionalized inorganic-organic hybrid materials: fabrications and applications Small 2016 12 6478 10.1002/smll.201601484 27510941 109. Tavakoli J Gascooke J Xie N Tang B Tang YH Enlightening freeze thaw process of physically cross-linked poly(vinyl alcohol) hydrogels by aggregation-induced emission fluorogens Acs Appl Polym Mater 2019 1 1390 10.1021/acsapm.9b00173 110. Gao M Xu GC Zhang RH Liu ZJ Xia HM Shao B Xue CH Li JC Miao SH Fu WL Zhang XS Zhou JJ Jiang XP Liang K Kong B Electrospinning superassembled mesoporous AlEgen-organosilica frameworks featuring diversified forms and superstability for wearable and washable solid-state fluorescence smart sensors Anal Chem 2021 93 2367 10.1021/acs.analchem.0c04226 33405892 111. Mao YY Akram M Shi JY Wen JX Yang C Jiang JP Lu ZG Zhou BP Tian YQ Optical oxygen sensors based on microfibers formed from fluorinated copolymers Sens Actuators B Chem 2019 282 885 10.1016/j.snb.2018.11.143 112. Zhang K Luo LL Li WJ Zhang HL Zhang Y Zhao JP Li Y High-performance dissolved oxygen sensors based on platinum(II) porphyrin embedded in polystyrene beads New J Chem 2017 41 6646 10.1039/C7NJ01025G 113. Mao YY Zhao Q Wu JC Pan TT Zhou BP Tian YQ A highly sensitive and fast-responding oxygen sensor based on POSS-containing hybrid copolymer films J Mater Chem C 2017 5 11395 10.1039/C7TC03606J 114. Sun ZJ Cai CX Guo F Ye CH Luo YW Ye SM Luo JC Zhu F Jiang CY Oxygen sensitive polymeric nanocapsules for optical dissolved oxygen sensors Nanotechnology 2018 29 145704 10.1088/1361-6528/aaa058 29219851 115. Huang TC Yeh LC Lai GH Huang BS Yang TI Hsu SC Lo AY Yeh JM Advanced superhydrophobic electroactive fluorinated polyimide and its application in anticorrosion coating Int J Green Energy 2017 14 113 10.1080/15435075.2016.1233495 116. Ren K Wang J Jia HL Electrospinning fibrous films doped with iridium complexes for high performance oxygen sensing: synthesis and characterization Sens Actuators B Chem 2017 240 697 10.1016/j.snb.2016.09.033 117. Cavaliere S Subianto S Savych I Jones DJ Roziere J Electrospinning: designed architectures for energy conversion and storage devices Energy Environ Sci 2011 4 4761 10.1039/c1ee02201f 118. Zaitsev NK, Melnikov PV, Alferov VA, Kopytin AV, German KE, editors. Stable optical oxygen sensing material based on perfluorinated polymer and fluorinated platinum(II) and palladium(II) porphyrins. 30th Eurosensors Conference; 2016 Sep 04–07; Budapest, HUNGARY; 2016. 119. Estella J Wencel D Moore JP Sourdaine M McDonagh C Fabrication and performance evaluation of highly sensitive hybrid sol-gel-derived oxygen sensor films based on a fluorinated precursor Anal Chim Acta 2010 666 83 10.1016/j.aca.2010.03.053 20433969 120. Mao YY Mei ZP Wen JY Li G Tian YH Zhou BP Tian YQ Honeycomb structured porous films from a platinum porphyrin-grafted poly(styrene-co-4-vinylpyridine) copolymer as an optical oxygen sensor Sens Actuators B Chem 2018 257 944 10.1016/j.snb.2017.11.042 121. Schoolaert E Hoogenboom R De Clerck K Colorimetric nanofibers as optical sensors Adv Funct Mater 2017 27 1702646 10.1002/adfm.201702646 122. Lagerwall JPF McCann JT Formo E Scalia G Xia YN Coaxial electrospinning of microfibres with liquid crystal in the core Chem Commun 2008 42 5420 5422 10.1039/b810450f 123. Wang JR Jakli A West JL Morphology tuning of electrospun liquid crystal/polymer fibers ChemPhysChem 2016 17 3080 10.1002/cphc.201600430 27398610 124. Wang D Park SY Kang IK Liquid crystals: emerging materials for use in real-time detection applications J Mater Chem C 2015 3 9038 10.1039/C5TC01321F 125. Cadwell KD Lockwood NA Nellis BA Alf ME Willis CR Abbott NL Detection of organophosphorous nerve agents using liquid crystals supported on chemically functionalized surfaces Sens Actuators B Chem 2007 128 91 10.1016/j.snb.2007.05.044 126. Reyes CG Sharma A Lagerwall JPF Non-electronic gas sensors from electrospun mats of liquid crystal core fibres for detecting volatile organic compounds at room temperature Liq Cryst 1986 2016 43 127. Wang JR Jakli A West JL Liquid crystal/polymer fiber mats as sensitive chemical sensors J Mol Liq 2018 267 490 10.1016/j.molliq.2018.01.051 128. Agra-Kooijman DM Robb C Guan Y Jakli A West JL Liquid crystal core polymer fiber mat electronic gas sensors Liq Cryst 2021 48 1880 1887 10.1080/02678292.2021.1904522 129. Cha JH Kim DH Choi SJ Koo WT Kim ID Sub-parts-per-million hydrogen sulfide colorimetric sensor: lead acetate anchored nanofibers toward halitosis diagnosis Anal Chem 2018 90 8769 10.1021/acs.analchem.8b01273 29790330 130. Cha XL Yu FF Fan Y Chen JF Wang LY Xiang Q Duan ZM Xu JQ Superhydrophilic ZnO nanoneedle array: controllable in situ growth on QCM transducer and enhanced humidity sensing properties and mechanism Sens Actuators B Chem 2018 263 436 10.1016/j.snb.2018.01.110 131. Wang LY Zhu Y Xiang Q Cheng ZX Chen Y Xu JQ One novel humidity-resistance formaldehyde molecular probe based hydrophobic diphenyl sulfone urea dry-gel: synthesis, sensing performance and mechanism Sens Actuators B Chem 2017 251 590 10.1016/j.snb.2017.05.074 132. Wang LY Metal-organic frameworks for QCM-based gas sensors: a review Sens Actuators A Phys 2020 307 11984 10.1016/j.sna.2020.111984 133. Boyadjiev SI Georgieva V Stefan N Stan GE Mihailescu N Visan A Mihailescu IN Besleaga C Szilagyi IM Characterization of PLD grown WO3 thin films for gas sensing Appl Surf Sci 2017 417 218 10.1016/j.apsusc.2017.03.212 134. Wang LY Xu J Wang XH Cheng ZX Xu JQ Facile preparation of N-rich functional polymer with porous framework as QCM sensing material for rapid humidity detection Sens Actuators B Chem 2019 288 289 10.1016/j.snb.2019.02.058 135. Yang MQ He JH Graphene oxide as quartz crystal microbalance sensing layers for detection of formaldehyde Sens Actuators B Chem 2016 228 486 10.1016/j.snb.2016.01.046 136. Kang ZJ Zhang DZ Li TT Liu XH Song XS Polydopamine-modified SnO2 nanofiber composite coated QCM gas sensor for high-performance formaldehyde sensing Sens Actuators B Chem 2021 345 130299 10.1016/j.snb.2021.130299 137. Yang MQ He JH A copper-manganese composite oxide as QCM sensing layers for detection of formaldehyde gas RSC Adv 2018 8 22 10.1039/C7RA11427C 138. Wang LY Yu YP Xiang Q Xu J Cheng ZX Xu JQ PODS-covered PDA film based formaldehyde sensor for avoiding humidity false response Sens Actuators B Chem 2018 255 2704 10.1016/j.snb.2017.09.082 139. El Yakhlifi S Ball V Polydopamine as a stable and functional nanomaterial Colloids Surf B Biointerfaces 2020 186 110719 10.1016/j.colsurfb.2019.110719 31846893 140. Zong J Zhang YS Zhu Y Zhao Y Zhang WJ Zhu YH Rapid and highly selective detection of formaldehyde in food using quartz crystal microbalance sensors based on biomimetic poly-dopamine functionalized hollow mesoporous silica spheres Sens Actuators B Chem 2018 271 311 10.1016/j.snb.2018.05.089 141. Diltemiz SE Ecevit K High-performance formaldehyde adsorption on CuO/ZnO composite nanofiber coated QCM sensors J Alloy Compd 2019 783 608 10.1016/j.jallcom.2018.12.237 142. Zhang HD, Yan X, Zhang ZH, Yu GF, Han WP, Zhang JC, Long YZ. Electrospun PEDOT: PSS/PVP nanofibers for CO gas sensing with quartz crystal microbalance technique. Int J Polym Sci. 2016, 2016. 143. Nagajyoti PC Lee KD Sreekanth TVM Heavy metals, occurrence and toxicity for plants: a review Environ Chem Lett 2010 8 199 10.1007/s10311-010-0297-8 144. Kumar V Parihar RD Sharma A Bakshi P Sidhu GPS Bali AS Karaouzas L Bhardwaj R Thukral AK Gyasi-Agyei Y Rodrigo-Comino J Global evaluation of heavy metal content in surface water bodies: a meta-analysis using heavy metal pollution indices and multivariate statistical analyses Chemosphere 2019 236 124364 10.1016/j.chemosphere.2019.124364 31326755 145. Vardhan KH Kumar PS Panda RC A review on heavy metal pollution, toxicity and remedial measures: current trends and future perspectives J Mol Liq 2019 290 111197 10.1016/j.molliq.2019.111197 146. Malik LA Bashir A Qureashi A Pandith AH Detection and removal of heavy metal ions: a review Environ Chem Lett 2019 17 1495 1521 10.1007/s10311-019-00891-z 147. Peters GR McCurdy RF Hindmarsh JT Environmental aspects of arsenic toxicity Crit Rev Clin Lab Sci 1996 33 457 10.3109/10408369609080055 8989506 148. Liu YC Tian X Cao SW Li Y Dong HJ Li YS Pollution characteristics and health risk assessment of arsenic transformed from feed additive organoarsenicals around chicken farms on the North China Plain Chemosphere 2021 278 130438 10.1016/j.chemosphere.2021.130438 34126682 149. Liu YH Xue Q Chang CW Wang R Wang Q Shan XH Highly efficient detection of Cd(II) ions by a stannum and cerium bimetal-modified laser-induced graphene electrode in water Chem Eng J 2022 433 133791 10.1016/j.cej.2021.133791 150. Yi YH Zhao Y Zhang ZM Wu YT Zhu GB Recent developments in electrochemical detection of cadmium Trends Environ Anal Chem 2022 33 e00152 10.1016/j.teac.2021.e00152 151. Wang CY Ma SJ Pan L Wu WR Wei YM Ou JJ Cyclized conjugated microporous polymer-coated silica nanospheres as fluorescent sensors for iron(III) and chromium(III) Chem Eng J 2022 435 134368 10.1016/j.cej.2021.134368 152. Wang GQ Zhang SR Cui JZ Gao WS Rong X Lu YX Gao CZ Preparation of nitrogen-doped carbon quantum dots from chelating agent and used as fluorescent probes for accurate detection of ClO- and Cr(VI) Anal Chim Acta 2022 1195 339478 10.1016/j.aca.2022.339478 35090652 153. Xu JM Liu MB Zhao WH Wang SQ Gui MF Li HB Yu RQ DNAzyme-based cascade signal amplification strategy for highly sensitive detection of lead ions in the environment J Hazard Mater 2022 429 128347 10.1016/j.jhazmat.2022.128347 35101754 154. Shu Y Ye QY Dai T Guan J Ji ZP Xu Q Hu XY Incorporation of perovskite nanocrystals into lanthanide metal-organic frameworks with enhanced stability for ratiometric and visual sensing of mercury in aqueous solution J Hazard Mater 2022 430 128360 10.1016/j.jhazmat.2022.128360 35152110 155. Stala L Ulatowska J Polowczyk I A review of polyampholytic ion scavengers for toxic metal ion removal from aqueous systems Water Res 2021 203 117523 10.1016/j.watres.2021.117523 34388492 156. Gong ZD Lei YS Wang ZW Zhang J Sun ZJ Li YY Huang JH Chan CC Ouyang X A taper-in-taper structured interferometric optical fiber sensor for Cu2+ ion detection Sensors 2022 22 2709 10.3390/s22072709 35408323 157. Leung AOW Duzgoren-Aydin NS Cheung KC Wong MH Heavy metals concentrations of surface dust from e-waste recycling and its human health implications in southeast China Environ Sci Technol 2008 42 2674 10.1021/es071873x 18505015 158. Gomes SIL de Boer TE van Gestel CAM van Straalen NM Soares A Roelofs D Amorim MJB Molecular mechanisms of zinc toxicity in the potworm Enchytraeus crypticus, analysed by high-throughput gene expression profiling Sci Total Environ 2022 825 153975 10.1016/j.scitotenv.2022.153975 35183636 159. Naveen MH Gurudatt NG Shim YB Applications of conducting polymer composites to electrochemical sensors: a review Appl Mater Today 2017 9 419 10.1016/j.apmt.2017.09.001 160. Kuila T Bose S Khanra P Mishra AK Kim NH Lee JH Recent advances in graphene-based biosensors Biosens Bioelectron 2011 26 4637 10.1016/j.bios.2011.05.039 21683572 161. Promphet N Rattanarat P Rangkupan R Chailapakul O Rodthongkum N An electrochemical sensor based on graphene/polyaniline/polystyrene nanoporous fibers modified electrode for simultaneous determination of lead and cadmium Sens Actuators B Chem 2015 207 526 10.1016/j.snb.2014.10.126 162. Huang H Zhu WC Gao XC Liu XY Ma HY Synthesis of a novel electrode material containing phytic acid-polyaniline nanofibers for simultaneous determination of cadmium and lead ions Anal Chim Acta 2016 947 32 10.1016/j.aca.2016.10.012 27846987 163. Zhao DL Wang TT Han D Rusinek C Steckl AJ Heineman WR Electrospun carbon nanofiber modified electrodes for stripping voltammetry Anal Chem 2015 87 9315 10.1021/acs.analchem.5b02017 26255824 164. Gao SS, Liu J, Luo J, Mamat X, Sambasivam S, Li YT, Hu X, Wagberg T, Hu GZ. Selective voltammetric determination of Cd(II) by using N, S-codoped porous carbon nanofibers. Microchim Acta 2018;185:282. 165. Tang Q Zhu GD Ge YX Yang JM Huang MH Liu JY AuNPs-polyaniline nanosheet array on carbon nanofiber for the determination of As(III) J Electroanal Chem 2020 873 114381 10.1016/j.jelechem.2020.114381 166. Yu XY Liu ZG Huang XJ Nanostructured metal oxides/hydroxides-based electrochemical sensor for monitoring environmental micropollutants Trends Environ Anal Chem 2014 3–4 28 10.1016/j.teac.2014.07.001 167. Oliveira VHB Rechotnek F da Silva EP Marques VD Rubira AF Silva R Lourenco SA Muniz EC A sensitive electrochemical sensor for Pb2+ ions based on ZnO nanofibers functionalized by l-cysteine J Mol Liq 2020 309 113041 10.1016/j.molliq.2020.113041 168. Girija S Sankar SS Thenrajan T Kundu S Wilson J Bi-metallic zeolite imidazole framework nanofibers for the selective determination of Cd2+ ions J Mater Chem B 2021 9 5656 10.1039/D1TB01170G 34190309 169. Teodoro KBR Shimizu FM Scagion VP Correa DS Ternary nanocomposites based on cellulose nanowhiskers, silver nanoparticles and electrospun nanofibers: use in an electronic tongue for heavy metal detection Sens Actuators B Chem 2019 290 387 10.1016/j.snb.2019.03.125 170. Li ZX Li HX Shi CX Zhang WY Zhou W Wei LH Yu MM Naked-eye-based highly selective sensing of Fe3+ and further for PPi with nano copolymer film Sens Actuators B Chem 2016 226 127 10.1016/j.snb.2015.11.105 171. Abedalwafa MA Li Y Li D Sanbhal N Yang JM Wang L Aminated polyacrylonitrile nanofibers with immobilized gold-silver core-shell nanoparticles for use in a colorimetric test strip for copper(II) Microchim Acta 2018 185 1 10 10.1007/s00604-018-2938-8 172. Zhang CD Li HD Yu QZ Jia L Wan LY Poly(aspartic acid) electrospun nanofiber hydrogel membrane-based reusable colorimetric sensor for Cu(II) and Fe(III) detection ACS Omega 2019 4 14633 10.1021/acsomega.9b02109 31528819 173. Wang L Li M Li WT Han Y Liu YJ Li Z Zhang BH Pan DY Rationally designed efficient dual-mode colorimetric/fluorescence sensor based on carbon dots for detection of pH and Cu2+ Ions Acs Sustain Chem Eng 2018 6 12668 10.1021/acssuschemeng.8b01625 174. Terra IAA Mercante LA Andre RS Correa DS Fluorescent and colorimetric electrospun nanofibers for heavy-metal sensing Biosensors (Basel) 2017 7 61 10.3390/bios7040061 29244741 175. Hu Y Chen L Jung H Zeng YY Lee S Swamy KMK Zhou X Kim MH Yoon J Effective strategy for colorimetric and fluorescence sensing of phosgene based on small organic dyes and nanofiber platforms ACS Appl Mater Interfaces 2016 8 22246 10.1021/acsami.6b07138 27498559 176. Wang W Li YP Sun MD Zhou C Zhang Y Li YX Yang QB Colorimetric and fluorescent nanofibrous film as a chemosensor for Hg2+ in aqueous solution prepared by electrospinning and host-guest interaction Chem Commun 2012 48 6040 10.1039/c2cc17664e 177. Parsaee Z Electrospun nanofibers decorated with bio-sonochemically synthesized gold nanoparticles as an ultrasensitive probe in amalgam-based mercury(II) detection system Ultrason Sonochem 2018 44 24 10.1016/j.ultsonch.2018.02.005 29680608 178. Guo PZ Sikdar D Huang XQ Si KJ Xiong W Gong S Yap LW Premaratne M Cheng WL Plasmonic core-shell nanoparticles for SERS detection of the pesticide thiram: size- and shape-dependent Raman enhancement Nanoscale 2015 7 2862 10.1039/C4NR06429A 25599516 179. Yang T Ma J Zhen SJ Huang CZ Electrostatic assemblies of well-dispersed AgNPs on the surface of electrospun nanofibers as highly active SERS substrates for wide range pH sensing ACS Appl Mater Interfaces 2016 8 14802 10.1021/acsami.6b03720 27214514 180. Xu SY Tang WQ Chase DB Sparks DL Rabolt JF A Highly sensitive, selective, and reproducible SERS sensor for detection of trace metalloids in the environment Acs Appl Nano Mater 2018 1 1257 10.1021/acsanm.7b00301 181. Zhang H Cao MH Wu W Xu HB Cheng S Fan LJ Polyacrylonitrile/noble metal/SiO2 nanofibers as substrates for the amplified detection of picomolar amounts of metal ions through plasmon-enhanced fluorescence Nanoscale 2015 7 1374 10.1039/C4NR05349D 25494487 182. Wang Q Xue Q Chen T Li JW Liu YH Shan XH Liu F Jia JB Recent advances in electrochemical sensors for antibiotics and their applications Chin Chem Lett 2021 32 609 10.1016/j.cclet.2020.10.025 183. Sun SM Korheina DKA Fu HT Ge XP Chronic exposure to dietary antibiotics affects intestinal health and antibiotic resistance gene abundance in oriental river prawn (Macrobrachium nipponense), and provokes human health risk Sci Total Environ 2020 720 137478 10.1016/j.scitotenv.2020.137478 32145616 184. Kumar M Jaiswal S Sodhi KK Shree P Singh DK Agrawal PK Shukla P Antibiotics bioremediation: perspectives on its ecotoxicity and resistance Environ Int 2019 124 448 10.1016/j.envint.2018.12.065 30684803 185. Huang FY An ZY Moran MJ Liu F Recognition of typical antibiotic residues in environmental media related to groundwater in China (2009–2019) J Hazard Mater 2020 399 122813 10.1016/j.jhazmat.2020.122813 32937691 186. Wang HX Wang N Wang B Zhao Q Fang H Fu CW Tang CX Jiang F Zhou Y Chen Y Jiang QW Antibiotics in drinking water in Shanghai and their contribution to antibiotic exposure of school children Environ Sci Technol 2016 50 2692 10.1021/acs.est.5b05749 26849047 187. Qiao M Ying GG Singer AC Zhu YG Review of antibiotic resistance in China and its environment Environ Int 2018 110 160 10.1016/j.envint.2017.10.016 29107352 188. Yang YY Song WJ Lin H Wang WB Du LN Xing W Antibiotics and antibiotic resistance genes in global lakes: a review and meta-analysis Environ Int 2018 116 60 10.1016/j.envint.2018.04.011 29653401 189. Zhang QQ Ying GG Pan CG Liu YS Zhao JL Comprehensive evaluation of antibiotics emission and fate in the river Basins of China: source analysis, multimedia modeling, and linkage to bacterial resistance Environ Sci Technol 2015 49 6772 10.1021/acs.est.5b00729 25961663 190. Ying GG He LY Ying AJ Zhang QQ Liu YS Zhao JL China must reduce its antibiotic use Environ Sci Technol 2017 51 1072 10.1021/acs.est.6b06424 28094517 191. Bu QW Wang B Huang J Deng SB Yu G Pharmaceuticals and personal care products in the aquatic environment in China: a review J Hazard Mater 2013 262 189 10.1016/j.jhazmat.2013.08.040 24036145 192. Huang FY Zou SZ Deng DD Lang H Liu F Antibiotics in a typical karst river system in China: spatiotemporal variation and environmental risks Sci Total Environ 2019 650 1348 10.1016/j.scitotenv.2018.09.131 30308821 193. Liu X Steele JC Meng XZ Usage, residue, and human health risk of antibiotics in Chinese aquaculture: a review Environ Pollut 2017 223 161 10.1016/j.envpol.2017.01.003 28131482 194. Li S Shi WZ Liu W Li HM Zhang W Hu JR Ke YC Sun WL Ni JR A duodecennial national synthesis of antibiotics in China's major rivers and seas (2005–2016) Sci Total Environ 2018 615 906 10.1016/j.scitotenv.2017.09.328 29017132 195. Baby JN Sriram B Wang SF George M Integration of samarium vanadate/carbon nanofiber through synergy: an electrochemical tool for sulfadiazine analysis J Hazard Mater 2021 408 124940 10.1016/j.jhazmat.2020.124940 33387714 196. Song JL Huang MH Lin XH Li SFY Jiang N Liu YB Guo HD Li YM Novel Fe-based metal-organic framework (MOF) modified carbon nanofiber as a highly selective and sensitive electrochemical sensor for tetracycline detection Chem Eng J 2022 427 130913 10.1016/j.cej.2021.130913 197. Ayankojo AG Reut J Opik A Furchner A Syritski V Hybrid molecularly imprinted polymer for amoxicillin detection Biosens Bioelectron 2018 118 102 10.1016/j.bios.2018.07.042 30056300 198. Yang GM Zhao FQ Molecularly imprinted polymer grown on multiwalled carbon nanotube surface for the sensitive electrochemical determination of amoxicillin Electrochim Acta 2015 174 33 10.1016/j.electacta.2015.05.156 199. Chen CL Lv XC Lei W Wu Y Feng SS Ding Y Lv JJ Hao QL Chen SM Amoxicillin on polyglutamic acid composite three-dimensional graphene modified electrode: reaction mechanism of amoxicillin insights by computational simulations Anal Chim Acta 2019 1073 22 10.1016/j.aca.2019.04.052 31146832 200. Zhang Y Tang WF Wang YQ Nian M Jiang F Zhang J Chen Q Environmental antibiotics exposure in school-age children in Shanghai and health risk assessment: a population-based representative investigation Sci Total Environ 2022 824 153859 10.1016/j.scitotenv.2022.153859 35176387 201. Wang HX Yang JQ Yu X Zhao GM Zhao Q Wang N Jiang YG Jiang F He GS Chen Y Zhou ZJ Jiang QW Exposure of adults to antibiotics in a shanghai suburban area and health risk assessment: a biomonitoring-based study Environ Sci Technol 2018 52 13942 10.1021/acs.est.8b03979 30388002 202. Tan XQ Liang YP Ye YY Liu ZH Meng JX Li FY Explainable deep learning-assisted fluorescence discrimination for aminoglycoside antibiotic identification Anal Chem 2022 94 829 836 10.1021/acs.analchem.1c03508 34978809 203. Pan Y Shan D Ding LL Yang XD Xu K Huang H Wang JF Ren HQ Developing a generally applicable electrochemical sensor for detecting macrolides in water with thiophene-based molecularly imprinted polymers Water Res 2021 205 117670 10.1016/j.watres.2021.117670 34583204 204. Omuferen LO Maseko B Olowoyo JO Occurrence of antibiotics in wastewater from hospital and convectional wastewater treatment plants and their impact on the effluent receiving rivers: current knowledge between 2010 and 2019 Environ Monit Assess 2022 194 1 25 10.1007/s10661-022-09846-4 205. Yin Y Zhu D Yang G Su JQ Duan GL Diverse antibiotic resistance genes and potential pathogens inhabit in the phyllosphere of fresh vegetables Sci Total Environ 2022 815 152851 10.1016/j.scitotenv.2021.152851 34990692 206. Wang J Teng XM Wang YS Si SX Ju J Pan W Wang JP Sun XB Wang WJ Carbon dots based fluorescence methods for the detections of pesticides and veterinary drugs: response mechanism, selectivity improvement and application Trac-Trends Anal Chem 2021 144 116430 10.1016/j.trac.2021.116430 207. Bu F Huang W Xian M Zhang XL Liang FB Liu XC Sun XY Feng DX Magnetic carboxyl-functionalized covalent organic frameworks for adsorption of quinolones with high capacities, fast kinetics and easy regeneration J Clean Prod 2022 336 130485 10.1016/j.jclepro.2022.130485 208. Wang HX Wang N Wang B Fang H Fu CW Tang CX Jiang F Zhou Y He GS Zhao Q Chen Y Jiang QW Antibiotics detected in urines and adipogenesis in school children Environ Int 2016 89–90 204 10.1016/j.envint.2016.02.005 209. Liu ZP Jin ML Lu H Yao JY Wang X Zhou GF Shui LL Molecularly imprinted polymer decorated 3D-framework of functionalized multi-walled carbon nanotubes for ultrasensitive electrochemical sensing of Norfloxacin in pharmaceutical formulations and rat plasma Sens Actuators B Chem 2019 288 363 10.1016/j.snb.2019.02.097 210. Feng LM Xue Q Liu F Cao QP Feng JJ Yang L Zhang FL Voltammetric determination of ofloxacin by using a laser-modified carbon glassy electrode Microchim Acta 2020 187 1 8 10.1007/s00604-019-4065-6 211. Fenzl C Nayak P Hirsch T Wolfbeis OS Alshareef HN Baeumner AJ Laser-scribed graphene electrodes for aptamer-based biosensing Acs Sens 2017 2 616 10.1021/acssensors.7b00066 28723173 212. Starzec K Cristea C Tertis M Feier B Wieczorek M Koscielniak P Kochana J Employment of electrostriction phenomenon for label-free electrochemical immunosensing of tetracycline Bioelectrochemistry 2020 132 107405 10.1016/j.bioelechem.2019.107405 31830671 213. Ni Y Liao Y Zheng MB Shao SJ In-situ growth of Co3O4 nanoparticles on mesoporous carbon nanofibers: a new nanocomposite for nonenzymatic amperometric sensing of H2O2 Microchim Acta 2017 184 3689 10.1007/s00604-017-2395-9 214. Kokulnathan T Sharma TSK Chen SM Han-Yu Y Synthesis and characterization of zirconium dioxide anchored carbon nanofiber composite for enhanced electrochemical determination of chloramphenicol in food samples J Electrochem Soc 2018 165 B281 10.1149/2.0861807jes 215. Qin LY Yang DZ Zhang M Zhao TY Luo Z Yu ZZ Superelastic and ultralight electrospun carbon nanofiber/MXene hybrid aerogels with anisotropic microchannels for pressure sensing and energy storage J Colloid Interface Sci 2021 589 264 10.1016/j.jcis.2020.12.102 33460857 216. Scagion VP Mercante LA Sakamoto KY Oliveira JE Fonseca FJ Mattoso LHC Ferreira MD Correa DS An electronic tongue based on conducting electrospun nanofibers for detecting tetracycline in milk samples RSC Adv 2016 6 103740 10.1039/C6RA21326J 217. Chen XY Hao SB Zong BY Liu CB Mao S Ultraselective antibiotic sensing with complementary strand DNA assisted aptamer/MoS2 field-effect transistors Biosens Bioelectron 2019 145 111711 10.1016/j.bios.2019.111711 31563801 218. Yin XH Hou T Huang BZ Yang LM Li F Aptamer recognition-trigged label-free homogeneous electrochemical strategy for an ultrasensitive cancer-derived exosome assay Chem Commun 2019 55 13705 10.1039/C9CC07253E 219. Li FL Wang XY Sun X Guo YM Multiplex electrochemical aptasensor for detecting multiple antibiotics residues based on carbon fiber and mesoporous carbon-gold nanoparticles Sens Actuators B Chem 2018 265 217 10.1016/j.snb.2018.03.042 220. Ramezani M Danesh NM Lavaee P Abnous K Taghdisi SM A selective and sensitive fluorescent aptasensor for detection of kanamycin based on catalytic recycling activity of exonuclease III and gold nanoparticles Sens Actuators B Chem 2016 222 1 10.1016/j.snb.2015.08.024 221. Yang Y Yang ZH Lv JJ Yuan R Chai YQ Thrombin aptasensor enabled by Pt nanoparticles-functionalized Co-based metal organic frameworks assisted electrochemical signal amplification Talanta 2017 169 44 10.1016/j.talanta.2017.03.037 28411820 222. Lebedev OI Millange F Serre C Van Tendeloo G Ferey G First direct imaging of giant pores of the metal-organic framework MIL-101 Chem Mater 2005 17 6525 10.1021/cm051870o 223. Liu Z Su RD Sun X Zhou WZ Gao BY Yue QY Li Q The obvious advantage of amino-functionalized metal-organic frameworks: as a persulfate activator for bisphenol F degradation Sci Total Environ 2020 741 140464 10.1016/j.scitotenv.2020.140464 32886982 224. Vafaye SE Rahman A Safaeian S Adabi M An electrochemical aptasensor based on electrospun carbon nanofiber mat and gold nanoparticles for the sensitive detection of Penicillin in milk J Food Meas Charact 2021 15 876 10.1007/s11694-020-00684-x 225. Liu Q Zheng YM Zhong LB Cheng XX Removal of tetracycline from aqueous solution by a Fe3O4 incorporated PAN electrospun nanofiber mat J Environ Sci 2015 28 29 10.1016/j.jes.2014.04.016 226. Mudabuka B Ogunlaja AS Tshentu ZR Torto N Electrospun poly(vinylbenzyl chloride) nanofibres functionalised with tris-(2,2′-pyridylimidazole)iron(III): a test strip for detection of ascorbic acid and dopamine Sens Actuators B Chem 2016 222 598 10.1016/j.snb.2015.08.081 227. Li Y Mohammed A Li D Wang L Test strips based on iron(III)-impregnated alginate/polyacrylonitrile nanofibers for naked eye screening of tetracycline Analyst 2018 143 3029 10.1039/C7AN02038D 29721556 228. Chen D Deng J Liang J Xie J Hu CH Huang KH A core-shell molecularly imprinted polymer grafted onto a magnetic glassy carbon electrode as a selective sensor for the determination of metronidazole Sens Actuators B Chem 2013 183 594 10.1016/j.snb.2013.04.050 229. Abedalwafa MA Li Y Ni CF Yang G Wang L Non-enzymatic colorimetric sensor strip based on melamine-functionalized gold nanoparticles assembled on polyamide nanofiber membranes for the detection of metronidazole Anal Methods 2019 11 3706 10.1039/C9AY01114E 230. Mao K Yang ZG Zhang H Li XQ Cooper JM Paper-based nanosensors to evaluate community-wide illicit drug use for wastewater-based epidemiology Water Res 2021 189 116559 10.1016/j.watres.2020.116559 33161329 231. Tang YZ Chen X Lv Y Wu ZY Chen F Chen ZG Excellent surface enhanced raman scattering of SiO2 fiber membrane embedded with Ag nanoparticles J Inorg Organomet Polym Mater 2018 28 251 10.1007/s10904-017-0703-9 232. Supraja P Tripathy S Vanjari SRK Singh V Singh SG Electrospun tin (IV) oxide nanofiber based electrochemical sensor for ultra-sensitive and selective detection of atrazine in water at trace levels Biosens Bioelectron 2019 141 111441 10.1016/j.bios.2019.111441 31229795 233. Yang XM Li X Zhang LZ Gong JM Electrospun template directed molecularly imprinted nanofibers incorporated with BiOI nanoflake arrays as photoactive electrode for photoelectrochemical detection of triphenyl phosphate Biosens Bioelectron 2017 92 61 10.1016/j.bios.2017.01.056 28187300 234. Yan X Li HX Su XG Review of optical sensors for pesticides Trac Trends Anal Chem 2018 103 1 10.1016/j.trac.2018.03.004 235. Watanabe E Baba K Highly sensitive quantification of pyrethroid insecticide etofenprox in vegetables with high-performance liquid chromatography and fluorescence detection J Chromatogr A 2015 1385 35 10.1016/j.chroma.2015.01.056 25662063 236. Jafari MT Saraji M Sherafatm H Polypyrrole/montmorillonite nanocomposite as a new solid phase microextraction fiber combined with gas chromatography-corona discharge ion mobility spectrometry for the simultaneous determination of diazinon and fenthion organophosphorus pesticides Anal Chim Acta 2014 814 69 10.1016/j.aca.2014.01.037 24528846 237. Rodrigues ET Pardal MA Salgueiro-Gonzalez N Muniategui-Lorenzo S Alpendurada MF A single-step pesticide extraction and clean-up multi-residue analytical method by selective pressurized liquid extraction followed by on-line solid phase extraction and ultra-high-performance liquid chromatography-tandem mass spectrometry for complex matrices J Chromatogr A 2016 1452 10 10.1016/j.chroma.2016.05.036 27234845 238. Feng K Zhai MY Wei YS Zong MH Wu H Han SY Fabrication of nano/micro-structured electrospun detection card for the detection of pesticide residues Foods 2021 10 889 10.3390/foods10040889 33921560 239. El-Moghazy AY Soliman EA Ibrahim HZ Marty JL Istamboulie G Noguer T Biosensor based on electrospun blended chitosan-poly (vinyl alcohol) nanofibrous enzymatically sensitized membranes for pirimiphosmethyl detection in olive oil Talanta 2016 155 258 10.1016/j.talanta.2016.04.018 27216682 240. Shao F Cao JY Ying Y Liu Y Wang D Guo XY Wu YP Wen Y Yang HF Preparation of hydrophobic film by electrospinning for rapid SERS detection of trace triazophos Sensors 2020 20 4120 10.3390/s20154120 32722113 241. Ji YF Shi YY Yang Y Yang PZ Wang L Lu JH Li JH Zhou L Ferronato C Chovelon JM Rethinking sulfate radical-based oxidation of nitrophenols: Formation of toxic polynitrophenols, nitrated biphenyls and diphenyl ethers J Hazard Mater 2019 361 152 10.1016/j.jhazmat.2018.08.083 30179786 242. Wang YA Li QC Zhang P O'Connor D Varma RS Yu M Hou DY One-pot green synthesis of bimetallic hollow palladium-platinum nanotubes for enhanced catalytic reduction of p-nitrophenol J Colloid Interface Sci 2019 539 161 10.1016/j.jcis.2018.12.053 30580171 243. Gerent GG Spinelli A Magnetite-platinum nanoparticles-modified glassy carbon electrode as electrochemical detector for nitrophenol isomers J Hazard Mater 2017 330 105 10.1016/j.jhazmat.2017.02.002 28214399 244. Wu H Guo LX Zhang JL Miao SL He CY Wang B Wu YQ Chen ZM Polyelectrolyte-free layer by layer self-assembled multilayer films of cationic phthalocyanine cobalt(II) and carbon nanotube for the efficient detection of 4-nitrophenol Sens Actuators B Chem 2016 230 359 10.1016/j.snb.2016.02.088 245. Han L Liu SG Liang JY Ju YJ Li NB Luo HQ pH-mediated reversible fluorescence nanoswitch based on inner filter effect induced fluorescence quenching for selective and visual detection of 4-nitrophenol J Hazard Mater 2019 362 45 10.1016/j.jhazmat.2018.09.025 30236941 246. Hu LM Zhang GS Wang Q Wang XJ Wang P Effect of microwave heating on persulfate activation for rapid degradation and mineralization of p-nitrophenol Acs Sustain Chem Eng 2019 7 11662 10.1021/acssuschemeng.9b01686 247. Zhu GD Tang Q Huang MH Yang JM Xu R Liu JY Polyaniline nanoconical array on carbon nanofiber for supersensitive determination of nitrophenol Sens Actuators B Chem 2020 320 128593 10.1016/j.snb.2020.128593 248. Antohe I Iordache I Antohe VA Socol G A polyaniline/platinum coated fiber optic surface plasmon resonance sensor for picomolar detection of 4-nitrophenol Sci Rep 2021 11 1 10 10.1038/s41598-021-89396-w 33414495 249. Jadhav SA Biji P Panthalingal MK Krishna CM Rajkumar S Joshi DS Sundaram N Development of integrated microfluidic platform coupled with surface-enhanced raman spectroscopy for diagnosis of COVID-19 Med Hypotheses 2021 146 110356 10.1016/j.mehy.2020.110356 33342643 250. Xu ZQ Fan GH Zheng T Lin CC Lin XC Xie ZH Aptamer-functionalized metal-organic framework-based electrospun nanofibrous composite coating fiber for specific recognition of ultratrace microcystin in water J Chromatogr A 2021 1656 462542 10.1016/j.chroma.2021.462542 34543883 251. Yang Y Zhang ZJ He YL Wang ZH Zhao YB Sun L Fabrication of Ag@TiO2 electrospinning nanofibrous felts as SERS substrate for direct and sensitive bacterial detection Sens Actuators B Chem 2018 273 600 10.1016/j.snb.2018.05.129 252. Niri AD Faridi-Majidi R Saber R Khosravani M Adabi M Electrospun carbon nanofiber-based electrochemical biosensor for the detection of hepatitis B virus Biointerface Res Appl Chem 2019 9 4022 10.33263/BRIAC94.022026 253. Arshad R Rhouati A Hayat A Nawaz MH Yameen MA Mujahid A Latif U MIP-based impedimetric sensor for detecting dengue fever biomarker Appl Biochem Biotechnol 2020 191 1384 10.1007/s12010-020-03285-y 32100233	0	PMC9734373	NO-CC CODE	2022-12-14 23:28:28	no	Adv Fiber Mater. 2022 Dec 5;:1-32	utf-8	Adv Fiber Mater	2,022	10.1007/s42765-022-00237-5	oa_other
==== Front Rev Environ Contam Toxicol Rev Environ Contam Toxicol Reviews of Environmental Contamination and Toxicology 0179-5953 2197-6554 Springer International Publishing Cham 14 10.1007/s44169-022-00014-w Review Occurrence of Pharmaceutical and Pesticide Transformation Products in Freshwater: Update on Environmental Levels, Toxicological Information and Future Challenges Rodrigues P. 123 Oliva-Teles L. 12 http://orcid.org/0000-0003-3360-3783 Guimarães L. [email protected] 12 Carvalho A. P. 12 1 grid.5808.5 0000 0001 1503 7226 CIIMAR - Interdisciplinary Centre of Marine and Environmental Research, University of Porto, Terminal de Cruzeiros do Porto de Leixões Av. General Norton de Matos S/n, 4450-208 Matosinhos, Portugal 2 grid.5808.5 0000 0001 1503 7226 Department of Biology, FCUP – Faculty of Sciences, University of Porto, Rua do Campo Alegre, S/N, 4169-007 Porto, Portugal 3 grid.5808.5 0000 0001 1503 7226 ICBAS/UP-Institute of Biomedical Sciences Abel Salazar, University of Porto, Rua de Jorge Viterbo Ferreira, 228, 4050-313 Porto, Portugal 7 12 2022 2022 260 1 149 5 2022 28 11 2022 © The Author(s), under exclusive license 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. Pharmaceuticals and pesticides are recognized micropollutants in freshwater systems. Their ever-increasing frequency of detection, levels found and little information available about their effects on non-target organisms, make them emerging contaminants. However, parental compounds are not the only substances of concern. Their metabolites and degradation products, hereby referred to as transformation products, are increasingly detected in freshwater samples and wastewater effluents. In the past years, a wealth of publications provided concentration levels detected in freshwater and some toxicological data, which required critical systematization. This review identified concentrations for 190 transformation products (92 from pesticides and 98 from pharmaceuticals) in water bodies and wastewater effluents. A concentration heatmap was produced to easily spot the substances found at higher levels and plan future research. The very limited available toxicological data link exposure to transformation products to adverse outcomes in humans (genotoxicity and alteration in detoxification processes) and aquatic species (mostly related to apical endpoints). Overall, environmental levels of these transformation products may pose a severe threat to aquatic organisms and need to be further investigated in sound experimental designs, testing for the effects of the single substances as well as of their mixtures. Such toxicological information is highly needed to improve both water treatment technologies and monitoring programmes. Supplementary Information The online version contains supplementary material available at 10.1007/s44169-022-00014-w. BiodivRestore 2020 ERA-NET CofundBioReset (DivRestore/0004/2020) Fundação para a Ciência e TecnologiaUIDB/04423/2020 UIDP/04423/2020 SFRH/BD/134518/2017 Rodrigues P. issue-copyright-statement© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 ==== Body pmcPesticide and Pharmaceutical Transformation Products as Environmental Contaminants Over the past decades, scientists produced a wealth of information about the toxic effects of pesticides and pharmaceuticals on freshwater species. Both groups of compounds are widely used in the world, with recognized benefits for human health and welfare (Santos et al. 2010; Mcknight et al. 2015). Moreover, their use is globally escalating, owing to i) today’s social habits, ii) the increase in life expectancy, iii) the human population growth and iv) the consequent increase in food demand. More so under the actual pandemia caused by the SARS-CoV-2 coronavirus. Despite the need for them, these classes of chemicals have also been associated with severe human and environmental health risks and are common micropollutants of freshwater systems (Corcoran et al. 2010; Santos et al. 2013; Reemtsma et al. 2013; Ortiz de García et al. 2014; Mcknight et al. 2015). Pesticides and pharmaceuticals are designed to have specific biological activity, exerting the desired effect before undergoing excretion and/or degradation. These same characteristics also make them persistent in the environment, ultimately causing toxicity to non-target fauna and flora (Fent et al. 2006; McKnight et al. 2015). Due to these characteristics and the still limited information available, they are the most representative classes included in the watch list of substances for Union-wide monitoring of the Water Framework Directive (WFD). Currently, they represent over 88% of the compounds listed in the WFD (European Union 2022). Pesticides occur in natural water, mainly by run-off from the agricultural fields where they are applied and through industrial wastewater. Although soils can store a good amount of pesticides due to the high affinity of these compounds to organic matter, surface water and groundwater are susceptible to pesticide contamination because of the existing soil–water interconnections, mainly adsorption (Sharma et al. 2019). Pesticides that are highly adsorbed to soil particles are less likely to infiltrate deep but can easily be carried via run-off of and reach surface water (Syafrudin et al. 2021). Due to their increased use, detection of pesticides in different water compartments is becoming more and more frequent (Corcoran et al. 2010; Reemtsma et al. 2013; Ortiz de García et al. 2014; Evgenidou et al. 2015; Vryzas, 2018). On the other hand, the distribution of pharmaceutical substances in the environment is predominantly made by aqueous transport of compounds contained in discharged wastewater effluents, which persisted through the conventional treatment processes (Khan et al. 2020). Contamination by pharmaceutical compounds may also occur by terrestrial run-off from agricultural fields and aquaculture activities (Hong et al. 2018). Sorption is also an important process for the transport of pharmaceuticals in an aquatic environment. This process is responsible for the partitioning of pharmaceuticals between the water and the sediment phase (Bavumiragira et al. 2022). Once pesticides and pharmaceuticals reach the aquatic environment, they undergo a series of abiotic and biotic transformation and degradation processes. Hydrolysis, photodegradation and biodegradation are considered the most important mechanisms involved in their transformation or degradation (Syafrudin et al. 2021; Khan et al. 2020). Hydrolysis is an abiotic degradation process that creates products more polar than the parental compounds. These reactions are mainly catalysed by hydrogen or hydroxide molecules (Bavumiragira et al. 2022). Photolysis or photochemical degradation of pesticides and pharmaceuticals occurs by decomposition of these compounds in the presence of ultraviolet (UV) light. When exposed to sunlight, pesticides and pharmaceuticals containing chemical functional groups able to absorb solar radiation are prone to photolysis. The reaction transforms parental compounds into transformation products that are usually more biodegradable and hydrolysable (Wilkinson et al. 2017; Bavumiragira et al. 2022). Biodegradation is a biotic process that can result in the partial or complete transformation of pesticides and pharmaceuticals by microorganisms, such as certain fungi, bacteria, protozoans and microalgae. These microorganisms are present in wastewater treatment plants (WWTPs) or occur naturally in suspended solids, sediments and within animals (i.e. gut microbiota) (Wilkinson et al. 2017; Jaffar et al. 2022). Microbial degradation is recognized in the literature as having an important role in the degradation of several pharmaceuticals in a wide range of water compartments (Christensen and Li, 2014). For pesticides, microbial degradation includes the mineralisation process, which consists in the break of a parental pesticide into carbon dioxide and co-metabolization where microbial-catalysed reactions break pesticides into other chemical forms (Syafrudin et al. 2021). Surface waters receiving wastewater effluents rich in microorganisms are usually prone to show higher biodegradation effectiveness. High rates of biodegradation are typically observed along the sediment–water interface in water bodies and wetlands (Li et al. 2016). Degradation effectiveness varies according to biotic and abiotic factors, such as temperature, pH, UV light, presence of dissolved organic matter, suspended material and micro- and macrobiota (Vryzas, 2018). Low turbidity, small depth, low total organic carbon content and sandy sediments favour the degradation of pesticides and pharmaceuticals (Baena-Nogueras et al. 2017). On the other hand, higher depths, low temperature and higher turbidity can lower the degradation effectiveness (Syafrudin et al. 2021; Bavumiragira et al. 2022). Nevertheless, the described processes originate transformation products that enter in natural water by a panoply of different sources. In recent years, several works have reported the detection of these transformation products in the range of ng to µg/L, sometimes at concentrations even higher than those found for the parental compounds (le Cor et al. 2021). However, the focus of the reports was primarily on the detection and quantification of the parental compounds. Concern about their transformation products, with the involvement of more groups in this research, took off mostly in the last decade, especially for pharmaceutical transformation products. Investigation about the occurrence and fate of transformation products in the aquatic environment skyrocketed in recent years, mainly due to advances reached in the chemical analytical methods (Fent et al. 2006; Valls-Cantenys et al. 2016). New instruments and methods with higher separation efficiencies, ability to find more polar compounds and deal with confounding matrix effects, appeared allowing scientists to detect trace concentrations in environmental compartments (Fent et al. 2006; Celiz et al. 2009; Valls-Cantenys et al. 2016). Previous excellent reviews have been dedicated to this topic, although mostly to pharmaceutical and personal care products or emerging contaminants of concern and less so to pesticides (La Farre et al. 2008; Celiz et al. 2009; Mompelat et al. 2009; Evgenidou et al. 2015; Picó & Barceló, 2015; le Cor et al. 2021; Ibáñez et al. 2021; Mosekiemang et al. 2021; Madikizela et al. 2022). Furthermore, the number of works produced about this theme suffered a remarkable increase in recent years. Many of these compounds, parental or transformation products, are however little known in terms of potential detrimental effects and not included in the regulatory monitoring frameworks. Hence, they are nowadays recognized as emerging contaminants of concern (Murray et al. 2010; Evgenidou et al. 2015; NORMAN network, www.norman-network.net). Pesticides and pharmaceuticals are the two main classes of chemicals continuously represented in the watch list of the WFD and are thus the focus of this review. The aim of this literature review was to identify ecotoxicological knowledge gaps limiting the risk assessment of transformation products of pesticides and pharmaceuticals found in aquatic samples. We present and discuss updated information about quantification methods, occurrence, fate and the effects of transformation products of these two classes of chemicals. Over recent years, information has been published that needed to be systematized and appraised to bring understanding about their potential impacts on human health and aquatic biota. An important aspect, still enigmatic, is whether these transformation products are more harmful to non-target organisms than their parental compounds and which other factors may influence their potential toxicity. Another problem is the concern raised by transformation products not only as sole compounds per se but also in complex mixtures; mixtures of different metabolites of the same substance and mixtures of different substances, including parental compounds and transformation products. Applied Methodology The literature review carried out focused on the global occurrence and fate of the target contaminants in freshwater (i.e. surface-, ground- and influent/effluent wastewater), as well as on the available toxicological and ecotoxicological data. It covers articles published between 1997 and 2022, which have been searched in SCOPUS, Web of Science, PubMed and Google Scholar databases. The terms “pesticides” or “pharmaceuticals” were searched for in combination with “transformation products” or “degradation products”, “metabolites”, “freshwater”, “quantification”, “human health” or “aquatic species”. The search fields were the “article title”, “abstract” and “keywords”. Criteria for inclusion of articles in the review were related to the detail provided by the studies (i.e. quantification of the transformation products identified, suitable information about the species employed in the biotests, the age of the exposed organisms, relevant exposure design and endpoints assessed), as well as authors’ awareness and control of essential experimental conditions that may bias the results. All analytical methods of quantification have been included, rather than focusing on the most widespread techniques. Adding to this, most articles available in the literature are directed to parental compounds. Some of these works identify a few metabolites. Others do not include terms related to transformation products in the search fields and so they may not been detected. Some articles identify transformation products but do not quantify them, preventing prediction of their concentration in environmental samples (i.e. Mosekiemang et al. 2021; Madikizela et al. 2022). Articles about degradation experiments of pesticides and pharmaceuticals under controlled conditions have also been included, since such transformation processes can occur in natural conditions. Sources and Fate of Environmental Contamination Transformation Products of Pesticides Pesticides have been used since ancient times. Most of them were mainly inorganic compounds or substances of natural origin. However, the development and synthesis of organic pesticides after the second world war increased exponentially its use, making pesticides the second most used group of substances in the environment, only behind fertilizers (Davis 2014; Stokstad and Grullon 2013). Millions of tons of pesticides are applied each year, predominantly in agriculture, which is the main activity responsible for the leaking of pesticides and their sub-products into freshwater ecosystems (Fenner et al. 2013). Although applied in the soil, pesticides can reach aquatic ecosystems through diffuse/non-point-source or point-source pollution sites (Vryzas 2018; Fig. 1).Fig. 1 Sources of pollution, as well as formation and fate of transformation products deriving from pesticides Diffuse or non-point-source pollution is related with the movement of pesticides from large areas across the watersheds that reach the aquatic environment. Point-source pollution is related to a specific identifiable source which can include chemical run-off during storage, loading, disposal, as well as the misapplication of pesticides to water bodies (Syafrudin et al. 2021). Groundwater is heavily impacted by pesticides, and their metabolites or degradation products (i.e. N,N-dimethylsulfamide; aminomethylphosphonic acid; 2,6-Dichlorobenzamide), mainly resulting from point-source pollution (Postigo and Barcelo 2015). Leaching of landfill and septic tanks or industrial leakage are amongst the main sources of groundwater contamination by pesticides and metabolites (Postigo and Barcelo 2015). Pesticide characteristics (i.e. solubility and vapour pressure) are responsible for higher or lower rates of leaching to groundwater (Park et al. 2020). They also play an essential influence on the degradation of pesticides and the formation of transformation products, including active metabolites. Sorption–desorption, volatilization, chemical and biological degradation, uptake by plants, soil infiltration and leaching are some processes responsible for the appearance of new metabolites and their transportation into groundwater (Arias-Estévez et al. 2007). Groundwater tends, however, to be less affected by contamination than other water bodies, due to the natural attenuation capacity of aquifers and their large capacities (Postigo and Barcelo 2015). Nonetheless, recent monitoring studies show that pesticides, mainly herbicides, and their metabolites/degradation products (i.e. desethylatrazine; cyanazine amide) are present in aquifers (Lapworth and Gooddy 2006; Reemtsma et al. 2013). Surface waters are mainly contaminated by diffuse pollution sources. Run-off of pesticides and metabolites from agricultural fields after heavy rains are the main pathways of transportation of these substances to surface waters (Vryzas 2018). Moreover, during rainfall pesticides and degradation products imprisoned in soil or sediments can reach surface waters due to movements of those sediments (Vryzas 2018). Application of pesticides using sprays or even the plantation of seeds can be a source of surface water contamination, thanks to wind dispersal (Vryzas 2018). In these waters, photodegradation is the main process responsible for the degradation of pesticides. The formation of such metabolites can reach higher concentrations and show higher toxicity than the parental compounds (Reddy and Kim 2015). Wastewater treatment plants are also amongst the main sources of point-source pollution. Pesticides applied in urban areas (i.e. in maintenance of green areas or ponds) tend to finish in WWTPs, where traditional wastewater treatment methods are ineffective for the removal of these compounds (Rousis et al. 2017; Munze et al. 2017). Moreover, WWTPs effluents show in some cases higher concentrations of pesticides and their transformation products, as well as more toxicity, than the influents. Owing to all this, pesticides and their metabolites can ultimately reach drinking water, exposing humans, as indicated by their detection in the serum and blood of some patients in clinical and scientific studies (Chau et al. 2015; Tyagi et al. 2015). Transformation Products of Pharmaceuticals According to Daughton (2016), the first studies regarding the presence of pharmaceuticals in the environment date back to the 1940s. Later on, between the 60 s and 70 s, several works were produced about the possibility of contamination of drinking and surface water by pharmaceuticals, through the discharge of wastewater effluents (i.e. Stumm-Zollinger and Fair 1965; Hignite and Azarnoff 1977). Nowadays, pharmaceutical products are continuously released into the environment, although in small quantities (Fig. 2).Fig. 2 Sources of pollution as well as formation and fate of transformation products deriving from pharmaceuticals After consumption by humans, pharmaceuticals pass through the liver where they are directly effluxed from the organism (phase 0) or enter phase I and phase II of drug metabolism (Fig. 3). In phase I, more polar metabolites, often still active, are produced through oxidation, reduction or hydrolysis reactions. These reactions are commonly mediated by different CYP450 genes (i.e. CYP1A; CYP2B; CYP3A). Many of these transformation products become substrates of phase II, where endogenous hydrophilic groups are added through methylation, glucuronidation, acetylation, sulfation or conjugation with glutathione or amino acids such as glycine, taurine and glutamic acid to form water-soluble inactive compounds that can be excreted by the body in phase III (Fig. 3). Phase III excretion is mediated by ABC transporters and different solute carriers. Due to such reactions, pharmaceuticals can thus be excreted by humans in different forms: unchanged (small proportion) or as active or inactive metabolites (Jjemba 2006; Brown et al. 2015).Fig. 3 Schematic representation of pharmaceutical’s biotransformation and excretion Hospital effluents and direct elimination (i.e. through inadequate sanitary disposal) of unused pharmaceuticals in sewage are therefore amongst the most important sources of water contamination (Santos et al. 2010) by pharmaceutical transformation products. Influents are treated in WWTPs (Wastewater Treatment Plants) by three main processes of pollutant removal. In the first treatment, the removal of suspended solids occurs. This treatment has a low degree of efficiency in the removal of micropollutants, like pharmaceuticals (parental compounds or metabolites). In the second treatment, several types of reactions occur, such as dilution, partition, biotic and abiotic transformation (Luo et al. 2014). In this treatment, the level of efficiency is variable depending on the substance or metabolite in question, as well as their physicochemical properties (Luo et al. 2014). The third treatment is related to health questions to humans or specific uses of the treated water. It consists in further removal of substances, like nitrogen or phosphorus, and it is not mandatory, in general (Guardabassi et al. 2002; Luo et al. 2014). After this processing, in some cases, the total load of pharmaceutical compounds or metabolites in the effluent can be higher than that in the influent (Luo et al. 2014). This can be explained by the degradation of parental compounds into several metabolites or degradation products and the transformation of metabolites back into the parental compounds that can occur during the biological treatment in the WWTPs (Luo et al. 2014). Parental compounds and metabolites can also be imprisoned in faecal matter and released into the water during the biological treatment, thus increasing the overall concentration of those substances (Luo et al. 2014). This shows that the treatments available in WWTPs are still not fully efficient in the removal of these micropollutants. Hence, discharge of contaminated effluents introduces into natural waters the parental compounds and many more metabolites or transformation products (i.e. venlafaxine, tramadol, O-desmethyltramadol) (Santos et al. 2010; Luo et al. 2014). Additionally, some of these pharmaceutical metabolites are expected to be more toxic than parental compounds and consequently more dangerous to the wildlife (Celiz et al. 2009). The use of medicines is not exclusive to humans. These are also used in agriculture and aquaculture to treat diseased animals. As in humans, they are also excreted mostly as metabolites in the urine and faeces of animals or through adsorption in dirt pounds and after tanks cleaning, thus entering the environment without any kind of treatment and contaminating the soil and water (Santos et al. 2010). This contamination contributes to further input of transformation products into natural waters via run-off and leaching from the affected soils (Kemper 2008). Other anthropological activities also act as sources of contamination. Industry discharges (sometimes illegally), the use of WWTPs sludge contaminated with all kinds of pharmaceutical compounds as fertilizer, or leakage of septic tanks from households still not connected to the sewage systems, are examples of these (Carrara et al. 2008; Santos et al. 2010). Detection of Pesticide and Pharmaceutical Transformation Products in Water Compartments Analytical Methods of Pesticide and Pharmaceutical Transformation Products As previously mentioned, knowledge about contamination of the aquatic environment by pesticides and pharmaceutical transformation products has increased, mainly due to advances reached in analytical methods (Fent et al. 2006; Valls-Cantenys et al. 2016). New methods, with higher separation efficiencies and the ability to find more polar compounds, appeared. This allowed scientists to detect concentrations in environmental compartments in the order of ng/L and μg/L and consequently raised awareness and concern about their potential hazardousness (Fent et al. 2006; Santos et al. 2010; Valls-Cantenys et al. 2016). However, these advances in analytical methods are not efficient if the correct sample preparation is not performed. The extraction of the analytes from an environmental water sample is a crucial step before the instrumental analysis. Extraction techniques are based on the passage of an analyte by different solvents, which must be the most suitable for the type of analytical tool to be employed (Rutkowska et al. 2019; Campanale et al. 2021). This step can highly influence the analytical process, mainly for quantitative analysis, since the analyte volume must be increased, whilst any interferences must be eliminated (Campanale et al. 2021). Several sample preparation techniques are already described in the literature. For analysis of water samples, SPE (solid-phase extraction) is the most extensively used technique (Dimpe and Nomngongo 2016; Campanale et al. 2021). This method uses columns or disks able to retain the active compounds present in water samples and posteriorly release them by washing with small quantities of suitable solvents (Dimpe and Nomngongo 2016; Campanale et al. 2021). This provides an extract with few interferences, suitable for different analytical methodologies, such as High-Pressure Liquid Chromatography-Mass Spectrometry (HPLC–MS) and Gas Chromatography-Mass Spectrometry (GC–MS). More recently, a new SPE-based approach has been tested: Solid-Phase MicroExtraction (SPME). This newer method is faster and requires fewer quantity of solvents and is well described as suitable for Gas chromatography (GC) analysis (Campanale et al. 2021). Liquid–liquid extraction (LLE) is a simple method widely used for water samples that is also applied in the analysis of pesticide and pharmaceuticals (Dimpe and Nomngongo 2016; Campanale et al. 2021). This method has the advantage to be well established amongst different governmental agencies, but it also is time-consuming and requires the use of organic solvents that are harmful to the environment and even the handler (Dimpe and Nomngongo 2016; Campanale et al. 2021). Though the techniques described are the most well established for pharmaceuticals and pesticides, they have some disadvantages. One is the loss of more volatile analytes during the extraction process, which can affect the result of the analysis; another is the use of toxic solvents (Dimpe and Nomngongo 2016; Campanale et al. 2021). Much more methods are available in the literature, although less widespread. The development of new cost-effective and green methodologies for fast extraction is the next challenge in need to be addressed to improve and analytical determination. Regarding the analytical methods and instrumentation, GC and/or liquid chromatography (LC) coupled to mass spectrometry (MS) are, nowadays, the most applied methods to detect pesticides and pharmaceutical compounds. For LC, some variations to this method are well established in the literature, such as high-performance liquid chromatography (HPLC) and ultra-performance liquid chromatography (UPLC) (Gumustas et al. 2013). Gas chromatography is the most suitable method to separate non-polar parental compounds and their transformation products (Sparkman et al. 2011). This happens because of the inclusion of a derivatization process during GC that increases volatility and sensitivity, but also increases the duration of the procedure (Subramaniam et al. 2013). Coupling the GC with the MS has the advantage to offer a specific mass spectrum for a certain compound when an electron ionization (EI) is also performed (Foltz et al. 2016). Polar pharmaceutical or pesticide and their transformation products are mainly separated by LC (Martín-Pozo et al. 2019). Most of the pesticides and pharmaceuticals in their unchanged form or as transformation products are usually quantified at low concentrations in environmental water samples (ng to μg/L). This makes liquid chromatography with tandem mass spectroscopy (LC–MS/MS) a widely used method for determination of these compounds. This is due to its higher discrimination between the analyte and matrix signal, coupled to robustness and relative ease of use (Kaufmann et al. 2012). However, the selectivity and sensitivity of the MS vary with the selected ionization. Electrospray ionization (ESI) is the most chosen technique for detecting pharmaceuticals, since it is the most potent ionization method for the target compounds (Huang et al. 2019). Levels in Different Water Compartments Transformation Products of Pesticides As mentioned above, advances in the detection techniques led to an increase in knowledge about the occurrence of pharmaceutical and pesticide transformation products in different water compartments. Nevertheless, the quantification of transformation products is still not a focus in scientific investigation, as often studies only present new methods of detection and their validation, but not the concentrations found in real samples, even for parental compounds (Wode et al. 2015; Boix et al. 2016). Overall, the search carried out in the scientific databases returned 87 articles providing concentrations of pesticide and pharmaceutical transformation products in environmental water samples. Only one article quantified both pesticide and pharmaceutical transformation products (Huntscha et al. 2012). Of these, 29 articles were dedicated to pesticides, presenting concentrations obtained for 92 transformation products resulting from 43 parental compounds (Fig. 4, Table S1 in the Support Information). The most assessed pesticide transformation products (69%) belonged to three functional classes: organochlorine and chloroacetanilide herbicides/pesticides; triazine herbicides; and organophosphate and carbamate pesticides (Fig. 4, Table S1). Data about the quantification of pesticide transformation products found in different water samples, and respective detection methods, are presented in Fig. 5 and Table S1 (supplementary data). Transformation products of triazine herbicides were frequently reported in different studies, with a special focus on atrazine and terbuthylazine (Fig. 5, Table S1). Concentrations of these varied widely from 0.046 µg/L for desethylatrazine to 124,01 µg/L for hydroxy-terbuthylazine. The high concentration found for hydroxy-terbuthylazine resulted from an experiment where the parental compound was applied in a constructed wetland planted with Typha latifolia (Papadopoulos et al. 2007). According to the authors, the maximum concentration of the metabolite was within the highest concentration range found for the parental compound. This is of main concern, given that these concentrations are in the order of µg/L. Though knowledge about the environmental impact of this transformation product is sparse, recent works highlighted negative effects on the early developmental stages of fish species even at concentrations found in natural water samples (Velisek et al. 2014).Fig. 4 Relative frequency of transformation products quantified in environmental water samples per functional class of pesticides or pharmaceuticals. Concentrations for 92 pesticide transformation products derived from 43 parental compounds and 98 pharmaceutical transformation products derived from 64 parental compounds were found in the scientific literature published between 2000 and 2020 Fig. 5 Maximum concentrations of pesticide (A) and pharmaceutical (B) transformation products found in different water compartments. The heatmaps were done with the log-transformed values (pg/L) (Tables S1 and S2). Grey squares represent situations for which no information could be found Chloroacetanilide herbicides are widely used for grass control in several crops. Compounds of this class are structurally similar and were extensively used from the mid-1990s until recently (Elsayed et al. 2015). The main transformation products ethane sulfonic acid (ESA) and oxalinic acid (OXA), alongside the parental compound, are easily transported to water bodies and usually detected in both surface and groundwater (Table S1), contributing to the degradation of water quality (Baran and Gourcy 2013). Another interesting observation is the concentration level of metolachlor OXA and ESA in relation to the parental compound. Studies reported that metabolites of metolachlor, mainly ESA, were found in groundwater at higher concentrations than the parental compound (White et al. 2009; Baran and Gourcy 2013). This may occur because metabolites adsorb less to soil particles, compared to the parental compound and are thus more prone to infiltration to aquifer recharge (Baran and Gourcy 2013). This highlights the importance of monitoring programmes not only for pesticides alone but also for their transformation products. One of the first mass-produced pesticides in the world was DDT (Dichlorodiphenyltrichloroethane). It is an inexpensive and highly efficient short-term insecticide, but in the long term, it is problematic to human and animal health (Kezios et al. 2013). This pesticide was systematically banned in developed countries since the 1970s and a global ban of DDT, for non-vector control use, was exerted in the Stockholm Convention on Persistent Organic Pollutants, which took effect in 2004. However, this substance as well as several of its transformation products are still found in natural water bodies (Table S1) and tissues of different organisms (Veljanoska-Sarafiloska et al. 2013). A study conducted in African lakes showed that 4,4-DDE, a DDT metabolite, was biomagnified in fish species of the lake (Deribe et al. 2013). This was worrying, as those fish were consumed by local populations, possibly impacting human health. It is also of high concern the fact that metabolites of DDT, as well as the parental compound, are still commonly found in the environment even after an almost total ban worldwide, showing the great persistence of this substance and its transformation products in the ecosystems. Carbofuran is one of the carbamate pesticides most toxic to vertebrates, including humans, but knowledge about its main transformation products is still sparse. Otieno and colleagues (2010) reported the presence of very high concentrations of 3-ketocarbofuran and carbofuran-3-hydroxy (Table S1) in surface waters highly impacted by agrochemical procedures. Concentrations found (> 890 μg/L) were well above the standard water concentrations allowed by the USA and European authorities for safe drinking and human use (Otieno et al. 2010). Also, these two compounds appeared to be more persistent and were detected in higher concentrations than the parental compound (Otieno et al. 2010). Considering this, it would be crucial to gain a higher level of knowledge about the possible effects of these substances on non-target organisms, including humans, that could help infer about the need for more strict monitoring routines aiming at minimizing potential impacts on water quality and populations’ health. Transformation Products of Pharmaceuticals Fifty-eight articles were dedicated to pharmaceuticals, presenting concentrations obtained for 98 transformation products resulting from 64 parental compounds (Fig. 4, Table S2 in the Support Information). The most investigated transformation products belonged also to three functional classes: psychotropic drugs; analgesics, antipyretics and opioid painkillers and anticonvulsants (Fig. 4, Table S2). Carbamazepine transformation products, alongside metabolites of selective monoamine reuptake inhibitors and of ibuprofen, were the ones most reported in the literature (Fig. 5). The range of concentrations found varied from < 0.50 ng/L to 462000 ng/L, showing that very high concentration values of pharmaceutical metabolites are already found in the natural environment. Acetaminophen metabolites were the ones with higher reported concentrations. Sunkara and Wells (2010) reported concentrations higher than 400000 ng/L for acetaminophen glucuronide and sulphate in WWTP effluents. Those values were obtained in samples collected after application of conventional treatment processes in WWTP, pointing out the inefficiency of these treatments for the removal of micropollutants. Moreover, the authors refer that sometimes, metabolite concentrations were higher in the effluent than in the influent and one of the reasons for that was the bioconversion that may occur during the biological treatment, as mentioned previously. However, following UV treatment, none of the metabolites was found. This could be soothing, but the UV treatment is not always applied in WWTP; it is an optional treatment used mainly in water for human consumption (Luo et al. 2014; Guardabassi et al. 2002). Water without UV treatment loaded with transformation products can thus re-enter the water cycle, potentially risking aquatic fauna and flora. Also, it can be reused in agricultural practices and therefore contaminate crops, making metabolites enter the food chain with risk to human health. Carboxy ibuprofen was also reported at a very high concentration, higher than 100000 ng/L, in WWTP influents according to Paíga and colleagues (2016). Samples were collected in a relatively small WWTP designed to serve a little less than 50,000 people. Receiving wastewaters were mainly domestic and conventional treatments with activated sludge were applied (Paíga et al. 2016). Carboxy ibuprofen is one of the most representative ibuprofen metabolites. Ibuprofen is a commonly used non-steroidal anti-inflammatory (NSAID) drug and in 2016 it was the most used NSAID in Portugal, where the study was conducted (Monteiro et al. 2017). It is thus important to have a stricter monitoring routine for these substances to better evaluate the possible effects of metabolites on human and non-human health. Also, carbamazepine-10,11-epoxide was reported to occur at concentrations higher than 10000 ng/L in WWTP influents (Gros et al. 2012) and municipal wastewater (Petrovic et al. 2014). This is one of the main carbamazepine metabolites and one of the most detected in natural water samples (Table S2). An interesting fact in the study of Petrovic et al. (2014) is that carbamazepine-10,11-epoxide was found at a much higher concentration than the parental compound. This was also reported previously by Lopez-Serna et al. (2012) in a study conducted in the Ebro River in Spain. Those data reinforce the necessity of an extensive assessment and monitoring routine for metabolites, once they can be more prevalent in water compartments, compared to their parental compounds. Risks of Pharmaceutical and Pesticide Transformation Products Human Health Although the available data are sparse, freshwater contamination does not affect only organisms living in those systems. Ultimately, humans can also suffer negative effects from exposure to transformation products. Humans are exposed to pesticide and pharmaceutical transformation products in different ways. Data presented in Tables S1 and S2 show levels of those transformation products detected in drinking water and groundwater as well, which is a common source of drinking water in cities around the world (Guimarães et al. 2019). As previously mentioned, exposure can occur via contaminated recreational water and/or consumption of contaminated freshwater organisms or other food produced with water originating from contaminated sites. Knowledge about human health risks caused by transformation products of pesticides and pharmaceuticals is still sparse, compared to parental compounds. Studies available in the scientific literature are presented in Table 1.Table 1 Toxicological studies about the human health risks of pesticide and pharmaceutical transformation products Transformation product [Parental compound] Concentrations Sample Exposure duration Endpoints Effects Reference Pesticides Chloroacetanilide, aniline; hydroxychloroacetanilide; and diethylquinoneimine [Alachlor] (Hill et al. 1997) 0; 0.03; 0.1; and 0.3 μM Lymphocyte cells 72 h Oncogenicity Induction of chromatid exchange at 0.1 μM for hydroxychloroacetanilide, 0.3 μM for chloroacetanilide and aniline Mitotane [DDT] ( Daffara et al. 2008) Distinct values for each sample Blood cells and saliva not applicable Hormonal levels and organ toxicity Inhibition of cortisol and DHEAS. Induction of thyroid function perturbations. Inhibition of testosterone secretion 2–4-dichlorophenol [2–4-D] (Bukowska, 2003) 10 to 500 ppm Blood cells 1 h Antioxidant enzymes Increase of superoxide dismutase and increase of gluthathione peroxidase activities DDE [DDT] (Perez-Maldonado et al. 2006) Distinct values for each sample Blood cells not applicable Genotoxicity Induction of peripheral blood mononuclear cells p-p' DDE [DDT] (Geric et al. 2012) 4.1 μg/ml Lymphocyte cells 1; 6 and 24 h Genotoxicity Induction of DNA damage p-p' DDE [DDT] (Geric et al. 2012) 3.9 μg/ml Lymphocyte cells 1; 6 and 24 h Genotoxicity Induction of DNA damage AMPA [glyphosate] (Benachour and Séralini, 2009) 18 concentrations from 10 ppm to 10% Embryonic kidney HUVEC primary neonate umbilical cord vein, embryonic kidney. and JEG3 placental cell lines 24 h Cytotoxicity Increased cellular mortality. Destruction of the membrane of all cell types AMPA [glyphosate] (Kwiatkowska et al. 2014)) 0.01–5 mM Erythrocytes 1, 4 and 24 h Haemolysis, haemoglobin oxidation, ROS formation and morphology Induction of haemolysis (0.05 to 5 mM) and haemoglobin oxidation (0.25 to 5 mM) at 24-h incubation. Increase in ROS production at concentrations starting from 0.25 Mm Methylsulphonic acid [glyphosate] (Kwiatkowska et al. 2014)) 0.01–5 mM Erythrocytes 1, 4 and 24 h Haemolysis, haemoglobin oxidation, ROS formation and morphology Induction of haemolysis (0.1 to 5 mM) and haemoglobin oxidation (0.5 to 5 mM) at 24-h incubation. Increase in ROS production at 0.5 and 5 mM Pharmaceuticals Gemfibrozil 1-O-β- glucoronide [gemfibrozil] (Ogilvie et al. 2006) 0.25 to 64 μM Liver microsomes 2 to 40 min CYP2C8 activity Potent inhibitor of CYP2C8 2-hidroxyestrone and 16-α hydroxyestrone [estrogens] (Eliassen et al. 2008) not applicable Blood cells not applicable Genotoxicity and mitogenicity Levels of 2-hydroxyestrone, and the ratio between 2-hydroxyestrone and 16-α hydroxyestrone were linked with certain types of breast cancer tumours in woman Morphine-3-glucoronide [morphine] (Dozio et al. 2022) 1, 10 and 100 μM Astrocytes 12, 24, 48 and 96 h Proteomics 96-h exposure lead to dysregulation of biological pathways linked with extracellular matrix organization, antigen presentation, cell adhesion and glutamate homeostasis Morphine-6-glucoronide [morphine] (Dozio et al. 2022) 1, 10 and 100 μM Astrocytes 12, 24, 48 and 96 h Proteomics Acute exposure increased the levels of proteins involved in cell adhesion and decreased the levels of extracellular matrix The adverse effects that pesticides can cause on human health are a long-known problem. This discussion gained bigger attention and impact since the publication of the book Silent Spring in 1962. In this publication, Rachel Carson described not only the environmental impacts coinciding with the widespread use of DDT in agriculture in the USA, but also the potential of DDT to cause cancer in exposed workers. In the book, other pesticides were also surveyed, such as 2,4-D (2,4-Dichlorophenoxyacetic acid), chlordane and heptachlor. More recently different environmental agencies, including EPA (United States Environmental Protection Agency) and ECHA (European Chemicals Agency) or international conventions are banning the use of some pesticides that were described as hazardous to human health. Amongst the pesticide metabolites that can elicit problems, mitotane was proven to be a selective toxicant to humans and is used as an adjuvant drug to treat adrenocortical tumours (Wajchenberg et al. 2000). Mitotane or o,p’-dichlorodiphenyldichloroethane (o-p’-DDD) is a DDT metabolite and apparently the only chemical able to inhibit corticoid synthesis and at the same time destroy cortical cells (Wajchenberg et al. 2000). However, despite the therapeutic use, mitotane was already reported in the literature to have side effects at hormonal levels in patients who were treated with this compound (Daffara et al. 2008). The authors analysed the blood cells and the saliva of the patients and found that mitotane treatment was linked to the inhibition of cortisol and DHEAS (Dehydroepiandrosterone sulphate). Also, perturbations of the thyroid function were described. Moreover, for males, an inhibition of testosterone secretion was also found. However, these side effects were usually reversible with the adequate treatment. Another DDT metabolite, DDE (dichlorodiphenyldichloroethylene) was reported to induce apoptosis of human peripheral blood mononuclear cells, both in vitro and in vivo (Perez-Maldonado et al. 2006). The authors studied blood collected from 61 healthy children during the year 2004 and from 57 children from southern Mexico. Exposure to both DDT, DDD and DDE was found in the tested children. However, significant correlations between apoptosis and exposure to pesticides were only found for DDE blood levels, (p = 0.010 and 0.040 for 2003 and 2004, respectively). This causes great concern since DDE is the most persistent DDT metabolite and thus exposure tends to be chronic, and apoptosis of the cells could result in an impairment of the immune system (Perez-Maldonado et al. 2006). Both p,p′-DDE chloroethane and p,p′-DDD (dichlorodiphenyldichloroethane) were reported to induce DNA damage in human lymphocytes, even at low concentrations (Geric et al. 2012). In this study, in vitro human lymphocytes were exposed for 1, 6 and 24 h to p,p’-DDE (4.1 μg/mL) or p,p’-DDD (3.9 μg/mL) and genotoxic effects were assessed using the cytokinesis-block micronucleus assay and the comet assay. Results showed an increase in the number of cells containing micronucleus, in relation to the control, in the 24-h exposures. Also, according to the comet assay, the percentage of DNA damages increased, in relation to the control. It is important to notice that the concentrations used are in the range found in human fluids, suggesting that these effects are already occurring in humans exposed to the metabolites (Geric et al. 2012). The metabolite 2,4-dichlorophenol, from the herbicide 2,4-D, was reported to cause effects on antioxidant enzymes and glutathione levels in human erythrocytes in vitro (Bukowska, 2003): the activity of superoxide dismutase decreased whilst that of glutathione peroxidase increased in a dose-dependent (10–500 ppm) manner. Moreover, exposure to 250-ppm 2,4-dichlorophenol also decreased the level of reduced glutathione in erythrocytes by 32%, in relation to the control. These effects are similar, though more pronounced, to those resulting from exposure to the parental compound 2,4-D, pointing to a major need for monitoring pesticide metabolites in natural samples. Dialkylquinoneimine metabolites of chloroacetanilide herbicides like alachlor and acetochlor were reported to induce in vitro sister chromatid exchanges in human lymphocytes (Hill et al. 1997). This study was performed to test the hypothesis that the oncogenicity of chloroacetanilide herbicides previously described was caused by genotoxic intermediates, like diethylbenzoquinoneimine, an alachlor metabolite. The investigation was done with cultured human peripheral lymphocytes, mostly T cells. At 0.3-µM high variability was observed, with effects elicited by N-dealkyl-alachlor, aniline metabolites and their 4-hydroxy derivatives and diethylbenzoquinone, in only half of the cases. At 0.1–0.3 µM the ratio between treated and control cells for sister chromatid exchange was always higher in exposures to diethylbenzoquinoneimine than to dimethyl- and ethylmethylbenzoquinoneimines. The study showed that all the compounds assessed were toxic to lymphocytes and provided the first evidence that metabolites of chloroacetanilide herbicides were genotoxic to humans and could significantly affect the immune system (Hill et al. 1997). Glyphosate metabolites were also reported to have cyto- and hematotoxicity in humans. Aminomethylphosphonic acid (AMPA) is the main metabolite of glyphosate. This transformation product is recognized to have similar levels of toxicity comparing to its parental compound, and human exposure was already described (Benachour and Séralini, 2009; Kwiatkowska et al. 2014). The embryonic kidney, HUVEC primary neonate umbilical cord vein and JEG3 placental cell lines were exposed to 18 different AMPA concentrations varying from 10 ppm to 10% for 24 h (Benachour and Séralini, 2009). The authors reported that AMPA exposure induced succinate dehydrogenase and adenylate kinase effects on human cells and thus mortality. AMPA exposure resulted in the destruction of the cell membrane, in all cell types. More recently, another study was performed to determine AMPA hematotoxicity in human erythrocytes (Kwiatkowska et al. 2014). The authors exposed human erythrocytes to 0.01–5 mM AMPA, during 1, 4 or 24 h and evaluated the exposure effects in haemolysis, haemoglobin oxidation, ROS formation and the erythrocytes morphology. Results showed that AMPA induced haemolysis at concentrations equal or higher than 0.05 mM and haemoglobin oxidation (≥ 0.25 mM) after 24 h of incubation. An increase in ROS production was also registered at concentrations starting from 0.25 mM. The same study also investigated the hematotoxic effects of other glyphosate metabolite: methylphosphonic acid. The results were similar to those obtained for AMPA, although at a different concentration range. Induction of haemolysis and haemoglobin oxidation occurred at concentrations ≥ 0.1 and 0.5 mM, respectively. In addition, ROS production was found at concentrations ≥ 0.5 mM (Kwiatkowska et al. 2014). Pharmaceutical metabolites are not usually expected to represent an exposure concern to humans. However, biotransformation and detoxification reactions can lead to the formation of active pharmaceutical metabolites potentially more toxic than the respective parental compounds (Celiz et al. 2009). For example, gemfibrozil 1-O-β-glucuronide, the major gemfibrozil metabolite, was found to be a more potent inhibitor of CYP2C8 than the parental compound in human liver microsomes (Ogilvie et al. 2006). Also, Ogilvie and colleagues found that gemfibrozil glucuronide, contrarily to the parental compound gemfibrozil, was found to be a CYP2C8 selective inhibitor acting in a metabolism-dependent way. To depict such differences, the authors evaluated both the parental compound and its main metabolites as inhibitors of the main drug metabolizing CYP450 enzymes (CYP1A2, 2B6, 2C8, 2C9, 2C19, 2D6 and 3A4) in human liver microsomes. Compounds inhibiting the activity of the CYP450 complex can affect the metabolism of other drugs and lead to accumulation and potential toxic effects, exerting an undesired effect in the exposed person (Ogilvie et al. 2006). In fact, the chemical reactivity of glucuronide metabolites has been linked with toxic properties. These metabolites can reach appreciable concentrations in human tissues and blood. They can also undergo hydrolysis and pH-dependent intramolecular acyl migration, irreversibly reacting with human tissues. This can cause chemical alterations leading to drug toxicity expressed by alterations in functional properties of the modified molecules or hypersensitivity and other immunotoxic reactions (Shipkova et al. 2003). Pharmaceutical endocrine disruptors have been linked to several adverse effects on human health (Safe 2000). A wide range of parental compounds have been associated with hazardous effects on human reproduction and cancer development, amongst others, and metabolites are not excluded. Estrogen metabolites are reported as possible mitogenic and genotoxic substances. Investigating blood samples collected between 1989 and 1990 in subjects taking oestrogens and in controls not taking them, Eliassen et al. (2008) found a significant positive association in women of the plasma levels of 2-hydroxyestrone, and the ratio between 2-hydroxyestrone and 16-α hydroxyestrone, with certain types of breast cancer tumours. The authors recognized, nevertheless, the need for replicating the study and increasing research about the relationship between estrogen metabolites and estrogen and progesterone receptors related to breast tumours. Morphine is a strong painkiller, which is widely prescribed worldwide. However, this opiate was described to be potentially toxic to humans, not only the parental compound but also its metabolites’ morphine-3-glucoronide and morphine-9-glucoronide (Dozio et al. 2022). In a recent study, Dozio and colleagues performed a deep proteomic study in human astrocytes to investigate the role of central nervous system glial cells in the mechanisms originating the side effects of morphine administration in humans. For that, they exposed astrocytes during 12, 24, 48 and 96 h to 1-, 10- and 100-μM morphine, morphine-3-glucoronide and morphine-6-glucoronide. The proteomic analysis showed the 96-h exposure to morphine-3-glucoronide lead to dysregulation of biological pathways linked with extracellular matrix organization, antigen presentation, cell adhesion and glutamate homeostasis. For morphine-6-glucoronide (12-24-h exposure), increased levels of proteins involved in cell adhesion and decreased levels of extracellular matrix were observed. Aquatic Biota Transformation Products of Pesticides Knowledge about toxic effects caused by pesticide transformation products is still sparse, compared to parental compounds. Studies available in the scientific literature are presented in Table 2.Table 2 Ecotoxicological studies about the effects of pesticide transformation products on aquatic species Transformation product [parental compound] Species Concentrations Exposure duration Endpoints Effects Reference o-p' DDT [DDT] (Donohoe and Curtis, 1996) Oncorhynchus mykiss 0, 0.1, 1, 5, 10 and 30 mg/kg 42 days Determination of vitellogenin levels Increased levels of vitellogenin in plasma and interaction with hepatic estrogenic binding sites in vivo o-p' DDE [DDT] (Donohoe and Curtis, 1996) Oncorhynchus mykiss 0, 0.1, 1, 5, 10 and 30 mg/kg 42 days Determination of vitellogenin levels Increased levels of vitellogenin in plasma and interaction with hepatic estrogenic binding sites in vivo p-p' DDE [DDT] (Donohoe and Curtis, 1996) Oncorhynchus mykiss 0, 0.1, 1, 5, 10 and 30 mg/kg 42 days Determination of vitellogenin levels No differences found in vitellogenin levels, relative to controls DDD [DDT] (Lotufo et al. 2000) Hyalella azteca 0.095, 0.178, 0.366, 0.692 and 1.381 µg/L 10 days Mortality and lethal residues in tissues DDD was less lethal than the parental compound (DDT) but its lethality was higher than that of the control at > 0.69 µg/L Diporeia spp. 0.944, 2.791, 7.420 and 17.056 µg/L 28 days No significant effects found DDE [DDT] (Lotufo et al. 2000) Hyalella azteca 1.117, 2.258, 4.947, 8.208 and 22.021 µg/L 10 days Mortality and lethal residues in tissues DDE was less lethal than the parental compound (DDT) but its lethality was higher than that of the control at > 2.258 µg/L Diporeia spp. 2.293, 4.726, 9.141 and 20.194 µg/L 28 days No significant effects found o-p' DDE [DDT] (Davis et al. 2009) Oreochromis mossambicus 5 µg/g 35 days Determination of Vitellogenin levels and hormone/insulin-like growth factor i-axis Increase in plasma levels of insulin growth factor 100 µg/g 5 days Increase in expression of both vitellogenin A and B, estrogen receptors α and β and also in insulin growth factor 3–4-dichloroaniline [diuron] (Scheil et al. 2009) Danio rerio 0.005, 0.01, 0.1 0.25, 0.5 and 1 mg/L 8 and 11 days Mortality and locomotor activity Locomotor activity and mortality were impaired at ≥ 0.5 mg/l 0.05, 0.1, 0.15, 0.2 and 0.2 5 mg/L 168 h Hsp70 levels A significant increase in relation to control was found at 0.25 mg/L 0.5, 0.7, 1, 1.5 and 2 mg/L 11 days Embryonic and larval development By-product caused larvae deformations at ≥ 0.25 mg/l 3–4-dichloroaniline [diuron] (Felicio et al. 2018) Oreochromis niloticus 40 and 200 ng/L 7 days Antioxidant and biotransformation biomarkers By-product caused significant alterations in antioxidant and biotransformation biomarkers, with ethoxyresorufin-O-deethylase (EROD) activity showing a dose-dependent response Deethylatrazine [atrazine] (Ralston-Hooper et al. 2009) Hyalella azteca 550, 1000, 2500, 5000, 10,000, 15000 µg/L 96 h, 21 and 42 days Mortality and sex ratio LC50 values were 5100 µg/L at 96 h and higher than 3000 µg/L at 21 days; no change in the sex ratio was found Diporeia spp. 0.03, 0.3, 3, 30, 300, 3000 µg/L 96 h, 21 and 42 days Mortality and sex ratio LC50 values were 7200 µg/L at 96 h and higher than 3000 µg/L at 21 days; no change in the sex ratio was found Pseudokirchneriella subcapitata No reported 96 h Growth inhibition Growth inhibition occurred at concentrations > 2000 µg/L Deisopropylatrazine [atrazine] (Ralston-Hooper et al. 2009) Hyalella azteca 550, 1000, 2500, 5000, 10,000, 15000 µg/L 96 h, 21 and 42 days Mortality and sex ratio LC50 values were > 3000 µg/L at 96 h and 330 µg/L at 21 days; no change in the sex ratio was found Diporeia spp. 0.03, 0.3, 3, 30, 300, 3000 µg/L 96 h, 21 and 42 days Mortality and sex ratio LC50 values were > 3000 µg/L at 96 h and 300ug/L at 21 days; no change in the sex ratio was found Pseudokirchneriella subcapitata No reported 96 h Growth inhibition Growth inhibition occurred for concentrations higher than 3000 µg/L Therbuthylazine-2-hydroxy [therbuthylazine] (Koutnik et al. 2017) Procambarus fallax f. virginalis 0.75, 75, 375 and 750 µg/L 62 days Mortality, growth, oxidative balance, antioxidant defences, ontogeny and histology Lower weight at 75 µg/L; delayed ontogenic development and lowered antioxidant defences in exposed animals Desethyl-terbuthylazine [therbuthylazine] (Velisek et al. 2016) Cyprinus carpio 1.80, 180, 900 and 1800 µg/L 7, 14, 20, 27 and 31 days Growth, LC50, histology, oxidative stress, mortality LC50 of 441.6 μg/L at 31 days; lower weight and length in fish exposed to 1800 μg/L for 7 days and 900 μg/L for 20 days; delayed ontogenetic development at > 1.8 µg/L; decreased antioxidant enzyme activity in all concentrations AMPA [glyphosate] (Guilherme et al. 2014) Anguilla anguilla 11.8 and 23. µg/L 1 and 3 days DNA and chromosome damage Significant genotoxic effect in relation to control group Fiprunil sulphide and sulfone [fipronil] (Weston and Lydy, 2014) 14 macroinvertebrate species 4 − 7 concentration steps separated by a factor of 2 48 and 96 h Mortality and ability to swim, cling or crawl, depending on the species Mean 96-h EC50 of 7 − 10 ng/L Fiprunil sulphide [fipronil] (Gong et al. 2021) Danio rerio 0.1 to 10 mg/L 72 h Mortality and oxidative stress LC50 = 0.36 mg/L. Significant decreased of SOD activity at 5 mg/L Chlorella pyrenoidosa Algae growth inhibition rate; content of pigment EC50: 0.10 mg/L; chlorophyll content significantly decreased in dose–response relationship; Fiprunil sulfone [fipronil] (Gong et al. 2021) Danio rerio 0.1 to 10 mg/L 72 h Mortality and oxidative stress LC50 = 0.21 mg/L. Significant decreased of SOD activity at 5 mg/L Chlorella pyrenoidosa Algae growth inhibition rate; content of pigment EC50: 0.13 mg/L; chlorophyll content significantly decreased in dose–response relationship; Fiprunil desulfinyl [fipronil] (Gong et al. 2021) Danio rerio 0.1 to 10 mg/L 72 h Mortality and oxidative stress LC50 = 1.13 mg/L. Significant decreased of SOD activity at 5 mg/L Chlorella pyrenoidosa Algae growth inhibition rate; content of pigment EC50: 0.43 mg/L; chlorophyll content significantly decreased in dose–- response relationship; Metolachlor OXA [metolachlor] (Velisek et al. 2018) Procambarus fallax f. virginalis 4.2, 42 and 420 µg/L 45 days Growth rate, behaviour, oxidative stress, histology and mortality Decreased growth and activity of antioxidant enzymes in all tested concentrations; delayed ontogenetic development and lower levels of reduced glutathione and lipid peroxidation Metolachlor OXA [metolachlor] (Rozmánková et al. 2020) Danio rerio 1, 30, 100 and 300 μg/L (single exposure); 1 and 30 μg/L (mixture) 120 h Mortality, hatching success, embryonic malformations, locomotion, spontaneous movements, heartbeat and gene expression Increased craniofacial, non-inflated gas bladder and yolk sac malformations at 100 μg/L or higher. Induction of p53 gene at 100 μg/L Metolachlor ESA [metolachlor] (Rozmánková et al. 2020) Danio rerio 1, 30, 100 and 300 μg/L (single exposure); 1 and 30 μg/L (mixture) 120 h Mortality, hatching success, embryonic malformations, locomotion, spontaneous movements, heartbeat and gene expression Increased craniofacial, non-inflated gas bladder and yolk sac malformations at 100 μg/L or higher. Induction of p53 gene at 100 μg/L. Induction of p53 and thyroid system regulation (dio2, thra, thrb) at 30 and 1 μg/L, respectively 3-trifluoromethyl-4-aminophenol [3-trifluoromethyl-4-nitrophenol] Huerta et al. 2020) Petromyzon marinus 0, 5, 50 and 200 µM Undefined exposure time Respiratory control ratio, mitochondrial oxygen consumption and mitochondrial transmembrane potential No significant effects found 4-nitro-3-methyl-phenol [3-trifluoromethyl-4-nitrophenol] (Huerta et al. 2020) Petromyzon marinus 0, 5, 50 and 200 µM Undefined exposure time Respiratory control ratio, mitochondrial oxygen consumption and mitochondrial transmembrane potential No significant effects found 4-amino-3-methylphenol [3-trifluoromethyl-4-nitrophenol] (Huerta et al. 2020) Petromyzon marinus 0, 5, 50 and 200 µM Undefined exposure time Respiratory control ratio, mitochondrial oxygen consumption and mitochondrial transmembrane potential Decreased respiratory control ratio at 50 µM; decreased oxygen consumption at 200 µM 4-nitroso-3-methyl-phenol [3-trifluoromethyl-4-nitrophenol] (Huerta et al. 2020) Petromyzon marinus 0, 5, 50 and 200 µM Undefined exposure time Respiratory control ratio, mitochondrial oxygen consumption and mitochondrial transmembrane potential No significant effects found 3-phenoxybenzyl alcohol [permethrin] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Moderately toxic (LC50 = 1.93 mg/L) Benzenesulfonamide [asulam] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Non-toxic (LC50 > 100 mg/L) benzimidazol [carbendazim] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Slightly toxic (LC50 = 66.19 mg/L) cyanoacetamide [DBNPA] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Slightly toxic (LC50 = 68 mg/L) cis-2,6-dimethylmorpholine [fenpropimorph] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Non-toxic (LC50 > 100 mg/L) ethiprole sulfone [ethiprole] (Gao et al. 2021) Danio rerio 100, 300, 800, 2000, 5000 μg/L 4 days Mortality; oxidative stress; development LC50 value was 1750 μg/L; induction of antioxidant enzymes and the developmental anomalies at 100 μg/L ethiprole sulphide [ethiprole] (Gao et al. 2021) Danio rerio 100, 110, 120, 150, 180 μg/L 4 days Mortality; oxidative stress; development LC50 value was 111 μg/L; induction of antioxidant enzymes and the developmental anomalies at 10 μg/L or higher rac-ethiprole amide [ethiprole] (Gao et al. 2021) Danio rerio 100, 500, 2500, 10,000, 50,000 μg/L 4 days Mortality; oxidative stress; development LC50 > 50,000 μg/L ethiprole sulfone amide [ethiprole] (Gao et al. 2021) Danio rerio 100, 500, 2500, 10,000, 50000 μg/L 4 days Mortality; oxidative stress; development LC50 > 50,000 μg/L desethylsulfinyl ethiprole [ethiprole] (Gao et al. 2021) Danio rerio 500, 800, 1500, 2500, 5000 μg/L 4 days Mortality; oxidative stress; development LC50 = 1728 μg/L One of the most controversial pesticides is DDT, which was reported to cause health issues to humans and living organisms in general. Moreover, studies are available in the literature linking exposure to DDT metabolites to negative effects on the health of aquatic organisms. Donohoe and Curtis (1996) injected juvenile rainbow trout with o,p’-DDT, o,p’-DDE or p,p’-DDE with doses ranging from 5 to 30 mg/kg at 0, 14 and 28 days and sampling was done at 14 and/or 42 days. They reported that o,p’-DDT and o,p’-DDE had estrogenic activity, because of the elevated plasma vitellogenin levels they can elicit in vivo and their interaction with hepatic estrogenic binding sites (Donohoe and Curtis, 1996). A study conducted in freshwater amphipods (Hyalella azteca and Diporeia spp.) reported that the metabolites DDD and DDE are less lethal than DDT (Lotufo et al. 2000). Hyalella azteca and Diporeia spp. were exposed to a wide range of concentrations of DDD for 10 days and DDT and DDE for 28 days. Besides mortality, median lethal residue (LR50), mean effect concentration (EC50) and mean effect residue (ER50) in tissues were also assessed. Although metabolites were less lethal, mortality of H. azteca was significantly higher in DDD and DDE treatments than in the control at 0.692 µg/L and 2.258 µg/L, respectively (Lotufo et al. 2000). This raises high concern, once concentrations of DDD in this range have already been reported in freshwater ecosystems. The endocrine-disrupting activity of o,p'-DDE was also evaluated more recently (Davis et al. 2009). In this study, the authors investigated the effects of this metabolite and other compounds on the expression of the vitellogenin gene from the tilapia Oreochromis mossambicus and the growth hormone insulin-like growth factor-I axis. Injection of 100 µg/g o,p'-DDE in fish increased the expression of vitellogenin A and B, as well as the transcription of estrogen receptors α and β and the expression of the putative somatolactin receptor and insulin-like growth factor (Davis et al. 2009). This once again reinforces the potential endocrine disruption that DDT metabolites may cause in freshwater fish. As previously mentioned, metabolites of triazine herbicides are amongst the most frequently found in freshwater systems. Moreover, there is evidence in the literature linking these substances to negative effects on living organisms. The main degradation product of diuron is 3,4-dichloroaniline for which the toxic potential towards freshwater organisms is described in the literature. In zebrafish, a sub-chronic exposure (11 days) to this metabolite caused deformations at ≥ 0.25 mg/l, whilst locomotor activity and mortality were impaired at ≥ 0.5 mg/l (Scheil et al. 2009). A recent work investigated the effects of 3,4-dichloroaniline on biotransformation enzymes and the oxidative stress response in the liver and gills of the Nile tilapia (Oreochromis niloticus) (Felicio et al. 2018). The authors found that in fish exposed for seven days to 40 and 200 ng/L the levels of several biotransformation and antioxidant enzymes were altered often in a non-monotonic response, except for ethoxyresorufin-O-deethylase (EROD) activity that exhibited a dose-dependent increase. Moreover, the multixenobiotic resistance (MXR) activity and the activity of glutathione S-transferase (GST) enzymes were decreased in gills after exposure to 3–4-dichloroaniline. Because the MXR mechanism is crucial for the protection of aquatic organisms against xenobiotics aggression (Ferreira et al. 2014), this suggests that exposure to this metabolite is endangering the health of fish and the contaminated aquatic systems. A reduction in this mechanism can lead to higher susceptibility of animals to xenobiotics by impairing homeostatic processes. The acute and chronic toxicity of deethylatrazine and deisopropylatrazine, metabolites of atrazine, were investigated in two amphipod species and in the microalgae Pseudokirchneriella subcapitata (Ralston-Hooper et al. 2009). Hyalella azteca and Diporeia spp. were exposed to concentrations ranging from 0.55 to 15 mg/L for 96 h and from 0.03 to 3000 µg/L for 21 days. Results showed the median lethal concentrations (LC50) and median growth inhibition concentration (IC50) for algae were ≥ 1.5 mg/L, i.e. higher than the levels found in the environment (Ralston-Hooper et al. 2009). In a recent study, marbled crayfish (Procambarus fallax f. virginalis) were exposed for 62 days to four concentrations of terbuthylazine-2-hydroxy: 0.75 μg/L (environmentally relevant), 75, 375 and 750 μg/L (Koutnik et al. 2017). Antioxidant defences, oxidative balance, histology, early ontogeny, growth and mortality were the parameters assessed to depict possible effects of this metabolite. Concentrations over 75 μg/L caused lower weight compared to the control group. The outcome of the study showed that terbuthylazine-2-hydroxy delayed ontogenetic development. Also, levels of thiobarbituric acid and antioxidant enzymes were significantly (p < 0.01) lower in groups exposed to the metabolite. This shows the potential danger of this metabolite to freshwater species, although the alterations found occurred in the groups exposed to non-environmental concentrations (Koutnik et al. 2017). The toxicity of terbuthylazine-desethyl, another metabolite of triazine herbicides, was assessed in the early stages of development of the common carp (Cyprinus carpio) (Velisek et al. 2016). Carp embryos were exposed to 1.80 μg/L (environmentally relevant), 180 μg/L, 900 μg/L and 1800 μg/L and samples were collected on days 7, 14, 20, 27 and 31. The 31d LC50 of terbuthylazine-desethyl was estimated to be 441.6 μg/L. Animals also exhibited lower weight and length at 7 (1800 μg/L) and 20 (900 μg/L) days of exposure. Terbuthylazine-desethyl at non-environmental concentrations also delayed the ontogenetic development, in relation to control. However, antioxidant enzyme activity was significantly lower in all test concentrations, including the environmentally relevant one, indicating that contamination by this metabolite should be compromising feral aquatic populations. The main metabolite of glyphosate, AMPA is one of the most controversial pesticides nowadays, due to its potential hazard to wildlife and human populations. Moreover, AMPA by itself was reported as hazardous to Anguilla anguilla by Guilherme et al. (2014). The eels were exposed for 1 and 3 days to environmentally relevant concentrations (11.8 and 23.6 μg/L) and genotoxicity was investigated by assessing damage to DNA through the Comet assay and erythrocytic nuclear abnormalities. These results showed a genotoxic effect of AMPA at concentrations already found in aquatic systems. About organophosphates, a recent study was conducted with the parasitic sea lamprey (Petromyzon marinus) to address possible effects on cardiac mitochondrial bioenergetics of the lampricide 3-trifluoromethyl-4-nitrophenol and its metabolite 3-trifluoromethyl-4-aminophenol, as well as 4-nitro-3-methyl-phenol (Huerta et al. 2020). The latter has a similar molecular structure and is a known transformation product of fenitrothion and its metabolites 4-amino-3-methylphenol and 4-nitroso-3-methyl-phenol. Mitochondria were extracted from the hearts of animals captured on the great lakes and incubated with 0, 5 and 50 µM of the test compounds to assess the respiratory control ratio and mitochondrial oxygen consumption or with 0, 5, 50 and 200 µM to assess the mitochondrial transmembrane potential. Results showed that 4-amino-3-methylphenol significantly lowered the respiratory control ratio (88% at 50 µM) and oxygen consumption by 64% (at 200 µM and with the addition of high concentrations of ADP) and by 45% (at 200 µM and addition of substrate for complex II). At last, for mitochondrial transmembrane potential, none of the tested transformation products caused significant alterations. Fipronil is a phenylpyrazole insecticide with crescent use in urban areas. The toxicity of its sulphide and sulfone metabolites was not recognized until 2014 when Weston and Ludy carried out a study determining EC50 values for 14 macroinvertebrate species. Results indicated a mean 96 h EC50 of 7−10 ng/L for fipronil metabolites in Chironomus dilutus (Weston and Lydy 2014). The same study also reported that creeks receiving urban stormwater run-off in California contained metabolite concentrations twice the EC50 found for C. dilutus and approximately one-third of the EC50 found for other aquatic macroinvertebrates (Weston and Lydy 2014). A recent study evaluated the toxicity of different fipronil metabolites: fipronil sulphide, fipronil sulphone and fipronil desulfinyl (Gong et al. 2021). In this work, the authors analysed the effects of 72-h exposure to these metabolites at concentrations ranging from 0.1 to 10 mg/L on zebrafish embryos and the green algae Chlorella pyrenoidosa. In zebrafish, LC50 values of 0.36, 0.31 and 1.13 mg/L were found for fipronil sulphide, sulfone and desulfinyl, respectively. Moreover, at 5 mg/L all metabolites significantly increased SOD activity, in relation to control. In C. pyrenoidosa growth inhibition, EC50 values of 0.10, 0.13 and 0.43 mg/L were found for fipronil sulphide, sulfone and desulfinyl, respectively. The metabolites investigated also caused a significant decrease in chlorophyll content, in relation to control, in a dose–response manner (Gong et al. 2021). Metabolites of chloroacetanilide herbicides are highly prevalent in aquatic ecosystems, mainly in oxalinic and endosulfonic acid forms. Metolachlor OXA was reported to negatively affect the early life stages of marbled crayfish (Velisek et al. 2018). Animals were exposed for 45 days to 4.2 μg/L (environmentally relevant), 42 μg/L and 420 μg/L and several endpoints were assessed. Metolachlor OXA caused significantly lower growth and decreased activity of antioxidant enzymes at all tested concentrations. The highest tested concentrations delayed ontogenetic development and decreased the levels of reduced glutathione and lipid peroxidation (Velisek et al. 2018). More recently, a study was performed to evaluate the impacts of single and combined exposure of metolachlor and its metabolites metolachlor ESA and metolachlor OXA on zebrafish embryos (Rozmánková et al. 2020). In this study, zebrafish embryos were exposed for 120 h to 1, 30, 100 and 300 μg/L of the single compounds or to 1 and 30 μg/L of a compound mixture and sublethal endpoints such as malformations, hatching rate, larval length, spontaneous movements, heartbeat and locomotion, as well as expression levels of eight genes linked to different critical pathways, were monitored. Increased craniofacial, non-inflated gas bladder and yolk sac malformations at 100 μg/L or higher were reported for both metabolites. For metolachlor OXA, a significant induction of p53 gene was found at 100 μg/L, compared to control, whilst for metolachlor ESA, a significant induction of p53 gene at 30 and 100 μg/L and thyroid system regulation (dio2, thra, thrb) was observed at 1 μg/L, in comparison to the control group. The disruption of the thyroid system represented a plausible danger for population maintenance, since it occurred at low environmental concentrations (Rozmánková et al. 2020). A recent study evaluated the acute toxicity of several biocide metabolites using the rainbow trout (Oncorhynchus mykiss) as a test model (Hernández-Moreno et al. 2022). The author exposed juvenile trout according to OECD TG203, for 96 h to 0.78, 3.15, 12.5, 50 and 100 mg/L of the following metabolites: 3-phenoxybenzyl alcohol, benzenesulfonamide, benzimidazole, cyanoacetamide and cis-2,6-dimethylmorpholine. The most toxic metabolite was 3-phenoxybenzyl alcohol, with an LC50 value of 1.93 mg/L, considered moderately toxic by the authors. Benzimidazole and cyanoacetamide with LC50 values of 66.19 and 68 mg/L, respectively, were reported as slightly toxic, whilst benzenesulfonamide and cis-2,6-dimethylmorpholine with LC50 values higher than 100 mg/L were considered non-toxic (Hernández-Moreno et al. 2022). Ethiprole is a non-systemic phenylpyrazole compound widely used as an insecticide. Recently, a study was performed to evaluate zebrafish embryotoxicity and effects on antioxidant enzymes (catalase, CAT and superoxide dismutase, SOD, activities) and oxidative stress (lipid peroxidation) of its main metabolites, i.e. ethiprole sulfone, ethiprole sulphide, ethiprole amide, ethiprole sulfone amide and desethylsulfinyl ethiprole (Gao et al. 2021). Results showed that only ethiprole sulfone and sulphide had effects on antioxidant defences and embryonic development. Ethiprole sulfone had an LC50 value of 1750 μg/L, induced antioxidant enzymes and increased developmental anomalies at 100 μg/L. Ethiprole sulphide had an LC50 value of 111 μg/L, induced antioxidant enzymes and increased developmental anomalies at 10 μg/L or higher. Rac-ethiprole amide and ethiprole sulfone amide had LC50 values higher than 5000 μg/L, whilst the LC50 value for desethylsulfinyl ethiprole was 1728 μg/L (Gao et al. 2021). Transformation Products of Pharmaceuticals Nowadays, one main challenge to the scientific community is to understand the effects of these substances on non-target organisms. There are, already, several reports about this topic. However, knowledge about the toxic effects caused by pharmaceutical transformation products is still scarce. A summary of the works found in the literature is shown in Table 3.Table 3 Ecotoxicological studies about the effects of pharmaceutical transformation products on freshwater species Transformation product [parental compound] Species Concentrations Exposure duration Endpoints Effects Reference Prednisone, dexamethasone and their undisclosed photodegradation products [prednisone, dexamethasone] (Della Greca et al. 2004) Brachionus calyciflorus 5 different test concentrations without known value. Results are reported as median effective concentrations in ppm 24 h Mortality 5-prednisone and 2-dexamethasone photoderivates had lower LC50 values than parent compounds but at levels not found in environmental samples (mg/L range) Thamnocephalus platyurus 24 h Mortality All photoderivates had lower LC50 values than parental compounds (higher toxicity), but at non environmentally relevant concentrations (> 710 ppm) Daphnia magna 24 h Mortality All photoderivates had lower EC50 values than parental compounds (higher toxicity), but at non environmentally relevant concentrations (mg/L range) Pseudokirchneriella subcapitata 72 h Growth inhibition Toxic effects similar to those found for the other species, except Ceriodaphnia dubia Ceriodaphnia dubia 7 days Population growth Both the photoderivatives of prednisolone and dexamethasone showed higher toxic effects on C. dubia growth after 7 days naproxen and its undisclosed photodegradation products [naproxen](Isidori et al. 2005) Brachionus calyciflorus Concentration values are not given. All test solutions were dissolved in DMSO (0.01% v/v). 5 different concentration were tested, as well as, a negative control 24 /48 h Mortality and reproduction All photoderivates had lower LC50 values than parental compounds, but at levels not found in environmental samples (mg/L range) for acute assay. In the chronic reproduction assay only one photoderivate was less toxic than the parental compound Thamnocephalus platyurus 24 h Mortality All photoderivates had lower LC50 values than parental compounds, but at levels not found in environmental samples (mg/L range) Ceriodaphnia dubia 24 h and 7 days Mortality and reproduction All photoderivates had lower LC50 values than parental compounds, but at levels not found in environmental samples (mg/L range). For reproduction, only one photoderivate was less toxic than the parental drug Pseudokirchneriella subcapitata 96 h Growth All photoderivatives of naproxen showed higher toxic effects on P. subcapitata growth diclofenac, ketoprofen, atenolol and their photodegradation products (undisclosed) [diclofenac, ketoprofen, atenolol] (Diniz et al. 2015) Danio rerio 1 mg/L 7 days Oxidative stress Diclofenac metabolites formed through UV photolysis treatments were more toxic than their parental compounds. Activity of antioxidant enzymes and lipid peroxidation levels were higher for by-products than the parental drugs. Overall, oxidative stress response causing toxicity was observed for all pharmaceuticals and by-products norfluoxetine [fluoxetine] (Stanley et al. 2007) Pimephales promelas 1 to 250 µg/L 7 days survival and growh The authors related higher toxicity in fish exposed to s-fluoxetine, which in mammals is expected to be more potent than R-norfluoxetine Daphnia magna 10 to 1000 µg/L 21 days immobilization, reproduction and grazing rate No observed effects Norfluoxetine [fluoxetine] (Fong and Molnar, 2008) Dreissena polymorpha 100 nM to 50 µM 4 h spawning Increased spawning in zebra mussels at 1–50 µM Mytilopsis leucophaeata 100 nM to 50 µM 4 h spawning Increased spawning in zebra mussels at 1–50 µM Sphaerium striatinum 100 nM to 10 µM 4 h parturition Significant increase in parturition induced at 10 µM norfluoxetine [fluoxetine] (Rodrigues et al. 2020) Danio rerio 0.64, 3.2, 16, 80 and 400 ng/L 80 h Embryonic development, gene expression and sensorimotor responses Increase of embryonic anomalies in relation to control, mainly for pigmentation. No effects found for gene expression and sensomotory response Norfluoxetine [fluoxetine] (Atzei et al. 2021) Danio rerio 0.03 to 10 µM 5 days Embryonic development, gene expression and light/dark movement Inhibition of light/dark, zebrafish locomotory activity, mainly in dark. Responses followed a dose–response relationship norfluoxetine [fluoxetine] (Rodrigues et al. 2022) Danio rerio 400 ng/L 80 h Embryonic development and gene expression Increase in pigmentation anomalies of embryos and larvae, relative to the parental compound n-desmethylsertraline [sertraline] (Lajeunesse et al. 2011) Salvelinus fontinalis WWTP water samples (undisclosed concentrations) 3 months Tissue bioaccumulation and Na/K-ATPase activity Bioaccumulation in several tissues, including (brain and liver). Na/K-ATPase activity negatively correlated with brain bioaccumulation desmethylsertraline-exposed brain tissue o-desmethylvenlafaxine [venlafaxine] (Stropnicky, 2017) Orconectes obscurus 0, 1 and 8 µg/L 14 days Aggressive behaviour Increase in the number of attacks per minute at the highest concentration tested Procambarus clarkii 0, 1 and 8 µg/L 14 days Aggressive behaviour Increase in the number of attacks per minute at the highest concentration tested o-desmethylvenlafaxine [venlafaxine] (Atzei et al. 2021) Danio rerio 0.03 to 300 µM 5 days Embryonic development, gene expression and light/dark movement Inhibition of light/dark, zebrafish locomotory activity, mainly in dark. Responses followed a dose–response relationship Clofibric acid [clofibrate] (Nunes et al. 2008) Gambusia holbrooki 176.4, 211.6, 253.92, 304.71 and 365.65 mg/L 96 h Oxidative damage Decrease in the amount of oxidized glutathione content in the liver and gills in exposed fish n- and o-desmethyltramadol [tramadol] (Zhuo et al. 2012) Danio rerio Intraperitoneal injection of tramadol (65 mg/kg) 1 h Weight, mitochondrial changes and behaviour Detection of n- (mostly) and o-desmethyltramadol in brain tissue. Fish exposed to tramadol exhibited weight loss, abnormal behaviour and mitochondrial structural changes, possibly mediated by its by-products Oxazepam [temazepam] (Huerta et al. 2016) Pimephales promelas 0.8, 4.7 and 30.6 µg/L 28 days Behaviour and bioaccumulation Brain was the tissue with higher accumulation rates; behavioural effects detected in the novel tank diving test were observed in fish exposed to 4.7 μg/L Oxazepam [temazepam] (Fahlman et al. 2021) Perca fluviatilis 15 μg/L 14 days anti-predator behaviour Stimulation of anti-predator behaviour (decreased activity, decreased distance to conspecifics and increased littoral habitat use) Oxcarbamazepine [carbamazepine] (Desbiolles et al. 2020) Lemna minor 27 ng/L 17 days Phytometabolites Increase in nitrogen compounds. Chlorophyll index was higher in relation to control Hydra circumcincta 900 ng/L 14 days Reproduction, morphological changes and oxidative stress biomarkers Single exposure impacted the total antioxidant capacity Acridine 9-carboxylic acid [oxcarbazepine] (Desbiolles et al. 2020) Lemna minor 27 ng/L 17 days Phytometabolites Alterations of the nitrogen balance and chlorophyll indices at environmental concentrations Oseltamivir carboxylate [osetalmivir] (Chen et al. 2020) Oryzias latipes 0, 0.06, 0.3, 90 and 300 µg/L 14, 21 and 56 days median survival, growth, reproduction and hatchability Long-term parental exposure to by-products affected the embryonic development of fish hatchability at 300 µg/L and development 90 µg/L Oseltamivir ethyl ester [osetalmivir] (Chen et al. 2020) Oryzias latipes 0, 0.06, 0.3, 90 and 300 µg/L 14, 21 and 56 days median survival, growth, reproduction and hatchability Long-term parental exposure to by-products affected the embryonic development of fish hatchability at 300 µg/L and development 90 µg/L Fenofibric acid [fenofibrate] (Jung et al. 2021) Danio rerio 5, 10, 20, 30 and 40 mg/L 72 h Mortality LC50 = 53.32 mg/L Carbamazepine-10,11-epoxide [carbamazepine] (Bars et al. 2021) Danio rerio 250 µg/L 120 h embryonic development Delay in swim bladder inflation at 120hpf 5-(4-hydroxyphenyl)-5-phenylhydantoin [phenytoin] (Bars et al. 2021) Danio rerio 250 µg/L 120 h embryonic development No effects found As mentioned above, metabolites can be formed during wastewater treatment in WWTPs. In fact, this situation is reported for photodegradation products of both prednisone and dexamethasone (DellaGreca et al. 2004). In this study, photoproducts of both pharmaceuticals were isolated, from an initial solution of 100 mL of both compounds mixed with 500 mL of water and their toxicity to different species was evaluated: the rotifer Brachionus calyciflorus and the crustaceans Thamnocephalus platyurus and Daphnia magna for acute toxicity and the microalgae Pseudokirchneriella subcapitata and the crustacean Ceriodaphnia dubia for chronic toxicity. Acute assays lasted for 24 h and were based on mortality (LC50). In chronic assays, growth inhibition was the endpoint assessed for algae (72-h duration) and population growth was the endpoint for C. dubia (7-day duration). Some photodegradation products of prednisone and dexamethasone were found to be more toxic than the parental compounds. However, the LC50 values obtained by the authors were considerably higher than the concentrations generally found in surface waters. The chronic exposures decreased the population growth in C. dubia (DellaGreca et al. 2004). A similar study was conducted for the non-steroidal anti-inflammatory drug naproxen and its photodegradation products (Isidori et al. 2005). In this work, acute toxicity tests were conducted with B. calyciflorus, T. platyurus and C. dubia. Chronic toxicity was assessed (reproduction and/or growth) in B. calyciflorus, C. dubia and the microalgae P. subcapitata. Results showed that photodegradation products were more acutely toxic than the parental compound, although at levels (mg/L range) well above those found in freshwater systems. Chronic exposure reduced the population growth in C. dubia at low concentrations (μg/L) for some photoproducts (Isidori et al. 2005). This situation warns of the need to improve treatment methodologies, for better removal of both the parental compounds and their transformation products. A more recent study also reported that diclofenac metabolites formed through UV photolysis treatments were more toxic than their parental compound (Diniz et al. 2015) (Table 2). Lienert and colleagues (2007) developed a study where the ecotoxicological risk of 42 pharmaceuticals and their metabolites was evaluated. In the study, both parental compounds and their respective metabolites were treated as a mixture of toxicants of similar action. When relevant data were not available in the literature, the authors estimated them from quantitative structure–activity relationships (QSAR). Moreover, from their known pharmaceutical information, they figured out the removal efficiency of these contaminants from urine. The results of this evaluation showed that mixtures of ibuprofen and its metabolites could represent an ecotoxicological risk for aquatic organisms. Likewise, acetylsalicylic acid, bezafibrate, carbamazepine, diclofenac, fenofibrate and paracetamol in a mixture with their respective metabolites could be of potential risk for aquatic organisms, however, to a lesser extent than ibuprofen. In Table S2, ibuprofen metabolites detected in environmental samples reach concentrations > 120 000 ng/l that, together with the results of Lienert et al. (2007), suggests that this contamination is jeopardizing affected aquatic ecosystems and their populations. Whilst QSAR models have some limitations that may generate not fully accurate data, the information presented by those authors established a relevant basis for highly needed subsequent research and risk assessment studies. Norfluoxetine, the main fluoxetine metabolite, was reported to cause enantiospecific sublethal effects in Pimephales promelas and Daphnia magna (Stanley et al. 2007). In this study, P. promelas juveniles were exposed for seven days to 1, 10, 50, 100 and 250 µg/L of R-, rac- and S-fluoxetine. The enantiomer S-fluoxetine showed higher toxicity to growth, survival and feeding rate. The authors related their results to the fact that S-norfluoxetine is more potent to mammals than R-fluoxetine. But this pattern was not found for D. magna. For this microcrustacean, a 21-day toxicity test was performed to determine immobilization, reproduction and grazing rate. Less than 24-hpf individuals were exposed to 10, 50, 100, 250, 500 and 1000 µg/L of R-, rac- and S-fluoxetine. The results obtained were similar for the three compounds, and the taxa differences were attributed to the higher homology between fish and mammals than between crustaceans and mammals. Norfluoxetine was also reported to induce spawning and parturition in bivalves (Fong and Molnar 2008). The authors exposed zebra mussels to 100 nM–50 µM, dark false mussels to 100 nM–50 µM and finger-nail clams to 100 nM–10 µM. Norfluoxetine increased spawning in both zebra mussels and dark false mussels, relative to the respective controls, at concentrations in the range of 1–50 µM. In finger-nail clams, norfluoxetine induced significant parturition only at 10 µM, relative to controls. Recently, Rodrigues and colleagues (2022) found that norfluoxetine could affect the embryonic development of zebrafish larvae. In the study, newly hatched embryos were exposed for 80hpf to norfluoxetine (0.0014 µM) and fluoxetine (0.0015 µM). Larvae exposed to norfluoxetine showed an increased frequency of pigmentation anomalies, in relation to the parental compound (Rodrigues et al. 2022). Still concerning the SSRI (selective serotonin reuptake inhibitors) type of depressants, the primary metabolite of sertraline, n-desmethylsertraline, was found to affect Na/K-ATPase activity in the trout brain (Lajeunesse et al. 2011). The authors studied the distribution of selected SSRI in several tissues of brook trout, as well as the Na/K-dependent ATPase pump activity in the brain. Fish were exposed for 3 months to a WWTP-treated effluent (primary treatment) before and after ozonation. The metabolite n-desmethylsertraline was one of the main substances found in various tissues. Also, Na/K-ATPase activity was negatively correlated with the accumulation of n-desmethylsertraline in the brain. Within the group of serotonin and norepinephrine reuptake inhibitors (SNRI), o-desmethylvenlafaxine (the active metabolite of venlafaxine) was implicated in behavioural changes of freshwater organisms (Stropnicky, 2017). The author exposed two species of crayfish, Orconectes obscurus and Procambarus clarkii to 0, 1 or 8 μg/L of o-desmethylvenlafaxine. The aggression behaviour of the crayfish, measured by the number of attacks per minute of exposed animals, was the endpoint assessed. An increase in the number of attacks was found for both species at 8 µg/L (Stropnicky, 2017). A more recent study related o-desmethylvenlafaxine exposure to behavioural changes in freshwater species (Atzei et al. 2021). The authors exposed zebrafish embryos to this metabolite in a concentration range of 0.03–300 µM, for 5 days. Embryonic development was monitored and a light/dark behavioural assay was performed. No significant developmental anomalies were elicited by o-desmethylvenlafaxine. However, a dose–response inhibition on locomotory function, mainly under dark conditions, was found (Atzei et al. 2021). Clofibric acid, a metabolite of clofibrate, is another metabolite with reported negative effects on fish species. This compound caused modifications of biomarkers related to antioxidant defences and oxidative stress in Gambusia holbrooki (Nunes et al. 2008). In their work, the authors exposed the fish for 96 h to 176.34, 211.60, 253.92, 304.71 and 365.65 mg/L of clofibric acid. This metabolite caused a decrease in the activity of several antioxidant enzymes and in particular the levels of oxidized glutathione, in both the liver and gills. The effects of chronic tramadol exposure were studied in the zebrafish brain (Zhuo et al. 2012). Following intramuscular injections (25 or 65 mg/kg), both n- and o-desmethyltramadol were detected in brain tissue, mainly n-desmethyltramadol. This is important, since fish chronically exposed to tramadol exhibited weight loss, abnormal behaviour and mitochondrial structural changes. Considering that the two metabolites were present in the brain tissue, it may be possible that both can exert their effects on the exposed animals. Nevertheless, further studies focused on their administration and specific effects are needed to support this. Oxazepam is one of the main metabolites of diazepam, a widely used benzodiazepine that is prescribed as an anticonvulsant, amongst other functions. In a recent study, specimens of Pimephales promelas were exposed to 0.8, 4.7 and 30.6 µg/L oxazepam for 28 days and the relationship between its internal concentrations and effects on fish behaviour was investigated with two types of tests: novel tank diving test and shelter-seeking test (Huerta et al. 2016b). The authors concluded the brain was the tissue with higher accumulation rates and significant behavioural effects in the novel tank diving test were observed in fish exposed to 4.7 μg/L. Although 4.7 μg/L is a concentration higher than found in freshwater bodies, it raises concern about the effects this metabolite can exert on fish behaviour and ultimately endanger populations impacted by this substance. Another study with the same compound revealed behavioural changes on Perca fluvialis (Fahlman et al. 2021). The results showed that anti-predation behaviour was stimulated in exposed animals, characterized by decreased activity and distance to conspecifics, as well as increased littoral habitat use (Fahlman et al. 2021). Carbamazepine is one of the most used anticonvulsants worldwide. Recently, some of its transformation products were a matter of study by Desbiolles et al. (2020). Their study focused on the chronic effects of oxcarbamazepine and acridine 9-carboxylic acid, in single or combined exposure with carbamazepine, in two different models: the duckweed Lemna minor and the cnidarian Hydra circumcinta. Tested concentrations were the same for both models; 600, 27 and 900 ng/L for carbamazepine, oxcarbamazepine and acridine 9-carboxylic acid, respectively. For L. minor, exposure lasted 17 days and different phytometabolites were monitored. Exposure to the transformation products separately and in a mixture with the parental compound caused alterations of nitrogen balance, namely an increase in nitrogen compounds. The chlorophyll index was also higher in oxcarbamazepine groups than in the control. Nevertheless, the phenols index varied deeply without any specific trend or alteration relative to the control group. Hydra circumcinta individuals were exposed to the compounds for 14 days and different endpoints were assessed, such as reproduction, morphological changes and evaluation of antioxidant and oxidative stress biomarkers. The results showed that oxcarbamazepine exposure had implications in the total antioxidant capacity of H. circumcincta increasing two-fold in relation to control. Exposure to acridine 9-carboxylic acid affected all tested endpoints, except the reproduction. Combined exposure assays resulted in an increase in malformations on cnidarians and a decrease in the budding rate (Desbiolles et al. 2020). Another carbamazepine metabolite (carbamazepine-10,11-epoxide) was recently addressed for its possible effects on zebrafish embryonic development (Bars et al. 2021). The authors exposed zebrafish embryos from ~ 3 to 120hpf to a concentration of 250 µg/L of this metabolite, i.e. considerably higher than the maximum concentration found in the environment. Embryonic development was monitored through the exposure period and anomalies were registered. Results showed that swim bladder inflation was significantly delayed in carbamazepine-10,11-epoxide-exposed larvae, compared to the control (Bars et al. 2021). This is important since inflation of the swim bladder allows larvae to stay in the water column and have more chances of survival. A recent study focused on the metabolites of the well-known antiviral oseltamivir (Tamiflu) and their chronic effects on the medaka Oryzias latipes (Chen et al. 2020). Results showed that long-term parental exposure to both oseltamivir carboxylate and oseltamivir ethyl ester affected embryonic development and fish hatchability at 300 µg/L and embryonic development at 90 µg/L. Fenofibric acid, a metabolite of the anti-lipidemic agent fenofibrate, was also evaluated for its toxicity to zebrafish embryos (Jung et al. 2021). An LC50 value of 53.32 mg/L was found at 72 h, which is considerably higher than the normally occurring concentration in the environment. The Way Forward This review gives an updated perspective on freshwater contamination by pharmaceuticals and pesticide transformation products and the available information about the toxicity of these substances. Detection of pharmaceuticals and pesticides is increasing in freshwater ecosystems, and concentrations in the range of ng to μg/L have been widely reported. Moreover, this same trend is described for their metabolites and transformation products. This occurrence made this field one of the most studied by the scientific community in the last years, with a number of published works addressing the potentially hazardous effects of such previously overlooked substances. The present research identified concentrations of 190 metabolites and transformation products (92 from pesticides and 98 from pharmaceuticals) in water bodies and wastewater effluents, none of them included in monitoring programmes set to achieve the good environmental status of freshwater ecosystems. Their formation processes, environmental fate in aquatic ecosystems and effects on humans and biota, summarized in Fig. 6, are varied and a considerable cause of concern. Reported concentrations are mainly in the order of ng to μg/L. The concentration heatmap produced in this work allows us to easily spot the substances found at higher levels.Fig. 6 Overall representation of pesticides and pharmaceutical transformation products aquatic contamination and risks for human and aquatic species Although the information presented herein about the quantification of pesticides and pharmaceutical transformation products is extensive (almost 200 compounds), this may just represent the tip of the iceberg. Worldwide there are more than 1500 pesticides approved for use in agriculture and about 4000 pharmaceutical compounds approved for human consumption (aus der Beek et al. 2016; Anagnostopoulou et al. 2022). These parental compounds can have one or several transformation products, which brutally increases the potential number of these pollutants in the aquatic environment. Also, transformation products of pesticides banned for several decades now are still found in freshwater. Transformation products are in several cases more stable in the environment and consequently reach concentrations higher than their parental compounds (Schuhmann et al. 2019; Celiz et al. 2009). All these numbers and characteristics reinforce the need to increase the monitoring of these compounds in aquatic systems and evaluate their impact on human and environmental health. The toxicological information available for the transformation products identified is very little and scattered, with no strategic approach underlying data collection for risk assessment and monitoring prioritization. Concerning the risk to humans, less than twenty metabolites (of the two groups combined) were investigated in in vitro studies. Several of these were found to elicit genotoxicity and effects on biotransformation and antioxidant processes. In aquatic organisms, only about 34% of the transformation products originating from pesticides and 14% of those originating from pharmaceuticals were evaluated for their potentially hazardous effects on biota. Most of these studies evaluated effects on only one (majority) or two trophic levels and more than half of them on vertebrates. Effects on plants and algae were rarely assessed. For pesticides, over 50% of the assessments were about acute and subacute toxicity effects, whilst for pharmaceuticals only about 20% of the assessments concerned chronic toxicity. Adding to this, for pharmaceutical metabolites various studies tested very high exposure levels, reporting effects at concentrations higher than those found in the environment. Nevertheless, for pesticide metabolites, several reports described a considerably wide range of negative effects on freshwater organisms, occurring at environmentally relevant concentrations. For pharmaceutical metabolites, different classes of drugs were proven to cause hazardous effects and jeopardize the homeostasis of freshwater species. All in all, the data presented herein clearly demonstrate that pesticide and pharmaceutical transformation products pose a threat to aquatic fauna and flora. Concerning the relative toxicity of transformation products, compared to the parental compounds, the available data prevent a clear global conclusion. In some cases, the transformation products are in fact less toxic. In other cases, some transformation products can be more active and toxic than the parental substance. Nowadays, there is increasing evidence that pesticide transformation products can be more toxic and persistent than their parental compounds (Iwafune 2018). In silico assays, performed with the ECOSAR (Ecological Structure Activity Relationships) software, which predicts the toxicity of different compounds, showed that the transformation products of several pesticides have a high toxicity potential to aquatic fauna and flora (Anagnostopoulou et al. 2022). Transformation products resulting from penoxsulam, pyrimethanil, imidacloprid, acetamiprid, thiacloprid and carbendazim were predicted to be more toxic than their parental compounds. In contrast, transformation products of fipronil present equal levels of toxicity, relative to fipronil itself (Anagnostopoulou et al. 2022). For pharmaceutical transformation products, there is a general idea that these compounds are less active and, consequently, less toxic than their parental compounds. However, there is evidence that some transformation products may be more toxic than the parental compounds. In humans, metabolites such as morphine and o-desmethyltramadol are more active than the parental compound (codeine and tramadol, respectively) (Rodieux et al. 2018). There are also reports of potential toxic effects elicited in patients, i.e. pethidine and dextroptopoxyphene (Coller et al. 2009). On the other hand, photodegradation products of prednisone, dexamethasone, naproxen, diclofenac, ketoprofen and atenolol formed in watercourses or even in WWTPs were reported to be toxic to different aquatic species at higher magnitude than their parental compounds (DellaGreca et al. 2004; Isidori et al. 2005; Diniz et al. 2015). Nonetheless, for most of the transformation products identified, the information is still scarce to draw sound conclusions. Something that is still not accounted for in most of the ecotoxicological works is the metabolism of parental substances in the test media. During exposure, parental compounds are metabolized and transformed by the exposed organisms. This is a process, influenced by media abiotic factors, which originates different transformation products. Such compounds can cause negative effects on the organisms, by themselves or in mixture with the respective parental compound. A previous study reported that fish exposed to tramadol exhibited weight loss, abnormal behaviour and structural mitochondrial changes that could be linked to the metabolites formed during the exposure, which accumulated in the animals’ brains and muscular tissue (Zhuo et al. 2012). The possibility that several negative impacts reported on aquatic species exposed to pharmaceuticals may derive not only from those compounds, but also from the mixture with their metabolites or even exclusively from the metabolites needs to be addressed soon. Overall, the results warn of the need to continue improving treatment methodologies, for better removal of transformation products, not only to avoid their discharge to the aquatic environment but also to assure a better quality for water reuse. From a toxicological viewpoint, it is also striking the lack of mechanistic information useful to improve predictive toxicology and the risk assessment of these chemicals. Most works focused on assessing classical apical endpoints employing standard testing approaches. Whilst this is always fruitful to obtain a quick grasp of the severity of a contamination scenario, more studies investigating the modes of action of these compounds are urgently needed. Also, the limited availability of reference standards for several transformation products makes it difficult to test the toxicity of these compounds to living organisms (Anagnostopoulou et al. 2022). However, this obstacle can be surpassed using in silico approaches, which reduce the need for animals and chemicals and can be valuable tools for toxicity and risk assessment. Future toxicological investigations should be based on the framework of Adverse Outcome Pathways (AOP) (Ankley et al. 2010). This concept identifies various key events and relationships between them, linking a molecular initiating event to an adverse outcome of significance to risk assessment. The adverse outcome is usually considered at the organ level or higher, preferably the ecological level. It indicates a morphological or physiological alteration occurring in an organism or its systems that elicits functional impairment or impairs its ability to compensate for chemical stress and achieve homeostasis. The AOP framework is recognized as useful to support regulatory decision-making and the prioritization of chemicals for risk assessment (Vinken et al. 2017; Perkins et al. 2019), a most important aspect for the contamination scenario described herein. Present-day high-throughput technologies (i.e. proteomic sequencing) allowing for the rapid and cost-effective generation of data should be used to identify key events and key event relationships through which the initiating event(s) will reflect on adverse outcomes to apical endpoints. Guidance documents for the development of AOPs were made available (OECD, 2013, 2018), as well as supporting databases and tools, such as the e.AOP.portal (http://aopkb.org), the AOP Wiki (http://aopwiki.org), the Effectopedia (http://effectopedia.org) and the Wikipathways (https://www.wikipathways.org/index.php/WikiPathways), the Harmonized Template 201: Intermediate effects (https://www.oecd.org/ehs/templates/harmonised-templates-intermediate-effects.htm) and the AOP Xplorer (http://datasciburgoon.github.io/aopxplorer. Collaborative networks based on resource and knowledge sharing, and rational effort application, should be made at a global level to establish and implement a structured strategy rapidly allowing to fulfil these gaps whilst avoiding unnecessary experimental redundancy (Martens et al. 2018). The present work emphasizes the need to reinforce the existing knowledge about contamination by pharmaceutical and pesticide transformation products in freshwater systems. This report compiled and analysed a significant amount of information linking exposure to transformation products to adverse outcomes in aquatic species and humans. Technological needs and knowledge gaps were identified and discussed, delineating future research steps on the topic, ultimately aiming at improving water management and monitoring programmes. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 117 KB) Acknowledgements This work was supported by project BioReset (DivRestore/0004/2020) funded by BiodivRestore 2020 ERA-NET Cofund (a joint action of Biodiversa+ and Water JPI) and Strategic Funding UIDB/04423/2020 and UIDP/04423/2020 (FCT, ERDF) and PR was supported by a FCT fellowship SFRH/BD/134518/2017. Data Availability Authors confirm that all relevant data are included in the article or its supplementary file. Declarations 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. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Anagnostopoulou K Nannou C Evgenidou E Lambropoulou D Overarching issues on relevant pesticide transformation products in the aquatic environment: A review Sci Total Environ 2022 10.1016/j.scitotenv.2021.152863 Ankley GT Bennett RS Erickson RJ Hoff DJ Hornung MW Johnson RD Mount DR Nichols JW Russom CL Schmieder PK Serrrano JA Tietge JE Villeneuve DL Adverse outcome pathways: a conceptual framework to support ecotoxicology research and risk assessment Environ Toxicol Chem 2010 29 730 741 10.1002/etc.34 20821501 Arias-Estévez M López-Periago E Martínez-Carballo E Simal-Gándara J Mejuto J-C García-Río L The mobility and degradation of pesticides in soils and the pollution of groundwater resources Agr Ecosyst Environ 2007 123 247 260 10.1016/j.agee.2007.07.011 Atzei A Jense I Zwart EP Legradi J Venhuis BJ van der Ven LTM Heusinkveld HJ Hessel EVS Developmental neurotoxicity of environmentally relevant pharmaceuticals and mixtures thereof in a zebrafish embryo behavioural test Int J Environ Res Public Health 2021 18 13 6717 10.3390/ijerph18136717 34206423 aus der Beek, T., Weber, F.A., Bergmann, A., Hickmann, S., Ebert, I., Hein, A., and Küster, A. Pharmaceuticals in the environment–Global occurrences and perspectives Environ Toxicol Chem 2016 35 823 835 10.1002/etc.3339 26666847 Baena-Nogueras RM Gonzalez-Mazo E Lara-Martin PA Degradation kinetics of pharmaceuticals and personal care products in surface waters: photolysis vs biodegradation Sci Total Environ 2017 10.1016/j.scitotenv.2017.03.015 Baran N Gourcy L Sorption and mineralization of S-metolachlor and its ionic metabolites in soils and vadose zone solids: consequences on groundwater quality in an alluvial aquifer (Ain Plain, France) J Contam Hydrol 2013 154 20 28 10.1016/j.jconhyd.2013.07.009 24055953 Bars C Hoyberghs J Valenzuela A Buyssens L Ayuso M Van Ginneken C Labro AJ Foubert K Van Cruchten SJ Developmental toxicity and biotransformation of two anti-epileptics in zebrafish embryos and early larvae Int J Mol Sci 2021 22 23 12696 10.3390/ijms222312696 34884510 Bavumiragira JP Ge J Yin H Fate and transport of pharmaceuticals in water systems: A processes review Sci Total Environ 2022 10.1016/j.scitotenv.2022.153635 Benachour N Séralini GE Glyphosate formulations induce apoptosis and necrosis in human umbilical, embryonic, and placental cells Chem Res Toxicol 2009 22 1 97 105 10.1021/tx800218n 19105591 Boix C Ibanez M Sancho JV Parsons JR Voogt P Hernandez F Biotransformation of pharmaceuticals in surface water and during waste water treatment: Identification and occurrence of transformation products J Hazard Mater 2016 302 175 187 10.1016/j.jhazmat.2015.09.053 26476304 Brown AK Challis JK Wong CS Hanson ML Selective serotonin reuptake inhibitors and beta-blocker transformation products may not pose a significant risk of toxicity to aquatic organisms in wastewater effluent-dominated receiving waters Integr Environ Assess Manag 2015 11 618 639 10.1002/ieam.1637 25820351 Bukowska B Effects of 2,4-D and its metabolite 2,4-dichlorophenol on antioxidant enzymes and level of glutathione in human erythrocytes Comp Biochem Physiol C: Toxicol Pharmacol 2003 135 435 441 10.1016/S1532-0456(03)00151-0 12965188 Campanale C Massarelli C Losacco D Bisaccia D Triozzi M Uricchio VF The monitoring of pesticides in water matrices and the analytical criticalities: A review TrAC Trend Anal Chem 2021 10.1016/j.trac.2021.116423 Carrara C Ptacek CJ Robertson WD Blowes DW Moncur MC Sverko E Backus S Fate of pharmaceutical and trace organic compounds in three septic system plumes, Ontario, Canada Environ Sci Technol 2008 42 2805 2811 10.1021/es070344q 18497127 Celiz MD Tso J Aga DS Pharmaceutical metabolites in the environment: Analytical challenges and ecological risks Environ Toxicol Chem 2009 28 2473 2484 10.1897/09-173.1 19663539 Chau ND Sebesvari Z Amelung W Renaud FG Pesticide pollution of multiple drinking water sources in the Mekong Delta, Vietnam: evidence from two provinces Environ Sci Pollut Res Int 2015 22 9042 9058 10.1007/s11356-014-4034-x 25572267 Chen WY Wu YT Lin HC Ieong MI Lee BH Impact of long-term parental exposure to Tamiflu metabolites on the development medaka offspring (Oryzias latipes) Environ Pollut 2020 261 114 146 10.1016/j.envpol.2020.114146 Christensen ER Li A Physical and Chemical Processes in the Aquatic Environment 2014 Hoboken Wiley & Sons Inc 255 259 Coller JK Christrup LL Somogyi AA Role of active metabolites in the use of opioids Eur J Clin Pharmacol 2009 65 121 139 10.1007/s00228-008-0570-y 18958460 Corcoran J Winter M Tyler C Pharmaceuticals in the aquatic environment: a critical review of the evidence for health effects in fish Crit Rev Toxicol 2010 40 287 304 10.3109/10408440903373590 20225984 Daffara F De Francia S Reimondo G Zaggia B Aroasio E Porpiglia F Volante M Termine A Di Carlo F Dogliotti L Angeli A Berruti A Terzolo M Prospective evaluation of mitotane toxicity in adrenocortical cancer patients treated adjuvantly Endocr Relat Cancer 2008 15 4 1043 1053 10.1677/ERC-08-0103 18824557 Daughton CG Pharmaceuticals and the Environment (PiE): Evolution and impact of the published literature revealed by bibliometric analysis Sci Total Environ 2016 562 391 426 10.1016/j.scitotenv.2016.03.109 27104492 Daughton CG Ternes TA Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Health Perspectives 1999 107 907 938 10.1289/ehp.99107s6907 Davis LK Visitacion N Riley LG Hiramatsu N Sullivan CV Hirano T Grau IE Effects of o, p'-DDE, heptachlor, and 17beta-estradiol on vitellogenin gene expression and the growth hormone/insulin-like growth factor-I axis in the tilapia, Oreochromis mossambicus Comp Biochem Physiol C: Toxicol Pharmacol 2009 149 507 514 10.1016/j.cbpc.2008.11.007 19101654 Davis, F. R. (2014). Banned: a history of pesticides and the science of toxicology. Yale University Press, London. https://www.jstor.org/stable/j.ctt13x1tbs. DellaGreca M Fiorentino A Isidori M Lavorgna M Previtera L Rubino M Temussi F Toxicity of prednisolone, dexamethasone and their photochemical derivatives on aquatic organisms Chemosphere 2004 54 629 637 10.1016/j.chemosphere.2003.09.008 14599508 Deribe E Rosseland BO Borgstrøm R Salbu B Gebremariam Z Dadebo E Skipperud L Eklo OM Biomagnification of DDT and its metabolites in four fish species of a tropical lake Ecotoxicol Environ Saf 2013 95 10 18 10.1016/j.ecoenv.2013.03.020 23790590 Desbiolles F Moreau X de Jong L Malleret L Grandet-Marchant Q Wong-Wah-Chung P Laffont-Schwob I Advances and limits of two model species for ecotoxicological assessment of carbamazepine, two transformation products and their mixture at environmental level in freshwater Water Res 2020 169 115267 10.1016/j.watres.2019.115267 31731244 Dimpe KM Nomngongo PN Current sample preparation methodologies for analysis of emerging pollutants in different environmental matrices TrAC Trends Anal Chem 2016 82 199 207 10.1016/j.trac.2016.05.023 Diniz MS Salgado R Pereira VJ Carvalho G Oehmen A Reis MA Noronha JP Ecotoxicity of ketoprofen, diclofenac, atenolol and their photolysis transformation products in zebrafish (Danio rerio) Sci Total Environ 2015 505 282 289 10.1016/j.scitotenv.2014.09.103 25461029 Donohoe RM Curtis LR Estrogenic activity of chlordecone, o, p′-DDT and o, p′-DDE in juvenile rainbow trout: induction of vitellogenesis and interaction with hepatic estrogen binding sites Aquat Toxicol 1996 36 1–2 31 52 10.1016/S0166-445X(96)00799-0 Dozio V Daali Y Desmeules J Sanchez JC Deep proteomics and phosphoproteomics reveal novel biological pathways perturbed by morphine, morphine-3-glucuronide and morphine-6-glucuronide in human astrocytes J Neurosci Res 2022 100 1 220 236 10.1002/jnr.24731 32954564 Eliassen AH Missmer SA Tworoger SS Hankinson SE Circulating 2-hydroxy- and 16alpha-hydroxy estrone levels and risk of breast cancer among postmenopausal women Cancer Epidemiol Biomarkers Prevention 2008 17 2029 2035 10.1158/1055-9965.epi-08-0262 Elsayed OF Maillard E Vuilleumier S Millet M Imfeld G Degradation of chloroacetanilide herbicides and bacterial community composition in lab-scale wetlands Sci Total Environ 2015 520 222 231 10.1016/j.scitotenv.2015.03.061 25817759 Evgenidou EN Konstantinou IK Lambropoulou DA Occurrence and removal of transformation products of PPCPs and illicit drugs in wastewaters: a review Sci Total Environ 2015 505 905 926 10.1016/j.scitotenv.2014.10.021 25461093 Fahlman J Hellstrom G Jonsson M Fick JB Rosvall M Klaminder J Impacts of oxazepam on Perch (Perca fluviatilis) behavior: fish familiarized to lake conditions do not show predicted anti-anxiety response Environ Sci Technol 2021 55 6 3624 3633 10.1021/acs.est.0c05587 33663207 Felicio AA Freitas JS Scarin JB de Souza Ondei L Teresa FB Schlenk D de Almeida EA Isolated and mixed effects of diuron and its metabolites on biotransformation enzymes and oxidative stress response of Nile tilapia (Oreochromis niloticus) Ecotoxicol Environ Saf 2018 149 248 256 10.1016/j.ecoenv.2017.12.009 29248837 Fenner K Canonica S Wackett LP Elsner M Evaluating pesticide degradation in the environment: blind spots and emerging opportunities Science 2013 341 6147 752 758 10.1126/science.1236281 23950532 Fent K Weston A Caminada D Ecotoxicology of human pharmaceuticals Aquat Toxicol 2006 76 122 159 10.1016/j.aquatox.2005.09.009 16257063 Ferreira M Costa J Reis-Henriques MA ABC transporters in fish species: a review Front Physiol 2014 5 266 10.3389/fphys.2014.00266 25101003 Foltz RL Andrenyak DM Crouch DJ Lindon J Tranter GE Koppenaal D Forensic science, applications of mass spectrometry Encyclopedia of Spectroscopy and Spectrometry 2016 Amsterdam Elsevier Ltd 707 711 Fong PP Molnar N Norfluoxetine induces spawning and parturition in estuarine and freshwater bivalves Bull Environ Contam Toxicol 2008 81 535 538 10.1007/s00128-008-9558-7 18787749 Gao J Wang F Cui J Zhang Q Wang P Liu D Zhou Z Assessment of toxicity and environmental behavior of chiral ethiprole and its metabolites using zebrafish model J Hazard Mater 2021 414 125492 10.1016/j.jhazmat.2021.125492 33647613 Geric M Ceraj-Ceric N Gajski G Vasilic Z Capuder Z Garaj-Vrhovac V Cytogenetic status of human lymphocytes after exposure to low concentrations of p, p'-DDT, and its metabolites (p, p'-DDE, and p, p'-DDD) in vitro Chemosphere 2012 87 1288 1294 10.1016/j.chemosphere.2012.01.037 22354074 Gong W Barrett H Hu Y Han J Wang F Wang W Zhou S Qu H Application of biochar: An approach to attenuate the pollution of the chiral pesticide fipronil and its metabolites in leachate from activated sludge Process Saf Environ Prot 2021 149 936 945 10.1016/j.psep.2021.03.044 Gros M Rodriguez-Mozaz S Barcelo D Fast and comprehensive multi-residue analysis of a broad range of human and veterinary pharmaceuticals and some of their metabolites in surface and treated waters by ultra-high-performance liquid chromatography coupled to quadrupole-linear ion trap tandem mass spectrometry J Chromatogr A 2012 1248 104 121 10.1016/j.chroma.2012.05.084 22704668 Guardabassi L Wong LF D.M., and Dalsgaard, A. The effects of tertiary wastewater treatment on the prevalence of antimicrobial resistant bacteria Water Res 2002 36 1955 1964 10.1016/s0043-1354(01)00429-8 12092570 Guilherme S Santos MA Gaivão I Pacheco M DNA and chromosomal damage induced in fish (Anguilla anguilla L.) by aminomethylphosphonic acid (AMPA)—the major environmental breakdown product of glyphosate Environ Sci Pollut Res 2014 21 14 8730 8739 10.1007/s11356-014-2803-1 Guimarães L Guilhermino L Afonso MJ Marques JM Chaminé I Assessment of urban groundwater: towards integrated hydrogeological and effects-based monitoring Sustain Water Resources Manag 2019 5 217 233 10.1007/s40899-019-00301-w Gumustas M Kurbanoglu S Uslu B Ozkan SA UPLC versus HPLC on drug analysis: advantageous, applications and their validation parameters Chromatographia 2013 76 1365 1427 10.1007/s10337-013-2477-8 Hernández-Moreno D Blázquez M Navas JM Fernández-Cruz ML Fish cell lines as screening tools to predict acute toxicity to fish of biocidal active substances and their relevant environmental metabolites Aquat Toxicol 2022 242 106020 10.1016/j.aquatox.2021.106020 34844051 Hignite C Azarnoff DL Drugs and drug metabolites as environmental contaminants: chlorophenoxyisobutyrate and salicyclic acid in sewage water effluent Life Sci 1977 20 2 337 341 10.1016/0024-3205(77)90329-0 839964 Hill AB Jefferies PR Quistad GB Casida JE Dialkylquinoneimine metabolites of chloroacetanilide herbicides induce sister chromatid exchanges in cultured human lymphocytes Mutation Res/genetic Toxicol Environ Mutagenesis 1997 395 159 171 10.1016/s1383-5718(97)00163-0 Hong B Lin Q Yu S Chen Y Chen Y Chiang P Urbanization gradient of selected pharmaceuticals in surface water at a watershed scale Sci Total Environ 2018 634 448 458 10.1016/j.scitotenv.2018.03.392 29631135 Huang Z Yao YN Li W Hu B Analytical properties of electrospray ionization mass spectrometry with solid substrates and nonpolar solvents Anal Chim Acta 2019 1050 105 112 10.1016/j.aca.2018.10.064 30661577 Huerta B Margiotta-Casaluci L Rodriguez-Mozaz S Scholze M Winter MJ Barcelo D Sumpter JP Anti-anxiety drugs and fish behavior: Establishing the link between internal concentrations of oxazepam and behavioral effects Environ Toxicol Chem 2016 35 2782 2790 10.1002/etc.3448 27061599 Huerta B Chung-Davidson YW Bussy U Zhang Y Bazil JN Li W Sea lamprey cardiac mitochondrial bioenergetics after exposure to TFM and its metabolites Aquat Toxicol 2020 219 105380 10.1016/j.aquatox.2019.105380 31855722 Huntscha S Singer HP McArdell CS Frank CE Hollender J Multiresidue analysis of 88 polar organic micropollutants in ground, surface and wastewater using online mixed-bed multilayer solid-phase extraction coupled to high performance liquid chromatography-tandem mass spectrometry J Chromatogr A 2012 1268 74 83 10.1016/j.chroma.2012.10.032 23137864 Ibáñez M Bijlsma L Pitarch E López FJ Hernández F Occurrence of pharmaceutical metabolites and transformation products in the aquatic environment of the Mediterranean area Trends Environ Anal Chem 2021 10.1016/j.teac.2021.e00118 Isidori M Lavorgna M Nardelli A Parrella A Previtera L Rubino M Ecotoxicity of naproxen and its phototransformation products Sci Total Environ 2005 348 93 101 10.1016/j.scitotenv.2004.12.068 16162316 Iwafune T Studies on the behavior and ecotoxicity of pesticides and their transformation products in a river J Pesticide Sci 2018 43 4 297 304 10.1584/jpestics.J18-01 Jaffar S Ahmad S Lu Y Contribution of insect gut microbiota and their associated enzymes in insect physiology and biodegradation of pesticides Front Microbiol 2022 13 979383 10.3389/fmicb.2022.979383 36187965 Jjemba PK Excretion and ecotoxicity of pharmaceutical and personal care products in the environment Ecotoxicol Environ Saf 2006 63 113 130 10.1016/j.ecoenv.2004.11.011 16399163 Jung F Thurn M Krollik K Gao GF Hering I Eilebrecht E Emara Y Weiler M Günday-Türeli N Türeli E Parnham MJ Wacker MG Predicting the environmental emissions arising from conventional and nanotechnology-related pharmaceutical drug products Environ Res 2021 192 110219 10.1016/j.envres.2020.110219 32980299 Kaufmann A Dvorak V Crüzer C Butcher P Maden K Walker S Widmer M Schürmann A Study of high-resolution mass spectrometry technology as a replacement for tandem mass spectrometry in the field of quantitative pesticide residue analysis J AOAC Int 2012 95 2 528 548 10.5740/jaoacint.11-074 22649942 Kemper N Veterinary antibiotics in the aquatic and terrestrial environment Ecol Ind 2008 8 1 13 10.1016/j.ecolind.2007.06.002 Kezios KL Liu X Cirillo PM Cohn BA Kalantzi OI Wang Y Factor-Litvak P Dichlorodiphenyltrichloroethane (DDT), DDT metabolites and pregnancy outcomes Reprod Toxicol 2013 35 156 164 10.1016/j.reprotox.2012.10.013 23142753 Khan HK Rehman MYA Malik RN Fate and toxicity of pharmaceuticals in water environment: An insight on their occurrence in South Asia J Environ Manage 2020 10.1016/j.jenvman.2020.111030 Koutnik D Stara A Zuskova E Kouba A Velisek J The chronic effects of terbuthylazine-2-hydroxy on early life stages of marbled crayfish (Procambarus fallax f. virginalis) Pestic Biochem Physiol 2017 136 29 33 10.1016/j.pestbp.2016.08.008 28187827 Kwiatkowska M Huras B Bukowska B The effect of metabolites and impurities of glyphosate on human erythrocytes (in vitro) Pestic Biochem Physiol 2014 109 34 43 10.1016/j.pestbp.2014.01.003 24581382 La Farre M Pérez S Kantiani L Barceló D Fate and toxicity of emerging pollutants, their metabolites and transformation products in the aquatic environment TrAC, Trends Anal Chem 2008 27 11 991 1007 10.1016/j.trac.2008.09.010 Lajeunesse A Gagnon C Gagne F Louis S Cejka P Sauve S Distribution of antidepressants and their metabolites in brook trout exposed to municipal wastewaters before and after ozone treatment—evidence of biological effects Chemosphere 2011 83 564 571 10.1016/j.chemosphere.2010.12.026 21211816 Lapworth DJ Gooddy DC Source and persistence of pesticides in a semi-confined chalk aquifer of southeast England Environ Pollut 2006 144 1031 1044 10.1016/j.envpol.2005.12.055 16750876 Le Cor F Slaby S Dufour V Iuretig A Feidt C Dauchy X Banas D Occurrence of pesticides and their transformation products in headwater streams: Contamination status and effect of ponds on contaminant concentrations Sci Total Environ 2021 10.1016/j.scitotenv.2021.147715 Li Z Sobek A Radke M Fate of pharmaceuticals and their transformation products in four small European rivers receiving treated wastewater Environ Sci Technol 2016 50 11 5614 5621 10.1021/acs.est.5b06327 27152425 Lienert J Güdel K Escher BI Screening method for ecotoxicological hazard assessment of 42 pharmaceuticals considering human metabolism and excretory routes Environ Sci Technol 2007 41 4471 4478 10.1021/es0627693 17626454 Lopez-Serna R Petrovic M Barcelo D Occurrence and distribution of multi-class pharmaceuticals and their active metabolites and transformation products in the Ebro river basin (NE Spain) Sci Total Environ 2012 440 280 289 10.1016/j.scitotenv.2012.06.027 22809787 Lotufo GR Landrum PF Gedeon ML Tigue EA Herche LR Comparative toxicity and toxicokinetics of ddt and its major metabolites in freshwater amphipods Environ Toxicol Chem 2000 19 368 379 10.1002/etc.5620190217 Luo Y Guo W Ngo HH Nghiem LD Hai FI Zhang J Liang S Wang XC A review on the occurrence of micropollutants in the aquatic environment and their fate and removal during wastewater treatment Sci Total Environ 2014 473–474 619 641 10.1016/j.scitotenv.2013.12.065 Madikizela LM Nuapia YB Chimuka L Ncube S Etale A Target and suspect screening of pharmaceuticals and their transformation products in the Klip river, South Africa, using ultra-high–performance liquid chromatography-mass spectrometry Environ Toxicol Chem 2022 41 437 447 10.1002/etc.5265 34888926 Martens M Verbruggen T Nymark P Grafström R Burgoon LD Aladjov H Torres Andón F Evelo CT Willighagen EL Introducing WikiPathways as a data-source to support adverse outcome pathways for regulatory risk assessment of chemicals and nanomaterials Front Genet 2018 9 661 10.3389/fgene.2018.00661 30622555 Martín-Pozo L de Alarcón-Gómez B Rodríguez-Gómez R García-Córcoles MT Çipa M Zafra-Gómez A Analytical methods for the determination of emerging contaminants in sewage sludge samples A Review Talanta 2019 192 508 533 10.1016/j.talanta.2018.09.056 30348425 McKnight US Rasmussen JJ Kronvang B Binning PJ Bjerg PL Sources, occurrence and predicted aquatic impact of legacy and contemporary pesticides in streams Environ Pollut 2015 200 64 76 10.1016/j.envpol.2015.02.015 25697475 Mompelat S Le Bot B Thomas O Occurrence and fate of pharmaceutical products and transformation products, from resource to drinking water Environ Int 2009 35 5 803 814 10.1016/j.envint.2008.10.008 19101037 Monteiro C Miranda CJ Brito F Fonseca C Araujo AR Consumption patterns of NSAIDs in central Portugal and the role of pharmacy professionals in promoting their rational use Drugs Therapy Perspectives 2017 33 32 40 10.1007/S40267-016-0352-Z Mosekiemang TT Stander MA de Villiers A Ultra-high pressure liquid chromatography coupled to travelling wave ion mobility-time of flight mass spectrometry for the screening of pharmaceutical metabolites in wastewater samples: Application to antiretrovirals J Chromatography A 2021 10.1016/j.chroma.2021.462650 Munze R Hannemann C Orlinskiy P Gunold R Paschke A Foit K Becker J Kaske O Paulsson E Peterson M Pesticides from wastewater treatment plant effluents affect invertebrate communities Sci Total Environ 2017 10.1016/j.scitotenv.2017.03.008 Murray KE Thomas SM Bodour AA Prioritizing research for trace pollutants and emerging contaminants in the freshwater environment Environ Pollut 2010 158 3462 3471 10.1016/j.envpol.2010.08.009 20828905 Nunes B Gaio AR Carvalho F Guilhermino L Behaviour and biomarkers of oxidative stress in Gambusia holbrooki after acute exposure to widely used pharmaceuticals and a detergent Ecotoxicol Environ Saf 2008 71 2 341 354 10.1016/j.ecoenv.2007.12.006 18243307 OECD OECD series on testing and assessment number 184: guidance document on developing and assessing adverse outcome pathways ENV/JM/MONO(2013) 6 2013 Paris OECD Publishing OECD (2018), "Users' Handbook supplement to the Guidance Document for developing and assessing Adverse Outcome Pathways", OECD Series on Adverse Outcome Pathways, No. 1, OECD Publishing, Paris. Ogilvie BW Zhang D Li W Rodrigues AD Gipson AE Holsapple J Toren P Parkinson A Glucuronidation converts gemfibrozil to a potent, metabolism-dependent inhibitor of CYP2C8: implications for drug-drug interactions Drug Metab Dispos 2006 34 191 197 10.1124/dmd.105.007633 16299161 Ortiz de García SA Pinto Pinto G García-Encina PA Irusta-Mata R Ecotoxicity and environmental risk assessment of pharmaceuticals and personal care products in aquatic environments and wastewater treatment plants Ecotoxicology 2014 23 1517 1533 10.1007/s10646-014-1293-8 25064485 Otieno PO Lalah JO Virani M Jondiko IO Schramm K-W Soil and water contamination with carbofuran residues in agricultural farmlands in Kenya following the application of the technical formulation Furadan J Environ Sci Health B 2010 45 137 144 10.1080/03601230903472058 20390943 Paiga P Santos L Ramos S Jorge S Silva JG Delerue-Matos C Presence of pharmaceuticals in the Lis river (Portugal): Sources, fate and seasonal variation Sci Total Environ 2016 573 164 177 10.1016/j.scitotenv.2016.08.089 27560998 Papadopoulos N Gikas E Zalidis G Tsarbopoulos A Simultaneous determination of terbuthylazine and its major hydroxy and dealkylated metabolites in wetland water samples using solid-phase extraction and high-performance liquid chromatography with diode-array detection J Agric Food Chem 2007 55 7270 7277 10.1021/jf0706777 17676755 Park WP Chang KM Hyun HN Boo KH Koo BJ Sorption and leaching characteristics of pesticides in volcanic ash soils of Jeju Island Korea Appl Biol Chem 2020 63 1 1 13 10.1186/s13765-020-00555-5 Perez-Maldonado IN Athanasiadou M Yanez L Gonzalez-Amaro R Bergman A Diaz-Barriga F DDE-induced apoptosis in children exposed to the DDT metabolite Sci Total Environ 2006 370 343 351 10.1016/j.scitotenv.2006.06.026 16904735 Perkins EJ Ashauer R Burgoon L Conolly R Landesmann B Mackay C Murphy CA Pollesch N Wheeler JR Zupanic A Scholz S Building and applying quantitative adverse outcome pathway models for chemical hazard and risk assessment Environ Toxicol Chem 2019 38 1850 1865 10.1002/etc.4505 31127958 Peters K Bundschuh M Schäfer RB Review on the effects of toxicants on freshwater ecosystem functions Environ Pollut 2013 180 324 329 10.1016/j.envpol.2013.05.025 23725857 Petrovic M Skrbic B Zivancev J Ferrando-Climent L Barcelo D Determination of 81 pharmaceutical drugs by high performance liquid chromatography coupled to mass spectrometry with hybrid triple quadrupole-linear ion trap in different types of water in Serbia Sci Total Environ 2014 468–469 415 428 10.1016/j.scitotenv.2013.08.079 Picó Y Barceló D Transformation products of emerging contaminants in the environment and high-resolution mass spectrometry: a new horizon Anal Bioanal Chem 2015 407 21 6257 6273 10.1007/s00216-015-8739-6 25990630 Planas C Puig A Rivera J Caixach J Analysis of pesticides and metabolites in Spanish surface waters by isotope dilution gas chromatography/mass spectrometry with previous automated solid-phase extraction Estimation of the uncertainty of the analytical results J Chromatogr A 2006 1131 242 252 10.1016/j.chroma.2006.07.091 16962600 Porazzi E Pardo Martinez M Fanelli R Benefati E GC–MS analysis of dichlobenil and its metabolites in groundwater Talanta 2005 68 1 146 154 10.1016/j.talanta.2005.04.044 18970297 Postigo C Barcelo D Synthetic organic compounds and their transformation products in groundwater: occurrence, fate and mitigation Sci Total Environ 2015 503–504 32 47 10.1016/j.scitotenv.2014.06.019 Ralston-Hooper K Hardy J Hahn L Ochoa-Acuna H Lee LS Mollenhauer R Sepulveda MS Acute and chronic toxicity of atrazine and its metabolites deethylatrazine and deisopropylatrazine on aquatic organisms Ecotoxicology 2009 18 899 905 10.1007/s10646-009-0351-0 19533344 Reddy PVL Kim K-H A review of photochemical approaches for the treatment of a wide range of pesticides J Hazard Mater 2015 285 325 335 10.1016/j.jhazmat.2014.11.036 25528231 Reemtsma T Alder L Banasiak U Emerging pesticide metabolites in groundwater and surface water as determined by the application of a multimethod for 150 pesticide metabolites Water Res 2013 47 5535 5545 10.1016/j.watres.2013.06.031 23863396 Rodieux F Vutskits L Posfay-Barbe KM Habre W Piguet V Desmeules JA Samer CF When the safe alternative is not that safe: tramadol prescribing in children Front Pharmacol 2018 5 9 148 10.3389/fphar.2018.00148 Rodrigues P Cunha V Oliva Teles L Ferreira M L., Guimarães. Norfluoxetine and venlafaxine in zebrafish larvae: Single and combined toxicity of two pharmaceutical products relevant for risk assessment J Hazard Mater 2020 400 123171 10.1016/j.jhazmat.2020.123171 32593945 Rodrigues P Cunha V Ferreira M Reis-Henriques MA Oliva Teles L Guimarães L Carvalho AP Differential molecular responses of zebrafish larvae to fluoxetine and norfluoxetine Water 2022 14 3 417 10.3390/w14030417 Rousis NI Bade R Bijlsma L Zuccato E Sancho JV Hernandez F Castiglioni S Monitoring a large number of pesticides and transformation products in water samples from Spain and Italy Environ Res 2017 156 31 38 10.1016/j.envres.2017.03.013 28314152 Rozmánková E Pípal M Bláhová L Njattuvetty-Chandran N Morin B Gonzalez P Bláha L Environmentally relevant mixture of S-metolachlor and its two metabolites affects thyroid metabolism in zebrafish embryos Aquat Toxicol 2020 221 105444 10.1016/j.aquatox.2020.105444 32078888 Rutkowska M Płotka-Wasylka J Sajid M Andruch V Liquid–phase microextraction: A review of reviews Microchem J 2019 149 103989 10.1016/j.microc.2019.103989 Safe SH Endocrine disruptors and human health–is there a problem? An update Environ Health Perspectives 2000 108 487 493 Santos LH Araujo AN Fachini A Pena A Delerue-Matos C Montenegro MC Ecotoxicological aspects related to the presence of pharmaceuticals in the aquatic environment J Hazard Mater 2010 175 45 95 10.1016/j.jhazmat.2009.10.100 19954887 Santos LH Paiga P Araujo AN Pena A Delerue-Matos C Montenegro MC Development of a simple analytical method for the simultaneous determination of paracetamol, paracetamol-glucuronide and p-aminophenol in river water J Chromatogr B 2013 930 75 81 10.1016/j.jchromb.2013.04.032 Scheil V Kienle C Osterauer R Gerhardt A Kohler HR Effects of 3,4-dichloroaniline and diazinon on different biological organisation levels of zebrafish (Danio rerio) embryos and larvae Ecotoxicology 2009 18 355 363 10.1007/s10646-008-0291-0 19093198 Schuhmann A Klammler G Weiss S Gans O Fank J Haberhauer G Gerzabek MH Degradation and leaching of bentazone, terbuthylazine and S-metolachlor and some of their metabolites: A long-term lysimeter experiment Plant Soil Environ 2019 65 5 273 281 10.17221/803/2018-pse Sharma A Kumar V Shahzad B Tanveer M Sidhu GPS Handa N Kohli SK Yadav P Bali AS Parihar RD Worldwide pesticide usage and its impacts on ecosystem SN Appl Sci 2019 10.1007/s42452-019-1485-1 Shipkova M Armstrong VW Oellerich M Wieland E Acyl glucuronide drug metabolites: toxicological and analytical implications Ther Drug Monit 2003 25 1 1 16 10.1097/00007691-200302000-00001 12548138 Sparkman OD Penton ZE Kitson FG Kitson FG Larsen BS McEwen CN Gas chromatography Gas Chromatography and Mass Spectrometry: A Practical Guide 2011 Amsterdam Elsevier 15 83 Stanley JK Ramirez AJ Chambliss CK Brooks BW Enantiospecific sublethal effects of the antidepressant fluoxetine to a model aquatic vertebrate and invertebrate Chemosphere 2007 69 9 16 10.1016/j.chemosphere.2007.04.080 17582462 Stokstad E Grullon G Infographic: Pesticide Planet. A global look at the uses, benefits, and drawbacks of pesticides Science 2013 341 730 731 10.1126/science.341.6147.730 23950524 Stropnicky, W. G. (2017). Effects of the Antidepressant O-Desmethylvenlafaxine on Crayfish Aggression.Theses and Dissertations (All). 1564 Stumm-Zollinger E Fair GM Biodegradation of Steroid Hormones Journal (water Pollution Control Federation) 1965 37 1506 1510 5849629 Subramaniam R Östin A Nilsson C Åstot C Direct derivatization and gas chromatography-tandem mass spectrometry identification of nerve agent biomarkers in urine samples J Chromatogr B 2013 928 98 105 10.1016/j.jchromb.2013.03.009 Sunkara M Wells MJM Phase II pharmaceutical metabolites acetaminophen glucuronide and acetaminophen sulfate in wastewater Environ Chem 2010 10.1071/EN09098 Syafrudin M Kristanti RA Yuniarto A Hadibarata T Rhee J Al-Onazi WA Algarni TS Almarri AH Al-Mohaimeed AM Pesticides in drinking water-a review Int J Environ Res Public Health 2021 10.3390/ijerph18020468 Tyagi V Garg N Mustafa MD Banerjee BD Guleria K Organochlorine pesticide levels in maternal blood and placental tissue with reference to preterm birth: a recent trend in North Indian population Environ Monit Assess 2015 187 471 10.1007/s10661-015-4369-x 26122123 Union E Commission Implementing Decision (EU) 2022/1307 of 22 July 2022: establishing a watch list of substances for Union-wide monitoring in the field of water policy pursuant to Directive 2008/105/EC Official J Eur Union, L 2022 197 117 121 European Union (2004). Directive 2004/27/EC of the European Parliament and of the Council of 31 March 2004 amending Directive 2001/83/EC on the Community code relating to medicinal products for human use. Official Journal of the European Union. L, 136, 34–57. Valls-Cantenys C Scheurer M Iglesias M Sacher F Brauch HJ Salvado V A sensitive multi-residue method for the determination of 35 micropollutants including pharmaceuticals, iodinated contrast media and pesticides in water Anal Bioanal Chem 2016 408 6189 6200 10.1007/s00216-016-9731-5 27382969 Velisek J Stara A Koutnik D Machova J Effect of terbuthylazine-2-hydroxy at environmental concentrations on early life stages of common carp (Cyprinus carpio L.) Biomed Res Int 2014 2014 621304 10.1155/2014/621304 24689046 Velisek J Koutnik D Zuskova E Stara A Effects of the terbuthylazine metabolite terbuthylazine-desethyl on common carp embryos and larvae Sci Total Environ 2016 539 214 220 10.1016/j.scitotenv.2015.08.152 26363394 Velisek J Stara A Zuskova E Kubec J Buric M Kouba A Chronic toxicity of metolachlor OA on growth, ontogenetic development, antioxidant biomarkers and histopathology of early life stages of marbled crayfish Sci Total Environ 2018 643 1456 1463 10.1016/j.scitotenv.2018.06.309 30189562 Veljanoska-Sarafiloska EM Jordanoski M Stafilov T Presence of DDT metabolites in water, sediment and fish muscle tissue from Lake Prespa, Republic of Macedonia J Environ Sci Health B 2013 48 548 558 10.1080/03601234.2013.774879 23581687 Vinken M Knapen D Vergauwen L Hengstler JG Angrish M Whelan M Adverse outcome pathways: a concise introduction for toxicologists Arch Toxicol 2017 91 3697 3707 10.1007/s00204-017-2020-z 28660287 Vryzas Z Pesticide fate in soil-sediment-water environment in relation to contamination preventing actions Curr Opinion Environ Sci Health 2018 4 5 9 10.1016/j.coesh.2018.03.001 Wajchenberg BL bergaria Pereira, M.A., Medonca, B.B., Latronico, A.C., Campos, C.P., Alves, V.A., Zerbini, M.C., Liberman, B., Carlos, G.G and Kirschner, M.A. Adrenocortical carcinoma: clinical and laboratory observations Cancer 2000 88 711 736 10.1002/(SICI)1097-0142(20000215)88:4<711::AID-CNCR1>3.0.CO;2-W 10679640 Weston DP Lydy MJ Toxicity of the insecticide fipronil and its degradates to benthic macroinvertebrates of urban streams Environ Sci Technol 2014 48 1290 1297 10.1021/es4045874 24368023 White PM Potter TL Bosch DD Joo H Schaffer B Muñoz-Carpena R Reduction in metolachlor and degradate concentrations in shallow groundwater through cover crop use J Agric Food Chem 2009 57 20 9658 9667 10.1021/jf9021527 19799423 Wilkinson J Hooda PS Barker J Barton S Swinden J Occurrence, fate and transformation of emerging contaminants in water: An overarching review of the field Environ Pollution 2017 231 954 970 10.1016/j.envpol.2017.08.032 Wode F van Baar P Dunnbier U Hecht F Taute T Jekel M Reemtsma T Search for over 2000 current and legacy micropollutants on a wastewater infiltration site with a UPLC-high resolution MS target screening method Water Res 2015 69 274 283 10.1016/j.watres.2014.11.034 25497426 Zhuo H-Q Huang L Huang H-Q Cai Z Effects of chronic tramadol exposure on the zebrafish brain: A proteomic study J Proteomics 2012 75 11 3351 3364 10.1016/j.jprot.2012.03.038 22507199	0	PMC9734374	NO-CC CODE	2022-12-14 23:28:28	no	Rev Environ Contam Toxicol. 2022 Dec 7; 260(1):14	utf-8	Rev Environ Contam Toxicol	2,022	10.1007/s44169-022-00014-w	oa_other
==== Front Rev Environ Contam Toxicol Rev Environ Contam Toxicol Reviews of Environmental Contamination and Toxicology 0179-5953 2197-6554 Springer International Publishing Cham 14 10.1007/s44169-022-00014-w Review Occurrence of Pharmaceutical and Pesticide Transformation Products in Freshwater: Update on Environmental Levels, Toxicological Information and Future Challenges Rodrigues P. 123 Oliva-Teles L. 12 http://orcid.org/0000-0003-3360-3783 Guimarães L. [email protected] 12 Carvalho A. P. 12 1 grid.5808.5 0000 0001 1503 7226 CIIMAR - Interdisciplinary Centre of Marine and Environmental Research, University of Porto, Terminal de Cruzeiros do Porto de Leixões Av. General Norton de Matos S/n, 4450-208 Matosinhos, Portugal 2 grid.5808.5 0000 0001 1503 7226 Department of Biology, FCUP – Faculty of Sciences, University of Porto, Rua do Campo Alegre, S/N, 4169-007 Porto, Portugal 3 grid.5808.5 0000 0001 1503 7226 ICBAS/UP-Institute of Biomedical Sciences Abel Salazar, University of Porto, Rua de Jorge Viterbo Ferreira, 228, 4050-313 Porto, Portugal 7 12 2022 2022 260 1 149 5 2022 28 11 2022 © The Author(s), under exclusive license 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. Pharmaceuticals and pesticides are recognized micropollutants in freshwater systems. Their ever-increasing frequency of detection, levels found and little information available about their effects on non-target organisms, make them emerging contaminants. However, parental compounds are not the only substances of concern. Their metabolites and degradation products, hereby referred to as transformation products, are increasingly detected in freshwater samples and wastewater effluents. In the past years, a wealth of publications provided concentration levels detected in freshwater and some toxicological data, which required critical systematization. This review identified concentrations for 190 transformation products (92 from pesticides and 98 from pharmaceuticals) in water bodies and wastewater effluents. A concentration heatmap was produced to easily spot the substances found at higher levels and plan future research. The very limited available toxicological data link exposure to transformation products to adverse outcomes in humans (genotoxicity and alteration in detoxification processes) and aquatic species (mostly related to apical endpoints). Overall, environmental levels of these transformation products may pose a severe threat to aquatic organisms and need to be further investigated in sound experimental designs, testing for the effects of the single substances as well as of their mixtures. Such toxicological information is highly needed to improve both water treatment technologies and monitoring programmes. Supplementary Information The online version contains supplementary material available at 10.1007/s44169-022-00014-w. BiodivRestore 2020 ERA-NET CofundBioReset (DivRestore/0004/2020) Fundação para a Ciência e TecnologiaUIDB/04423/2020 UIDP/04423/2020 SFRH/BD/134518/2017 Rodrigues P. issue-copyright-statement© The Author(s), under exclusive license to Springer Nature Switzerland AG 2022 ==== Body pmcPesticide and Pharmaceutical Transformation Products as Environmental Contaminants Over the past decades, scientists produced a wealth of information about the toxic effects of pesticides and pharmaceuticals on freshwater species. Both groups of compounds are widely used in the world, with recognized benefits for human health and welfare (Santos et al. 2010; Mcknight et al. 2015). Moreover, their use is globally escalating, owing to i) today’s social habits, ii) the increase in life expectancy, iii) the human population growth and iv) the consequent increase in food demand. More so under the actual pandemia caused by the SARS-CoV-2 coronavirus. Despite the need for them, these classes of chemicals have also been associated with severe human and environmental health risks and are common micropollutants of freshwater systems (Corcoran et al. 2010; Santos et al. 2013; Reemtsma et al. 2013; Ortiz de García et al. 2014; Mcknight et al. 2015). Pesticides and pharmaceuticals are designed to have specific biological activity, exerting the desired effect before undergoing excretion and/or degradation. These same characteristics also make them persistent in the environment, ultimately causing toxicity to non-target fauna and flora (Fent et al. 2006; McKnight et al. 2015). Due to these characteristics and the still limited information available, they are the most representative classes included in the watch list of substances for Union-wide monitoring of the Water Framework Directive (WFD). Currently, they represent over 88% of the compounds listed in the WFD (European Union 2022). Pesticides occur in natural water, mainly by run-off from the agricultural fields where they are applied and through industrial wastewater. Although soils can store a good amount of pesticides due to the high affinity of these compounds to organic matter, surface water and groundwater are susceptible to pesticide contamination because of the existing soil–water interconnections, mainly adsorption (Sharma et al. 2019). Pesticides that are highly adsorbed to soil particles are less likely to infiltrate deep but can easily be carried via run-off of and reach surface water (Syafrudin et al. 2021). Due to their increased use, detection of pesticides in different water compartments is becoming more and more frequent (Corcoran et al. 2010; Reemtsma et al. 2013; Ortiz de García et al. 2014; Evgenidou et al. 2015; Vryzas, 2018). On the other hand, the distribution of pharmaceutical substances in the environment is predominantly made by aqueous transport of compounds contained in discharged wastewater effluents, which persisted through the conventional treatment processes (Khan et al. 2020). Contamination by pharmaceutical compounds may also occur by terrestrial run-off from agricultural fields and aquaculture activities (Hong et al. 2018). Sorption is also an important process for the transport of pharmaceuticals in an aquatic environment. This process is responsible for the partitioning of pharmaceuticals between the water and the sediment phase (Bavumiragira et al. 2022). Once pesticides and pharmaceuticals reach the aquatic environment, they undergo a series of abiotic and biotic transformation and degradation processes. Hydrolysis, photodegradation and biodegradation are considered the most important mechanisms involved in their transformation or degradation (Syafrudin et al. 2021; Khan et al. 2020). Hydrolysis is an abiotic degradation process that creates products more polar than the parental compounds. These reactions are mainly catalysed by hydrogen or hydroxide molecules (Bavumiragira et al. 2022). Photolysis or photochemical degradation of pesticides and pharmaceuticals occurs by decomposition of these compounds in the presence of ultraviolet (UV) light. When exposed to sunlight, pesticides and pharmaceuticals containing chemical functional groups able to absorb solar radiation are prone to photolysis. The reaction transforms parental compounds into transformation products that are usually more biodegradable and hydrolysable (Wilkinson et al. 2017; Bavumiragira et al. 2022). Biodegradation is a biotic process that can result in the partial or complete transformation of pesticides and pharmaceuticals by microorganisms, such as certain fungi, bacteria, protozoans and microalgae. These microorganisms are present in wastewater treatment plants (WWTPs) or occur naturally in suspended solids, sediments and within animals (i.e. gut microbiota) (Wilkinson et al. 2017; Jaffar et al. 2022). Microbial degradation is recognized in the literature as having an important role in the degradation of several pharmaceuticals in a wide range of water compartments (Christensen and Li, 2014). For pesticides, microbial degradation includes the mineralisation process, which consists in the break of a parental pesticide into carbon dioxide and co-metabolization where microbial-catalysed reactions break pesticides into other chemical forms (Syafrudin et al. 2021). Surface waters receiving wastewater effluents rich in microorganisms are usually prone to show higher biodegradation effectiveness. High rates of biodegradation are typically observed along the sediment–water interface in water bodies and wetlands (Li et al. 2016). Degradation effectiveness varies according to biotic and abiotic factors, such as temperature, pH, UV light, presence of dissolved organic matter, suspended material and micro- and macrobiota (Vryzas, 2018). Low turbidity, small depth, low total organic carbon content and sandy sediments favour the degradation of pesticides and pharmaceuticals (Baena-Nogueras et al. 2017). On the other hand, higher depths, low temperature and higher turbidity can lower the degradation effectiveness (Syafrudin et al. 2021; Bavumiragira et al. 2022). Nevertheless, the described processes originate transformation products that enter in natural water by a panoply of different sources. In recent years, several works have reported the detection of these transformation products in the range of ng to µg/L, sometimes at concentrations even higher than those found for the parental compounds (le Cor et al. 2021). However, the focus of the reports was primarily on the detection and quantification of the parental compounds. Concern about their transformation products, with the involvement of more groups in this research, took off mostly in the last decade, especially for pharmaceutical transformation products. Investigation about the occurrence and fate of transformation products in the aquatic environment skyrocketed in recent years, mainly due to advances reached in the chemical analytical methods (Fent et al. 2006; Valls-Cantenys et al. 2016). New instruments and methods with higher separation efficiencies, ability to find more polar compounds and deal with confounding matrix effects, appeared allowing scientists to detect trace concentrations in environmental compartments (Fent et al. 2006; Celiz et al. 2009; Valls-Cantenys et al. 2016). Previous excellent reviews have been dedicated to this topic, although mostly to pharmaceutical and personal care products or emerging contaminants of concern and less so to pesticides (La Farre et al. 2008; Celiz et al. 2009; Mompelat et al. 2009; Evgenidou et al. 2015; Picó & Barceló, 2015; le Cor et al. 2021; Ibáñez et al. 2021; Mosekiemang et al. 2021; Madikizela et al. 2022). Furthermore, the number of works produced about this theme suffered a remarkable increase in recent years. Many of these compounds, parental or transformation products, are however little known in terms of potential detrimental effects and not included in the regulatory monitoring frameworks. Hence, they are nowadays recognized as emerging contaminants of concern (Murray et al. 2010; Evgenidou et al. 2015; NORMAN network, www.norman-network.net). Pesticides and pharmaceuticals are the two main classes of chemicals continuously represented in the watch list of the WFD and are thus the focus of this review. The aim of this literature review was to identify ecotoxicological knowledge gaps limiting the risk assessment of transformation products of pesticides and pharmaceuticals found in aquatic samples. We present and discuss updated information about quantification methods, occurrence, fate and the effects of transformation products of these two classes of chemicals. Over recent years, information has been published that needed to be systematized and appraised to bring understanding about their potential impacts on human health and aquatic biota. An important aspect, still enigmatic, is whether these transformation products are more harmful to non-target organisms than their parental compounds and which other factors may influence their potential toxicity. Another problem is the concern raised by transformation products not only as sole compounds per se but also in complex mixtures; mixtures of different metabolites of the same substance and mixtures of different substances, including parental compounds and transformation products. Applied Methodology The literature review carried out focused on the global occurrence and fate of the target contaminants in freshwater (i.e. surface-, ground- and influent/effluent wastewater), as well as on the available toxicological and ecotoxicological data. It covers articles published between 1997 and 2022, which have been searched in SCOPUS, Web of Science, PubMed and Google Scholar databases. The terms “pesticides” or “pharmaceuticals” were searched for in combination with “transformation products” or “degradation products”, “metabolites”, “freshwater”, “quantification”, “human health” or “aquatic species”. The search fields were the “article title”, “abstract” and “keywords”. Criteria for inclusion of articles in the review were related to the detail provided by the studies (i.e. quantification of the transformation products identified, suitable information about the species employed in the biotests, the age of the exposed organisms, relevant exposure design and endpoints assessed), as well as authors’ awareness and control of essential experimental conditions that may bias the results. All analytical methods of quantification have been included, rather than focusing on the most widespread techniques. Adding to this, most articles available in the literature are directed to parental compounds. Some of these works identify a few metabolites. Others do not include terms related to transformation products in the search fields and so they may not been detected. Some articles identify transformation products but do not quantify them, preventing prediction of their concentration in environmental samples (i.e. Mosekiemang et al. 2021; Madikizela et al. 2022). Articles about degradation experiments of pesticides and pharmaceuticals under controlled conditions have also been included, since such transformation processes can occur in natural conditions. Sources and Fate of Environmental Contamination Transformation Products of Pesticides Pesticides have been used since ancient times. Most of them were mainly inorganic compounds or substances of natural origin. However, the development and synthesis of organic pesticides after the second world war increased exponentially its use, making pesticides the second most used group of substances in the environment, only behind fertilizers (Davis 2014; Stokstad and Grullon 2013). Millions of tons of pesticides are applied each year, predominantly in agriculture, which is the main activity responsible for the leaking of pesticides and their sub-products into freshwater ecosystems (Fenner et al. 2013). Although applied in the soil, pesticides can reach aquatic ecosystems through diffuse/non-point-source or point-source pollution sites (Vryzas 2018; Fig. 1).Fig. 1 Sources of pollution, as well as formation and fate of transformation products deriving from pesticides Diffuse or non-point-source pollution is related with the movement of pesticides from large areas across the watersheds that reach the aquatic environment. Point-source pollution is related to a specific identifiable source which can include chemical run-off during storage, loading, disposal, as well as the misapplication of pesticides to water bodies (Syafrudin et al. 2021). Groundwater is heavily impacted by pesticides, and their metabolites or degradation products (i.e. N,N-dimethylsulfamide; aminomethylphosphonic acid; 2,6-Dichlorobenzamide), mainly resulting from point-source pollution (Postigo and Barcelo 2015). Leaching of landfill and septic tanks or industrial leakage are amongst the main sources of groundwater contamination by pesticides and metabolites (Postigo and Barcelo 2015). Pesticide characteristics (i.e. solubility and vapour pressure) are responsible for higher or lower rates of leaching to groundwater (Park et al. 2020). They also play an essential influence on the degradation of pesticides and the formation of transformation products, including active metabolites. Sorption–desorption, volatilization, chemical and biological degradation, uptake by plants, soil infiltration and leaching are some processes responsible for the appearance of new metabolites and their transportation into groundwater (Arias-Estévez et al. 2007). Groundwater tends, however, to be less affected by contamination than other water bodies, due to the natural attenuation capacity of aquifers and their large capacities (Postigo and Barcelo 2015). Nonetheless, recent monitoring studies show that pesticides, mainly herbicides, and their metabolites/degradation products (i.e. desethylatrazine; cyanazine amide) are present in aquifers (Lapworth and Gooddy 2006; Reemtsma et al. 2013). Surface waters are mainly contaminated by diffuse pollution sources. Run-off of pesticides and metabolites from agricultural fields after heavy rains are the main pathways of transportation of these substances to surface waters (Vryzas 2018). Moreover, during rainfall pesticides and degradation products imprisoned in soil or sediments can reach surface waters due to movements of those sediments (Vryzas 2018). Application of pesticides using sprays or even the plantation of seeds can be a source of surface water contamination, thanks to wind dispersal (Vryzas 2018). In these waters, photodegradation is the main process responsible for the degradation of pesticides. The formation of such metabolites can reach higher concentrations and show higher toxicity than the parental compounds (Reddy and Kim 2015). Wastewater treatment plants are also amongst the main sources of point-source pollution. Pesticides applied in urban areas (i.e. in maintenance of green areas or ponds) tend to finish in WWTPs, where traditional wastewater treatment methods are ineffective for the removal of these compounds (Rousis et al. 2017; Munze et al. 2017). Moreover, WWTPs effluents show in some cases higher concentrations of pesticides and their transformation products, as well as more toxicity, than the influents. Owing to all this, pesticides and their metabolites can ultimately reach drinking water, exposing humans, as indicated by their detection in the serum and blood of some patients in clinical and scientific studies (Chau et al. 2015; Tyagi et al. 2015). Transformation Products of Pharmaceuticals According to Daughton (2016), the first studies regarding the presence of pharmaceuticals in the environment date back to the 1940s. Later on, between the 60 s and 70 s, several works were produced about the possibility of contamination of drinking and surface water by pharmaceuticals, through the discharge of wastewater effluents (i.e. Stumm-Zollinger and Fair 1965; Hignite and Azarnoff 1977). Nowadays, pharmaceutical products are continuously released into the environment, although in small quantities (Fig. 2).Fig. 2 Sources of pollution as well as formation and fate of transformation products deriving from pharmaceuticals After consumption by humans, pharmaceuticals pass through the liver where they are directly effluxed from the organism (phase 0) or enter phase I and phase II of drug metabolism (Fig. 3). In phase I, more polar metabolites, often still active, are produced through oxidation, reduction or hydrolysis reactions. These reactions are commonly mediated by different CYP450 genes (i.e. CYP1A; CYP2B; CYP3A). Many of these transformation products become substrates of phase II, where endogenous hydrophilic groups are added through methylation, glucuronidation, acetylation, sulfation or conjugation with glutathione or amino acids such as glycine, taurine and glutamic acid to form water-soluble inactive compounds that can be excreted by the body in phase III (Fig. 3). Phase III excretion is mediated by ABC transporters and different solute carriers. Due to such reactions, pharmaceuticals can thus be excreted by humans in different forms: unchanged (small proportion) or as active or inactive metabolites (Jjemba 2006; Brown et al. 2015).Fig. 3 Schematic representation of pharmaceutical’s biotransformation and excretion Hospital effluents and direct elimination (i.e. through inadequate sanitary disposal) of unused pharmaceuticals in sewage are therefore amongst the most important sources of water contamination (Santos et al. 2010) by pharmaceutical transformation products. Influents are treated in WWTPs (Wastewater Treatment Plants) by three main processes of pollutant removal. In the first treatment, the removal of suspended solids occurs. This treatment has a low degree of efficiency in the removal of micropollutants, like pharmaceuticals (parental compounds or metabolites). In the second treatment, several types of reactions occur, such as dilution, partition, biotic and abiotic transformation (Luo et al. 2014). In this treatment, the level of efficiency is variable depending on the substance or metabolite in question, as well as their physicochemical properties (Luo et al. 2014). The third treatment is related to health questions to humans or specific uses of the treated water. It consists in further removal of substances, like nitrogen or phosphorus, and it is not mandatory, in general (Guardabassi et al. 2002; Luo et al. 2014). After this processing, in some cases, the total load of pharmaceutical compounds or metabolites in the effluent can be higher than that in the influent (Luo et al. 2014). This can be explained by the degradation of parental compounds into several metabolites or degradation products and the transformation of metabolites back into the parental compounds that can occur during the biological treatment in the WWTPs (Luo et al. 2014). Parental compounds and metabolites can also be imprisoned in faecal matter and released into the water during the biological treatment, thus increasing the overall concentration of those substances (Luo et al. 2014). This shows that the treatments available in WWTPs are still not fully efficient in the removal of these micropollutants. Hence, discharge of contaminated effluents introduces into natural waters the parental compounds and many more metabolites or transformation products (i.e. venlafaxine, tramadol, O-desmethyltramadol) (Santos et al. 2010; Luo et al. 2014). Additionally, some of these pharmaceutical metabolites are expected to be more toxic than parental compounds and consequently more dangerous to the wildlife (Celiz et al. 2009). The use of medicines is not exclusive to humans. These are also used in agriculture and aquaculture to treat diseased animals. As in humans, they are also excreted mostly as metabolites in the urine and faeces of animals or through adsorption in dirt pounds and after tanks cleaning, thus entering the environment without any kind of treatment and contaminating the soil and water (Santos et al. 2010). This contamination contributes to further input of transformation products into natural waters via run-off and leaching from the affected soils (Kemper 2008). Other anthropological activities also act as sources of contamination. Industry discharges (sometimes illegally), the use of WWTPs sludge contaminated with all kinds of pharmaceutical compounds as fertilizer, or leakage of septic tanks from households still not connected to the sewage systems, are examples of these (Carrara et al. 2008; Santos et al. 2010). Detection of Pesticide and Pharmaceutical Transformation Products in Water Compartments Analytical Methods of Pesticide and Pharmaceutical Transformation Products As previously mentioned, knowledge about contamination of the aquatic environment by pesticides and pharmaceutical transformation products has increased, mainly due to advances reached in analytical methods (Fent et al. 2006; Valls-Cantenys et al. 2016). New methods, with higher separation efficiencies and the ability to find more polar compounds, appeared. This allowed scientists to detect concentrations in environmental compartments in the order of ng/L and μg/L and consequently raised awareness and concern about their potential hazardousness (Fent et al. 2006; Santos et al. 2010; Valls-Cantenys et al. 2016). However, these advances in analytical methods are not efficient if the correct sample preparation is not performed. The extraction of the analytes from an environmental water sample is a crucial step before the instrumental analysis. Extraction techniques are based on the passage of an analyte by different solvents, which must be the most suitable for the type of analytical tool to be employed (Rutkowska et al. 2019; Campanale et al. 2021). This step can highly influence the analytical process, mainly for quantitative analysis, since the analyte volume must be increased, whilst any interferences must be eliminated (Campanale et al. 2021). Several sample preparation techniques are already described in the literature. For analysis of water samples, SPE (solid-phase extraction) is the most extensively used technique (Dimpe and Nomngongo 2016; Campanale et al. 2021). This method uses columns or disks able to retain the active compounds present in water samples and posteriorly release them by washing with small quantities of suitable solvents (Dimpe and Nomngongo 2016; Campanale et al. 2021). This provides an extract with few interferences, suitable for different analytical methodologies, such as High-Pressure Liquid Chromatography-Mass Spectrometry (HPLC–MS) and Gas Chromatography-Mass Spectrometry (GC–MS). More recently, a new SPE-based approach has been tested: Solid-Phase MicroExtraction (SPME). This newer method is faster and requires fewer quantity of solvents and is well described as suitable for Gas chromatography (GC) analysis (Campanale et al. 2021). Liquid–liquid extraction (LLE) is a simple method widely used for water samples that is also applied in the analysis of pesticide and pharmaceuticals (Dimpe and Nomngongo 2016; Campanale et al. 2021). This method has the advantage to be well established amongst different governmental agencies, but it also is time-consuming and requires the use of organic solvents that are harmful to the environment and even the handler (Dimpe and Nomngongo 2016; Campanale et al. 2021). Though the techniques described are the most well established for pharmaceuticals and pesticides, they have some disadvantages. One is the loss of more volatile analytes during the extraction process, which can affect the result of the analysis; another is the use of toxic solvents (Dimpe and Nomngongo 2016; Campanale et al. 2021). Much more methods are available in the literature, although less widespread. The development of new cost-effective and green methodologies for fast extraction is the next challenge in need to be addressed to improve and analytical determination. Regarding the analytical methods and instrumentation, GC and/or liquid chromatography (LC) coupled to mass spectrometry (MS) are, nowadays, the most applied methods to detect pesticides and pharmaceutical compounds. For LC, some variations to this method are well established in the literature, such as high-performance liquid chromatography (HPLC) and ultra-performance liquid chromatography (UPLC) (Gumustas et al. 2013). Gas chromatography is the most suitable method to separate non-polar parental compounds and their transformation products (Sparkman et al. 2011). This happens because of the inclusion of a derivatization process during GC that increases volatility and sensitivity, but also increases the duration of the procedure (Subramaniam et al. 2013). Coupling the GC with the MS has the advantage to offer a specific mass spectrum for a certain compound when an electron ionization (EI) is also performed (Foltz et al. 2016). Polar pharmaceutical or pesticide and their transformation products are mainly separated by LC (Martín-Pozo et al. 2019). Most of the pesticides and pharmaceuticals in their unchanged form or as transformation products are usually quantified at low concentrations in environmental water samples (ng to μg/L). This makes liquid chromatography with tandem mass spectroscopy (LC–MS/MS) a widely used method for determination of these compounds. This is due to its higher discrimination between the analyte and matrix signal, coupled to robustness and relative ease of use (Kaufmann et al. 2012). However, the selectivity and sensitivity of the MS vary with the selected ionization. Electrospray ionization (ESI) is the most chosen technique for detecting pharmaceuticals, since it is the most potent ionization method for the target compounds (Huang et al. 2019). Levels in Different Water Compartments Transformation Products of Pesticides As mentioned above, advances in the detection techniques led to an increase in knowledge about the occurrence of pharmaceutical and pesticide transformation products in different water compartments. Nevertheless, the quantification of transformation products is still not a focus in scientific investigation, as often studies only present new methods of detection and their validation, but not the concentrations found in real samples, even for parental compounds (Wode et al. 2015; Boix et al. 2016). Overall, the search carried out in the scientific databases returned 87 articles providing concentrations of pesticide and pharmaceutical transformation products in environmental water samples. Only one article quantified both pesticide and pharmaceutical transformation products (Huntscha et al. 2012). Of these, 29 articles were dedicated to pesticides, presenting concentrations obtained for 92 transformation products resulting from 43 parental compounds (Fig. 4, Table S1 in the Support Information). The most assessed pesticide transformation products (69%) belonged to three functional classes: organochlorine and chloroacetanilide herbicides/pesticides; triazine herbicides; and organophosphate and carbamate pesticides (Fig. 4, Table S1). Data about the quantification of pesticide transformation products found in different water samples, and respective detection methods, are presented in Fig. 5 and Table S1 (supplementary data). Transformation products of triazine herbicides were frequently reported in different studies, with a special focus on atrazine and terbuthylazine (Fig. 5, Table S1). Concentrations of these varied widely from 0.046 µg/L for desethylatrazine to 124,01 µg/L for hydroxy-terbuthylazine. The high concentration found for hydroxy-terbuthylazine resulted from an experiment where the parental compound was applied in a constructed wetland planted with Typha latifolia (Papadopoulos et al. 2007). According to the authors, the maximum concentration of the metabolite was within the highest concentration range found for the parental compound. This is of main concern, given that these concentrations are in the order of µg/L. Though knowledge about the environmental impact of this transformation product is sparse, recent works highlighted negative effects on the early developmental stages of fish species even at concentrations found in natural water samples (Velisek et al. 2014).Fig. 4 Relative frequency of transformation products quantified in environmental water samples per functional class of pesticides or pharmaceuticals. Concentrations for 92 pesticide transformation products derived from 43 parental compounds and 98 pharmaceutical transformation products derived from 64 parental compounds were found in the scientific literature published between 2000 and 2020 Fig. 5 Maximum concentrations of pesticide (A) and pharmaceutical (B) transformation products found in different water compartments. The heatmaps were done with the log-transformed values (pg/L) (Tables S1 and S2). Grey squares represent situations for which no information could be found Chloroacetanilide herbicides are widely used for grass control in several crops. Compounds of this class are structurally similar and were extensively used from the mid-1990s until recently (Elsayed et al. 2015). The main transformation products ethane sulfonic acid (ESA) and oxalinic acid (OXA), alongside the parental compound, are easily transported to water bodies and usually detected in both surface and groundwater (Table S1), contributing to the degradation of water quality (Baran and Gourcy 2013). Another interesting observation is the concentration level of metolachlor OXA and ESA in relation to the parental compound. Studies reported that metabolites of metolachlor, mainly ESA, were found in groundwater at higher concentrations than the parental compound (White et al. 2009; Baran and Gourcy 2013). This may occur because metabolites adsorb less to soil particles, compared to the parental compound and are thus more prone to infiltration to aquifer recharge (Baran and Gourcy 2013). This highlights the importance of monitoring programmes not only for pesticides alone but also for their transformation products. One of the first mass-produced pesticides in the world was DDT (Dichlorodiphenyltrichloroethane). It is an inexpensive and highly efficient short-term insecticide, but in the long term, it is problematic to human and animal health (Kezios et al. 2013). This pesticide was systematically banned in developed countries since the 1970s and a global ban of DDT, for non-vector control use, was exerted in the Stockholm Convention on Persistent Organic Pollutants, which took effect in 2004. However, this substance as well as several of its transformation products are still found in natural water bodies (Table S1) and tissues of different organisms (Veljanoska-Sarafiloska et al. 2013). A study conducted in African lakes showed that 4,4-DDE, a DDT metabolite, was biomagnified in fish species of the lake (Deribe et al. 2013). This was worrying, as those fish were consumed by local populations, possibly impacting human health. It is also of high concern the fact that metabolites of DDT, as well as the parental compound, are still commonly found in the environment even after an almost total ban worldwide, showing the great persistence of this substance and its transformation products in the ecosystems. Carbofuran is one of the carbamate pesticides most toxic to vertebrates, including humans, but knowledge about its main transformation products is still sparse. Otieno and colleagues (2010) reported the presence of very high concentrations of 3-ketocarbofuran and carbofuran-3-hydroxy (Table S1) in surface waters highly impacted by agrochemical procedures. Concentrations found (> 890 μg/L) were well above the standard water concentrations allowed by the USA and European authorities for safe drinking and human use (Otieno et al. 2010). Also, these two compounds appeared to be more persistent and were detected in higher concentrations than the parental compound (Otieno et al. 2010). Considering this, it would be crucial to gain a higher level of knowledge about the possible effects of these substances on non-target organisms, including humans, that could help infer about the need for more strict monitoring routines aiming at minimizing potential impacts on water quality and populations’ health. Transformation Products of Pharmaceuticals Fifty-eight articles were dedicated to pharmaceuticals, presenting concentrations obtained for 98 transformation products resulting from 64 parental compounds (Fig. 4, Table S2 in the Support Information). The most investigated transformation products belonged also to three functional classes: psychotropic drugs; analgesics, antipyretics and opioid painkillers and anticonvulsants (Fig. 4, Table S2). Carbamazepine transformation products, alongside metabolites of selective monoamine reuptake inhibitors and of ibuprofen, were the ones most reported in the literature (Fig. 5). The range of concentrations found varied from < 0.50 ng/L to 462000 ng/L, showing that very high concentration values of pharmaceutical metabolites are already found in the natural environment. Acetaminophen metabolites were the ones with higher reported concentrations. Sunkara and Wells (2010) reported concentrations higher than 400000 ng/L for acetaminophen glucuronide and sulphate in WWTP effluents. Those values were obtained in samples collected after application of conventional treatment processes in WWTP, pointing out the inefficiency of these treatments for the removal of micropollutants. Moreover, the authors refer that sometimes, metabolite concentrations were higher in the effluent than in the influent and one of the reasons for that was the bioconversion that may occur during the biological treatment, as mentioned previously. However, following UV treatment, none of the metabolites was found. This could be soothing, but the UV treatment is not always applied in WWTP; it is an optional treatment used mainly in water for human consumption (Luo et al. 2014; Guardabassi et al. 2002). Water without UV treatment loaded with transformation products can thus re-enter the water cycle, potentially risking aquatic fauna and flora. Also, it can be reused in agricultural practices and therefore contaminate crops, making metabolites enter the food chain with risk to human health. Carboxy ibuprofen was also reported at a very high concentration, higher than 100000 ng/L, in WWTP influents according to Paíga and colleagues (2016). Samples were collected in a relatively small WWTP designed to serve a little less than 50,000 people. Receiving wastewaters were mainly domestic and conventional treatments with activated sludge were applied (Paíga et al. 2016). Carboxy ibuprofen is one of the most representative ibuprofen metabolites. Ibuprofen is a commonly used non-steroidal anti-inflammatory (NSAID) drug and in 2016 it was the most used NSAID in Portugal, where the study was conducted (Monteiro et al. 2017). It is thus important to have a stricter monitoring routine for these substances to better evaluate the possible effects of metabolites on human and non-human health. Also, carbamazepine-10,11-epoxide was reported to occur at concentrations higher than 10000 ng/L in WWTP influents (Gros et al. 2012) and municipal wastewater (Petrovic et al. 2014). This is one of the main carbamazepine metabolites and one of the most detected in natural water samples (Table S2). An interesting fact in the study of Petrovic et al. (2014) is that carbamazepine-10,11-epoxide was found at a much higher concentration than the parental compound. This was also reported previously by Lopez-Serna et al. (2012) in a study conducted in the Ebro River in Spain. Those data reinforce the necessity of an extensive assessment and monitoring routine for metabolites, once they can be more prevalent in water compartments, compared to their parental compounds. Risks of Pharmaceutical and Pesticide Transformation Products Human Health Although the available data are sparse, freshwater contamination does not affect only organisms living in those systems. Ultimately, humans can also suffer negative effects from exposure to transformation products. Humans are exposed to pesticide and pharmaceutical transformation products in different ways. Data presented in Tables S1 and S2 show levels of those transformation products detected in drinking water and groundwater as well, which is a common source of drinking water in cities around the world (Guimarães et al. 2019). As previously mentioned, exposure can occur via contaminated recreational water and/or consumption of contaminated freshwater organisms or other food produced with water originating from contaminated sites. Knowledge about human health risks caused by transformation products of pesticides and pharmaceuticals is still sparse, compared to parental compounds. Studies available in the scientific literature are presented in Table 1.Table 1 Toxicological studies about the human health risks of pesticide and pharmaceutical transformation products Transformation product [Parental compound] Concentrations Sample Exposure duration Endpoints Effects Reference Pesticides Chloroacetanilide, aniline; hydroxychloroacetanilide; and diethylquinoneimine [Alachlor] (Hill et al. 1997) 0; 0.03; 0.1; and 0.3 μM Lymphocyte cells 72 h Oncogenicity Induction of chromatid exchange at 0.1 μM for hydroxychloroacetanilide, 0.3 μM for chloroacetanilide and aniline Mitotane [DDT] ( Daffara et al. 2008) Distinct values for each sample Blood cells and saliva not applicable Hormonal levels and organ toxicity Inhibition of cortisol and DHEAS. Induction of thyroid function perturbations. Inhibition of testosterone secretion 2–4-dichlorophenol [2–4-D] (Bukowska, 2003) 10 to 500 ppm Blood cells 1 h Antioxidant enzymes Increase of superoxide dismutase and increase of gluthathione peroxidase activities DDE [DDT] (Perez-Maldonado et al. 2006) Distinct values for each sample Blood cells not applicable Genotoxicity Induction of peripheral blood mononuclear cells p-p' DDE [DDT] (Geric et al. 2012) 4.1 μg/ml Lymphocyte cells 1; 6 and 24 h Genotoxicity Induction of DNA damage p-p' DDE [DDT] (Geric et al. 2012) 3.9 μg/ml Lymphocyte cells 1; 6 and 24 h Genotoxicity Induction of DNA damage AMPA [glyphosate] (Benachour and Séralini, 2009) 18 concentrations from 10 ppm to 10% Embryonic kidney HUVEC primary neonate umbilical cord vein, embryonic kidney. and JEG3 placental cell lines 24 h Cytotoxicity Increased cellular mortality. Destruction of the membrane of all cell types AMPA [glyphosate] (Kwiatkowska et al. 2014)) 0.01–5 mM Erythrocytes 1, 4 and 24 h Haemolysis, haemoglobin oxidation, ROS formation and morphology Induction of haemolysis (0.05 to 5 mM) and haemoglobin oxidation (0.25 to 5 mM) at 24-h incubation. Increase in ROS production at concentrations starting from 0.25 Mm Methylsulphonic acid [glyphosate] (Kwiatkowska et al. 2014)) 0.01–5 mM Erythrocytes 1, 4 and 24 h Haemolysis, haemoglobin oxidation, ROS formation and morphology Induction of haemolysis (0.1 to 5 mM) and haemoglobin oxidation (0.5 to 5 mM) at 24-h incubation. Increase in ROS production at 0.5 and 5 mM Pharmaceuticals Gemfibrozil 1-O-β- glucoronide [gemfibrozil] (Ogilvie et al. 2006) 0.25 to 64 μM Liver microsomes 2 to 40 min CYP2C8 activity Potent inhibitor of CYP2C8 2-hidroxyestrone and 16-α hydroxyestrone [estrogens] (Eliassen et al. 2008) not applicable Blood cells not applicable Genotoxicity and mitogenicity Levels of 2-hydroxyestrone, and the ratio between 2-hydroxyestrone and 16-α hydroxyestrone were linked with certain types of breast cancer tumours in woman Morphine-3-glucoronide [morphine] (Dozio et al. 2022) 1, 10 and 100 μM Astrocytes 12, 24, 48 and 96 h Proteomics 96-h exposure lead to dysregulation of biological pathways linked with extracellular matrix organization, antigen presentation, cell adhesion and glutamate homeostasis Morphine-6-glucoronide [morphine] (Dozio et al. 2022) 1, 10 and 100 μM Astrocytes 12, 24, 48 and 96 h Proteomics Acute exposure increased the levels of proteins involved in cell adhesion and decreased the levels of extracellular matrix The adverse effects that pesticides can cause on human health are a long-known problem. This discussion gained bigger attention and impact since the publication of the book Silent Spring in 1962. In this publication, Rachel Carson described not only the environmental impacts coinciding with the widespread use of DDT in agriculture in the USA, but also the potential of DDT to cause cancer in exposed workers. In the book, other pesticides were also surveyed, such as 2,4-D (2,4-Dichlorophenoxyacetic acid), chlordane and heptachlor. More recently different environmental agencies, including EPA (United States Environmental Protection Agency) and ECHA (European Chemicals Agency) or international conventions are banning the use of some pesticides that were described as hazardous to human health. Amongst the pesticide metabolites that can elicit problems, mitotane was proven to be a selective toxicant to humans and is used as an adjuvant drug to treat adrenocortical tumours (Wajchenberg et al. 2000). Mitotane or o,p’-dichlorodiphenyldichloroethane (o-p’-DDD) is a DDT metabolite and apparently the only chemical able to inhibit corticoid synthesis and at the same time destroy cortical cells (Wajchenberg et al. 2000). However, despite the therapeutic use, mitotane was already reported in the literature to have side effects at hormonal levels in patients who were treated with this compound (Daffara et al. 2008). The authors analysed the blood cells and the saliva of the patients and found that mitotane treatment was linked to the inhibition of cortisol and DHEAS (Dehydroepiandrosterone sulphate). Also, perturbations of the thyroid function were described. Moreover, for males, an inhibition of testosterone secretion was also found. However, these side effects were usually reversible with the adequate treatment. Another DDT metabolite, DDE (dichlorodiphenyldichloroethylene) was reported to induce apoptosis of human peripheral blood mononuclear cells, both in vitro and in vivo (Perez-Maldonado et al. 2006). The authors studied blood collected from 61 healthy children during the year 2004 and from 57 children from southern Mexico. Exposure to both DDT, DDD and DDE was found in the tested children. However, significant correlations between apoptosis and exposure to pesticides were only found for DDE blood levels, (p = 0.010 and 0.040 for 2003 and 2004, respectively). This causes great concern since DDE is the most persistent DDT metabolite and thus exposure tends to be chronic, and apoptosis of the cells could result in an impairment of the immune system (Perez-Maldonado et al. 2006). Both p,p′-DDE chloroethane and p,p′-DDD (dichlorodiphenyldichloroethane) were reported to induce DNA damage in human lymphocytes, even at low concentrations (Geric et al. 2012). In this study, in vitro human lymphocytes were exposed for 1, 6 and 24 h to p,p’-DDE (4.1 μg/mL) or p,p’-DDD (3.9 μg/mL) and genotoxic effects were assessed using the cytokinesis-block micronucleus assay and the comet assay. Results showed an increase in the number of cells containing micronucleus, in relation to the control, in the 24-h exposures. Also, according to the comet assay, the percentage of DNA damages increased, in relation to the control. It is important to notice that the concentrations used are in the range found in human fluids, suggesting that these effects are already occurring in humans exposed to the metabolites (Geric et al. 2012). The metabolite 2,4-dichlorophenol, from the herbicide 2,4-D, was reported to cause effects on antioxidant enzymes and glutathione levels in human erythrocytes in vitro (Bukowska, 2003): the activity of superoxide dismutase decreased whilst that of glutathione peroxidase increased in a dose-dependent (10–500 ppm) manner. Moreover, exposure to 250-ppm 2,4-dichlorophenol also decreased the level of reduced glutathione in erythrocytes by 32%, in relation to the control. These effects are similar, though more pronounced, to those resulting from exposure to the parental compound 2,4-D, pointing to a major need for monitoring pesticide metabolites in natural samples. Dialkylquinoneimine metabolites of chloroacetanilide herbicides like alachlor and acetochlor were reported to induce in vitro sister chromatid exchanges in human lymphocytes (Hill et al. 1997). This study was performed to test the hypothesis that the oncogenicity of chloroacetanilide herbicides previously described was caused by genotoxic intermediates, like diethylbenzoquinoneimine, an alachlor metabolite. The investigation was done with cultured human peripheral lymphocytes, mostly T cells. At 0.3-µM high variability was observed, with effects elicited by N-dealkyl-alachlor, aniline metabolites and their 4-hydroxy derivatives and diethylbenzoquinone, in only half of the cases. At 0.1–0.3 µM the ratio between treated and control cells for sister chromatid exchange was always higher in exposures to diethylbenzoquinoneimine than to dimethyl- and ethylmethylbenzoquinoneimines. The study showed that all the compounds assessed were toxic to lymphocytes and provided the first evidence that metabolites of chloroacetanilide herbicides were genotoxic to humans and could significantly affect the immune system (Hill et al. 1997). Glyphosate metabolites were also reported to have cyto- and hematotoxicity in humans. Aminomethylphosphonic acid (AMPA) is the main metabolite of glyphosate. This transformation product is recognized to have similar levels of toxicity comparing to its parental compound, and human exposure was already described (Benachour and Séralini, 2009; Kwiatkowska et al. 2014). The embryonic kidney, HUVEC primary neonate umbilical cord vein and JEG3 placental cell lines were exposed to 18 different AMPA concentrations varying from 10 ppm to 10% for 24 h (Benachour and Séralini, 2009). The authors reported that AMPA exposure induced succinate dehydrogenase and adenylate kinase effects on human cells and thus mortality. AMPA exposure resulted in the destruction of the cell membrane, in all cell types. More recently, another study was performed to determine AMPA hematotoxicity in human erythrocytes (Kwiatkowska et al. 2014). The authors exposed human erythrocytes to 0.01–5 mM AMPA, during 1, 4 or 24 h and evaluated the exposure effects in haemolysis, haemoglobin oxidation, ROS formation and the erythrocytes morphology. Results showed that AMPA induced haemolysis at concentrations equal or higher than 0.05 mM and haemoglobin oxidation (≥ 0.25 mM) after 24 h of incubation. An increase in ROS production was also registered at concentrations starting from 0.25 mM. The same study also investigated the hematotoxic effects of other glyphosate metabolite: methylphosphonic acid. The results were similar to those obtained for AMPA, although at a different concentration range. Induction of haemolysis and haemoglobin oxidation occurred at concentrations ≥ 0.1 and 0.5 mM, respectively. In addition, ROS production was found at concentrations ≥ 0.5 mM (Kwiatkowska et al. 2014). Pharmaceutical metabolites are not usually expected to represent an exposure concern to humans. However, biotransformation and detoxification reactions can lead to the formation of active pharmaceutical metabolites potentially more toxic than the respective parental compounds (Celiz et al. 2009). For example, gemfibrozil 1-O-β-glucuronide, the major gemfibrozil metabolite, was found to be a more potent inhibitor of CYP2C8 than the parental compound in human liver microsomes (Ogilvie et al. 2006). Also, Ogilvie and colleagues found that gemfibrozil glucuronide, contrarily to the parental compound gemfibrozil, was found to be a CYP2C8 selective inhibitor acting in a metabolism-dependent way. To depict such differences, the authors evaluated both the parental compound and its main metabolites as inhibitors of the main drug metabolizing CYP450 enzymes (CYP1A2, 2B6, 2C8, 2C9, 2C19, 2D6 and 3A4) in human liver microsomes. Compounds inhibiting the activity of the CYP450 complex can affect the metabolism of other drugs and lead to accumulation and potential toxic effects, exerting an undesired effect in the exposed person (Ogilvie et al. 2006). In fact, the chemical reactivity of glucuronide metabolites has been linked with toxic properties. These metabolites can reach appreciable concentrations in human tissues and blood. They can also undergo hydrolysis and pH-dependent intramolecular acyl migration, irreversibly reacting with human tissues. This can cause chemical alterations leading to drug toxicity expressed by alterations in functional properties of the modified molecules or hypersensitivity and other immunotoxic reactions (Shipkova et al. 2003). Pharmaceutical endocrine disruptors have been linked to several adverse effects on human health (Safe 2000). A wide range of parental compounds have been associated with hazardous effects on human reproduction and cancer development, amongst others, and metabolites are not excluded. Estrogen metabolites are reported as possible mitogenic and genotoxic substances. Investigating blood samples collected between 1989 and 1990 in subjects taking oestrogens and in controls not taking them, Eliassen et al. (2008) found a significant positive association in women of the plasma levels of 2-hydroxyestrone, and the ratio between 2-hydroxyestrone and 16-α hydroxyestrone, with certain types of breast cancer tumours. The authors recognized, nevertheless, the need for replicating the study and increasing research about the relationship between estrogen metabolites and estrogen and progesterone receptors related to breast tumours. Morphine is a strong painkiller, which is widely prescribed worldwide. However, this opiate was described to be potentially toxic to humans, not only the parental compound but also its metabolites’ morphine-3-glucoronide and morphine-9-glucoronide (Dozio et al. 2022). In a recent study, Dozio and colleagues performed a deep proteomic study in human astrocytes to investigate the role of central nervous system glial cells in the mechanisms originating the side effects of morphine administration in humans. For that, they exposed astrocytes during 12, 24, 48 and 96 h to 1-, 10- and 100-μM morphine, morphine-3-glucoronide and morphine-6-glucoronide. The proteomic analysis showed the 96-h exposure to morphine-3-glucoronide lead to dysregulation of biological pathways linked with extracellular matrix organization, antigen presentation, cell adhesion and glutamate homeostasis. For morphine-6-glucoronide (12-24-h exposure), increased levels of proteins involved in cell adhesion and decreased levels of extracellular matrix were observed. Aquatic Biota Transformation Products of Pesticides Knowledge about toxic effects caused by pesticide transformation products is still sparse, compared to parental compounds. Studies available in the scientific literature are presented in Table 2.Table 2 Ecotoxicological studies about the effects of pesticide transformation products on aquatic species Transformation product [parental compound] Species Concentrations Exposure duration Endpoints Effects Reference o-p' DDT [DDT] (Donohoe and Curtis, 1996) Oncorhynchus mykiss 0, 0.1, 1, 5, 10 and 30 mg/kg 42 days Determination of vitellogenin levels Increased levels of vitellogenin in plasma and interaction with hepatic estrogenic binding sites in vivo o-p' DDE [DDT] (Donohoe and Curtis, 1996) Oncorhynchus mykiss 0, 0.1, 1, 5, 10 and 30 mg/kg 42 days Determination of vitellogenin levels Increased levels of vitellogenin in plasma and interaction with hepatic estrogenic binding sites in vivo p-p' DDE [DDT] (Donohoe and Curtis, 1996) Oncorhynchus mykiss 0, 0.1, 1, 5, 10 and 30 mg/kg 42 days Determination of vitellogenin levels No differences found in vitellogenin levels, relative to controls DDD [DDT] (Lotufo et al. 2000) Hyalella azteca 0.095, 0.178, 0.366, 0.692 and 1.381 µg/L 10 days Mortality and lethal residues in tissues DDD was less lethal than the parental compound (DDT) but its lethality was higher than that of the control at > 0.69 µg/L Diporeia spp. 0.944, 2.791, 7.420 and 17.056 µg/L 28 days No significant effects found DDE [DDT] (Lotufo et al. 2000) Hyalella azteca 1.117, 2.258, 4.947, 8.208 and 22.021 µg/L 10 days Mortality and lethal residues in tissues DDE was less lethal than the parental compound (DDT) but its lethality was higher than that of the control at > 2.258 µg/L Diporeia spp. 2.293, 4.726, 9.141 and 20.194 µg/L 28 days No significant effects found o-p' DDE [DDT] (Davis et al. 2009) Oreochromis mossambicus 5 µg/g 35 days Determination of Vitellogenin levels and hormone/insulin-like growth factor i-axis Increase in plasma levels of insulin growth factor 100 µg/g 5 days Increase in expression of both vitellogenin A and B, estrogen receptors α and β and also in insulin growth factor 3–4-dichloroaniline [diuron] (Scheil et al. 2009) Danio rerio 0.005, 0.01, 0.1 0.25, 0.5 and 1 mg/L 8 and 11 days Mortality and locomotor activity Locomotor activity and mortality were impaired at ≥ 0.5 mg/l 0.05, 0.1, 0.15, 0.2 and 0.2 5 mg/L 168 h Hsp70 levels A significant increase in relation to control was found at 0.25 mg/L 0.5, 0.7, 1, 1.5 and 2 mg/L 11 days Embryonic and larval development By-product caused larvae deformations at ≥ 0.25 mg/l 3–4-dichloroaniline [diuron] (Felicio et al. 2018) Oreochromis niloticus 40 and 200 ng/L 7 days Antioxidant and biotransformation biomarkers By-product caused significant alterations in antioxidant and biotransformation biomarkers, with ethoxyresorufin-O-deethylase (EROD) activity showing a dose-dependent response Deethylatrazine [atrazine] (Ralston-Hooper et al. 2009) Hyalella azteca 550, 1000, 2500, 5000, 10,000, 15000 µg/L 96 h, 21 and 42 days Mortality and sex ratio LC50 values were 5100 µg/L at 96 h and higher than 3000 µg/L at 21 days; no change in the sex ratio was found Diporeia spp. 0.03, 0.3, 3, 30, 300, 3000 µg/L 96 h, 21 and 42 days Mortality and sex ratio LC50 values were 7200 µg/L at 96 h and higher than 3000 µg/L at 21 days; no change in the sex ratio was found Pseudokirchneriella subcapitata No reported 96 h Growth inhibition Growth inhibition occurred at concentrations > 2000 µg/L Deisopropylatrazine [atrazine] (Ralston-Hooper et al. 2009) Hyalella azteca 550, 1000, 2500, 5000, 10,000, 15000 µg/L 96 h, 21 and 42 days Mortality and sex ratio LC50 values were > 3000 µg/L at 96 h and 330 µg/L at 21 days; no change in the sex ratio was found Diporeia spp. 0.03, 0.3, 3, 30, 300, 3000 µg/L 96 h, 21 and 42 days Mortality and sex ratio LC50 values were > 3000 µg/L at 96 h and 300ug/L at 21 days; no change in the sex ratio was found Pseudokirchneriella subcapitata No reported 96 h Growth inhibition Growth inhibition occurred for concentrations higher than 3000 µg/L Therbuthylazine-2-hydroxy [therbuthylazine] (Koutnik et al. 2017) Procambarus fallax f. virginalis 0.75, 75, 375 and 750 µg/L 62 days Mortality, growth, oxidative balance, antioxidant defences, ontogeny and histology Lower weight at 75 µg/L; delayed ontogenic development and lowered antioxidant defences in exposed animals Desethyl-terbuthylazine [therbuthylazine] (Velisek et al. 2016) Cyprinus carpio 1.80, 180, 900 and 1800 µg/L 7, 14, 20, 27 and 31 days Growth, LC50, histology, oxidative stress, mortality LC50 of 441.6 μg/L at 31 days; lower weight and length in fish exposed to 1800 μg/L for 7 days and 900 μg/L for 20 days; delayed ontogenetic development at > 1.8 µg/L; decreased antioxidant enzyme activity in all concentrations AMPA [glyphosate] (Guilherme et al. 2014) Anguilla anguilla 11.8 and 23. µg/L 1 and 3 days DNA and chromosome damage Significant genotoxic effect in relation to control group Fiprunil sulphide and sulfone [fipronil] (Weston and Lydy, 2014) 14 macroinvertebrate species 4 − 7 concentration steps separated by a factor of 2 48 and 96 h Mortality and ability to swim, cling or crawl, depending on the species Mean 96-h EC50 of 7 − 10 ng/L Fiprunil sulphide [fipronil] (Gong et al. 2021) Danio rerio 0.1 to 10 mg/L 72 h Mortality and oxidative stress LC50 = 0.36 mg/L. Significant decreased of SOD activity at 5 mg/L Chlorella pyrenoidosa Algae growth inhibition rate; content of pigment EC50: 0.10 mg/L; chlorophyll content significantly decreased in dose–response relationship; Fiprunil sulfone [fipronil] (Gong et al. 2021) Danio rerio 0.1 to 10 mg/L 72 h Mortality and oxidative stress LC50 = 0.21 mg/L. Significant decreased of SOD activity at 5 mg/L Chlorella pyrenoidosa Algae growth inhibition rate; content of pigment EC50: 0.13 mg/L; chlorophyll content significantly decreased in dose–response relationship; Fiprunil desulfinyl [fipronil] (Gong et al. 2021) Danio rerio 0.1 to 10 mg/L 72 h Mortality and oxidative stress LC50 = 1.13 mg/L. Significant decreased of SOD activity at 5 mg/L Chlorella pyrenoidosa Algae growth inhibition rate; content of pigment EC50: 0.43 mg/L; chlorophyll content significantly decreased in dose–- response relationship; Metolachlor OXA [metolachlor] (Velisek et al. 2018) Procambarus fallax f. virginalis 4.2, 42 and 420 µg/L 45 days Growth rate, behaviour, oxidative stress, histology and mortality Decreased growth and activity of antioxidant enzymes in all tested concentrations; delayed ontogenetic development and lower levels of reduced glutathione and lipid peroxidation Metolachlor OXA [metolachlor] (Rozmánková et al. 2020) Danio rerio 1, 30, 100 and 300 μg/L (single exposure); 1 and 30 μg/L (mixture) 120 h Mortality, hatching success, embryonic malformations, locomotion, spontaneous movements, heartbeat and gene expression Increased craniofacial, non-inflated gas bladder and yolk sac malformations at 100 μg/L or higher. Induction of p53 gene at 100 μg/L Metolachlor ESA [metolachlor] (Rozmánková et al. 2020) Danio rerio 1, 30, 100 and 300 μg/L (single exposure); 1 and 30 μg/L (mixture) 120 h Mortality, hatching success, embryonic malformations, locomotion, spontaneous movements, heartbeat and gene expression Increased craniofacial, non-inflated gas bladder and yolk sac malformations at 100 μg/L or higher. Induction of p53 gene at 100 μg/L. Induction of p53 and thyroid system regulation (dio2, thra, thrb) at 30 and 1 μg/L, respectively 3-trifluoromethyl-4-aminophenol [3-trifluoromethyl-4-nitrophenol] Huerta et al. 2020) Petromyzon marinus 0, 5, 50 and 200 µM Undefined exposure time Respiratory control ratio, mitochondrial oxygen consumption and mitochondrial transmembrane potential No significant effects found 4-nitro-3-methyl-phenol [3-trifluoromethyl-4-nitrophenol] (Huerta et al. 2020) Petromyzon marinus 0, 5, 50 and 200 µM Undefined exposure time Respiratory control ratio, mitochondrial oxygen consumption and mitochondrial transmembrane potential No significant effects found 4-amino-3-methylphenol [3-trifluoromethyl-4-nitrophenol] (Huerta et al. 2020) Petromyzon marinus 0, 5, 50 and 200 µM Undefined exposure time Respiratory control ratio, mitochondrial oxygen consumption and mitochondrial transmembrane potential Decreased respiratory control ratio at 50 µM; decreased oxygen consumption at 200 µM 4-nitroso-3-methyl-phenol [3-trifluoromethyl-4-nitrophenol] (Huerta et al. 2020) Petromyzon marinus 0, 5, 50 and 200 µM Undefined exposure time Respiratory control ratio, mitochondrial oxygen consumption and mitochondrial transmembrane potential No significant effects found 3-phenoxybenzyl alcohol [permethrin] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Moderately toxic (LC50 = 1.93 mg/L) Benzenesulfonamide [asulam] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Non-toxic (LC50 > 100 mg/L) benzimidazol [carbendazim] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Slightly toxic (LC50 = 66.19 mg/L) cyanoacetamide [DBNPA] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Slightly toxic (LC50 = 68 mg/L) cis-2,6-dimethylmorpholine [fenpropimorph] (Hernández-Moreno et al. 2022) Oncorhynchus mykiss 0.78, 3.15, 12.5, 50 and 100 mg/L 96 h Mortality Non-toxic (LC50 > 100 mg/L) ethiprole sulfone [ethiprole] (Gao et al. 2021) Danio rerio 100, 300, 800, 2000, 5000 μg/L 4 days Mortality; oxidative stress; development LC50 value was 1750 μg/L; induction of antioxidant enzymes and the developmental anomalies at 100 μg/L ethiprole sulphide [ethiprole] (Gao et al. 2021) Danio rerio 100, 110, 120, 150, 180 μg/L 4 days Mortality; oxidative stress; development LC50 value was 111 μg/L; induction of antioxidant enzymes and the developmental anomalies at 10 μg/L or higher rac-ethiprole amide [ethiprole] (Gao et al. 2021) Danio rerio 100, 500, 2500, 10,000, 50,000 μg/L 4 days Mortality; oxidative stress; development LC50 > 50,000 μg/L ethiprole sulfone amide [ethiprole] (Gao et al. 2021) Danio rerio 100, 500, 2500, 10,000, 50000 μg/L 4 days Mortality; oxidative stress; development LC50 > 50,000 μg/L desethylsulfinyl ethiprole [ethiprole] (Gao et al. 2021) Danio rerio 500, 800, 1500, 2500, 5000 μg/L 4 days Mortality; oxidative stress; development LC50 = 1728 μg/L One of the most controversial pesticides is DDT, which was reported to cause health issues to humans and living organisms in general. Moreover, studies are available in the literature linking exposure to DDT metabolites to negative effects on the health of aquatic organisms. Donohoe and Curtis (1996) injected juvenile rainbow trout with o,p’-DDT, o,p’-DDE or p,p’-DDE with doses ranging from 5 to 30 mg/kg at 0, 14 and 28 days and sampling was done at 14 and/or 42 days. They reported that o,p’-DDT and o,p’-DDE had estrogenic activity, because of the elevated plasma vitellogenin levels they can elicit in vivo and their interaction with hepatic estrogenic binding sites (Donohoe and Curtis, 1996). A study conducted in freshwater amphipods (Hyalella azteca and Diporeia spp.) reported that the metabolites DDD and DDE are less lethal than DDT (Lotufo et al. 2000). Hyalella azteca and Diporeia spp. were exposed to a wide range of concentrations of DDD for 10 days and DDT and DDE for 28 days. Besides mortality, median lethal residue (LR50), mean effect concentration (EC50) and mean effect residue (ER50) in tissues were also assessed. Although metabolites were less lethal, mortality of H. azteca was significantly higher in DDD and DDE treatments than in the control at 0.692 µg/L and 2.258 µg/L, respectively (Lotufo et al. 2000). This raises high concern, once concentrations of DDD in this range have already been reported in freshwater ecosystems. The endocrine-disrupting activity of o,p'-DDE was also evaluated more recently (Davis et al. 2009). In this study, the authors investigated the effects of this metabolite and other compounds on the expression of the vitellogenin gene from the tilapia Oreochromis mossambicus and the growth hormone insulin-like growth factor-I axis. Injection of 100 µg/g o,p'-DDE in fish increased the expression of vitellogenin A and B, as well as the transcription of estrogen receptors α and β and the expression of the putative somatolactin receptor and insulin-like growth factor (Davis et al. 2009). This once again reinforces the potential endocrine disruption that DDT metabolites may cause in freshwater fish. As previously mentioned, metabolites of triazine herbicides are amongst the most frequently found in freshwater systems. Moreover, there is evidence in the literature linking these substances to negative effects on living organisms. The main degradation product of diuron is 3,4-dichloroaniline for which the toxic potential towards freshwater organisms is described in the literature. In zebrafish, a sub-chronic exposure (11 days) to this metabolite caused deformations at ≥ 0.25 mg/l, whilst locomotor activity and mortality were impaired at ≥ 0.5 mg/l (Scheil et al. 2009). A recent work investigated the effects of 3,4-dichloroaniline on biotransformation enzymes and the oxidative stress response in the liver and gills of the Nile tilapia (Oreochromis niloticus) (Felicio et al. 2018). The authors found that in fish exposed for seven days to 40 and 200 ng/L the levels of several biotransformation and antioxidant enzymes were altered often in a non-monotonic response, except for ethoxyresorufin-O-deethylase (EROD) activity that exhibited a dose-dependent increase. Moreover, the multixenobiotic resistance (MXR) activity and the activity of glutathione S-transferase (GST) enzymes were decreased in gills after exposure to 3–4-dichloroaniline. Because the MXR mechanism is crucial for the protection of aquatic organisms against xenobiotics aggression (Ferreira et al. 2014), this suggests that exposure to this metabolite is endangering the health of fish and the contaminated aquatic systems. A reduction in this mechanism can lead to higher susceptibility of animals to xenobiotics by impairing homeostatic processes. The acute and chronic toxicity of deethylatrazine and deisopropylatrazine, metabolites of atrazine, were investigated in two amphipod species and in the microalgae Pseudokirchneriella subcapitata (Ralston-Hooper et al. 2009). Hyalella azteca and Diporeia spp. were exposed to concentrations ranging from 0.55 to 15 mg/L for 96 h and from 0.03 to 3000 µg/L for 21 days. Results showed the median lethal concentrations (LC50) and median growth inhibition concentration (IC50) for algae were ≥ 1.5 mg/L, i.e. higher than the levels found in the environment (Ralston-Hooper et al. 2009). In a recent study, marbled crayfish (Procambarus fallax f. virginalis) were exposed for 62 days to four concentrations of terbuthylazine-2-hydroxy: 0.75 μg/L (environmentally relevant), 75, 375 and 750 μg/L (Koutnik et al. 2017). Antioxidant defences, oxidative balance, histology, early ontogeny, growth and mortality were the parameters assessed to depict possible effects of this metabolite. Concentrations over 75 μg/L caused lower weight compared to the control group. The outcome of the study showed that terbuthylazine-2-hydroxy delayed ontogenetic development. Also, levels of thiobarbituric acid and antioxidant enzymes were significantly (p < 0.01) lower in groups exposed to the metabolite. This shows the potential danger of this metabolite to freshwater species, although the alterations found occurred in the groups exposed to non-environmental concentrations (Koutnik et al. 2017). The toxicity of terbuthylazine-desethyl, another metabolite of triazine herbicides, was assessed in the early stages of development of the common carp (Cyprinus carpio) (Velisek et al. 2016). Carp embryos were exposed to 1.80 μg/L (environmentally relevant), 180 μg/L, 900 μg/L and 1800 μg/L and samples were collected on days 7, 14, 20, 27 and 31. The 31d LC50 of terbuthylazine-desethyl was estimated to be 441.6 μg/L. Animals also exhibited lower weight and length at 7 (1800 μg/L) and 20 (900 μg/L) days of exposure. Terbuthylazine-desethyl at non-environmental concentrations also delayed the ontogenetic development, in relation to control. However, antioxidant enzyme activity was significantly lower in all test concentrations, including the environmentally relevant one, indicating that contamination by this metabolite should be compromising feral aquatic populations. The main metabolite of glyphosate, AMPA is one of the most controversial pesticides nowadays, due to its potential hazard to wildlife and human populations. Moreover, AMPA by itself was reported as hazardous to Anguilla anguilla by Guilherme et al. (2014). The eels were exposed for 1 and 3 days to environmentally relevant concentrations (11.8 and 23.6 μg/L) and genotoxicity was investigated by assessing damage to DNA through the Comet assay and erythrocytic nuclear abnormalities. These results showed a genotoxic effect of AMPA at concentrations already found in aquatic systems. About organophosphates, a recent study was conducted with the parasitic sea lamprey (Petromyzon marinus) to address possible effects on cardiac mitochondrial bioenergetics of the lampricide 3-trifluoromethyl-4-nitrophenol and its metabolite 3-trifluoromethyl-4-aminophenol, as well as 4-nitro-3-methyl-phenol (Huerta et al. 2020). The latter has a similar molecular structure and is a known transformation product of fenitrothion and its metabolites 4-amino-3-methylphenol and 4-nitroso-3-methyl-phenol. Mitochondria were extracted from the hearts of animals captured on the great lakes and incubated with 0, 5 and 50 µM of the test compounds to assess the respiratory control ratio and mitochondrial oxygen consumption or with 0, 5, 50 and 200 µM to assess the mitochondrial transmembrane potential. Results showed that 4-amino-3-methylphenol significantly lowered the respiratory control ratio (88% at 50 µM) and oxygen consumption by 64% (at 200 µM and with the addition of high concentrations of ADP) and by 45% (at 200 µM and addition of substrate for complex II). At last, for mitochondrial transmembrane potential, none of the tested transformation products caused significant alterations. Fipronil is a phenylpyrazole insecticide with crescent use in urban areas. The toxicity of its sulphide and sulfone metabolites was not recognized until 2014 when Weston and Ludy carried out a study determining EC50 values for 14 macroinvertebrate species. Results indicated a mean 96 h EC50 of 7−10 ng/L for fipronil metabolites in Chironomus dilutus (Weston and Lydy 2014). The same study also reported that creeks receiving urban stormwater run-off in California contained metabolite concentrations twice the EC50 found for C. dilutus and approximately one-third of the EC50 found for other aquatic macroinvertebrates (Weston and Lydy 2014). A recent study evaluated the toxicity of different fipronil metabolites: fipronil sulphide, fipronil sulphone and fipronil desulfinyl (Gong et al. 2021). In this work, the authors analysed the effects of 72-h exposure to these metabolites at concentrations ranging from 0.1 to 10 mg/L on zebrafish embryos and the green algae Chlorella pyrenoidosa. In zebrafish, LC50 values of 0.36, 0.31 and 1.13 mg/L were found for fipronil sulphide, sulfone and desulfinyl, respectively. Moreover, at 5 mg/L all metabolites significantly increased SOD activity, in relation to control. In C. pyrenoidosa growth inhibition, EC50 values of 0.10, 0.13 and 0.43 mg/L were found for fipronil sulphide, sulfone and desulfinyl, respectively. The metabolites investigated also caused a significant decrease in chlorophyll content, in relation to control, in a dose–response manner (Gong et al. 2021). Metabolites of chloroacetanilide herbicides are highly prevalent in aquatic ecosystems, mainly in oxalinic and endosulfonic acid forms. Metolachlor OXA was reported to negatively affect the early life stages of marbled crayfish (Velisek et al. 2018). Animals were exposed for 45 days to 4.2 μg/L (environmentally relevant), 42 μg/L and 420 μg/L and several endpoints were assessed. Metolachlor OXA caused significantly lower growth and decreased activity of antioxidant enzymes at all tested concentrations. The highest tested concentrations delayed ontogenetic development and decreased the levels of reduced glutathione and lipid peroxidation (Velisek et al. 2018). More recently, a study was performed to evaluate the impacts of single and combined exposure of metolachlor and its metabolites metolachlor ESA and metolachlor OXA on zebrafish embryos (Rozmánková et al. 2020). In this study, zebrafish embryos were exposed for 120 h to 1, 30, 100 and 300 μg/L of the single compounds or to 1 and 30 μg/L of a compound mixture and sublethal endpoints such as malformations, hatching rate, larval length, spontaneous movements, heartbeat and locomotion, as well as expression levels of eight genes linked to different critical pathways, were monitored. Increased craniofacial, non-inflated gas bladder and yolk sac malformations at 100 μg/L or higher were reported for both metabolites. For metolachlor OXA, a significant induction of p53 gene was found at 100 μg/L, compared to control, whilst for metolachlor ESA, a significant induction of p53 gene at 30 and 100 μg/L and thyroid system regulation (dio2, thra, thrb) was observed at 1 μg/L, in comparison to the control group. The disruption of the thyroid system represented a plausible danger for population maintenance, since it occurred at low environmental concentrations (Rozmánková et al. 2020). A recent study evaluated the acute toxicity of several biocide metabolites using the rainbow trout (Oncorhynchus mykiss) as a test model (Hernández-Moreno et al. 2022). The author exposed juvenile trout according to OECD TG203, for 96 h to 0.78, 3.15, 12.5, 50 and 100 mg/L of the following metabolites: 3-phenoxybenzyl alcohol, benzenesulfonamide, benzimidazole, cyanoacetamide and cis-2,6-dimethylmorpholine. The most toxic metabolite was 3-phenoxybenzyl alcohol, with an LC50 value of 1.93 mg/L, considered moderately toxic by the authors. Benzimidazole and cyanoacetamide with LC50 values of 66.19 and 68 mg/L, respectively, were reported as slightly toxic, whilst benzenesulfonamide and cis-2,6-dimethylmorpholine with LC50 values higher than 100 mg/L were considered non-toxic (Hernández-Moreno et al. 2022). Ethiprole is a non-systemic phenylpyrazole compound widely used as an insecticide. Recently, a study was performed to evaluate zebrafish embryotoxicity and effects on antioxidant enzymes (catalase, CAT and superoxide dismutase, SOD, activities) and oxidative stress (lipid peroxidation) of its main metabolites, i.e. ethiprole sulfone, ethiprole sulphide, ethiprole amide, ethiprole sulfone amide and desethylsulfinyl ethiprole (Gao et al. 2021). Results showed that only ethiprole sulfone and sulphide had effects on antioxidant defences and embryonic development. Ethiprole sulfone had an LC50 value of 1750 μg/L, induced antioxidant enzymes and increased developmental anomalies at 100 μg/L. Ethiprole sulphide had an LC50 value of 111 μg/L, induced antioxidant enzymes and increased developmental anomalies at 10 μg/L or higher. Rac-ethiprole amide and ethiprole sulfone amide had LC50 values higher than 5000 μg/L, whilst the LC50 value for desethylsulfinyl ethiprole was 1728 μg/L (Gao et al. 2021). Transformation Products of Pharmaceuticals Nowadays, one main challenge to the scientific community is to understand the effects of these substances on non-target organisms. There are, already, several reports about this topic. However, knowledge about the toxic effects caused by pharmaceutical transformation products is still scarce. A summary of the works found in the literature is shown in Table 3.Table 3 Ecotoxicological studies about the effects of pharmaceutical transformation products on freshwater species Transformation product [parental compound] Species Concentrations Exposure duration Endpoints Effects Reference Prednisone, dexamethasone and their undisclosed photodegradation products [prednisone, dexamethasone] (Della Greca et al. 2004) Brachionus calyciflorus 5 different test concentrations without known value. Results are reported as median effective concentrations in ppm 24 h Mortality 5-prednisone and 2-dexamethasone photoderivates had lower LC50 values than parent compounds but at levels not found in environmental samples (mg/L range) Thamnocephalus platyurus 24 h Mortality All photoderivates had lower LC50 values than parental compounds (higher toxicity), but at non environmentally relevant concentrations (> 710 ppm) Daphnia magna 24 h Mortality All photoderivates had lower EC50 values than parental compounds (higher toxicity), but at non environmentally relevant concentrations (mg/L range) Pseudokirchneriella subcapitata 72 h Growth inhibition Toxic effects similar to those found for the other species, except Ceriodaphnia dubia Ceriodaphnia dubia 7 days Population growth Both the photoderivatives of prednisolone and dexamethasone showed higher toxic effects on C. dubia growth after 7 days naproxen and its undisclosed photodegradation products [naproxen](Isidori et al. 2005) Brachionus calyciflorus Concentration values are not given. All test solutions were dissolved in DMSO (0.01% v/v). 5 different concentration were tested, as well as, a negative control 24 /48 h Mortality and reproduction All photoderivates had lower LC50 values than parental compounds, but at levels not found in environmental samples (mg/L range) for acute assay. In the chronic reproduction assay only one photoderivate was less toxic than the parental compound Thamnocephalus platyurus 24 h Mortality All photoderivates had lower LC50 values than parental compounds, but at levels not found in environmental samples (mg/L range) Ceriodaphnia dubia 24 h and 7 days Mortality and reproduction All photoderivates had lower LC50 values than parental compounds, but at levels not found in environmental samples (mg/L range). For reproduction, only one photoderivate was less toxic than the parental drug Pseudokirchneriella subcapitata 96 h Growth All photoderivatives of naproxen showed higher toxic effects on P. subcapitata growth diclofenac, ketoprofen, atenolol and their photodegradation products (undisclosed) [diclofenac, ketoprofen, atenolol] (Diniz et al. 2015) Danio rerio 1 mg/L 7 days Oxidative stress Diclofenac metabolites formed through UV photolysis treatments were more toxic than their parental compounds. Activity of antioxidant enzymes and lipid peroxidation levels were higher for by-products than the parental drugs. Overall, oxidative stress response causing toxicity was observed for all pharmaceuticals and by-products norfluoxetine [fluoxetine] (Stanley et al. 2007) Pimephales promelas 1 to 250 µg/L 7 days survival and growh The authors related higher toxicity in fish exposed to s-fluoxetine, which in mammals is expected to be more potent than R-norfluoxetine Daphnia magna 10 to 1000 µg/L 21 days immobilization, reproduction and grazing rate No observed effects Norfluoxetine [fluoxetine] (Fong and Molnar, 2008) Dreissena polymorpha 100 nM to 50 µM 4 h spawning Increased spawning in zebra mussels at 1–50 µM Mytilopsis leucophaeata 100 nM to 50 µM 4 h spawning Increased spawning in zebra mussels at 1–50 µM Sphaerium striatinum 100 nM to 10 µM 4 h parturition Significant increase in parturition induced at 10 µM norfluoxetine [fluoxetine] (Rodrigues et al. 2020) Danio rerio 0.64, 3.2, 16, 80 and 400 ng/L 80 h Embryonic development, gene expression and sensorimotor responses Increase of embryonic anomalies in relation to control, mainly for pigmentation. No effects found for gene expression and sensomotory response Norfluoxetine [fluoxetine] (Atzei et al. 2021) Danio rerio 0.03 to 10 µM 5 days Embryonic development, gene expression and light/dark movement Inhibition of light/dark, zebrafish locomotory activity, mainly in dark. Responses followed a dose–response relationship norfluoxetine [fluoxetine] (Rodrigues et al. 2022) Danio rerio 400 ng/L 80 h Embryonic development and gene expression Increase in pigmentation anomalies of embryos and larvae, relative to the parental compound n-desmethylsertraline [sertraline] (Lajeunesse et al. 2011) Salvelinus fontinalis WWTP water samples (undisclosed concentrations) 3 months Tissue bioaccumulation and Na/K-ATPase activity Bioaccumulation in several tissues, including (brain and liver). Na/K-ATPase activity negatively correlated with brain bioaccumulation desmethylsertraline-exposed brain tissue o-desmethylvenlafaxine [venlafaxine] (Stropnicky, 2017) Orconectes obscurus 0, 1 and 8 µg/L 14 days Aggressive behaviour Increase in the number of attacks per minute at the highest concentration tested Procambarus clarkii 0, 1 and 8 µg/L 14 days Aggressive behaviour Increase in the number of attacks per minute at the highest concentration tested o-desmethylvenlafaxine [venlafaxine] (Atzei et al. 2021) Danio rerio 0.03 to 300 µM 5 days Embryonic development, gene expression and light/dark movement Inhibition of light/dark, zebrafish locomotory activity, mainly in dark. Responses followed a dose–response relationship Clofibric acid [clofibrate] (Nunes et al. 2008) Gambusia holbrooki 176.4, 211.6, 253.92, 304.71 and 365.65 mg/L 96 h Oxidative damage Decrease in the amount of oxidized glutathione content in the liver and gills in exposed fish n- and o-desmethyltramadol [tramadol] (Zhuo et al. 2012) Danio rerio Intraperitoneal injection of tramadol (65 mg/kg) 1 h Weight, mitochondrial changes and behaviour Detection of n- (mostly) and o-desmethyltramadol in brain tissue. Fish exposed to tramadol exhibited weight loss, abnormal behaviour and mitochondrial structural changes, possibly mediated by its by-products Oxazepam [temazepam] (Huerta et al. 2016) Pimephales promelas 0.8, 4.7 and 30.6 µg/L 28 days Behaviour and bioaccumulation Brain was the tissue with higher accumulation rates; behavioural effects detected in the novel tank diving test were observed in fish exposed to 4.7 μg/L Oxazepam [temazepam] (Fahlman et al. 2021) Perca fluviatilis 15 μg/L 14 days anti-predator behaviour Stimulation of anti-predator behaviour (decreased activity, decreased distance to conspecifics and increased littoral habitat use) Oxcarbamazepine [carbamazepine] (Desbiolles et al. 2020) Lemna minor 27 ng/L 17 days Phytometabolites Increase in nitrogen compounds. Chlorophyll index was higher in relation to control Hydra circumcincta 900 ng/L 14 days Reproduction, morphological changes and oxidative stress biomarkers Single exposure impacted the total antioxidant capacity Acridine 9-carboxylic acid [oxcarbazepine] (Desbiolles et al. 2020) Lemna minor 27 ng/L 17 days Phytometabolites Alterations of the nitrogen balance and chlorophyll indices at environmental concentrations Oseltamivir carboxylate [osetalmivir] (Chen et al. 2020) Oryzias latipes 0, 0.06, 0.3, 90 and 300 µg/L 14, 21 and 56 days median survival, growth, reproduction and hatchability Long-term parental exposure to by-products affected the embryonic development of fish hatchability at 300 µg/L and development 90 µg/L Oseltamivir ethyl ester [osetalmivir] (Chen et al. 2020) Oryzias latipes 0, 0.06, 0.3, 90 and 300 µg/L 14, 21 and 56 days median survival, growth, reproduction and hatchability Long-term parental exposure to by-products affected the embryonic development of fish hatchability at 300 µg/L and development 90 µg/L Fenofibric acid [fenofibrate] (Jung et al. 2021) Danio rerio 5, 10, 20, 30 and 40 mg/L 72 h Mortality LC50 = 53.32 mg/L Carbamazepine-10,11-epoxide [carbamazepine] (Bars et al. 2021) Danio rerio 250 µg/L 120 h embryonic development Delay in swim bladder inflation at 120hpf 5-(4-hydroxyphenyl)-5-phenylhydantoin [phenytoin] (Bars et al. 2021) Danio rerio 250 µg/L 120 h embryonic development No effects found As mentioned above, metabolites can be formed during wastewater treatment in WWTPs. In fact, this situation is reported for photodegradation products of both prednisone and dexamethasone (DellaGreca et al. 2004). In this study, photoproducts of both pharmaceuticals were isolated, from an initial solution of 100 mL of both compounds mixed with 500 mL of water and their toxicity to different species was evaluated: the rotifer Brachionus calyciflorus and the crustaceans Thamnocephalus platyurus and Daphnia magna for acute toxicity and the microalgae Pseudokirchneriella subcapitata and the crustacean Ceriodaphnia dubia for chronic toxicity. Acute assays lasted for 24 h and were based on mortality (LC50). In chronic assays, growth inhibition was the endpoint assessed for algae (72-h duration) and population growth was the endpoint for C. dubia (7-day duration). Some photodegradation products of prednisone and dexamethasone were found to be more toxic than the parental compounds. However, the LC50 values obtained by the authors were considerably higher than the concentrations generally found in surface waters. The chronic exposures decreased the population growth in C. dubia (DellaGreca et al. 2004). A similar study was conducted for the non-steroidal anti-inflammatory drug naproxen and its photodegradation products (Isidori et al. 2005). In this work, acute toxicity tests were conducted with B. calyciflorus, T. platyurus and C. dubia. Chronic toxicity was assessed (reproduction and/or growth) in B. calyciflorus, C. dubia and the microalgae P. subcapitata. Results showed that photodegradation products were more acutely toxic than the parental compound, although at levels (mg/L range) well above those found in freshwater systems. Chronic exposure reduced the population growth in C. dubia at low concentrations (μg/L) for some photoproducts (Isidori et al. 2005). This situation warns of the need to improve treatment methodologies, for better removal of both the parental compounds and their transformation products. A more recent study also reported that diclofenac metabolites formed through UV photolysis treatments were more toxic than their parental compound (Diniz et al. 2015) (Table 2). Lienert and colleagues (2007) developed a study where the ecotoxicological risk of 42 pharmaceuticals and their metabolites was evaluated. In the study, both parental compounds and their respective metabolites were treated as a mixture of toxicants of similar action. When relevant data were not available in the literature, the authors estimated them from quantitative structure–activity relationships (QSAR). Moreover, from their known pharmaceutical information, they figured out the removal efficiency of these contaminants from urine. The results of this evaluation showed that mixtures of ibuprofen and its metabolites could represent an ecotoxicological risk for aquatic organisms. Likewise, acetylsalicylic acid, bezafibrate, carbamazepine, diclofenac, fenofibrate and paracetamol in a mixture with their respective metabolites could be of potential risk for aquatic organisms, however, to a lesser extent than ibuprofen. In Table S2, ibuprofen metabolites detected in environmental samples reach concentrations > 120 000 ng/l that, together with the results of Lienert et al. (2007), suggests that this contamination is jeopardizing affected aquatic ecosystems and their populations. Whilst QSAR models have some limitations that may generate not fully accurate data, the information presented by those authors established a relevant basis for highly needed subsequent research and risk assessment studies. Norfluoxetine, the main fluoxetine metabolite, was reported to cause enantiospecific sublethal effects in Pimephales promelas and Daphnia magna (Stanley et al. 2007). In this study, P. promelas juveniles were exposed for seven days to 1, 10, 50, 100 and 250 µg/L of R-, rac- and S-fluoxetine. The enantiomer S-fluoxetine showed higher toxicity to growth, survival and feeding rate. The authors related their results to the fact that S-norfluoxetine is more potent to mammals than R-fluoxetine. But this pattern was not found for D. magna. For this microcrustacean, a 21-day toxicity test was performed to determine immobilization, reproduction and grazing rate. Less than 24-hpf individuals were exposed to 10, 50, 100, 250, 500 and 1000 µg/L of R-, rac- and S-fluoxetine. The results obtained were similar for the three compounds, and the taxa differences were attributed to the higher homology between fish and mammals than between crustaceans and mammals. Norfluoxetine was also reported to induce spawning and parturition in bivalves (Fong and Molnar 2008). The authors exposed zebra mussels to 100 nM–50 µM, dark false mussels to 100 nM–50 µM and finger-nail clams to 100 nM–10 µM. Norfluoxetine increased spawning in both zebra mussels and dark false mussels, relative to the respective controls, at concentrations in the range of 1–50 µM. In finger-nail clams, norfluoxetine induced significant parturition only at 10 µM, relative to controls. Recently, Rodrigues and colleagues (2022) found that norfluoxetine could affect the embryonic development of zebrafish larvae. In the study, newly hatched embryos were exposed for 80hpf to norfluoxetine (0.0014 µM) and fluoxetine (0.0015 µM). Larvae exposed to norfluoxetine showed an increased frequency of pigmentation anomalies, in relation to the parental compound (Rodrigues et al. 2022). Still concerning the SSRI (selective serotonin reuptake inhibitors) type of depressants, the primary metabolite of sertraline, n-desmethylsertraline, was found to affect Na/K-ATPase activity in the trout brain (Lajeunesse et al. 2011). The authors studied the distribution of selected SSRI in several tissues of brook trout, as well as the Na/K-dependent ATPase pump activity in the brain. Fish were exposed for 3 months to a WWTP-treated effluent (primary treatment) before and after ozonation. The metabolite n-desmethylsertraline was one of the main substances found in various tissues. Also, Na/K-ATPase activity was negatively correlated with the accumulation of n-desmethylsertraline in the brain. Within the group of serotonin and norepinephrine reuptake inhibitors (SNRI), o-desmethylvenlafaxine (the active metabolite of venlafaxine) was implicated in behavioural changes of freshwater organisms (Stropnicky, 2017). The author exposed two species of crayfish, Orconectes obscurus and Procambarus clarkii to 0, 1 or 8 μg/L of o-desmethylvenlafaxine. The aggression behaviour of the crayfish, measured by the number of attacks per minute of exposed animals, was the endpoint assessed. An increase in the number of attacks was found for both species at 8 µg/L (Stropnicky, 2017). A more recent study related o-desmethylvenlafaxine exposure to behavioural changes in freshwater species (Atzei et al. 2021). The authors exposed zebrafish embryos to this metabolite in a concentration range of 0.03–300 µM, for 5 days. Embryonic development was monitored and a light/dark behavioural assay was performed. No significant developmental anomalies were elicited by o-desmethylvenlafaxine. However, a dose–response inhibition on locomotory function, mainly under dark conditions, was found (Atzei et al. 2021). Clofibric acid, a metabolite of clofibrate, is another metabolite with reported negative effects on fish species. This compound caused modifications of biomarkers related to antioxidant defences and oxidative stress in Gambusia holbrooki (Nunes et al. 2008). In their work, the authors exposed the fish for 96 h to 176.34, 211.60, 253.92, 304.71 and 365.65 mg/L of clofibric acid. This metabolite caused a decrease in the activity of several antioxidant enzymes and in particular the levels of oxidized glutathione, in both the liver and gills. The effects of chronic tramadol exposure were studied in the zebrafish brain (Zhuo et al. 2012). Following intramuscular injections (25 or 65 mg/kg), both n- and o-desmethyltramadol were detected in brain tissue, mainly n-desmethyltramadol. This is important, since fish chronically exposed to tramadol exhibited weight loss, abnormal behaviour and mitochondrial structural changes. Considering that the two metabolites were present in the brain tissue, it may be possible that both can exert their effects on the exposed animals. Nevertheless, further studies focused on their administration and specific effects are needed to support this. Oxazepam is one of the main metabolites of diazepam, a widely used benzodiazepine that is prescribed as an anticonvulsant, amongst other functions. In a recent study, specimens of Pimephales promelas were exposed to 0.8, 4.7 and 30.6 µg/L oxazepam for 28 days and the relationship between its internal concentrations and effects on fish behaviour was investigated with two types of tests: novel tank diving test and shelter-seeking test (Huerta et al. 2016b). The authors concluded the brain was the tissue with higher accumulation rates and significant behavioural effects in the novel tank diving test were observed in fish exposed to 4.7 μg/L. Although 4.7 μg/L is a concentration higher than found in freshwater bodies, it raises concern about the effects this metabolite can exert on fish behaviour and ultimately endanger populations impacted by this substance. Another study with the same compound revealed behavioural changes on Perca fluvialis (Fahlman et al. 2021). The results showed that anti-predation behaviour was stimulated in exposed animals, characterized by decreased activity and distance to conspecifics, as well as increased littoral habitat use (Fahlman et al. 2021). Carbamazepine is one of the most used anticonvulsants worldwide. Recently, some of its transformation products were a matter of study by Desbiolles et al. (2020). Their study focused on the chronic effects of oxcarbamazepine and acridine 9-carboxylic acid, in single or combined exposure with carbamazepine, in two different models: the duckweed Lemna minor and the cnidarian Hydra circumcinta. Tested concentrations were the same for both models; 600, 27 and 900 ng/L for carbamazepine, oxcarbamazepine and acridine 9-carboxylic acid, respectively. For L. minor, exposure lasted 17 days and different phytometabolites were monitored. Exposure to the transformation products separately and in a mixture with the parental compound caused alterations of nitrogen balance, namely an increase in nitrogen compounds. The chlorophyll index was also higher in oxcarbamazepine groups than in the control. Nevertheless, the phenols index varied deeply without any specific trend or alteration relative to the control group. Hydra circumcinta individuals were exposed to the compounds for 14 days and different endpoints were assessed, such as reproduction, morphological changes and evaluation of antioxidant and oxidative stress biomarkers. The results showed that oxcarbamazepine exposure had implications in the total antioxidant capacity of H. circumcincta increasing two-fold in relation to control. Exposure to acridine 9-carboxylic acid affected all tested endpoints, except the reproduction. Combined exposure assays resulted in an increase in malformations on cnidarians and a decrease in the budding rate (Desbiolles et al. 2020). Another carbamazepine metabolite (carbamazepine-10,11-epoxide) was recently addressed for its possible effects on zebrafish embryonic development (Bars et al. 2021). The authors exposed zebrafish embryos from ~ 3 to 120hpf to a concentration of 250 µg/L of this metabolite, i.e. considerably higher than the maximum concentration found in the environment. Embryonic development was monitored through the exposure period and anomalies were registered. Results showed that swim bladder inflation was significantly delayed in carbamazepine-10,11-epoxide-exposed larvae, compared to the control (Bars et al. 2021). This is important since inflation of the swim bladder allows larvae to stay in the water column and have more chances of survival. A recent study focused on the metabolites of the well-known antiviral oseltamivir (Tamiflu) and their chronic effects on the medaka Oryzias latipes (Chen et al. 2020). Results showed that long-term parental exposure to both oseltamivir carboxylate and oseltamivir ethyl ester affected embryonic development and fish hatchability at 300 µg/L and embryonic development at 90 µg/L. Fenofibric acid, a metabolite of the anti-lipidemic agent fenofibrate, was also evaluated for its toxicity to zebrafish embryos (Jung et al. 2021). An LC50 value of 53.32 mg/L was found at 72 h, which is considerably higher than the normally occurring concentration in the environment. The Way Forward This review gives an updated perspective on freshwater contamination by pharmaceuticals and pesticide transformation products and the available information about the toxicity of these substances. Detection of pharmaceuticals and pesticides is increasing in freshwater ecosystems, and concentrations in the range of ng to μg/L have been widely reported. Moreover, this same trend is described for their metabolites and transformation products. This occurrence made this field one of the most studied by the scientific community in the last years, with a number of published works addressing the potentially hazardous effects of such previously overlooked substances. The present research identified concentrations of 190 metabolites and transformation products (92 from pesticides and 98 from pharmaceuticals) in water bodies and wastewater effluents, none of them included in monitoring programmes set to achieve the good environmental status of freshwater ecosystems. Their formation processes, environmental fate in aquatic ecosystems and effects on humans and biota, summarized in Fig. 6, are varied and a considerable cause of concern. Reported concentrations are mainly in the order of ng to μg/L. The concentration heatmap produced in this work allows us to easily spot the substances found at higher levels.Fig. 6 Overall representation of pesticides and pharmaceutical transformation products aquatic contamination and risks for human and aquatic species Although the information presented herein about the quantification of pesticides and pharmaceutical transformation products is extensive (almost 200 compounds), this may just represent the tip of the iceberg. Worldwide there are more than 1500 pesticides approved for use in agriculture and about 4000 pharmaceutical compounds approved for human consumption (aus der Beek et al. 2016; Anagnostopoulou et al. 2022). These parental compounds can have one or several transformation products, which brutally increases the potential number of these pollutants in the aquatic environment. Also, transformation products of pesticides banned for several decades now are still found in freshwater. Transformation products are in several cases more stable in the environment and consequently reach concentrations higher than their parental compounds (Schuhmann et al. 2019; Celiz et al. 2009). All these numbers and characteristics reinforce the need to increase the monitoring of these compounds in aquatic systems and evaluate their impact on human and environmental health. The toxicological information available for the transformation products identified is very little and scattered, with no strategic approach underlying data collection for risk assessment and monitoring prioritization. Concerning the risk to humans, less than twenty metabolites (of the two groups combined) were investigated in in vitro studies. Several of these were found to elicit genotoxicity and effects on biotransformation and antioxidant processes. In aquatic organisms, only about 34% of the transformation products originating from pesticides and 14% of those originating from pharmaceuticals were evaluated for their potentially hazardous effects on biota. Most of these studies evaluated effects on only one (majority) or two trophic levels and more than half of them on vertebrates. Effects on plants and algae were rarely assessed. For pesticides, over 50% of the assessments were about acute and subacute toxicity effects, whilst for pharmaceuticals only about 20% of the assessments concerned chronic toxicity. Adding to this, for pharmaceutical metabolites various studies tested very high exposure levels, reporting effects at concentrations higher than those found in the environment. Nevertheless, for pesticide metabolites, several reports described a considerably wide range of negative effects on freshwater organisms, occurring at environmentally relevant concentrations. For pharmaceutical metabolites, different classes of drugs were proven to cause hazardous effects and jeopardize the homeostasis of freshwater species. All in all, the data presented herein clearly demonstrate that pesticide and pharmaceutical transformation products pose a threat to aquatic fauna and flora. Concerning the relative toxicity of transformation products, compared to the parental compounds, the available data prevent a clear global conclusion. In some cases, the transformation products are in fact less toxic. In other cases, some transformation products can be more active and toxic than the parental substance. Nowadays, there is increasing evidence that pesticide transformation products can be more toxic and persistent than their parental compounds (Iwafune 2018). In silico assays, performed with the ECOSAR (Ecological Structure Activity Relationships) software, which predicts the toxicity of different compounds, showed that the transformation products of several pesticides have a high toxicity potential to aquatic fauna and flora (Anagnostopoulou et al. 2022). Transformation products resulting from penoxsulam, pyrimethanil, imidacloprid, acetamiprid, thiacloprid and carbendazim were predicted to be more toxic than their parental compounds. In contrast, transformation products of fipronil present equal levels of toxicity, relative to fipronil itself (Anagnostopoulou et al. 2022). For pharmaceutical transformation products, there is a general idea that these compounds are less active and, consequently, less toxic than their parental compounds. However, there is evidence that some transformation products may be more toxic than the parental compounds. In humans, metabolites such as morphine and o-desmethyltramadol are more active than the parental compound (codeine and tramadol, respectively) (Rodieux et al. 2018). There are also reports of potential toxic effects elicited in patients, i.e. pethidine and dextroptopoxyphene (Coller et al. 2009). On the other hand, photodegradation products of prednisone, dexamethasone, naproxen, diclofenac, ketoprofen and atenolol formed in watercourses or even in WWTPs were reported to be toxic to different aquatic species at higher magnitude than their parental compounds (DellaGreca et al. 2004; Isidori et al. 2005; Diniz et al. 2015). Nonetheless, for most of the transformation products identified, the information is still scarce to draw sound conclusions. Something that is still not accounted for in most of the ecotoxicological works is the metabolism of parental substances in the test media. During exposure, parental compounds are metabolized and transformed by the exposed organisms. This is a process, influenced by media abiotic factors, which originates different transformation products. Such compounds can cause negative effects on the organisms, by themselves or in mixture with the respective parental compound. A previous study reported that fish exposed to tramadol exhibited weight loss, abnormal behaviour and structural mitochondrial changes that could be linked to the metabolites formed during the exposure, which accumulated in the animals’ brains and muscular tissue (Zhuo et al. 2012). The possibility that several negative impacts reported on aquatic species exposed to pharmaceuticals may derive not only from those compounds, but also from the mixture with their metabolites or even exclusively from the metabolites needs to be addressed soon. Overall, the results warn of the need to continue improving treatment methodologies, for better removal of transformation products, not only to avoid their discharge to the aquatic environment but also to assure a better quality for water reuse. From a toxicological viewpoint, it is also striking the lack of mechanistic information useful to improve predictive toxicology and the risk assessment of these chemicals. Most works focused on assessing classical apical endpoints employing standard testing approaches. Whilst this is always fruitful to obtain a quick grasp of the severity of a contamination scenario, more studies investigating the modes of action of these compounds are urgently needed. Also, the limited availability of reference standards for several transformation products makes it difficult to test the toxicity of these compounds to living organisms (Anagnostopoulou et al. 2022). However, this obstacle can be surpassed using in silico approaches, which reduce the need for animals and chemicals and can be valuable tools for toxicity and risk assessment. Future toxicological investigations should be based on the framework of Adverse Outcome Pathways (AOP) (Ankley et al. 2010). This concept identifies various key events and relationships between them, linking a molecular initiating event to an adverse outcome of significance to risk assessment. The adverse outcome is usually considered at the organ level or higher, preferably the ecological level. It indicates a morphological or physiological alteration occurring in an organism or its systems that elicits functional impairment or impairs its ability to compensate for chemical stress and achieve homeostasis. The AOP framework is recognized as useful to support regulatory decision-making and the prioritization of chemicals for risk assessment (Vinken et al. 2017; Perkins et al. 2019), a most important aspect for the contamination scenario described herein. Present-day high-throughput technologies (i.e. proteomic sequencing) allowing for the rapid and cost-effective generation of data should be used to identify key events and key event relationships through which the initiating event(s) will reflect on adverse outcomes to apical endpoints. Guidance documents for the development of AOPs were made available (OECD, 2013, 2018), as well as supporting databases and tools, such as the e.AOP.portal (http://aopkb.org), the AOP Wiki (http://aopwiki.org), the Effectopedia (http://effectopedia.org) and the Wikipathways (https://www.wikipathways.org/index.php/WikiPathways), the Harmonized Template 201: Intermediate effects (https://www.oecd.org/ehs/templates/harmonised-templates-intermediate-effects.htm) and the AOP Xplorer (http://datasciburgoon.github.io/aopxplorer. Collaborative networks based on resource and knowledge sharing, and rational effort application, should be made at a global level to establish and implement a structured strategy rapidly allowing to fulfil these gaps whilst avoiding unnecessary experimental redundancy (Martens et al. 2018). The present work emphasizes the need to reinforce the existing knowledge about contamination by pharmaceutical and pesticide transformation products in freshwater systems. This report compiled and analysed a significant amount of information linking exposure to transformation products to adverse outcomes in aquatic species and humans. Technological needs and knowledge gaps were identified and discussed, delineating future research steps on the topic, ultimately aiming at improving water management and monitoring programmes. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 117 KB) Acknowledgements This work was supported by project BioReset (DivRestore/0004/2020) funded by BiodivRestore 2020 ERA-NET Cofund (a joint action of Biodiversa+ and Water JPI) and Strategic Funding UIDB/04423/2020 and UIDP/04423/2020 (FCT, ERDF) and PR was supported by a FCT fellowship SFRH/BD/134518/2017. Data Availability Authors confirm that all relevant data are included in the article or its supplementary file. Declarations 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. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Anagnostopoulou K Nannou C Evgenidou E Lambropoulou D Overarching issues on relevant pesticide transformation products in the aquatic environment: A review Sci Total Environ 2022 10.1016/j.scitotenv.2021.152863 Ankley GT Bennett RS Erickson RJ Hoff DJ Hornung MW Johnson RD Mount DR Nichols JW Russom CL Schmieder PK Serrrano JA Tietge JE Villeneuve DL Adverse outcome pathways: a conceptual framework to support ecotoxicology research and risk assessment Environ Toxicol Chem 2010 29 730 741 10.1002/etc.34 20821501 Arias-Estévez M López-Periago E Martínez-Carballo E Simal-Gándara J Mejuto J-C García-Río L The mobility and degradation of pesticides in soils and the pollution of groundwater resources Agr Ecosyst Environ 2007 123 247 260 10.1016/j.agee.2007.07.011 Atzei A Jense I Zwart EP Legradi J Venhuis BJ van der Ven LTM Heusinkveld HJ Hessel EVS Developmental neurotoxicity of environmentally relevant pharmaceuticals and mixtures thereof in a zebrafish embryo behavioural test Int J Environ Res Public Health 2021 18 13 6717 10.3390/ijerph18136717 34206423 aus der Beek, T., Weber, F.A., Bergmann, A., Hickmann, S., Ebert, I., Hein, A., and Küster, A. Pharmaceuticals in the environment–Global occurrences and perspectives Environ Toxicol Chem 2016 35 823 835 10.1002/etc.3339 26666847 Baena-Nogueras RM Gonzalez-Mazo E Lara-Martin PA Degradation kinetics of pharmaceuticals and personal care products in surface waters: photolysis vs biodegradation Sci Total Environ 2017 10.1016/j.scitotenv.2017.03.015 Baran N Gourcy L Sorption and mineralization of S-metolachlor and its ionic metabolites in soils and vadose zone solids: consequences on groundwater quality in an alluvial aquifer (Ain Plain, France) J Contam Hydrol 2013 154 20 28 10.1016/j.jconhyd.2013.07.009 24055953 Bars C Hoyberghs J Valenzuela A Buyssens L Ayuso M Van Ginneken C Labro AJ Foubert K Van Cruchten SJ Developmental toxicity and biotransformation of two anti-epileptics in zebrafish embryos and early larvae Int J Mol Sci 2021 22 23 12696 10.3390/ijms222312696 34884510 Bavumiragira JP Ge J Yin H Fate and transport of pharmaceuticals in water systems: A processes review Sci Total Environ 2022 10.1016/j.scitotenv.2022.153635 Benachour N Séralini GE Glyphosate formulations induce apoptosis and necrosis in human umbilical, embryonic, and placental cells Chem Res Toxicol 2009 22 1 97 105 10.1021/tx800218n 19105591 Boix C Ibanez M Sancho JV Parsons JR Voogt P Hernandez F Biotransformation of pharmaceuticals in surface water and during waste water treatment: Identification and occurrence of transformation products J Hazard Mater 2016 302 175 187 10.1016/j.jhazmat.2015.09.053 26476304 Brown AK Challis JK Wong CS Hanson ML Selective serotonin reuptake inhibitors and beta-blocker transformation products may not pose a significant risk of toxicity to aquatic organisms in wastewater effluent-dominated receiving waters Integr Environ Assess Manag 2015 11 618 639 10.1002/ieam.1637 25820351 Bukowska B Effects of 2,4-D and its metabolite 2,4-dichlorophenol on antioxidant enzymes and level of glutathione in human erythrocytes Comp Biochem Physiol C: Toxicol Pharmacol 2003 135 435 441 10.1016/S1532-0456(03)00151-0 12965188 Campanale C Massarelli C Losacco D Bisaccia D Triozzi M Uricchio VF The monitoring of pesticides in water matrices and the analytical criticalities: A review TrAC Trend Anal Chem 2021 10.1016/j.trac.2021.116423 Carrara C Ptacek CJ Robertson WD Blowes DW Moncur MC Sverko E Backus S Fate of pharmaceutical and trace organic compounds in three septic system plumes, Ontario, Canada Environ Sci Technol 2008 42 2805 2811 10.1021/es070344q 18497127 Celiz MD Tso J Aga DS Pharmaceutical metabolites in the environment: Analytical challenges and ecological risks Environ Toxicol Chem 2009 28 2473 2484 10.1897/09-173.1 19663539 Chau ND Sebesvari Z Amelung W Renaud FG Pesticide pollution of multiple drinking water sources in the Mekong Delta, Vietnam: evidence from two provinces Environ Sci Pollut Res Int 2015 22 9042 9058 10.1007/s11356-014-4034-x 25572267 Chen WY Wu YT Lin HC Ieong MI Lee BH Impact of long-term parental exposure to Tamiflu metabolites on the development medaka offspring (Oryzias latipes) Environ Pollut 2020 261 114 146 10.1016/j.envpol.2020.114146 Christensen ER Li A Physical and Chemical Processes in the Aquatic Environment 2014 Hoboken Wiley & Sons Inc 255 259 Coller JK Christrup LL Somogyi AA Role of active metabolites in the use of opioids Eur J Clin Pharmacol 2009 65 121 139 10.1007/s00228-008-0570-y 18958460 Corcoran J Winter M Tyler C Pharmaceuticals in the aquatic environment: a critical review of the evidence for health effects in fish Crit Rev Toxicol 2010 40 287 304 10.3109/10408440903373590 20225984 Daffara F De Francia S Reimondo G Zaggia B Aroasio E Porpiglia F Volante M Termine A Di Carlo F Dogliotti L Angeli A Berruti A Terzolo M Prospective evaluation of mitotane toxicity in adrenocortical cancer patients treated adjuvantly Endocr Relat Cancer 2008 15 4 1043 1053 10.1677/ERC-08-0103 18824557 Daughton CG Pharmaceuticals and the Environment (PiE): Evolution and impact of the published literature revealed by bibliometric analysis Sci Total Environ 2016 562 391 426 10.1016/j.scitotenv.2016.03.109 27104492 Daughton CG Ternes TA Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Health Perspectives 1999 107 907 938 10.1289/ehp.99107s6907 Davis LK Visitacion N Riley LG Hiramatsu N Sullivan CV Hirano T Grau IE Effects of o, p'-DDE, heptachlor, and 17beta-estradiol on vitellogenin gene expression and the growth hormone/insulin-like growth factor-I axis in the tilapia, Oreochromis mossambicus Comp Biochem Physiol C: Toxicol Pharmacol 2009 149 507 514 10.1016/j.cbpc.2008.11.007 19101654 Davis, F. R. (2014). Banned: a history of pesticides and the science of toxicology. Yale University Press, London. https://www.jstor.org/stable/j.ctt13x1tbs. DellaGreca M Fiorentino A Isidori M Lavorgna M Previtera L Rubino M Temussi F Toxicity of prednisolone, dexamethasone and their photochemical derivatives on aquatic organisms Chemosphere 2004 54 629 637 10.1016/j.chemosphere.2003.09.008 14599508 Deribe E Rosseland BO Borgstrøm R Salbu B Gebremariam Z Dadebo E Skipperud L Eklo OM Biomagnification of DDT and its metabolites in four fish species of a tropical lake Ecotoxicol Environ Saf 2013 95 10 18 10.1016/j.ecoenv.2013.03.020 23790590 Desbiolles F Moreau X de Jong L Malleret L Grandet-Marchant Q Wong-Wah-Chung P Laffont-Schwob I Advances and limits of two model species for ecotoxicological assessment of carbamazepine, two transformation products and their mixture at environmental level in freshwater Water Res 2020 169 115267 10.1016/j.watres.2019.115267 31731244 Dimpe KM Nomngongo PN Current sample preparation methodologies for analysis of emerging pollutants in different environmental matrices TrAC Trends Anal Chem 2016 82 199 207 10.1016/j.trac.2016.05.023 Diniz MS Salgado R Pereira VJ Carvalho G Oehmen A Reis MA Noronha JP Ecotoxicity of ketoprofen, diclofenac, atenolol and their photolysis transformation products in zebrafish (Danio rerio) Sci Total Environ 2015 505 282 289 10.1016/j.scitotenv.2014.09.103 25461029 Donohoe RM Curtis LR Estrogenic activity of chlordecone, o, p′-DDT and o, p′-DDE in juvenile rainbow trout: induction of vitellogenesis and interaction with hepatic estrogen binding sites Aquat Toxicol 1996 36 1–2 31 52 10.1016/S0166-445X(96)00799-0 Dozio V Daali Y Desmeules J Sanchez JC Deep proteomics and phosphoproteomics reveal novel biological pathways perturbed by morphine, morphine-3-glucuronide and morphine-6-glucuronide in human astrocytes J Neurosci Res 2022 100 1 220 236 10.1002/jnr.24731 32954564 Eliassen AH Missmer SA Tworoger SS Hankinson SE Circulating 2-hydroxy- and 16alpha-hydroxy estrone levels and risk of breast cancer among postmenopausal women Cancer Epidemiol Biomarkers Prevention 2008 17 2029 2035 10.1158/1055-9965.epi-08-0262 Elsayed OF Maillard E Vuilleumier S Millet M Imfeld G Degradation of chloroacetanilide herbicides and bacterial community composition in lab-scale wetlands Sci Total Environ 2015 520 222 231 10.1016/j.scitotenv.2015.03.061 25817759 Evgenidou EN Konstantinou IK Lambropoulou DA Occurrence and removal of transformation products of PPCPs and illicit drugs in wastewaters: a review Sci Total Environ 2015 505 905 926 10.1016/j.scitotenv.2014.10.021 25461093 Fahlman J Hellstrom G Jonsson M Fick JB Rosvall M Klaminder J Impacts of oxazepam on Perch (Perca fluviatilis) behavior: fish familiarized to lake conditions do not show predicted anti-anxiety response Environ Sci Technol 2021 55 6 3624 3633 10.1021/acs.est.0c05587 33663207 Felicio AA Freitas JS Scarin JB de Souza Ondei L Teresa FB Schlenk D de Almeida EA Isolated and mixed effects of diuron and its metabolites on biotransformation enzymes and oxidative stress response of Nile tilapia (Oreochromis niloticus) Ecotoxicol Environ Saf 2018 149 248 256 10.1016/j.ecoenv.2017.12.009 29248837 Fenner K Canonica S Wackett LP Elsner M Evaluating pesticide degradation in the environment: blind spots and emerging opportunities Science 2013 341 6147 752 758 10.1126/science.1236281 23950532 Fent K Weston A Caminada D Ecotoxicology of human pharmaceuticals Aquat Toxicol 2006 76 122 159 10.1016/j.aquatox.2005.09.009 16257063 Ferreira M Costa J Reis-Henriques MA ABC transporters in fish species: a review Front Physiol 2014 5 266 10.3389/fphys.2014.00266 25101003 Foltz RL Andrenyak DM Crouch DJ Lindon J Tranter GE Koppenaal D Forensic science, applications of mass spectrometry Encyclopedia of Spectroscopy and Spectrometry 2016 Amsterdam Elsevier Ltd 707 711 Fong PP Molnar N Norfluoxetine induces spawning and parturition in estuarine and freshwater bivalves Bull Environ Contam Toxicol 2008 81 535 538 10.1007/s00128-008-9558-7 18787749 Gao J Wang F Cui J Zhang Q Wang P Liu D Zhou Z Assessment of toxicity and environmental behavior of chiral ethiprole and its metabolites using zebrafish model J Hazard Mater 2021 414 125492 10.1016/j.jhazmat.2021.125492 33647613 Geric M Ceraj-Ceric N Gajski G Vasilic Z Capuder Z Garaj-Vrhovac V Cytogenetic status of human lymphocytes after exposure to low concentrations of p, p'-DDT, and its metabolites (p, p'-DDE, and p, p'-DDD) in vitro Chemosphere 2012 87 1288 1294 10.1016/j.chemosphere.2012.01.037 22354074 Gong W Barrett H Hu Y Han J Wang F Wang W Zhou S Qu H Application of biochar: An approach to attenuate the pollution of the chiral pesticide fipronil and its metabolites in leachate from activated sludge Process Saf Environ Prot 2021 149 936 945 10.1016/j.psep.2021.03.044 Gros M Rodriguez-Mozaz S Barcelo D Fast and comprehensive multi-residue analysis of a broad range of human and veterinary pharmaceuticals and some of their metabolites in surface and treated waters by ultra-high-performance liquid chromatography coupled to quadrupole-linear ion trap tandem mass spectrometry J Chromatogr A 2012 1248 104 121 10.1016/j.chroma.2012.05.084 22704668 Guardabassi L Wong LF D.M., and Dalsgaard, A. The effects of tertiary wastewater treatment on the prevalence of antimicrobial resistant bacteria Water Res 2002 36 1955 1964 10.1016/s0043-1354(01)00429-8 12092570 Guilherme S Santos MA Gaivão I Pacheco M DNA and chromosomal damage induced in fish (Anguilla anguilla L.) by aminomethylphosphonic acid (AMPA)—the major environmental breakdown product of glyphosate Environ Sci Pollut Res 2014 21 14 8730 8739 10.1007/s11356-014-2803-1 Guimarães L Guilhermino L Afonso MJ Marques JM Chaminé I Assessment of urban groundwater: towards integrated hydrogeological and effects-based monitoring Sustain Water Resources Manag 2019 5 217 233 10.1007/s40899-019-00301-w Gumustas M Kurbanoglu S Uslu B Ozkan SA UPLC versus HPLC on drug analysis: advantageous, applications and their validation parameters Chromatographia 2013 76 1365 1427 10.1007/s10337-013-2477-8 Hernández-Moreno D Blázquez M Navas JM Fernández-Cruz ML Fish cell lines as screening tools to predict acute toxicity to fish of biocidal active substances and their relevant environmental metabolites Aquat Toxicol 2022 242 106020 10.1016/j.aquatox.2021.106020 34844051 Hignite C Azarnoff DL Drugs and drug metabolites as environmental contaminants: chlorophenoxyisobutyrate and salicyclic acid in sewage water effluent Life Sci 1977 20 2 337 341 10.1016/0024-3205(77)90329-0 839964 Hill AB Jefferies PR Quistad GB Casida JE Dialkylquinoneimine metabolites of chloroacetanilide herbicides induce sister chromatid exchanges in cultured human lymphocytes Mutation Res/genetic Toxicol Environ Mutagenesis 1997 395 159 171 10.1016/s1383-5718(97)00163-0 Hong B Lin Q Yu S Chen Y Chen Y Chiang P Urbanization gradient of selected pharmaceuticals in surface water at a watershed scale Sci Total Environ 2018 634 448 458 10.1016/j.scitotenv.2018.03.392 29631135 Huang Z Yao YN Li W Hu B Analytical properties of electrospray ionization mass spectrometry with solid substrates and nonpolar solvents Anal Chim Acta 2019 1050 105 112 10.1016/j.aca.2018.10.064 30661577 Huerta B Margiotta-Casaluci L Rodriguez-Mozaz S Scholze M Winter MJ Barcelo D Sumpter JP Anti-anxiety drugs and fish behavior: Establishing the link between internal concentrations of oxazepam and behavioral effects Environ Toxicol Chem 2016 35 2782 2790 10.1002/etc.3448 27061599 Huerta B Chung-Davidson YW Bussy U Zhang Y Bazil JN Li W Sea lamprey cardiac mitochondrial bioenergetics after exposure to TFM and its metabolites Aquat Toxicol 2020 219 105380 10.1016/j.aquatox.2019.105380 31855722 Huntscha S Singer HP McArdell CS Frank CE Hollender J Multiresidue analysis of 88 polar organic micropollutants in ground, surface and wastewater using online mixed-bed multilayer solid-phase extraction coupled to high performance liquid chromatography-tandem mass spectrometry J Chromatogr A 2012 1268 74 83 10.1016/j.chroma.2012.10.032 23137864 Ibáñez M Bijlsma L Pitarch E López FJ Hernández F Occurrence of pharmaceutical metabolites and transformation products in the aquatic environment of the Mediterranean area Trends Environ Anal Chem 2021 10.1016/j.teac.2021.e00118 Isidori M Lavorgna M Nardelli A Parrella A Previtera L Rubino M Ecotoxicity of naproxen and its phototransformation products Sci Total Environ 2005 348 93 101 10.1016/j.scitotenv.2004.12.068 16162316 Iwafune T Studies on the behavior and ecotoxicity of pesticides and their transformation products in a river J Pesticide Sci 2018 43 4 297 304 10.1584/jpestics.J18-01 Jaffar S Ahmad S Lu Y Contribution of insect gut microbiota and their associated enzymes in insect physiology and biodegradation of pesticides Front Microbiol 2022 13 979383 10.3389/fmicb.2022.979383 36187965 Jjemba PK Excretion and ecotoxicity of pharmaceutical and personal care products in the environment Ecotoxicol Environ Saf 2006 63 113 130 10.1016/j.ecoenv.2004.11.011 16399163 Jung F Thurn M Krollik K Gao GF Hering I Eilebrecht E Emara Y Weiler M Günday-Türeli N Türeli E Parnham MJ Wacker MG Predicting the environmental emissions arising from conventional and nanotechnology-related pharmaceutical drug products Environ Res 2021 192 110219 10.1016/j.envres.2020.110219 32980299 Kaufmann A Dvorak V Crüzer C Butcher P Maden K Walker S Widmer M Schürmann A Study of high-resolution mass spectrometry technology as a replacement for tandem mass spectrometry in the field of quantitative pesticide residue analysis J AOAC Int 2012 95 2 528 548 10.5740/jaoacint.11-074 22649942 Kemper N Veterinary antibiotics in the aquatic and terrestrial environment Ecol Ind 2008 8 1 13 10.1016/j.ecolind.2007.06.002 Kezios KL Liu X Cirillo PM Cohn BA Kalantzi OI Wang Y Factor-Litvak P Dichlorodiphenyltrichloroethane (DDT), DDT metabolites and pregnancy outcomes Reprod Toxicol 2013 35 156 164 10.1016/j.reprotox.2012.10.013 23142753 Khan HK Rehman MYA Malik RN Fate and toxicity of pharmaceuticals in water environment: An insight on their occurrence in South Asia J Environ Manage 2020 10.1016/j.jenvman.2020.111030 Koutnik D Stara A Zuskova E Kouba A Velisek J The chronic effects of terbuthylazine-2-hydroxy on early life stages of marbled crayfish (Procambarus fallax f. virginalis) Pestic Biochem Physiol 2017 136 29 33 10.1016/j.pestbp.2016.08.008 28187827 Kwiatkowska M Huras B Bukowska B The effect of metabolites and impurities of glyphosate on human erythrocytes (in vitro) Pestic Biochem Physiol 2014 109 34 43 10.1016/j.pestbp.2014.01.003 24581382 La Farre M Pérez S Kantiani L Barceló D Fate and toxicity of emerging pollutants, their metabolites and transformation products in the aquatic environment TrAC, Trends Anal Chem 2008 27 11 991 1007 10.1016/j.trac.2008.09.010 Lajeunesse A Gagnon C Gagne F Louis S Cejka P Sauve S Distribution of antidepressants and their metabolites in brook trout exposed to municipal wastewaters before and after ozone treatment—evidence of biological effects Chemosphere 2011 83 564 571 10.1016/j.chemosphere.2010.12.026 21211816 Lapworth DJ Gooddy DC Source and persistence of pesticides in a semi-confined chalk aquifer of southeast England Environ Pollut 2006 144 1031 1044 10.1016/j.envpol.2005.12.055 16750876 Le Cor F Slaby S Dufour V Iuretig A Feidt C Dauchy X Banas D Occurrence of pesticides and their transformation products in headwater streams: Contamination status and effect of ponds on contaminant concentrations Sci Total Environ 2021 10.1016/j.scitotenv.2021.147715 Li Z Sobek A Radke M Fate of pharmaceuticals and their transformation products in four small European rivers receiving treated wastewater Environ Sci Technol 2016 50 11 5614 5621 10.1021/acs.est.5b06327 27152425 Lienert J Güdel K Escher BI Screening method for ecotoxicological hazard assessment of 42 pharmaceuticals considering human metabolism and excretory routes Environ Sci Technol 2007 41 4471 4478 10.1021/es0627693 17626454 Lopez-Serna R Petrovic M Barcelo D Occurrence and distribution of multi-class pharmaceuticals and their active metabolites and transformation products in the Ebro river basin (NE Spain) Sci Total Environ 2012 440 280 289 10.1016/j.scitotenv.2012.06.027 22809787 Lotufo GR Landrum PF Gedeon ML Tigue EA Herche LR Comparative toxicity and toxicokinetics of ddt and its major metabolites in freshwater amphipods Environ Toxicol Chem 2000 19 368 379 10.1002/etc.5620190217 Luo Y Guo W Ngo HH Nghiem LD Hai FI Zhang J Liang S Wang XC A review on the occurrence of micropollutants in the aquatic environment and their fate and removal during wastewater treatment Sci Total Environ 2014 473–474 619 641 10.1016/j.scitotenv.2013.12.065 Madikizela LM Nuapia YB Chimuka L Ncube S Etale A Target and suspect screening of pharmaceuticals and their transformation products in the Klip river, South Africa, using ultra-high–performance liquid chromatography-mass spectrometry Environ Toxicol Chem 2022 41 437 447 10.1002/etc.5265 34888926 Martens M Verbruggen T Nymark P Grafström R Burgoon LD Aladjov H Torres Andón F Evelo CT Willighagen EL Introducing WikiPathways as a data-source to support adverse outcome pathways for regulatory risk assessment of chemicals and nanomaterials Front Genet 2018 9 661 10.3389/fgene.2018.00661 30622555 Martín-Pozo L de Alarcón-Gómez B Rodríguez-Gómez R García-Córcoles MT Çipa M Zafra-Gómez A Analytical methods for the determination of emerging contaminants in sewage sludge samples A Review Talanta 2019 192 508 533 10.1016/j.talanta.2018.09.056 30348425 McKnight US Rasmussen JJ Kronvang B Binning PJ Bjerg PL Sources, occurrence and predicted aquatic impact of legacy and contemporary pesticides in streams Environ Pollut 2015 200 64 76 10.1016/j.envpol.2015.02.015 25697475 Mompelat S Le Bot B Thomas O Occurrence and fate of pharmaceutical products and transformation products, from resource to drinking water Environ Int 2009 35 5 803 814 10.1016/j.envint.2008.10.008 19101037 Monteiro C Miranda CJ Brito F Fonseca C Araujo AR Consumption patterns of NSAIDs in central Portugal and the role of pharmacy professionals in promoting their rational use Drugs Therapy Perspectives 2017 33 32 40 10.1007/S40267-016-0352-Z Mosekiemang TT Stander MA de Villiers A Ultra-high pressure liquid chromatography coupled to travelling wave ion mobility-time of flight mass spectrometry for the screening of pharmaceutical metabolites in wastewater samples: Application to antiretrovirals J Chromatography A 2021 10.1016/j.chroma.2021.462650 Munze R Hannemann C Orlinskiy P Gunold R Paschke A Foit K Becker J Kaske O Paulsson E Peterson M Pesticides from wastewater treatment plant effluents affect invertebrate communities Sci Total Environ 2017 10.1016/j.scitotenv.2017.03.008 Murray KE Thomas SM Bodour AA Prioritizing research for trace pollutants and emerging contaminants in the freshwater environment Environ Pollut 2010 158 3462 3471 10.1016/j.envpol.2010.08.009 20828905 Nunes B Gaio AR Carvalho F Guilhermino L Behaviour and biomarkers of oxidative stress in Gambusia holbrooki after acute exposure to widely used pharmaceuticals and a detergent Ecotoxicol Environ Saf 2008 71 2 341 354 10.1016/j.ecoenv.2007.12.006 18243307 OECD OECD series on testing and assessment number 184: guidance document on developing and assessing adverse outcome pathways ENV/JM/MONO(2013) 6 2013 Paris OECD Publishing OECD (2018), "Users' Handbook supplement to the Guidance Document for developing and assessing Adverse Outcome Pathways", OECD Series on Adverse Outcome Pathways, No. 1, OECD Publishing, Paris. Ogilvie BW Zhang D Li W Rodrigues AD Gipson AE Holsapple J Toren P Parkinson A Glucuronidation converts gemfibrozil to a potent, metabolism-dependent inhibitor of CYP2C8: implications for drug-drug interactions Drug Metab Dispos 2006 34 191 197 10.1124/dmd.105.007633 16299161 Ortiz de García SA Pinto Pinto G García-Encina PA Irusta-Mata R Ecotoxicity and environmental risk assessment of pharmaceuticals and personal care products in aquatic environments and wastewater treatment plants Ecotoxicology 2014 23 1517 1533 10.1007/s10646-014-1293-8 25064485 Otieno PO Lalah JO Virani M Jondiko IO Schramm K-W Soil and water contamination with carbofuran residues in agricultural farmlands in Kenya following the application of the technical formulation Furadan J Environ Sci Health B 2010 45 137 144 10.1080/03601230903472058 20390943 Paiga P Santos L Ramos S Jorge S Silva JG Delerue-Matos C Presence of pharmaceuticals in the Lis river (Portugal): Sources, fate and seasonal variation Sci Total Environ 2016 573 164 177 10.1016/j.scitotenv.2016.08.089 27560998 Papadopoulos N Gikas E Zalidis G Tsarbopoulos A Simultaneous determination of terbuthylazine and its major hydroxy and dealkylated metabolites in wetland water samples using solid-phase extraction and high-performance liquid chromatography with diode-array detection J Agric Food Chem 2007 55 7270 7277 10.1021/jf0706777 17676755 Park WP Chang KM Hyun HN Boo KH Koo BJ Sorption and leaching characteristics of pesticides in volcanic ash soils of Jeju Island Korea Appl Biol Chem 2020 63 1 1 13 10.1186/s13765-020-00555-5 Perez-Maldonado IN Athanasiadou M Yanez L Gonzalez-Amaro R Bergman A Diaz-Barriga F DDE-induced apoptosis in children exposed to the DDT metabolite Sci Total Environ 2006 370 343 351 10.1016/j.scitotenv.2006.06.026 16904735 Perkins EJ Ashauer R Burgoon L Conolly R Landesmann B Mackay C Murphy CA Pollesch N Wheeler JR Zupanic A Scholz S Building and applying quantitative adverse outcome pathway models for chemical hazard and risk assessment Environ Toxicol Chem 2019 38 1850 1865 10.1002/etc.4505 31127958 Peters K Bundschuh M Schäfer RB Review on the effects of toxicants on freshwater ecosystem functions Environ Pollut 2013 180 324 329 10.1016/j.envpol.2013.05.025 23725857 Petrovic M Skrbic B Zivancev J Ferrando-Climent L Barcelo D Determination of 81 pharmaceutical drugs by high performance liquid chromatography coupled to mass spectrometry with hybrid triple quadrupole-linear ion trap in different types of water in Serbia Sci Total Environ 2014 468–469 415 428 10.1016/j.scitotenv.2013.08.079 Picó Y Barceló D Transformation products of emerging contaminants in the environment and high-resolution mass spectrometry: a new horizon Anal Bioanal Chem 2015 407 21 6257 6273 10.1007/s00216-015-8739-6 25990630 Planas C Puig A Rivera J Caixach J Analysis of pesticides and metabolites in Spanish surface waters by isotope dilution gas chromatography/mass spectrometry with previous automated solid-phase extraction Estimation of the uncertainty of the analytical results J Chromatogr A 2006 1131 242 252 10.1016/j.chroma.2006.07.091 16962600 Porazzi E Pardo Martinez M Fanelli R Benefati E GC–MS analysis of dichlobenil and its metabolites in groundwater Talanta 2005 68 1 146 154 10.1016/j.talanta.2005.04.044 18970297 Postigo C Barcelo D Synthetic organic compounds and their transformation products in groundwater: occurrence, fate and mitigation Sci Total Environ 2015 503–504 32 47 10.1016/j.scitotenv.2014.06.019 Ralston-Hooper K Hardy J Hahn L Ochoa-Acuna H Lee LS Mollenhauer R Sepulveda MS Acute and chronic toxicity of atrazine and its metabolites deethylatrazine and deisopropylatrazine on aquatic organisms Ecotoxicology 2009 18 899 905 10.1007/s10646-009-0351-0 19533344 Reddy PVL Kim K-H A review of photochemical approaches for the treatment of a wide range of pesticides J Hazard Mater 2015 285 325 335 10.1016/j.jhazmat.2014.11.036 25528231 Reemtsma T Alder L Banasiak U Emerging pesticide metabolites in groundwater and surface water as determined by the application of a multimethod for 150 pesticide metabolites Water Res 2013 47 5535 5545 10.1016/j.watres.2013.06.031 23863396 Rodieux F Vutskits L Posfay-Barbe KM Habre W Piguet V Desmeules JA Samer CF When the safe alternative is not that safe: tramadol prescribing in children Front Pharmacol 2018 5 9 148 10.3389/fphar.2018.00148 Rodrigues P Cunha V Oliva Teles L Ferreira M L., Guimarães. Norfluoxetine and venlafaxine in zebrafish larvae: Single and combined toxicity of two pharmaceutical products relevant for risk assessment J Hazard Mater 2020 400 123171 10.1016/j.jhazmat.2020.123171 32593945 Rodrigues P Cunha V Ferreira M Reis-Henriques MA Oliva Teles L Guimarães L Carvalho AP Differential molecular responses of zebrafish larvae to fluoxetine and norfluoxetine Water 2022 14 3 417 10.3390/w14030417 Rousis NI Bade R Bijlsma L Zuccato E Sancho JV Hernandez F Castiglioni S Monitoring a large number of pesticides and transformation products in water samples from Spain and Italy Environ Res 2017 156 31 38 10.1016/j.envres.2017.03.013 28314152 Rozmánková E Pípal M Bláhová L Njattuvetty-Chandran N Morin B Gonzalez P Bláha L Environmentally relevant mixture of S-metolachlor and its two metabolites affects thyroid metabolism in zebrafish embryos Aquat Toxicol 2020 221 105444 10.1016/j.aquatox.2020.105444 32078888 Rutkowska M Płotka-Wasylka J Sajid M Andruch V Liquid–phase microextraction: A review of reviews Microchem J 2019 149 103989 10.1016/j.microc.2019.103989 Safe SH Endocrine disruptors and human health–is there a problem? An update Environ Health Perspectives 2000 108 487 493 Santos LH Araujo AN Fachini A Pena A Delerue-Matos C Montenegro MC Ecotoxicological aspects related to the presence of pharmaceuticals in the aquatic environment J Hazard Mater 2010 175 45 95 10.1016/j.jhazmat.2009.10.100 19954887 Santos LH Paiga P Araujo AN Pena A Delerue-Matos C Montenegro MC Development of a simple analytical method for the simultaneous determination of paracetamol, paracetamol-glucuronide and p-aminophenol in river water J Chromatogr B 2013 930 75 81 10.1016/j.jchromb.2013.04.032 Scheil V Kienle C Osterauer R Gerhardt A Kohler HR Effects of 3,4-dichloroaniline and diazinon on different biological organisation levels of zebrafish (Danio rerio) embryos and larvae Ecotoxicology 2009 18 355 363 10.1007/s10646-008-0291-0 19093198 Schuhmann A Klammler G Weiss S Gans O Fank J Haberhauer G Gerzabek MH Degradation and leaching of bentazone, terbuthylazine and S-metolachlor and some of their metabolites: A long-term lysimeter experiment Plant Soil Environ 2019 65 5 273 281 10.17221/803/2018-pse Sharma A Kumar V Shahzad B Tanveer M Sidhu GPS Handa N Kohli SK Yadav P Bali AS Parihar RD Worldwide pesticide usage and its impacts on ecosystem SN Appl Sci 2019 10.1007/s42452-019-1485-1 Shipkova M Armstrong VW Oellerich M Wieland E Acyl glucuronide drug metabolites: toxicological and analytical implications Ther Drug Monit 2003 25 1 1 16 10.1097/00007691-200302000-00001 12548138 Sparkman OD Penton ZE Kitson FG Kitson FG Larsen BS McEwen CN Gas chromatography Gas Chromatography and Mass Spectrometry: A Practical Guide 2011 Amsterdam Elsevier 15 83 Stanley JK Ramirez AJ Chambliss CK Brooks BW Enantiospecific sublethal effects of the antidepressant fluoxetine to a model aquatic vertebrate and invertebrate Chemosphere 2007 69 9 16 10.1016/j.chemosphere.2007.04.080 17582462 Stokstad E Grullon G Infographic: Pesticide Planet. A global look at the uses, benefits, and drawbacks of pesticides Science 2013 341 730 731 10.1126/science.341.6147.730 23950524 Stropnicky, W. G. (2017). Effects of the Antidepressant O-Desmethylvenlafaxine on Crayfish Aggression.Theses and Dissertations (All). 1564 Stumm-Zollinger E Fair GM Biodegradation of Steroid Hormones Journal (water Pollution Control Federation) 1965 37 1506 1510 5849629 Subramaniam R Östin A Nilsson C Åstot C Direct derivatization and gas chromatography-tandem mass spectrometry identification of nerve agent biomarkers in urine samples J Chromatogr B 2013 928 98 105 10.1016/j.jchromb.2013.03.009 Sunkara M Wells MJM Phase II pharmaceutical metabolites acetaminophen glucuronide and acetaminophen sulfate in wastewater Environ Chem 2010 10.1071/EN09098 Syafrudin M Kristanti RA Yuniarto A Hadibarata T Rhee J Al-Onazi WA Algarni TS Almarri AH Al-Mohaimeed AM Pesticides in drinking water-a review Int J Environ Res Public Health 2021 10.3390/ijerph18020468 Tyagi V Garg N Mustafa MD Banerjee BD Guleria K Organochlorine pesticide levels in maternal blood and placental tissue with reference to preterm birth: a recent trend in North Indian population Environ Monit Assess 2015 187 471 10.1007/s10661-015-4369-x 26122123 Union E Commission Implementing Decision (EU) 2022/1307 of 22 July 2022: establishing a watch list of substances for Union-wide monitoring in the field of water policy pursuant to Directive 2008/105/EC Official J Eur Union, L 2022 197 117 121 European Union (2004). Directive 2004/27/EC of the European Parliament and of the Council of 31 March 2004 amending Directive 2001/83/EC on the Community code relating to medicinal products for human use. Official Journal of the European Union. L, 136, 34–57. Valls-Cantenys C Scheurer M Iglesias M Sacher F Brauch HJ Salvado V A sensitive multi-residue method for the determination of 35 micropollutants including pharmaceuticals, iodinated contrast media and pesticides in water Anal Bioanal Chem 2016 408 6189 6200 10.1007/s00216-016-9731-5 27382969 Velisek J Stara A Koutnik D Machova J Effect of terbuthylazine-2-hydroxy at environmental concentrations on early life stages of common carp (Cyprinus carpio L.) Biomed Res Int 2014 2014 621304 10.1155/2014/621304 24689046 Velisek J Koutnik D Zuskova E Stara A Effects of the terbuthylazine metabolite terbuthylazine-desethyl on common carp embryos and larvae Sci Total Environ 2016 539 214 220 10.1016/j.scitotenv.2015.08.152 26363394 Velisek J Stara A Zuskova E Kubec J Buric M Kouba A Chronic toxicity of metolachlor OA on growth, ontogenetic development, antioxidant biomarkers and histopathology of early life stages of marbled crayfish Sci Total Environ 2018 643 1456 1463 10.1016/j.scitotenv.2018.06.309 30189562 Veljanoska-Sarafiloska EM Jordanoski M Stafilov T Presence of DDT metabolites in water, sediment and fish muscle tissue from Lake Prespa, Republic of Macedonia J Environ Sci Health B 2013 48 548 558 10.1080/03601234.2013.774879 23581687 Vinken M Knapen D Vergauwen L Hengstler JG Angrish M Whelan M Adverse outcome pathways: a concise introduction for toxicologists Arch Toxicol 2017 91 3697 3707 10.1007/s00204-017-2020-z 28660287 Vryzas Z Pesticide fate in soil-sediment-water environment in relation to contamination preventing actions Curr Opinion Environ Sci Health 2018 4 5 9 10.1016/j.coesh.2018.03.001 Wajchenberg BL bergaria Pereira, M.A., Medonca, B.B., Latronico, A.C., Campos, C.P., Alves, V.A., Zerbini, M.C., Liberman, B., Carlos, G.G and Kirschner, M.A. Adrenocortical carcinoma: clinical and laboratory observations Cancer 2000 88 711 736 10.1002/(SICI)1097-0142(20000215)88:4<711::AID-CNCR1>3.0.CO;2-W 10679640 Weston DP Lydy MJ Toxicity of the insecticide fipronil and its degradates to benthic macroinvertebrates of urban streams Environ Sci Technol 2014 48 1290 1297 10.1021/es4045874 24368023 White PM Potter TL Bosch DD Joo H Schaffer B Muñoz-Carpena R Reduction in metolachlor and degradate concentrations in shallow groundwater through cover crop use J Agric Food Chem 2009 57 20 9658 9667 10.1021/jf9021527 19799423 Wilkinson J Hooda PS Barker J Barton S Swinden J Occurrence, fate and transformation of emerging contaminants in water: An overarching review of the field Environ Pollution 2017 231 954 970 10.1016/j.envpol.2017.08.032 Wode F van Baar P Dunnbier U Hecht F Taute T Jekel M Reemtsma T Search for over 2000 current and legacy micropollutants on a wastewater infiltration site with a UPLC-high resolution MS target screening method Water Res 2015 69 274 283 10.1016/j.watres.2014.11.034 25497426 Zhuo H-Q Huang L Huang H-Q Cai Z Effects of chronic tramadol exposure on the zebrafish brain: A proteomic study J Proteomics 2012 75 11 3351 3364 10.1016/j.jprot.2012.03.038 22507199	0	PMC9734376	NO-CC CODE	2022-12-14 23:28:28	no	Gynäkologie. 2022 Dec 5; 55(12):888-889	latin-1	null	null	null	oa_other
==== Front Int J Cardiovasc Imaging Int J Cardiovasc Imaging The International Journal of Cardiovascular Imaging 1569-5794 1875-8312 Springer Netherlands Dordrecht 36494505 2759 10.1007/s10554-022-02759-w Letter to the Editor Abnormal echocardiographic findings after COVID-19 infection: looking for an appropriate balance Garcia-Zamora Sebastián [email protected] 12 Liblik Kiera 3 Priotti Mauricio 1 Picco José Miguel 4 Lepori Augusto José 5 Merlo Pablo Martín 2 Gastaldello Natalio 2 1 Delta Clinic, Rosario, Argentina 2 Argentine Association of Critical Ultrasonography, ASARUC, Buenos Aires, Argentina 3 grid.410356.5 0000 0004 1936 8331 Queen’s University, Kingston, ON Canada 4 Institute of Cardiology and Sports Medicine Wolff, Mendoza, Argentina 5 Institute of Cardiology and Cardiovascular Surgery, Posadas,, Argentina 10 12 2022 12 12 11 2022 15 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. ==== Body pmcTo the editor, We read the editorial by Singh et al. about our registry with great interest. We thank the authors for their interest in our project and publication [1]. We would first like to respond to the inquiry about performing echocardiography on asymptomatic patients. As per our methods section, patients underwent echocardiography at our center for a variety of reasons [2]. Due to the COVID-19 lockdowns during the registry development, patients were past 14 days of confirmed COVID-19 infection by a polymerase chain reaction (PCR) test before they had an echocardiogram. In Argentina and Brazil, there was no indication to repeat a PCR for COVID-19 after an initial positive test, given that this strategy had not shown any advantage in prognosis and management. Accordingly, the time interval between the COVID-19 infection and the echocardiographic examination was influenced by physician referral time of the test. Given that we did not intervene in the decision to request any study for the patients included in our registry, we performed a non-probabilistic sampling, as described previously. Concerning the objective of our registry, our original publication stated that “we aimed to explore the prevalence of echocardiographic cardiac abnormalities in ambulatory patients after recovery of a first documented COVID-19 infection”. That is why we included patients with different disease severity, as most groups have done previously. Another misconception seems to be the lack of demographic information about the included patients. Beyond the baseline characteristic we described in Table 1, we detailed that “we included 595 participants who recovered from a COVID-19 infection with an average age of 45.5 ± 14.9 years, of which 50.8% were female. The majority of patients (82.5%) had the disease at home […]”. We also gave the precise information that “the mean body mass index in the patients was 26.8 ± 4.8 kg/m2, and 61.7% of the participants denied any relevant medical history. The most frequent comorbidity was arterial hypertension (28.6%)”. The supplementary data also reported the time elapsed from the patient’s recovery after COVID-19 infection and echocardiographic examination. Maybe the most provocative discussion is about the role of echocardiography after a COVID-19 infection. We agree with Singh et al. that the echocardiogram is a relatively expensive examination and not widely available in many regions of the world. In this line, the electrocardiogram is a well-known cost-effective tool to assess cardiovascular abnormalities in many diseases, even COVID-19. However, it is critical to note that in most diseases, the accuracy of the electrocardiogram to detect abnormalities is influenced by the knowledge and skills of those who evaluate it. During the first wave of COVID-19, there was great uncertainty about cardiovascular involvement after the disease. Thus, some initial research stated that most patients had some cardiovascular involvement after COVID-19 infection, even asymptomatic patients. That led many groups, like ours, to explore this possibility with different methods. In this scenario, we believe that the echocardiogram with global longitudinal strain represents a valuable balance between the sensitivity and specificity to detect cardiovascular abnormalities at a relatively low cost compared with more sophisticated tests. Our study showed that less than one in eleven patients had any echocardiographic abnormality. The most frequent was reduced left or right ventricle global longitudinal strain value, representing a subclinical affectation. We believe this observation is in line with daily practice. Finally, we have not advocated for performing an echocardiogram on each patient after COVID-19 infection. Our study suggests that cardiovascular abnormalities after a COVID-19 infection are infrequent and usually mild. We are convinced that in the post-COVID era, perhaps the most significant challenge we are facing is to strike the balance between over-testing and underdiagnosing our patients. Declarations Disclosure of Conflict of interest Nothing to declare. Disclosure of funding source The present study has not received any grants or financial support. Compliance with Ethics Guidelines This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors. Ethics approval This article is based on previously conducted studies and does not contain any new 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. Singh N, Hani ZB, AlRemeithi R. Abnormal echocardiographic findings after COVID-19 infection: a multicenter registry. Int J Cardiovasc Imaging. 2022. 10.1007/s10554-022-02732-7. [Online ahead of print] 2. Garcia-Zamora S, Picco JM, Lepori AJ, Galello MI, Saad AK, Ayon M, et al. Abnormal echocardiographic findings after COVID-19 infection: a multicenter registry. Int J Cardiovasc Imaging. 2022. 10.1007/s10554-022-02706-9. [Online ahead of print]	36494505	PMC9734378	NO-CC CODE	2022-12-14 23:29:53	no	Int J Cardiovasc Imaging. 2022 Dec 10;:1-2	utf-8	Int J Cardiovasc Imaging	2,022	10.1007/s10554-022-02759-w	oa_other
==== Front Intern Emerg Med Intern Emerg Med Internal and Emergency Medicine 1828-0447 1970-9366 Springer International Publishing Cham 36469247 3134 10.1007/s11739-022-03134-2 CE-Research Letter to the Editor Not so mild: emergency department utilization after index COVID infection stratified by disease severity http://orcid.org/0000-0002-8299-7689 Bastani Aveh [email protected] 1 Homayouni Ramin 2 Heinrich Kevin 3 Nair Girish Balachandran 4 1 grid.461921.9 0000 0004 0460 1081 Department of Emergency Medicine, Beaumont Health, Troy, MI USA 2 grid.261277.7 0000 0001 2219 916X Department of Foundational Medical Studies, Oakland University William Beaumont School of Medicine, Rochester, MI USA 3 Quire Inc., Memphis, TN USA 4 grid.461921.9 0000 0004 0460 1081 Department of Pulmonary and Critical Care, Beaumont Health, Royal Oak, MI USA 5 12 2022 14 5 8 2022 12 10 2022 © The Author(s), under exclusive licence to Società Italiana di Medicina Interna (SIMI) 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 pmcDear Editor, As of February 1st, 2022, over 419 million cases of COVID-19 had been confirmed worldwide with over 5 million deaths [1]. In the United States alone there have been over 79 million cases of COVID-19 with over 950,000 deaths [2]. Despite efforts to mitigate the stress on the entire healthcare system, including lockdowns and vaccine mandates, significant concern remains regarding our ability to handle the downstream effects of the COVID-19 pandemic. One of these concerns revolves around lingering health issues and healthcare utilization of COVID-19 patients after their index infection. To date no large-scale study has described the rate of ED utilization for patients after their index COVID-19 infection with respect to the severity of their illness, specifically comparing “mild” vs. “severe” cases. Our objective was to describe the prevalence of patients who returned to the Emergency Department after their COVID-19 index infection as stratified by disease severity. Furthermore, we sought to compare the quantity and quality of those visits for patients who initially presented with mild (“non-hospitalized”) infection compared with those who had severe (“hospitalized”) infection. Our study was conducted across an eight-hospital health system in southeast Michigan, including a large academic tertiary care center and multiple large and small community hospitals. We conducted a retrospective analysis of all patients who presented to the health system with a positive COVID-19 PCR test. Given the ubiquitous and all-encompassing nature of COVID-19 during the study period, we did not have a non-COVID-19 comparison group. Furthermore, given this study design, significant restrictions were placed on our initial cohort to confirm that these patients were active within our healthcare system prior to their index COVID-19 diagnosis. Our inclusion criteria required that within one year prior to their index COVID-19 infection, patients had: (1) At least one office visit, (2) A documented problem list, and (3) A medical history documented in the health system EHR. Patients were also only included in the analysis if they had their index COVID-19 infection before April 30, 2021. This allowed for: (1) At least a six-month follow-up period and (2) Limited the time of the cohort to when patients did not readily have home testing available. Finally, patients were separated into two groups: (1) Mild cases, defined by those who were not hospitalized within 14 days of their initial COVID-19 diagnosis; (2) Severe cases, defined by those who were hospitalized within 14 days of their initial COVID-19 diagnosis. Expired patients were excluded from the final analysis. Finally, we wanted to specifically tease out the impact of COVID-19 between the mild and severe cohorts. To that end, we extracted all existing ICD-10 diagnosis codes (for any time and any encounter) for each patient prior to their index COVID-19 diagnosis and compared this information to the ED encounter ICD-10 diagnoses codes within the six-month follow-up period. Chronic diseases were derived based on the CMS definitions for each patient using ICD-10 codes extracted from their problem list and medical histories at any time prior to their index diagnosis [3]. Our primary outcome was the percentage of patients who came back to the Emergency Department after their COVID-19 diagnosis stratified by the severity of their initial infection (Mild vs. Severe). Among the patients who returned to the ED after their index diagnosis, we examined only the new ICD-10 codes associated with their subsequent visits, understanding that ICD-10 codes are biased toward more severe symptomology. New codes were defined as those which only appeared in ED encounter billing diagnosis post-index COVID-19 diagnosis and that were not documented previously at any point in the patients EHR. Specifically, we counted all new ICD-10 codes which were present in greater than five percent of either the mild or severe cohort and represented the common systems affected by COVID-19 (Pulmonary, Cardiac, Renal, and Constitutional). Secondary outcomes include the difference in ED utilization broken down by initial disease severity (Outpatient vs Inpatient), demographics, and co-morbidities as defined by the number of Chronic Conditions listed in their EMR. Baseline characteristics and clinical outcomes were compared between patients with mild versus severe COVID-19. Normal or approximately normal variables were reported using the mean (standard deviation), whereas skewed variables were reported with the median (interquartile range [IQR]). Categorical variables were compared using the Chi-square test or Fisher exact test. Normal variables were compared using a 2-sided Student t test and ordinal variables used the Kruskal–Wallis test. To evaluate the association between baseline comorbid conditions and severity of disease, an age-adjusted logistic regression analysis was performed, with continuous variables as co-variables and categorical variables as factors. All p values were 2-sided and a p < 0.05 was considered to indicate statistical significance. Statistical analysis was performed using SAS 9.4 (SAS Institute, Cary, North Carolina). The authors of this IRB exempt study have no relavent conflicts of interest. Our study population included a total of 6,838 adult patients with a PCR-confirmed diagnosis of COVID-19. Among these, 65% (n = 4,467) had mild disease, who did not require hospitalization within 14 days of the index diagnosis. Table 1 reports demographics and for mild and severe COVID-19 groups. In the mild cohort, 879 out of 4,467 (19.7%) of the patients had at least one return ED visit, while 1087 of the 2,371 (45.8%) of the severe patients had a return ED visit (Fig. 1). These counts represented 2.27 higher odds (95% CL 1.987, 2.58, p < 0.0001) of repeat ED visits in severe compared to mild cases. As expected, those with increased chronic disease count had 1.09 higher odds of developing a severe infection with each additional chronic condition. To determine the possible reasons for the return ED visits, we filtered the ED billing diagnosis codes that were never previously recorded in the EHR of each patient. Table 2 shows the new ICD-10 codes, broken down by organ systems for severe and mild cases. With regard to the pulmonary system, cough and shortness of breath were more common in the mild group while respiratory failure, viral pneumonia and oxygen dependence were more common in the severe cohort. For the cardiovascular system, chest pain and new anticoagulant use were more common in the mild group while long-term anti-coagulant use was more common in the severe cohort. With regard to the renal system, dehydration, hypovolemia and hyperosmolality/hypernatremia were the most common for the mild cohort, while multiple electrolyte imbalances and evidence of acute and chronic kidney failure were common for the severe cohort. Finally, both cohorts had evidence of new constitutional symptoms including weakness, fatigue and malaise. The incidence of headaches was higher (8.2%) in the mild cases compared to the severe cases (3.3%).Table 1 Baseline characteristics of COVID-19 population seeking hospital-level care after initial infection Mild disease (n = 4467) Severe disease (n = 2371) Significance Age in years 51.4 (17.1) 63.8 (15.7) < 0.0001 Female 2917(55.5%) 1318 (65.3%) < .0001 Race African American 975 (21.8%) 766 (32.2%) Caucasian 3032 (67.9%) 1415 (59.7%) < .0001 Asian 112(2.5%) 51 (2.2%) More than one 91(2.0%) 44(1.9%) Other 257 (5.8%) 95 (4.0%) Chronic Disease Count 4(5) 7(5) < 0.0001 Age is presented as mean (standard deviation). Chronic disease count presented as median (IQR). Other numbers represent n (%) Fig. 1 ED visits after mild (a) and severe (b) COVID-19 infection. a 4467 total “Mild” COVID-19 Cases Broken Down by # of Return ED Visits (19.7% had at least one return ED visit). b 2371 total “Severe” COVID-19 Cases Broken Down by # of Return ED Visits (45.8% had at least one return ED visit) Table 2 New ICD-10 Codes for patients presenting to the Emergency Department after index COVID-19 infection, stratified by severity of the disease (Mild vs. Severe) System Severe Mild p value (Fisher’s exact) Patient Count ED Percentage (%) Patient Count ED Percentage (%) Pulmonary Shortness of Breath (R06.02) 47 4.3 81 9 0.5709 Cough (R05) 40 3.7 74 8.2 0.9212 Acute Respiratory Failure with hypoxia (J96.01) 226 20.7 15 1.7 < 0.0001 Acute and chronic respiratory failure with hypoxia (J96.21) 51 4 < 0.0001 Chronic respiratory failure with hypoxia (J96.11) 24 4 < 0.0001 Viral Pneumonia (J12.89) 178 16.3 22 2.4 < 0.0001 Oxygen Dependence (Z99.81) 57 5.2 31 3.4 < 0.0001 Cardiac Other chest pain(R07.89) 34 3.1 75 8.3 0.7556 Chest pain unspecified (R07.9) 0.1507 Tachycardia, unspecified (R00.0) 69 6.3 60 6.7 < 0.0001 New ASA use (Z79.82) 84 7.7 46 5.1 < 0.0001 New Anticoagulant use (Z79.01) 96 8.8 23 2.5 < 0.001 Renal Dehydration (E86.0) 120 11 47 5.2 < 0.0001 Hyperosmolality/hypernatremia (E87.0) 50 4.6 47 5.2 < 0.0001 Hypovolemia (E86.1) 36 3.3 47 5.2 < 0.0001 Hyperkalemia (E87.5) 59 5.4 17 1.9 < 0.0001 Hypokalemia (E87.6) 128 11.7 42 4.7 < 0.0001 Hypomagnesemia (E83.42) 89 8.2 20 2.2 < 0.0001 Acute Kidney Failure, unspecified (N17.9) 118 10.8 35 3.9 < 0.0001 Acidosis (E87.2) 114 10.4 31 3.4 < 0.0001 Hypo-osmolality/hyponatremia (E87.1) 104 9.5 25 2.8 < 0.0001 Hypertensive Kidney Failure (I12.9) 73 6.7 21 2.3 < 0.0001 Chronic Kidney Disease, unspecified (N18.9) 71 6.5 20 2.2 < 0.0001 Constitutional Weakness (R53.1) 67 6.2 33 3.7 < 0.0001 Fatigue (R53.83) 51 4.7 56 6.2 0.0074 Malaise (R53.81) 71 6.5 21 2.3 < 0.0001 Headache (R51.9) 36 3.3 74 8.2 0.6879 As both severe and mild COVID-19 patients present back to the emergency department after their initial infection, an understanding of the quantity and characteristics of those visits is vital in delivering quality care. In addition, the ubiquitous presence of COVID-19 testing would render a similar analysis describing the outcomes of mild cases impossible. We present the largest cohort to date of both mild and severe covid patients who presented back to the ED at least 14 days after their index infection. In describing these patients, we have identified that both severe and mild cases have repeat ED visits after index infection. Although, patients with an initial severe case present back to the emergency department and are subsequently admitted more often than their mild counterparts; the mild cohort still represents a significant burden to the healthcare system with 19.7% of those patients returning to the emergency department within six months. Finally, both in our cohort and relevant scientific literature regarding COVID-19, mild cases represent significant morbidity with patients mirroring the symptomatology of their severe counterparts [4, 5]. Symptoms such as fatigue, weakness, headache, and malaise are well represented within our cohorts and are known to persist despite the initial severity of the illness [6]. It is important to note that other cofactors including vaccination status and treatments provided that would directly affect our primary outcome were not assessed in this analysis. Though most of the acute COVID-19 cases are classified as mild, 19.7% of those patients came back to the emergency department and mirrored the symptomatology of their severe counterparts who returned at a rate of 46.1%. As emergency department volumes continue to return to pre-pandemic levels, understanding the prevalence of symptoms and characteristics of these patients will provide a foundation for optimizing management, conducting relevant research and distributing resources. Declarations Conflict of Interest The authors declare that they have no conflict of interest. Human and animal rights This article does not contain any studies directly involving human participants, as it is a review of data already collected in a COVID database. Informed consent For this type of retrospective study, formal consent is not required. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. https://www.worldometers.info/coronavirus/. Accessed Jan 2022 2. https://www.worldometers.info/coronavirus/country/us/. Accessed Jan 2022 3. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Chronic-Conditions/CC_Main. Accessed Jan 2022 4. Dennis A Wamil M Kapur S Multi-organ impairment in low-risk individuals with long COVID MedRxiv 2020 10.1101/2020.10.14.20212555 33052357 5. Crook H Raza S Nowell J Young M Edison P Long covid—mechanisms, risk factors, and management BMJ 2021 374 1648 10.1136/bmj.n1648 6. UK Office for National Statistics. Prevalence of long COVID symptoms and COVID-19 complications (2020) https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/healthandlifeexpectancies/datasets/prevalenceoflongcovidsymptomsandcovid19complications. Accessed Jan 2022	36469247	PMC9734382	NO-CC CODE	2022-12-14 23:28:28	no	Intern Emerg Med. 2022 Dec 5;:1-4	utf-8	Intern Emerg Med	2,022	10.1007/s11739-022-03134-2	oa_other
==== Front Reg Environ Change Reg Environ Change Regional Environmental Change 1436-3798 1436-378X Springer Berlin Heidelberg Berlin/Heidelberg 1994 10.1007/s10113-022-01994-0 Original Article Factors associated with farmers’ use of indigenous and scientific climate forecasts in Rwenzori region, Western Uganda http://orcid.org/0000-0002-6434-9100 Nkuba Michael Robert [email protected] 1 Chanda Raban 1 Mmopelwa Gagoitseope 1 Kato Edward 2 Mangheni Margaret Najjingo 3 Lesolle David 1 Adedoyin Akintayo 4 Mujuni Godfrey 5 1 grid.7621.2 0000 0004 0635 5486 Department of Environmental Sciences, University of Botswana, 4775 Notwane Road, Private Bag 00704, Gaborone, Botswana 2 grid.419346.d 0000 0004 0480 4882 International Food Policy and Research Institute, Washington, D.C USA 3 grid.11194.3c 0000 0004 0620 0548 Department of Extension and Innovation Studies, College of Agricultural and Environmental Sciences, Makerere University, Kampala, Uganda 4 grid.7621.2 0000 0004 0635 5486 Department of Physics, University of Botswana, Gaborone, Botswana 5 grid.463702.4 Uganda National Meteorological Authority, Kampala, Uganda Communicated by Shuaib Lwasa 7 12 2022 2023 23 1 42 9 2021 20 10 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. Although scientific climate forecast (SF) distribution by national climate services has improved over time, farmers seem not to make good use of climate forecasts, a likely contributing factor to vulnerability to climate change. This study investigated factors associated with farmers’ use of SFs and indigenous forecasts (IFs) for agricultural use in the Rwenzori region, western Uganda. Household survey gathered data on demographic characteristics, climate information use and livelihood choices from 580 farmers. Data was analysed using the probit model. Results showed that significant factors associated with using both IFs and SFs were farm size, education, age, reception of scientific forecasts in local languages, agricultural extension access, short-mature crop access, farmer-to-farmer network and accessing forecasts through radio. This study shows that IFs were used complementarily with SFs. On the other hand, significant factors associated with using IFs only were livelihood choices such as tuber and goat production, access to government interventions on climate change adaptations, agro-ecological zone and social capital. Climate risks and climate risk perceptions negatively influenced the use of scientific forecasts. Co-production of climate information, capacity-building and active engagement of stakeholders in dissemination mechanisms can improve climate forecast use. Investments in more weather stations in various districts will therefore be a key factor in obtaining more accurate scientific forecasts and could lead to increased use of scientific climate forecasts. Governments in developing countries, the private sector, global and regional development partners should support investments in weather stations and capacity building of national meteorological systems. Supplementary Information The online version contains supplementary material available at 10.1007/s10113-022-01994-0. Keywords Scientific climate forecasts Indigenous forecasts Farmers Co-production Cognitive bias Uganda issue-copyright-statement© Springer-Verlag GmbH Germany, part of Springer Nature 2023 ==== Body pmcIntroduction Although climate services in the developing world have improved over time, scientific climate forecasts’ (SFs) use has remained low (Nkuba et al. 2021a). SFs are accessible from radio, television, mobile phones (Jost et al. 2015a, b), the internet, agricultural extension agents and farmer-to-farmer networks (Goddard et al. 2010). Development agencies involved in disaster risk reduction and in rural and agriculture development such as World Vision, Care International, Save the Children and Oxfam have also played a role in the dissemination of SFs (ACCRA 2014). SFs are communicated in local languages in Senegal, Malawi and Tanzania (Hampson et al. 2014; Lo and Dieng 2015). Using appropriate local languages to disseminate SFs improves access and use among farmers (Jost et al. 2015a, b). The main challenge in using SFs by stakeholders such as farmers is their probabilistic nature (Nicholls 1999): what has been predicted may or may not happen. Decision-making based on forecasts with shortcomings raises serious concerns among arable farmers (hereafter referred to as farmers) (Goddard et al. 2001). The limited predictive accuracy negatively influences use of SFs by farmers. Goddard et al. (2001) found that obstacles to using SFs include meteorologists’ failure to provide full information about the predictive accuracy or reliability of the forecasts, and their interpretation as deterministic forecasts. SFs lack the spatial and temporal specificity that farmers are most interested in. The SFs are disseminated in terms of regions or districts which have wide geographical coverage. Thus, predicted weather events may occur in some areas but not in all. Simply producing and disseminating climate forecasts do not make them valuable to farmers. However, Patt et al. (2005) reported that the use of SFs improved crop yields in Zimbabwe. This implies that there is improved farmers’ welfare through utility of SFs. Farmers’ limited use of SFs has been associated with lack of saliency, credibility, trust and legitimacy (Patt and Gwata 2002; Cash et al. 2003, 2006; McNie 2007). Furthermore, farmers have raised as major shortcomings of SFs poor spatial and temporal resolution with regard to failure to provide forecasts on onset and cessation of rains (Kalanda-Joshua et al. 2011; Nkomwa et al. 2014). One of the factors that has led to low use of SFs in Africa is low meteorological station density (Medany et al. 2006; UNECA 2011), which has resulted in poor spatial resolution with negative credibility and trust implications. The meteorological station density in Africa is eight times lower than the World Meteorological Organisation’s (WMO) minimum recommended level (Medany et al. 2006); the distribution of the meteorological stations in Africa does not match the existing agro-ecological systems. Meteorological station density is one of the cornerstones of index-based weather insurance (IBWI) (Akter et al. 2016). Increased investment in rural weather station infrastructure not only improves the use of SFs but also increases the uptake of IBWI (Amare et al. 2019) meaning lending institutions may not be in be position to provide much needed insurance and credit facilities to African farmers due to lack of climate information. To address the poor station density in Africa, the United Nations Development Programme (UNDP) has supported a number of African countries by providing them with automatic weather stations (Snow et al. 2016). Besides station density, other challenges of climate services in Africa include dysfunctional stations, or poorly maintained, or low data quality due to low national budget allocation to national meteorological services in many developing countries (UNECA 2011; Snow et al. 2016). The Uganda National Meteorological Authority (UNMA) provides biannual climate forecasts on the onset and cessation of rains, 10-day rainfall forecasts and rainfall distribution (UNMA 2017b). With support from UNDP, there has been increased investment in weather infrastructure in Uganda (Snow et al. 2016). Alongside the high access to SFs (Jost et al. 2015a, b), the Rwenzori region has a low meteorological station density (Fig. 1), with negative implications for the use of SFs. The meteorological authority collaborates effectively with other agencies involved in weather data generation such as agricultural research stations, wildlife protected areas, national universities and private tea and sugar plantations in rural areas (Snow et al. 2016). Only 5% of rain gauges in Uganda are functional and another 921 rain gauges are needed to attain an optimum level (Isabirye 2017).Fig. 1 Study area map There are multiple sources of indigenous forecasts (IFs) (Roncoli et al. 2002; Kalanda-Joshua et al. 2011; Kolawole et al. 2014; Nkomwa et al. 2014). Sources of IFs include farmer-to-farmer extension network, farmers’ organisation, elderly farmers and farmer’s own observations (Appendix Table 3). There are 10 (abiotic and biotic environmental) indicators that are observed to provide weather and climate information (Nkuba et al., 2020b). These indicators are relevant to farmers in their localities providing forecasts at high spatial resolution ( Roncoli et al., 2002; Orlove et al. 2010; Nkomwa et al. 2014). The temporal dimension of IF provides relevant information on onset and cessation of rains, which is very important in farmers’ adaptation to changes in rainfall seasons (Roncoli et al. 2002; Nkomwa et al. 2014). This suggests that IF has high credibility, trust and legitimacy among farmers. Farmers who use IFs only have expressed interest in using SFs after interaction with meteorologists (Orlove et al. 2010; Mpandeli and Maponya 2013). Climate variability negatively influences the use of IFs (Roncoli et al. 2002; Speranza et al. 2010). In light of the challenges of using IFs exclusively or SFs exclusively, some farmers have resorted to using a combination of SFs and IFs since SFs complements IFs (Ziervogel and Opere 2010; Roncoli et al. 2008; Mpandeli and Maponya 2013). Research has shown that some farmers begin with using IFs and then revise their forecast partially or completely after receiving an SF (Lybbert et al. 2007). There are farmers who revise their IFs after access to SFs and there are farmers who use IFs only and who do not revise their forecasts even after accessing SFs. There is high access to SFs (89%) in the study area. This is attributed to the proliferation of FM radio stations and the use of local languages. Climate forecasts are used in farmers’ decision-making related to when to plant, harvest and what crops to grow for a given cropping season. The climate forecasts are essential in climate change adaptation (Nkuba et al. 2020a). Farmers use climate forecasts to reduce their vulnerability to the impacts of climate risks such as droughts and floods (Hansen 2002). In bimodal rainfall regions of East Africa, onset determines which cereals are to be grown, with maize being grown during periods of early onsets, while sorghum and millet are grown during periods of late onsets (Mugalavai et al. 2008). Forecasts thus influence the choice of crop enterprises for coming farming seasons. Maize is very sensitive to water stress. For sweet potatoes (Ipomoea batatas), potatoes (Solanum tuberosum) and cassava (Manihot esculenta), water stress leads to poor tuber formation (El-Sharkawy and Cadavid 2002; MacKerron and Jefferies 1986; Yooyongwech et al. 2013), sweet potato weevil infestation (Ebregt et al. 2004, 2007), fruit cracking and blotchy ripening for vegetables (Steduto et al. 2012). The farmers’ use of forecasts for variations in the crop-water requirements greatly influences choice of crop. Some studies have investigated factors associated with climate forecast use. A study done in Northern Kenya and Southern Ethiopia (Luseno et al. 2003) found that location, education level and access to radio are significant factors in the use of SFs. Another study done in Botswana (Kolawole et al. 2014) reported that age and education level are associated with the use of climate forecasts. The two studies did not include factors such as the language in which SFs are received, livelihood choices, the agro-ecological zone or access to agricultural extension, which the current study takes into account. Access to forecasts seems to have gender dimensions, with men reportedly having more access due to their control of the radio in the rural households (Jost et al. 2015a, b). Jost et al. (2015a, b) found that women in Uganda preferred to receive forecasts from gatherings and community radios, while television was the preferred option for rural women in Bangladesh. A study in south-eastern Kenya (Muema et al. 2018) showed that age, gender, drought frequency, radio and access to improved varieties were factors associated with the use of SFs. A study in eight African countries1 in East and West Africa (Oyekale 2015) showed that education level, climate shocks, radio and gender were factors associated with the use of SFs. These studies did not look at factors associated with farmers’ use of IFs only and both IFs and SFs, hence the need for this study. The above literature review has highlighted factors associated with use of climate forecasts by farmers. Nevertheless, knowledge gap regarding the effect of crop type, agro-ecological zones and climate risk perceptions on the use of IFs only or both IFs and SFs in farming still exists. The objective of this paper is to investigate factors associated with farmers’ use of IFs only or both IFs and SFs in the Rwenzori region in western Uganda. We answer the question regarding what factors are associated with farmers’ use of IF singly or the combination of both IF and SF. This gap has not been adequately addressed in the international literature. The consensus globally, regionally and nationally is that climate change is occurring and its negative effects have been acknowledged. Access to climate and weather information therefore becomes imperative to the farming communities which experience the direct vulnerabilities from climate change. In this study, we investigate what can be done to increase use of SFs among rural farmers in Uganda, where there is also a tendency of farmers to use their own non-scientifically based indigenous forecasts (IFs). This study has highlighted to policymakers, climate scientists, climate change activists and government funding partners that farmers who exclusively use SF forecast are very few, with majority of farmers still depending on IFs which are inaccurate and arbitrary. This might have the consequence of farmers not being able to respond to climate change risks appropriately and timely if there is still over reliance on IF forecasts. This study identifies the factors that can be targeted by government and its stakeholders to turn around this situation and increase usage of SF forecasts. This study has also revealed that possibly it would be hard to promote use of SFs only in rural communities but rather a hybrid between SFs and IFs might lead to wider usage of SFs. Theoretical framework This study investigated decision-making under uncertainty. Some scholars emphasise the consequentialist approach to decision-making, implying that decision choices have repercussions. The farmers who are the recipients of SFs may or may not use them. National meteorological systems are the principal providers of SFs. The farmers receive SFs which are used to examine decisions under uncertainty. In this study, UNMA disseminates SFs to the farming community. Some of the farmers may not use SFs even after receiving them but use IFs only. There is a budget allocation by the Ugandan government for gathering and dissemination of SFs and early-warning information by funding UNMA. Farmers who use information from meteorological stations are using posterior beliefs which may be objective. Information from meteorology stations can facilitate farmers to change their knowledge (prior beliefs) about climate. Farmers may update their prior beliefs (indigenous knowledge) with information from meteorologists (posterior beliefs) and may follow the recommendations from the experts. Alternatively, the farmers may not believe in the accuracy of the information from the meteorologists, and hence, they may not update their prior beliefs, resulting in their not following the recommendations from the experts. Farmers receiving SFs may change or revise their prior beliefs based on indigenous knowledge, underscoring the importance of confidence in forecast information received (Luseno et al. 2003). The meteorological stations make use of historical rainfall data and other climate parameters to forecast what the rainfall pattern will be in the coming days and months (seasonal climate forecast). Meteorologists go further in recommending what actions the farmers ought to take. However, the final decision on what actions to take lies within the farmers. The value and use of climate forecasts by farmers are based on the range of actions and their capability to respond. This study seeks to fill the knowledge gaps regarding the factors associated with farmers’ decision-making under uncertainty based on the climate forecasts (SFs and IFs) they used. The paper contributes to the current debate on influence of climate information on decision-making under uncertainty. Materials and methods The study area The study was conducted in the Rwenzori region of western Uganda. In terms of climate, the region experiences bimodal rainfall, which contributes to having two cropping seasons with the first season running from February to May and the second season from August to December. The temperature ranges from 12 to 24 °C, annual rainfall ranges from 800 to 3000 mm, and elevation ranges from 500 to 5000 m above sea level. Most of the farmers have completed primary education (Nkuba et al. 2020a). The farmers’ ability to comprehend climate information can be enhanced through stakeholder engagement by Uganda National Meteorological Authority (Nkuba et al. 2019). The soils are mainly sandy loam and sand clay loam which support production of cereals, tubers and vegetables (Nkuba et al. 2020a). Tree planting for commercial crop production of coffee, cocoa, fruit trees and commercial woodlots is very vibrant in the study area (Nkuba et al. 2020a). Access to pastures supports climate change adaptation measures for livestock farmers such as livestock migration, herd mobility and livestock diversification (Nkuba et al. 2021c). The region was selected because it has several agro-ecological zones—mountainous, lowland, mountainous and forested, wetland and forested—in order to investigate the effect of agro-climate on climate forecast use. Wildlife protected areas (WPA) in the study area (Kibale, Toro-Semiliki and Mount Rwenzori National parks) (Fig. 1) and Rwebitaba Zonal Agricultural Research Institute (in Kabarole district) have meteorological stations which provide climate data used in meteorological forecasts for the region. Weather stations are in government-aided establishments such as research stations and WPAs. Farmers in remote areas tend to find SF predictions to be inaccurate. Farming is a major livelihood source in the region. Access to SFs in local languages is due to the spread of FM radio and television stations. The region is inhabited by many ethnic groups, all of which use indigenous knowledge in their day-to-day lives. Data collection and analysis procedures Data was gathered from August to October 2015. A respondent survey using questionnaires was used to collect data on socio-economic factors, farm and institutional characteristics by trained research assistants. The respondent survey information was triangulated with data from focus group discussions (FGDs) and interviews. FGDs were used to get farmers’ views on the use of IFs and SFs. FGDs were held in Kyenjojo and Kabarole districts to get the farmers’ views on the use of IF and SF. A female FGD of 15 members and a male FGD of 16 members were carried out in Kyejonjo district. In Kabarole district, a female FGD of 15 members and male FGD of 17 members were conducted. The members were farmers who use IFs and SFs in their farming activities. To ensure that members of FGDs expressed their views freely, gender-segregated focus groups were used. The use of the mixed methods approach was for triangulation of the data obtained to enhance its validity and reliability. Data was analysed using Stata 16 statistical software. This study is nested in an earlier study (Nkuba et al. 2020a). A multi-stage stratified approach was used in the sampling of respondents. The first stage involved districts, the second stage counties, the third stage sub-counties and the smallest unit the household. The Uganda Bureau of Statistics disseminates the population data of households according to district, county, sub-county and household. The selection criteria were (i) farming systems, namely arable farming, pastoralism and agro-pastoralism and (ii) agro-ecological systems such as forested, lowlands, mountainous and wetlands. The statistically selected sample size of farmers was allocated to particular sub-counties in the selected districts using proportional allocation to size, where the size represents the number of households in the district. Based on a population of the study area of 102,496 households, according to the Uganda population census report of 2014, a sample size of 778 was selected with 95% confidence level and a margin of error of 3.5%. To allow for replacement in the sample of respondents who might drop out of the study, 19% of the statistically selected sample was included, giving a total study sample of 924. This was also to ensure a good sample size for sub-samples (for those who use both IFs and SFs, and IFs only). After data cleaning, 17 questionnaires were excluded due to incomplete responses. Of the 907 respondents, 580 were farmers, 270 pastoralists and 57 agro-pastoralists. This paper limits itself to the 580 farmers in the sample. Theoretical model The probit regression was used to analyse farmers’ use of forecasts. The probit model was used for examining the likelihood of a future climate event (for example rain onset and cessation) happening at a particular time using climate forecasts. There is no complete certainty regarding the occurrence of future climate outcomes that have been predicted. The predicted outcome may or may not happen. The predicted climate outcomes (dependent variables) in this study address use or non-use of IFs and SFs. IF and SF are specified as dichotomous outcomes with yes (use) and no (no-use), coded as 1 and 0 respectively. The probit model is a non-linear model that estimates with probabilistic maximum likelihood, often used for binary outcomes (Gujarati 2013). The regression results in this study are based on probit regression estimates. The probit model was estimated using Stata 16 software. There is potential for presence of self-selection in the use or non-use of forecasts. To overcome issues of selectivity convoluting the estimates, we have controlled for several conditioning factors that could be correlated with use of forecast and at the same time also influence outcomes of interest. We think this approach helps to minimise self-selection bias due to forecast use. We also tried to estimate models with and without forecasts to have a feel for the extent to which the estimates are being affected by selectivity. The estimates did not change when forecasts were excluded in the models. With these two approaches, we think that the possible threat from self-selection is not very serious in our analysis. We could use the instrumental variable approach because it was difficult to find strong excluded instruments that would satisfy all the instrument validity requirements. Empirical model The empirical model used in the analysis was specified as follows.1 Yij=β1Householdcharacteristics+β2Informationsources+β3Farmcharacteristics+β4Agro-ecologicalzone+β5SocialCapital+β6Percerptionofclimaterisk+β7Climaterisk+β8Wealth+ε For use of IF only2 Zij=β1Householdcharacteristics+β2Informationsources+β3Farmcharacteristics+β4Agro-ecologicalzone+β5SocialCapital+β6Percerptionofclimaterisk+β7Climaterisk+β8Wealth+ε For use of IF and SF where Yij(j = 1,2,3,4) representing the four models of using only IF and Zij(j = 1,2,3,4) representing the four models of using both IF and SF (Appendix Table 4). In this study, we control for several factors that could influence use of information as well as those factors that have been identified as affecting adoption of agricultural technologies in developing countries. Past empirical and theoretical literature has guided our theoretical choice of the factors that we include in our econometric models as explained below. Factors associated with farmers’ use of climate forecasts include education level, age, gender (Roncoli et al. 2002), livelihood choices (Vogel 2000; Ingrama et al. 2002; Patt and Gwata 2002; Ziervogel and Calder 2003; Crane et al. 2010; Haigh et al. 2015; Klemm and McPherson 2017), translation of SFs into local languages (Ingrama et al. 2002; Ziervogel and Downing 2004), source of forecasts for onset and cessation (Roncoli et al. 2002; Haigh et al. 2015), access to credit (Vogel 2000; Ingrama et al. 2002), access to non-farm enterprises (Ingrama et al. 2002; Ziervogel and Calder 2003; Crane et al. 2010; Klemm and McPherson 2017), access to agricultural extension (Vogel 2000; Ziervogel and Downing 2004; Coles and Scott 2009; Crane et al. 2010; Haigh et al. 2015; Klemm and McPherson 2017), access to improved crop varieties (Ingrama et al. 2002), climate risks (Ingrama et al. 2002), agro-ecological zone (Ingrama et al. 2002; Ziervogel 2004; Klemm and McPherson 2017), perception of climate risks (Vogel 2000; Ingrama et al. 2002; Ziervogel 2004; Haigh et al. 2015) and farm size (Vogel 2000; Coles and Scott 2009). Climate risks and perception of climate risks are important factors associated with farmers’ use of forecasts (Vogel 2000; Ingrama et al. 2002; Haigh et al. 2015). Farmers’ cognitive biases2 have an effect on their risk perceptions which influences climate forecasts use (Waldman et al. 2019). Unlike pastoralists, who practice herd mobility as an adaptive mechanism, farmers cannot migrate their crop fields. Changes in the onset and cessation of rains progress slowly towards their final manifestation in damage to crops (Slovic 2000a). The explanatory variables include Respondent characteristics (H): level of education, age, gender and farming experience; farm characteristics (F): type of crops grown and type of livestock kept; institutional characteristics (I): sources of IF, sources of SF, access to agricultural extension, credit access, improved crop access, non-farm access; agro-ecological system (A): forested, lowland, mountainous, wetland, mountainous and forested; social capital (S): membership of farmers’ organisation, farmer-to-farmer extension networks; perception of climate risks (P) drought increase, flood increase; climate risks (C): flood experience, drought experience; wealth (W): farm size. Variables and expected signs for use of IFs only and both IFs and SFs are shown in Appendix Table 5. Results Descriptive overview: use of forecasts Results show that almost half (49%) of the farmers used IFs only and half (50%) used both SFs and IFs (Appendix Table 6). Only one respondent used SF only. Over half (54%) of the respondents were male (Appendix Table 7). There were no significant differences in the use of either both IF and SF or IFS only between men and women respondents. Forecasts for rain onset and cessation were very crucial in farming. Considering the sub-sample of those who used IFs only, all the respondents used indigenous knowledge in predicting the onset of rains compared to 77% for the sub-sample of those who used both IFs and SFs. Most of the farmers (97%) reported that IF was reliable, compared to 43% for SF. Farmers who used both IFs and SFs reported that they used SFs to confirm the IFs based on what they had observed and learnt from elders. The most important crops grown were cereals and tubers (Appendix Table 3). Farmers’ organisations and farmer-to-farmer networks also played a crucial role in providing climate information (Appendix Table 3). Radio and non-government organisations were important SF dissemination channels (Appendix Table 8). Local FM stations provide SFs in local languages (Appendix Table 7), which has greatly improved access even to farmers with no formal education. Agricultural extension workers played a minimal role in the dissemination of SFs and information on rain season duration. Agricultural extension has been impacted by the involvement of Uganda Peoples Defence Force under Operation Wealth Creation, which plays a major role in the government distribution scheme3 for agricultural inputs such as improved seeds, seedlings and livestock during onset of rains. The results show that there were significant differences between farmers who used both IFs and SFs and those who used IFs only (Appendix Tables 3, 7, and 9 at 5% level of significance). Farmers from mountainous forested areas and bare mountainous areas were using IF only significantly more than both IFs and SFs (Appendix Table 7). Location and terrain appear to affect use of SFs, possibly because of differences in the level of infrastructural development between mountainous areas and lowlands. Farmers in lowlands were more likely to use SFs, while their counterparts in mountains were more likely to use IFs. This suggests constraints linked to quality of infrastructural development limiting access to SFs. Farmers growing cereals were using both IFs and SFs significantly less than IF only (Appendix Table 3). Cereals are very sensitive to water stress and are grown on a commercial scale in the study area. This increases the reliance on rain-fed agriculture in rural areas where there is limited use of irrigation, resulting in an increase in the use of SFs. Factors influencing farmers’ use of indigenous forecasts only The results show that factors positively and significantly associated with using IFs only among farmers were as follows: livelihood choices such as tubers and goat production, agro-ecology such as being resident in mountainous areas, reception of information from farmers’ organisations about the onset and cessation of rains access to government programme interventions on climate change adaptations and perceptions of climate variability and change as increases in floods and drought (Table 1). Factors negatively and significantly associated with using IFs only were non-farm access and livelihoods that depend on maize production.Table 1 Marginal effects of use of IFs only for arable farmers Variable Onset dy/dx Cessation dy/dx 5 day dy/dx Seasonal dy/dx Tubers 0.054(0.051) 0.214*(0.077) 0.201(0.079) 0.161**(0.074) Mountainous and forested 0.200(0.125) Mountainous 0.156(0.086) Farmers’ organisation as source of onset and cessation 0.264*(0.063) 0.306(0.082) 0.322*(0.095) 0.164(0.094) Credit access − 0.120(0.073) Non-farm access − 0.200*(0.078) Agricultural extension access − 0.170(0.095) Rainfall increase 0.108(0.083) Rainfall season change 0.120(0.079) Droughts increase 0.134(0.074) Floods increase 0.218(0.110) 0.244(0.116) 0.322(0.115) Access govt interventions on climate change adaptations 0.116(0.058) 0.176*(0.088) ln maize production − 0.017(0.010) − 0.025(0.015) − 0.036(0.016) − 0.022(0.015) ln local goat production 0.053(0.044) 0.073(0.046) 0.107(0.044) Log likelihood − 281.134 − 130.445 − 125.016 − 125.204 Pseudo R2 0.045 0.103 0.136 0.109 N 425 210 210 210 (Source: survey data 2015 overall title of the research project was ‘The Use of Indigenous and Scientific Forecasts in adaptation to climate variability and Change. A case study of Rwenzori region, Western Uganda’). Figures in parentheses are standard errors. *, and * denote that significance at the 1%, 5% and 10% levels respectively Government programme interventions on climate change adaptations include the provision of improved crop varieties and tree-planting materials to farmers, whose adoption depends on forecasts for the onset of rains. The agricultural inputs, such as seeds and seedlings, were delivered by National Agricultural Advisory Services, Operation Wealth Creation under the Office of the President and the Uganda Forestry Authority. Farmers had confidence in IFs and hence used it for predicting the onset of rains to effectively participate in government adaptation programmes. Cereals (especially maize), tubers and tree crops are sensitive to water stress, especially during the vegetative stage, and farmers’ confidence in IFs influenced their decisions on choice of crop enterprises and allocation of resources, depending on the onset and cessation forecasts. Farmers preferred to grow maize in the second rainy season, which is long enough and runs from August to November. Short-maturing cereals like sorghum and millet were preferred for the first rainy season, which is short and runs from March to May. A key informant reported that winds in the mountainous and forested areas on the windward side of Mount Rwenzori have high moisture content which easily reaches saturation point resulting in rainfall. Farmers in such areas rely a great deal on observing IF indicators like clouds and wind to get reliable forecasts. Factors influencing farmers’ farmers’ use of both indigenous forecasts and scientific forecasts The findings show that factors positively and significantly associated with using both IFs and SFs were reception of SFs in local language and English, attainment of higher education (diploma or advanced secondary school education), access to SFs through radio and TV, access to short-maturing crop varieties, availability of agricultural extension services, age and fellow farmer as source of SFs (Table 2).4 Factors negatively and significantly associated with using both IFs and SFs were livelihood that depend on vegetables, tubers and maize, being residents in mountainous forested areas, primary school education, Non-Governmental Organisation as source of onset and cessation forecasts and drought experience, and perceiving climate variability and change as seasonal rainfall change and drought increase (Table 2). Vegetable growers seemed not to have enough confidence in SFs as a basis for decision-making, probably because of vegetable crops’ water requirements. Command of English has a higher impact than local language on the use of SFs. Farmers who are comfortable with receiving SFs in English have fewer constraints than those who receive SFs in local languages. English may be more used by elite and well-to-do farmers. Farmers using local languages have less ability to access SFs compared to their counterparts who also use English.Table 2 Marginal effects of use of both IFs and SFs for arable farmers Variable Onset dy/dx Cessation dy/dx 5 day dy/dx Seasonal dy/dx Male 0.050(0.038) Vegetables − 0.107(0.051) − 0.131(0.053) − 0.134(0.044) Tubers − 0.127(0.045) Age 0.003(0.002) 0.004(0.002) Local language 0.464*(0.045) 0.402(0.058) 0.214(0.056) 0.201(0.047) English 0.577(0.063) 0.611(0.082) 0.367(0.115) 0.317(0.127) Mountainous and forested − 0.229(0.054) − 0.210(0.065) -0.089(0.047) Primary education − 0.135(0.053) -0.046(0.039) Higher education 0.375(0.163) 0.414*(0.144) Radio as source of scientific forecasts 0.249(0.090) 0.227(0.056) TV as source of scientific forecasts 0.113(0.090) 0.391(0.159) Fellow farmer as source of scientific forecasts 0.142(0.061) 0.152(0.066) 0.161*(0.060) NGO as source of onset and cessation forecasts − 0.339(0.040) − 0.264(0.052) − 0.208(0.047) Drought experience -0.098(0.057) − 0.126(0.071) Short mature crop access 0.156(0.063) 0.109(0.067) 0.094(0.053) Agricultural extension access 0.149(0.077) 0.180(0.089) Rainfall increase -0.058(0.041) Rainfall season change − 0.083(0.048) − 0.098(0.052) -0.053(0.040) Droughts increase − 0.127(0.054) − 0.151*(0.046) Unpredictable trains 0.051(0.041) ln maize production − 0.023(0.011) 0.017(0.010) -0.016(0.008) ln farm size 0.040(0.023) 0.027(0.026) 0.024(0.023) 0.019(0.019) ln likelihood − 270.210 − 194.653 − 195.332 -168.930 Pseudo R2 0.205 0.248 0.173 0.144 N 507 394 394 394 (Source: survey data 2015). Significant at 1%, significant at 5%, *significant at 10%. Figures in parentheses are standard errors There are few weather stations in the region (Fig. 1) whose rainfall data is used to make predictions, resulting in poor predictive accuracy and wide spatial variation for forecasts from UNMA. The rainfall data from the weather station in the mountainous area of Kabarole was used to make forecasts for the mountainous and forested area, quite far from the station.5 A key informant pointed out that there was apparently no weather station for rainfall data in the mountainous and forested area of Bundibugyo. This casts doubt on the reliability of SFs for seasonal forecasts for mountainous forested areas using extrapolated rainfall data from weather stations located in distant ecological zones. Forecasts were made in terms of regions and districts, yet farmers are interested in information at parish or village level. Receiving SFs in local languages made dissemination accessible to farmers, even to those with no formal schooling. Mountainous and forested areas have high precipitation in the form of mist and fog, which contributes positively to soil moisture availability. Climate risks like drought are negatively associated with the use SFs. It is probable that climate risks like droughts can influence farmers’ cognitive biases such as availability bias, anchoring effect and confirmation bias, which could impair farmers’ judgement in using SFs. Farmers can overestimate or underestimate the likelihood of droughts due to over-confidence in their recent memory of the severity and frequency of the occurrences. It is also possible to consider long dry spells as droughts. Climate risk perception such as drought increase and seasonal rainfall change was negatively associated with use of SFs. Climate risk perceptions can influence farmer cognitive biases such as framing effect, hindsight bias, anchoring effect and confirmation bias, which could impair farmers’ judgement in using SFs. The framing effect created by the limited coverage of dissemination mechanisms such as media about the effects of change in rain onset and cessation can lead farmers to underestimate the start and end of the rainfall season. Most media and meteorologists tend to focus more on rainfall distribution in terms of ‘above normal’ or ‘below normal’ in their dissemination of the seasonal climate forecast for the farming season, rather than duration of the rainy season. Farmers’ experiences can make them over-confident and consequently impair their judgement of frequency and severity of droughts. But there were also respondents who perceive that SFs are not credible, as exemplified by the following observation by a female participant in the FGD: ‘For me, I don’t think it’s truth because there is when they told us to prepare for heavy hail storms but they didn’t come’ (i.e., scientific forecasts are not reliable). There is also a lack of local specificity in SFs, rendering them less credible to the farmers. One FGD participant, for instance, remarked that ‘You can see that it [rain] has not happened here but when in another place it has’. Some doubt was also cast on the legitimacy of SFs, as indicated by one participant who declared that ‘They also just guess’. This indicates that farmers do not trust the processes that meteorologists use to make forecasts, and consequently, they distrust the forecasts as well. Farmers attached high credibility to IFs because of the long tradition of using and depending on them, as is evidenced by the following statement from a FGD participant: ‘from when we were born until we have grown to now it’s what we found our elders using. The elders before us this [IFs] is what they were using, their knowledge of long ago. For that we also grow up following it’. IFs knowledge is based on the long-term observation of nature and is passed on orally from one generation to another. There were 15 indicators used in IF identified in the study area (Nkuba et al. 2020b) implying that there is a wide variety of sources of forecasts, resulting in increased legitimacy. The indicators are biotic factors such as plants, insects and birds and abiotic factors such as wind and clouds. The abiotic factors are similar to what meteorologists use in their predictions. Farmers’ organisations and elderly farmers were social capital that provide forecast information within their specific local area. Religion also plays a role in the use of IFs; for example, many respondents said that the onset of rains corresponds to the day of Mother Mary, namely 18 August. One participant in the FGD reported that ‘Indigenous knowledge is a very important thing because that’s the knowledge God created us with. Because even what is taught us [scientific forecasts]… you would learn and by the time you reach home you have forgotten it. But indigenous knowledge is better. Because with scientific forecasts, someone will guess what’s not there’. Radio is the most widely used method of disseminating information about SFs. Radio stations broadcast in both English and local languages. TV is widely used by elite and well-to-do farmers. TV stations like National TV and Uganda Broadcasting Services are widely used as sources of forecasts by wealthier farmers. A key informant in the civil service stated that agricultural extension workers receive SFs from UNMA through the Ministry of Agriculture on a regular basis. There are factors that influence the use of both IFs and SFs, and IFs only, differently (Tables 1 and 2). For instance, access to improved crop varieties and agricultural extension services is positively associated with using both IFs and SFs and negatively associated with using IFs only. Access to agricultural extension is closely associated with SFs. Maize production was positively associated with use of both IFs and SFs but negatively associated with IFs for the 5-day forecasts. Short-range SFs such as 5-day forecasts provide reliable climate information (dry spells, or optimum planting and harvesting days) which is relevant to farmers’ involvement in water stress crops like maize. Being a resident in mountainous and forested areas, and perceiving climate variability and change as drought increase and seasonal rainfall change, was positively associated with using IFs only and negatively associated with using IFs and SFs. Mountainous and forested areas are positively associated with high precipitation due to orographic effects on cloud formation and condensation, contributing to less variability in rainfall. Having a higher level of education (secondary education and diploma) was positively associated with use of IFs and SFs. Increase in education increases uptake of SFs. However, primary education was negatively associated with use of IFs and SFs. Primary education probably causes the farmers to overestimate or underestimate due to over-confidence based on their recent past experiences. Reception of SFs from fellow farmers were positively associated with using IFs and SFs, while reception of onset and cessation of rain forecasts from farmers’ organisations was positively associated with using IFs only. Group decisions were more trusted in using IFs only. Confidence in using IFs only is built up through social gatherings of like-minded people. Information disseminated through farmer-to-farmer networks was more trusted in using SFs. This could arise from the confidence that the farmer-to-farmer networks built over time in using SFs with good outcomes. Discussion Results showed that significant factors associated with using both IFs and SFs were farm size, education, age, reception of scientific forecasts in local languages, agricultural extension access, short-mature crop access, farmer-to-farmer network and accessing forecasts through mass media. This study shows that IFs were used complementarily with SFs. On the other hand, significant factors associated with using IFs only were livelihood choices such as tuber and goat production, access to government interventions on climate change adaptations, agro-ecological zone and social capital. Research has shown that dissemination of SFs through local languages is the most preferred among rural households (Antwi-Agyei et al. 2021). Dissemination in local languages is usually facilitated by media such as local FM radio stations. The current study has also confirmed that radio and television play a crucial role in the dissemination of SFs. Radio has been reported to be an important dissemination mechanism for SFs in rural regions (Jost et al 2015a, b). This indicates that widening FM radio station broadcasting coverage in Uganda could provide an opportunity for increasing the utilisation of SFs in rural areas. The main shortcoming of using such media as radio and television for dissemination is that there is usually no possibility for feedback from farmers. This is perhaps one of the reasons that farmers seem to have greater confidence in forecasts from their fellow farmers: there is feedback and discussion of experiences and the implications of using such forecasts. National meteorological services need to employ dissemination mechanisms that provide feedback from farmers for improved effectiveness in the use of forecasts. The findings of the current study also show that the onset and cessation of rains forecasts from farmer-to-farmer extension networks had mixed results, with a positive association with use of IFs only and a negative association with use of both IFs and SFs. This shows that farmers had more confidence in IFs than in SFs because IFs have local specificity while SFs have broad spatial and temporal variations (Speranza et al. 2010). Farmers’ indigenous knowledge of climate forecasts is more trusted than forecasts by climate scientists (Kolawole et al. 2014; Roudier et al. 2014). The results of the current study have also shown that farmers have negative associations with use of forecasts from NGOs. The challenge is that NGO staff operating in rural areas may not be able to answer all the farmers’ questions about forecasts, which may engender a lack of trust in the forecasts among farmers (Ofoegbu and New 2021). This study shows that farmers involved in tuber production are more associated with use of IFs only. Farmers involved in tuber and maize production do not have confidence in SFs because of past experiences of poor spatial and temporal specificity and, therefore, in their eyes, of false forecasts (Ziervogel 2004). Our study shows mixed results regarding the association of education with climate forecast use. Post primary education was positively associated with use of SFs. Luseno et al. (2003) study in Kenya and Ethiopia showed that formal education improved confidence in and access to SFs. Similarly, a study done in Botswana showed that the more educated a farmer is, the more likely he/she will be to use SFs received through the media (Kolawole et al. 2014). However, primary education was negatively associated with the use of SFs. A study done is Zambia reported that one additional year in education reduced farmers’ perception of change in rain onset due to psychological factors (Waldman et al. 2019). This calls for further research on the effect of psychological factors on climate forecast use and a multidisciplinary approach for climate services research. Climate change perceptions have a mixed effect on the use of forecasts, with a positive association with use of IFs and a negative association with use of both IFs and SFs. Experienced farmers make adjustments in their livelihoods after experiencing a drought, or they make changes in their responses to the onset and cessation of rains because of the likelihood of a recurrence (Slovic et al. 2000b). Farmers experience high crop losses due the change in rainfall duration and severe droughts, thus increasing their confidence in using IFs only. The range of alternatives for farmers is limited, which increases the level of catastrophe to their livelihoods (Slovic et al. 2000c). This also influences the use of IFs only, in order to avoid or to minimise experiencing such risks as loss of seeds due to late onset of rains, or poor harvests due to early cessation of rainfall. Farmers’ assessment of risks, which influences their use of IFs, is related to climate risks that reduce household welfare due to crop yield losses (Slovic 2000, 2000a, 2000b, 2000c). However, the perception of unpredictable rains as climate change is associated with using both. This is consistent with Partey et al. (2018),who found that increased rainfall variability positively influences the use of climate information in Ghana. The seasonal climate and short-range forecasts from UNMA provide predictions of rainfall distribution which influence risk perceptions (UNMA 2017a), as illustrated by a participant in a FGD, who said that ‘Now like this season when they say the rains are going to be a lot, we become cautious of the floods’. This indicates that SFs are also positively associated with risk perceptions and climate change perceptions. The results from the FGDs show that SFs are associated with increased crop yields which is consistent with findings from studies done in Burkina Faso and Zimbabwe (Patt et al. 2005; Roncoli et al. 2008). Research shows that farmers who feel they have adaptive capacity for a particular climate risk (availability heuristic) tend to have lower the climate risk perception (Duinen et al. 2015) which can lead underestimating its effects (Tversky and Kahneman 1979). Farmers with drought experiences have hindsight bias leading to increased confidence in their degree of prediction of future droughts without using SFs. Drought increase is a dreaded risk whose calamities are known through loss of crop yields and change in rainfall season is a delayed risk with a slow manifestation of harm. The recent memory of climate risks like rainfall seasonal change and drought could influence the use of SFs due to hindsight bias and anchoring effect. The framing effect from various stakeholders and media can result in making rainfall season change look less dangerous than gradual onset of droughts, leading to bias in using SFs. Research has shown that maize farmers have biases related to rain onset (Waldman et al. 2017), which is in agreement with our findings. Research has shown that farmers’ climate risk perception and past experience can lead to poor decision-making related to climate forecast use, due to cognitive biases (Waldman et al. 2019). Farmers from mountainous forested areas, and mountainous areas, are positively associated with use of IFs only. Due to orographic factors, the wide spatial variations in rainfall in these agro-ecological zones cause farmers to have more confidence in IFs only due to IFs’ local specificity compared to SFs, with SFs’ low spatial coverage because of low meteorological station density, especially in the mountainous areas (Sultan et al. 2020). Furthermore, Leroux (2001) found that ‘with increasing altitude the nature of precipitation changes; raindrops become smaller, showers give way to more continuous rain, and thence to rain, drizzle, mist and fog’ (p. 50). Forests act as windbreakers which disturb surface circulation and have a greenhouse effect that reduces temperature variations (Leroux 2001). Relative humidity in forested areas is high and this consequently leads to forests having a strong influence on rainfall over wide areas (Leroux 2001). The use of both SFs and IFs can be enhanced by experts engaging in constant dialogue with end-users. Studies in Mozambique, Kenya, Ivory Coast, Senegal, Tanzania and Benin have shown that confidence in SFs greatly improved when national meteorological services received constructive feedback from stakeholders such as farmers (Ziervogel and Opere 2010; Lo and Dieng 2015). Research has also shown that the legitimacy, salience and credibility of SFs have been improved by the active engagement of farmers, resulting in farmers’ increased use of SFs (Bouroncle et al. 2019). Integration of SFs and IFs is critical for the effective utilisation of forecasts in rural livelihoods based on farming. Ziervogel and Opere (2010) have provided information on projects in Africa that had integrated to good effect meteorological and indigenous-based seasonal forecasts in the agricultural sector. Stakeholder engagement can be in form of farm rainfall data generation by farmers. Botswana’s Department of Meteorological services has installed rain-gauges for farmers who send daily rainfall data via telephone. Farmers involved in farm rainfall data generation get excited when rainfall data from their farms is read during the news bulletin on national television station. Botswana meteorologists conducted meteorology extension services during the agricultural shows and held feedback sessions with farmers involved in rainfall data generation before the COVID-19 pandemic. A pilot study on citizen science in Namibia and South Africa has shown that a farmer involved farm rainfall data generation had much trust in SF, constructive engagement with meteorologists, reduced vulnerability to climate change, improved crop yields, livestock sales and household incomes (Landman et al. 2020). The results of the study are applicable to humid tropical zones and may not be applicable to a semi-arid area. This calls for further research regarding factors associated with use of indigenous and scientific forecasts in semi-arid areas. Conclusion and recommendations This study has established that farming in the Rwenzori region is informed by the use of both IFs and SFs or IFs only. Factors that promote the use of SFs include post-primary school education, access to appropriate rural institutions and dissemination of forecasts through radio using local languages. However, primary education was negatively associated with use of SFs. This is attributed to farmers’ cognitive biases. This calls for further research on the effect of psychological factors on climate forecast use. Research on climate information use requires a transdisciplinary approach involving environmental psychologists, social scientists, agronomists and meteorologists. Farmer-to-farmer extension networks are positively associated with use of IFs only and negatively associated with the use of both IFs and SFs. This calls for active stakeholders’ engagement by national meteorological services. This will improve the salience, legitimacy and credibility of meteorological forecasts in rural areas. Climate risk perceptions and climate risks are negatively associated with the use of SFs. Addressing farmers’ cognitive biases associated with climate risks and climate risk perceptions by national meteorological systems actively engaging stakeholders in climate forecast dissemination mechanisms could improve uptake of SFs. Investments in ddissemination of weather forecasts should be maintained and or increased, as the study has shown that television and radio have a positive impact. SFs were found to be complementary to IFs. SFs reinforce IFs. This calls for co-production of climate information in order to promote an increased use of forecasts in rural areas. Indigenous forecasts will continue to play a major role in influencing land-use interventions among farmers in Uganda and sub-Saharan Africa. We do not contest the use of SFs among farmers because trust in SFs among farmers is likely to improve with an increase in the number and density of weather stations in rural areas. Investment in more weather stations (automatic weather stations and rain gauges) in farming areas is a key factor in obtaining more spatially specific and accurate SFs. This could result in the improved use of SFs which might lead to improved food security and reduce vulnerability to climate change. Governments in developing countries, the private sector and the global and regional development partners should support investments in weather stations and capacity building of national meteorological systems. These possibilities are of course influenced not only by the use of forecasts but also by a number of other equally important factors such as access to inputs, agricultural extension services and credit. A longitudinal study on the validity of indigenous forecasts should be explored. In places with bimodal rainfall distribution, data has to be collected for both seasons in a period of 3 years or more to get meaningful outcomes. For comparative purposes, similar research should be conducted in other agro-ecological zones such as semi-arid areas and temperate areas, which were not covered in this study. The study has also established that farmers use scientific and indigenous forecasts in making decisions under uncertainty. Some farmers receive SFs but do not use them, instead using IFs only. Some receive SFs and update them with IFs, resulting in using both indigenous and scientific forecasts. Supplementary Information Below is the link to the electronic supplementary material. ESM 1 (DOC 135 kb) 1 Burkina Faso, Senegal, Mali, Niger, Ghana, Kenya, Tanzania and Ethiopia. 2 Cognitive biases include hindsight bias, confirmation bias, framing effect, anchoring effect, availability bias and decision regret effect (Tversky and Kahneman 1974; Kahneman and Tversky 1982; Nicholls 1999). 3 Media reports have indicated farmers rejecting agricultural inputs because they were supplied without taking into account rain onset. 4 The independent variables in the two models (Tables 1 and 2) differ because of precision concerns. Variables with standard errors larger than the coefficients indicate a poor estimation (Gujarati 2013) and were therefore not included in the models. 5 The distance from Kabarole to Bundibugyo is 24 km. Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References ACCRA (2014) Planning for the future and adapting to climate change in Uganda. Lessons from ACCRA In A. C. C. R. Alliance (Ed.), (pp. 4). Kampala, Uganda Africa Climate Change Resilience Alliance Akter S, Krupnik TJ, Rossi F, Khanam F (2016) The influence of gender and product design on farmers’ preferences for weather-indexed crop insurance. Glob Environ Chang 38:217–229. 10.1016/j.gloenvcha.2016.03.010 Amare A, Simane B, Nyangaga J, Defisa A, Hamza D, et al (2019) Index-based livestock insurance to manage climate risks in Borena zone of southern Oromia, Ethiopia. Clim Risk Manag 25:100191. 10.1016/j.crm.2019.100191 Antwi-Agyei P, Dougill AJ, Abaidoo RC (2021) Opportunities and barriers for using climate information for building resilient agricultural systems in Sudan savannah agro-ecological zone of north-eastern Ghana. Clim Serv 22.10.1016/j.cliser.2021.100226 Bouroncle C Müller A Giraldo D Rios D Imbach P Girón E Portillo F Boni A van Etten J Ramirez-Villegas J A systematic approach to assess climate information products applied to agriculture and food security in Guatemala and Colombia Climate Services 2019 16 October 100137 10.1016/j.cliser.2019.100137 Cash DW, Borck JC, Patt AG (2006) Countering the loading-dock approach to linking science and decision making: comparative analysis of El Niño/Southern Oscillation (ENSO) forecasting systems. Sci Technol Human Values 31(4):465–494. 10.1177/0162243906287547 Cash DW, Clark WC, Alcock F, Dickson NM, Eckley N, et al (2003) Knowledge systems for sustainable development. Proc Natl Acad Sci 100(14):8086–8091. 10.1073/pnas.1231332100 Coles AR, Scott CA (2009) Vulnerability and adaptation to climate change and variability in semi‐arid rural southeastern Arizona, USA. Nat Resour Forum 33(4):297–309. https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1477-8947.2009.01253.x Crane TA Roncoli C Paz J Breuer N Broad K Ingram KT Hoogenboom G Forecast skill and farmers’ skills: seasonal climate forecasts and agricultural risk management in the southeastern United States Weather, Clim Soc 2010 2 1 44 59 10.1175/2009WCAS1006.1 Duinen R Filatova T Geurts P van der Veen A Empirical analysis of farmers’ drought risk perception: objective factors, personal circumstances, and social influence Risk Anal 2015 35 4 741 755 10.1111/risa.12299 25514996 Ebregt E Struik PC Abidin PE Odongo B Farmers' information on sweet potato production and millipede infestation in north-eastern Uganda. I. Associations between spatial and temporal crop diversity and the level of pest infestation NJAS - Wageningen J Life Sci 2004 52 1 47 68 10.1016/S1573-5214(04)80029-0 Ebregt E Struik PC Odongo B Abidin PE Piecemeal versus one-time harvesting of sweet potato in north-eastern Uganda with special reference to pest damage NJAS - Wageningen J Life Sci 2007 55 1 75 92 10.1016/S1573-5214(07)80005-4 El-Sharkawy MA, Cadavid L (2002) Response of cassava to prolonged water stress imposed at different stages of growth. Exp Agric 38(03):333–350. 10.1017/S001447970200306X Goddard L, Mason SJ, Zebiak SE, Ropelewski CF, Basher R, et al (2001) Current approaches to seasonal to interannual climate predictions. Int J Climatol 21(9):1111–1152. 10.1002/joc.636 Goddard L, Aitchellouche Y, Baethgen W, Dettinger M, Graham R, et al (2010) Providing seasonal-to-interannual climate information for risk management and decision-making. Procedia Environ Sci 1:81–101. 10.1016/j.proenv.2010.09.007 Gujarati DN Basic Econometrics. Boston 2013 United States of America McGraw-Hill companies Haigh T Takle E Andresen J Widhalm M Carlton JS Angel J Mapping the decision points and climate information use of agricultural producers across the US Corn Belt Clim Risk Manag 2015 7 20 30 10.1016/j.crm.2015.01.004 Hampson KJ, Chapota R, Emmanuel J, Tall A, Huggins-Rao S, et al (2014) Delivering climate services for farmers and pastoralists through interactive radio: scoping report for the GFCS Adaptation Programme in Africa. Retrieved from CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS)Copenhagen, Denmark. https://hdl.handle.net/10568/65728 Hansen JW Realizing the potential benefits of climate prediction to agriculture: Issues, approaches, challenges Agric Syst 2002 74 3 309 330 10.1016/S0308-521X(02)00043-4 Ingrama KT, Roncolia MC, Kirshen PH (2002) Opportunities and constraints for farmers of west Africa to use seasonal precipitation forecasts with Burkina Faso as a case study. Agric Syst 74:331–349. 10.1016/S0308-521X(02)00044-6 Isabirye P (2017) Factoring spatial variability of rainfall in the design of optimum rainguage network for Uganda. Thesis for Master of science for climate change and Disater Management. Busitema University, Uganda. https://ir.busitema.ac.ug/bitstream/handle/20.500.12283/1898/Isabirye%20-%20Abstract.pdf?sequence=2&isAllowed=y Jost C, Kyazze F, Naab J, Neelormi S, Kinyangi J, et al (2015a) Understanding gender dimensions of agriculture and climate change in smallholder farming communities. Clim Dev 8(3):133-144.10.1080/17565529.2015.1050978 Jost C, Kyazze F, Naab J, Neelormi S, Kinyangi J, et al (2015b) Understanding gender dimensions of agriculture and climate change in smallholder farming communities. Clim Dev 8(2):1-12.10.1080/17565529.2015.1050978 Kahneman D, Tversky A (1982) The psychology of preferences. Scientific American 246:160–73 Kalanda-Joshua M Ngongondo C Chipeta L Mpembeka F Integrating indigenous knowledge with conventional science: enhancing localised climate and weather forecasts in Nessa, Mulanje, Malawi Phys Chem Earth, Parts a/b/c 2011 36 14–15 996 1003 10.1016/j.pce.2011.08.001 Klemm T McPherson RA The development of seasonal climate forecasting for agricultural producers Agric for Meteorol 2017 232 384 399 10.1016/j.agrformet.2016.09.005 Kolawole OD, Wolski P, Ngwenya B, Mmopelwa G (2014) Ethno-meteorology and scientific weather forecasting: small farmers and scientists’ perspectives on climate variability in the Okavango Delta, Botswana. Clim Risk Manag 4:43–58. 10.1016/j.crm.2014.08.002 Landman WA Archer ERM Tadross MA Citizen science for the prediction of climate extremes in South Africa and Namibia Frontiers in Climate 2020 2 1 8 10.3389/fclim.2020.00005 Leroux M The meteorology and climate of tropical Africa 2001 UK Springer Science & Business Media, UK Liebenehm S, Waibel H (2014) Simultaneous estimation of risk and time preferences among small-scale cattle farmers in West Africa. Am J Agr Econ 96(5):1420–1438. 10.1093/ajae/aau056 Lo H, Dieng M (2015) Impact assessment of communicating seasonal climate forecasts in Kaffrine, Diourbel, Louga, Thies and Fatick (Niakhar) regions in Senegal: Final Report for CCAFS West Africa Regional Program. Retrieved from Copenhagen, Denmark. http://bit.ly/1ifHC3P Luseno WK Mcpeak JG Barrett CB Little PD Gebru G Assessing the value of climate forecast information for pastoralists: evidence from Southern Ethiopia and Northern Kenya World Dev 2003 31 9 1477 1494 10.1016/S0305-750X(03)00113-X Lybbert TJ Barrett CB Mcpeak JG Luseno WK Bayesian Herders: Updating of Rainfall Beliefs in Response to External forecasts World Dev 2007 35 3 480 497 10.1016/j.worlddev.2006.04.004 MacKerron D, Jefferies R (1986) The influence of early soil moisture stress on tuber numbers in potato. Potato Res 29(3):299–312. 10.1007/BF02359959 McNie EC (2007) Reconciling the supply of scientific information with user demands: an analysis of the problem and review of the literature. Environ Sci Policy 10(1):17–38. 10.1016/j.envsci.2006.10.004 Medany M Niang-Diop I Nyong T Tabo R Background paper on impacts, vulnerability and adaptation to climate change in Africa 2006 Ghana Paper presented at the UNFCCC Convention Mpandeli S, Maponya P (2013) The Use of Climate Forecasts Information by Farmers in Limpopo Province, South Africa. J Agric Sci 5(2):47. 10.5539/jas.v5n2p47 Muema E, Mburu J, Coulibaly J, Mutune J (2018) Determinants of access and utilisation of seasonal climate information services among smallholder farmers in Makueni County, Kenya. Heliyon 4(11):e00889. 10.1016/j.heliyon.2018.e00889 Mugalavai EM Kipkorir EC Raes D Rao MS Analysis of rainfall onset, cessation and length of growing season for western Kenya Agric for Meteorol 2008 148 6–7 1123 1135 10.1016/j.agrformet.2008.02.013 Nicholls N (1999) Cognitive illusions, heuristics, and climate prediction. Bull Am Meteor Soc 80(7):1385. 10.1175/1520-0477 Nkomwa EC, Kalanda JM, Ngongondo C, Monjerezi M, Chipungu F (2014) Assessing indigenous knowledge systems and climate change adaptation strategies in agriculture: a case study of Chagaka Village, Chikhwawa, Southern Malawi. Phys Chem Earth, Parts a/b/c 67:164–172. 10.1016/j.pce.2013.10.002 Nkuba M Chanda R Mmopelwa G Mangheni MN Lesolle D Kato E Do indigenous forecasts and scientific forecasts influence arable farmers’ and agro-pastoralists’ estimation of onset and cessation of rains? Empirical evidence from Rwenzori region, Western Uganda Agric Forest Meteorol 2019 278 107667 10.1016/j.agrformet.2019.107667 Nkuba M Chanda R Mmopelwa G Kato E Mangheni MN Lesolle D Influence of indigenous knowledge and scientific climate forecasts on arable farmers’ climate adaptation methods in the Rwenzori region, Western Uganda Environ Manag 2020 10.1007/s00267-020-01264-x Nkuba M Chanda R Mmopelwa G Mangheni MN Lesolle D Kato E Indigenous knowledge systems and indicators of rain: evidence from Rwenzori Region, Western Uganda Weather, Climate, and Society 2020 12 2 213 234 10.1175/wcas-d-19-0027.1 Nkuba M Chanda R Mmopelwa G Mangheni MN Lesolle D Adedoyin A Mujuni G Determinants of pastoralists’ use of indigenous knowledge and scientific forecasts in Rwenzoriregion, WesternUganda Clim Serv 2021 23 100242 10.1016/j.cliser.2021a.100242 Nkuba M, Chanda R, Mmopelwa G, Kato E, Mangheni MN, et al (2021b) Indigenous and scientific forecasts on climate change perceptions of arable farmers: Rwenzori region, Western Uganda. Chapter 82. pg 1685–1703. In W.Leal Filho, N.Ogugu, L.Adelake,D.Ayal and I.Da Silva (Eds), African Handbook of climate change adaptation. Vol 2. Climate Change, Technologies, and Resources Management. Springer Nature Switzerland. 10.1007/978-3-030-45106-6_113 Nkuba M, Chanda R, Mmopelwa G, Kato E, Mangheni MN, et al (2021c) Barriers to Climate change adaptation among Pastoralists: Rwenzori region, Western Uganda. Chapter 35. pg. 659–676. In W Leal Filho, N Ogugu, L Adelake,D Ayal and I Da Silva (Eds), African Handbook of climate change adaptation. Vol 2. Climate Change, Technologies, and Resources Management. Springer Nature Switzerland. 10.1007/978-3-030-45106-6_99 Ofoegbu C New M The role of farmers and organizational networks in climate information communication: the case of Ghana Int J Clim Change Strat Manag 2021 13 1 19 34 10.1108/IJCCSM-04-2020-0030 Orlove B Roncoli C Kabugo M Majugu A Indigenous climate knowledge in southern Uganda: the multiple components of a dynamic regional system Clim Change 2010 100 243 265 10.1007/s10584-009-9586-2 Oyekale AS Factors explaining farm households’ access to and utilization of extreme climate forecasts in Sub-Saharan Africa (SSA) Environ Econ 2015 6 1 91 103 Partey ST Dakorah AD Zougmoré RB Ouédraogo M Nyasimi M Nikoi GK Huyer S Gender and climate risk management: evidence of climate information use in Ghana Clim Change 2018 158 61 75 10.1007/s10584-018-2239-6 Patt A, Gwata C (2002) Effective seasonal climate forecast applications: examining constraints for subsistence farmers in Zimbabwe. Glob Environ Chang 12(3):185–195. 10.1016/S0959-3780(02)00013-4 Patt A Suarez P Gwata C Effects of seasonal climate forecasts and participatory workshops among subsistence farmers in Zimbabwe Proc Natl Acad Sci USA 2005 102 35 12623 12628 10.1073/pnas.0506125102 16116076 Roncoli C Ingram K Kirshen P Reading the rains: local knowledge and rainfall forecasting in Burkina Faso Soc Nat Resour 2002 15 5 409 427 10.1080/08941920252866774 Roncoli C, Jost C, Kirshen P, Sanon M, Ingram KT, et al (2008) From accessing to assessing forecasts: an end-to-end study of participatory climate forecast dissemination in Burkina Faso (West Africa). Clim Change 92(3-4):433-460.10.1007/s10584-008-9445-6 Roudier P Muller B d’Aquino P Roncoli C Soumaré MA Batté L Sultan B The role of climate forecasts in smallholder agriculture: lessons from participatory research in two communities in Senegal Clim Risk Manag 2014 2 42 55 10.1016/j.crm.2014.02.001 Slovic P Slovic P Perception of risk The perception of risk 2000 London, UK Earthscan publications 220 231 Slovic P, Fischhoff B, Lichtenstein S (2000a) Cognitive processes and societal risk taking The Perception of Risk. Earthscan, London, UK, pp 32–50 Slovic P, Fischhoff B, Lichtenstein S (2000b) Facts and fears: understanding perceived risk. In: Slovic P (ed) Perception of Risk. Earthscan, London, UK, pp 137–153 Slovic P, White G, Kunreuther H (2000c) Decision processes, rationality, and adjustments to natural hazards. In: Slovic P (ed) Perception of Risk. Earthscan, London, UK, pp 1–31 Snow JT, Biagini B, Benchwick G, George G, Hoedjes J, et al (2016) A new vision for weather and climate services in Africa. Retrieved from New York Speranza CI Kiteme B Ambenje P Urs W Makali S Indigenous knowledge related to climate variability and change: insights from droughts in semi-arid areas of former Makueni District, Kenya Clim Change 2010 100 295 315 10.1007/s10584-009-9713-0 Steduto P, Hsiao TC, Fereres E, and Raes D (2012) Crop yield response to water: FAO Roma Tversky A, Kahneman D (1974) Judgment under uncertainty: heuristics and biases. Science 185(4157):1124–1131. 10.1126/science.185.4157.1124 Tversky A, Kahneman D (1979) Prospect theory: an analysis of decision under risk. Econometrica 47(2):263–291. 10.2307/1914185 UNECA (2011) Assessment of Africa’s climatic records and recording networks including strategic for rescuing of climatic data. Retrieved from Adisa Baba, United Nations Economic Commission for Africa; African Climate Policy Centre UNMA (2017a) March to May 2017a seasonal rainfall outlook over Uganda Kampala Uganda; Uganda National Meteorogical Authority Uganda National Meteorogical Authority UNMA (2017b) September to December 2017b Seasonal rainfall outlook over Uganda Kampala, Uganda Uganda National Meteorogical Authority Uganda National Meteorogical Authority Vogel C Usable science: an assessment of long-term seasonal forecasts amongst farmers in rural areas of Sourth Africa S Afr Geogr J 2000 82 2 107 116 10.1080/03736245.2000.9713700 Waldman KB Blekking JP Attari SZ Evans TP Maize seed choice and perceptions of climate variability among smallholder farmers Glob Environ Chang 2017 47 April 51 63 10.1016/j.gloenvcha.2017.09.007 Waldman KB Vergopolan N Attari SZ Sheffield J Estes LD Caylor KK Evans TP Cognitive biases about climate variability in smallholder farming systems in Zambia Weather, Climate, and Society 2019 11 2 369 383 10.1175/WCAS-D-18-0050.1 Yooyongwech S Theerawitaya C Samphumphuang T Cha-um S Water-deficit tolerant identification in sweet potato genotypes (Ipomoea batatas (L.) Lam.) in vegetative developmental stage using multivariate physiological indices Sci Hortic 2013 162 242 251 10.1016/j.scienta.2013.07.041 Ziervogel G (2004) Targeting seasonal climate forecasts for integration into household level decisions: the case of smallholder farmers in Lesotho. Geogr J 170(1):6–21. 10.1111/j.0016-7398.2004.05002.x Ziervogel G Calder R Climate variability and rural livelihoods: assessing the impact of seasonal climate forecasts in Lesotho Area 2003 35 4 403 417 10.1111/j.0004-0894.2003.00190.x Ziervogel G, Downing TE (2004) Stakeholder networks: improving seasonal climate forecasts. Clim Change 65(1–2):73–101. 10.1023/B:CLIM.0000037492.18679.9e Ziervogel G, Opere A (2010) Integrating meteorological and indigenous knowledge-based seasonal climate forecasts for the agricultural sector. http://hdl.handle.net/10625/46185	0	PMC9734385	NO-CC CODE	2022-12-14 23:28:28	no	Reg Environ Change. 2023 Dec 7; 23(1):4	utf-8	Reg Environ Change	2,022	10.1007/s10113-022-01994-0	oa_other
==== Front Reg Environ Change Reg Environ Change Regional Environmental Change 1436-3798 1436-378X Springer Berlin Heidelberg Berlin/Heidelberg 1994 10.1007/s10113-022-01994-0 Original Article Factors associated with farmers’ use of indigenous and scientific climate forecasts in Rwenzori region, Western Uganda http://orcid.org/0000-0002-6434-9100 Nkuba Michael Robert [email protected] 1 Chanda Raban 1 Mmopelwa Gagoitseope 1 Kato Edward 2 Mangheni Margaret Najjingo 3 Lesolle David 1 Adedoyin Akintayo 4 Mujuni Godfrey 5 1 grid.7621.2 0000 0004 0635 5486 Department of Environmental Sciences, University of Botswana, 4775 Notwane Road, Private Bag 00704, Gaborone, Botswana 2 grid.419346.d 0000 0004 0480 4882 International Food Policy and Research Institute, Washington, D.C USA 3 grid.11194.3c 0000 0004 0620 0548 Department of Extension and Innovation Studies, College of Agricultural and Environmental Sciences, Makerere University, Kampala, Uganda 4 grid.7621.2 0000 0004 0635 5486 Department of Physics, University of Botswana, Gaborone, Botswana 5 grid.463702.4 Uganda National Meteorological Authority, Kampala, Uganda Communicated by Shuaib Lwasa 7 12 2022 2023 23 1 42 9 2021 20 10 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. Although scientific climate forecast (SF) distribution by national climate services has improved over time, farmers seem not to make good use of climate forecasts, a likely contributing factor to vulnerability to climate change. This study investigated factors associated with farmers’ use of SFs and indigenous forecasts (IFs) for agricultural use in the Rwenzori region, western Uganda. Household survey gathered data on demographic characteristics, climate information use and livelihood choices from 580 farmers. Data was analysed using the probit model. Results showed that significant factors associated with using both IFs and SFs were farm size, education, age, reception of scientific forecasts in local languages, agricultural extension access, short-mature crop access, farmer-to-farmer network and accessing forecasts through radio. This study shows that IFs were used complementarily with SFs. On the other hand, significant factors associated with using IFs only were livelihood choices such as tuber and goat production, access to government interventions on climate change adaptations, agro-ecological zone and social capital. Climate risks and climate risk perceptions negatively influenced the use of scientific forecasts. Co-production of climate information, capacity-building and active engagement of stakeholders in dissemination mechanisms can improve climate forecast use. Investments in more weather stations in various districts will therefore be a key factor in obtaining more accurate scientific forecasts and could lead to increased use of scientific climate forecasts. Governments in developing countries, the private sector, global and regional development partners should support investments in weather stations and capacity building of national meteorological systems. Supplementary Information The online version contains supplementary material available at 10.1007/s10113-022-01994-0. Keywords Scientific climate forecasts Indigenous forecasts Farmers Co-production Cognitive bias Uganda issue-copyright-statement© Springer-Verlag GmbH Germany, part of Springer Nature 2023 ==== Body pmcIntroduction Although climate services in the developing world have improved over time, scientific climate forecasts’ (SFs) use has remained low (Nkuba et al. 2021a). SFs are accessible from radio, television, mobile phones (Jost et al. 2015a, b), the internet, agricultural extension agents and farmer-to-farmer networks (Goddard et al. 2010). Development agencies involved in disaster risk reduction and in rural and agriculture development such as World Vision, Care International, Save the Children and Oxfam have also played a role in the dissemination of SFs (ACCRA 2014). SFs are communicated in local languages in Senegal, Malawi and Tanzania (Hampson et al. 2014; Lo and Dieng 2015). Using appropriate local languages to disseminate SFs improves access and use among farmers (Jost et al. 2015a, b). The main challenge in using SFs by stakeholders such as farmers is their probabilistic nature (Nicholls 1999): what has been predicted may or may not happen. Decision-making based on forecasts with shortcomings raises serious concerns among arable farmers (hereafter referred to as farmers) (Goddard et al. 2001). The limited predictive accuracy negatively influences use of SFs by farmers. Goddard et al. (2001) found that obstacles to using SFs include meteorologists’ failure to provide full information about the predictive accuracy or reliability of the forecasts, and their interpretation as deterministic forecasts. SFs lack the spatial and temporal specificity that farmers are most interested in. The SFs are disseminated in terms of regions or districts which have wide geographical coverage. Thus, predicted weather events may occur in some areas but not in all. Simply producing and disseminating climate forecasts do not make them valuable to farmers. However, Patt et al. (2005) reported that the use of SFs improved crop yields in Zimbabwe. This implies that there is improved farmers’ welfare through utility of SFs. Farmers’ limited use of SFs has been associated with lack of saliency, credibility, trust and legitimacy (Patt and Gwata 2002; Cash et al. 2003, 2006; McNie 2007). Furthermore, farmers have raised as major shortcomings of SFs poor spatial and temporal resolution with regard to failure to provide forecasts on onset and cessation of rains (Kalanda-Joshua et al. 2011; Nkomwa et al. 2014). One of the factors that has led to low use of SFs in Africa is low meteorological station density (Medany et al. 2006; UNECA 2011), which has resulted in poor spatial resolution with negative credibility and trust implications. The meteorological station density in Africa is eight times lower than the World Meteorological Organisation’s (WMO) minimum recommended level (Medany et al. 2006); the distribution of the meteorological stations in Africa does not match the existing agro-ecological systems. Meteorological station density is one of the cornerstones of index-based weather insurance (IBWI) (Akter et al. 2016). Increased investment in rural weather station infrastructure not only improves the use of SFs but also increases the uptake of IBWI (Amare et al. 2019) meaning lending institutions may not be in be position to provide much needed insurance and credit facilities to African farmers due to lack of climate information. To address the poor station density in Africa, the United Nations Development Programme (UNDP) has supported a number of African countries by providing them with automatic weather stations (Snow et al. 2016). Besides station density, other challenges of climate services in Africa include dysfunctional stations, or poorly maintained, or low data quality due to low national budget allocation to national meteorological services in many developing countries (UNECA 2011; Snow et al. 2016). The Uganda National Meteorological Authority (UNMA) provides biannual climate forecasts on the onset and cessation of rains, 10-day rainfall forecasts and rainfall distribution (UNMA 2017b). With support from UNDP, there has been increased investment in weather infrastructure in Uganda (Snow et al. 2016). Alongside the high access to SFs (Jost et al. 2015a, b), the Rwenzori region has a low meteorological station density (Fig. 1), with negative implications for the use of SFs. The meteorological authority collaborates effectively with other agencies involved in weather data generation such as agricultural research stations, wildlife protected areas, national universities and private tea and sugar plantations in rural areas (Snow et al. 2016). Only 5% of rain gauges in Uganda are functional and another 921 rain gauges are needed to attain an optimum level (Isabirye 2017).Fig. 1 Study area map There are multiple sources of indigenous forecasts (IFs) (Roncoli et al. 2002; Kalanda-Joshua et al. 2011; Kolawole et al. 2014; Nkomwa et al. 2014). Sources of IFs include farmer-to-farmer extension network, farmers’ organisation, elderly farmers and farmer’s own observations (Appendix Table 3). There are 10 (abiotic and biotic environmental) indicators that are observed to provide weather and climate information (Nkuba et al., 2020b). These indicators are relevant to farmers in their localities providing forecasts at high spatial resolution ( Roncoli et al., 2002; Orlove et al. 2010; Nkomwa et al. 2014). The temporal dimension of IF provides relevant information on onset and cessation of rains, which is very important in farmers’ adaptation to changes in rainfall seasons (Roncoli et al. 2002; Nkomwa et al. 2014). This suggests that IF has high credibility, trust and legitimacy among farmers. Farmers who use IFs only have expressed interest in using SFs after interaction with meteorologists (Orlove et al. 2010; Mpandeli and Maponya 2013). Climate variability negatively influences the use of IFs (Roncoli et al. 2002; Speranza et al. 2010). In light of the challenges of using IFs exclusively or SFs exclusively, some farmers have resorted to using a combination of SFs and IFs since SFs complements IFs (Ziervogel and Opere 2010; Roncoli et al. 2008; Mpandeli and Maponya 2013). Research has shown that some farmers begin with using IFs and then revise their forecast partially or completely after receiving an SF (Lybbert et al. 2007). There are farmers who revise their IFs after access to SFs and there are farmers who use IFs only and who do not revise their forecasts even after accessing SFs. There is high access to SFs (89%) in the study area. This is attributed to the proliferation of FM radio stations and the use of local languages. Climate forecasts are used in farmers’ decision-making related to when to plant, harvest and what crops to grow for a given cropping season. The climate forecasts are essential in climate change adaptation (Nkuba et al. 2020a). Farmers use climate forecasts to reduce their vulnerability to the impacts of climate risks such as droughts and floods (Hansen 2002). In bimodal rainfall regions of East Africa, onset determines which cereals are to be grown, with maize being grown during periods of early onsets, while sorghum and millet are grown during periods of late onsets (Mugalavai et al. 2008). Forecasts thus influence the choice of crop enterprises for coming farming seasons. Maize is very sensitive to water stress. For sweet potatoes (Ipomoea batatas), potatoes (Solanum tuberosum) and cassava (Manihot esculenta), water stress leads to poor tuber formation (El-Sharkawy and Cadavid 2002; MacKerron and Jefferies 1986; Yooyongwech et al. 2013), sweet potato weevil infestation (Ebregt et al. 2004, 2007), fruit cracking and blotchy ripening for vegetables (Steduto et al. 2012). The farmers’ use of forecasts for variations in the crop-water requirements greatly influences choice of crop. Some studies have investigated factors associated with climate forecast use. A study done in Northern Kenya and Southern Ethiopia (Luseno et al. 2003) found that location, education level and access to radio are significant factors in the use of SFs. Another study done in Botswana (Kolawole et al. 2014) reported that age and education level are associated with the use of climate forecasts. The two studies did not include factors such as the language in which SFs are received, livelihood choices, the agro-ecological zone or access to agricultural extension, which the current study takes into account. Access to forecasts seems to have gender dimensions, with men reportedly having more access due to their control of the radio in the rural households (Jost et al. 2015a, b). Jost et al. (2015a, b) found that women in Uganda preferred to receive forecasts from gatherings and community radios, while television was the preferred option for rural women in Bangladesh. A study in south-eastern Kenya (Muema et al. 2018) showed that age, gender, drought frequency, radio and access to improved varieties were factors associated with the use of SFs. A study in eight African countries1 in East and West Africa (Oyekale 2015) showed that education level, climate shocks, radio and gender were factors associated with the use of SFs. These studies did not look at factors associated with farmers’ use of IFs only and both IFs and SFs, hence the need for this study. The above literature review has highlighted factors associated with use of climate forecasts by farmers. Nevertheless, knowledge gap regarding the effect of crop type, agro-ecological zones and climate risk perceptions on the use of IFs only or both IFs and SFs in farming still exists. The objective of this paper is to investigate factors associated with farmers’ use of IFs only or both IFs and SFs in the Rwenzori region in western Uganda. We answer the question regarding what factors are associated with farmers’ use of IF singly or the combination of both IF and SF. This gap has not been adequately addressed in the international literature. The consensus globally, regionally and nationally is that climate change is occurring and its negative effects have been acknowledged. Access to climate and weather information therefore becomes imperative to the farming communities which experience the direct vulnerabilities from climate change. In this study, we investigate what can be done to increase use of SFs among rural farmers in Uganda, where there is also a tendency of farmers to use their own non-scientifically based indigenous forecasts (IFs). This study has highlighted to policymakers, climate scientists, climate change activists and government funding partners that farmers who exclusively use SF forecast are very few, with majority of farmers still depending on IFs which are inaccurate and arbitrary. This might have the consequence of farmers not being able to respond to climate change risks appropriately and timely if there is still over reliance on IF forecasts. This study identifies the factors that can be targeted by government and its stakeholders to turn around this situation and increase usage of SF forecasts. This study has also revealed that possibly it would be hard to promote use of SFs only in rural communities but rather a hybrid between SFs and IFs might lead to wider usage of SFs. Theoretical framework This study investigated decision-making under uncertainty. Some scholars emphasise the consequentialist approach to decision-making, implying that decision choices have repercussions. The farmers who are the recipients of SFs may or may not use them. National meteorological systems are the principal providers of SFs. The farmers receive SFs which are used to examine decisions under uncertainty. In this study, UNMA disseminates SFs to the farming community. Some of the farmers may not use SFs even after receiving them but use IFs only. There is a budget allocation by the Ugandan government for gathering and dissemination of SFs and early-warning information by funding UNMA. Farmers who use information from meteorological stations are using posterior beliefs which may be objective. Information from meteorology stations can facilitate farmers to change their knowledge (prior beliefs) about climate. Farmers may update their prior beliefs (indigenous knowledge) with information from meteorologists (posterior beliefs) and may follow the recommendations from the experts. Alternatively, the farmers may not believe in the accuracy of the information from the meteorologists, and hence, they may not update their prior beliefs, resulting in their not following the recommendations from the experts. Farmers receiving SFs may change or revise their prior beliefs based on indigenous knowledge, underscoring the importance of confidence in forecast information received (Luseno et al. 2003). The meteorological stations make use of historical rainfall data and other climate parameters to forecast what the rainfall pattern will be in the coming days and months (seasonal climate forecast). Meteorologists go further in recommending what actions the farmers ought to take. However, the final decision on what actions to take lies within the farmers. The value and use of climate forecasts by farmers are based on the range of actions and their capability to respond. This study seeks to fill the knowledge gaps regarding the factors associated with farmers’ decision-making under uncertainty based on the climate forecasts (SFs and IFs) they used. The paper contributes to the current debate on influence of climate information on decision-making under uncertainty. Materials and methods The study area The study was conducted in the Rwenzori region of western Uganda. In terms of climate, the region experiences bimodal rainfall, which contributes to having two cropping seasons with the first season running from February to May and the second season from August to December. The temperature ranges from 12 to 24 °C, annual rainfall ranges from 800 to 3000 mm, and elevation ranges from 500 to 5000 m above sea level. Most of the farmers have completed primary education (Nkuba et al. 2020a). The farmers’ ability to comprehend climate information can be enhanced through stakeholder engagement by Uganda National Meteorological Authority (Nkuba et al. 2019). The soils are mainly sandy loam and sand clay loam which support production of cereals, tubers and vegetables (Nkuba et al. 2020a). Tree planting for commercial crop production of coffee, cocoa, fruit trees and commercial woodlots is very vibrant in the study area (Nkuba et al. 2020a). Access to pastures supports climate change adaptation measures for livestock farmers such as livestock migration, herd mobility and livestock diversification (Nkuba et al. 2021c). The region was selected because it has several agro-ecological zones—mountainous, lowland, mountainous and forested, wetland and forested—in order to investigate the effect of agro-climate on climate forecast use. Wildlife protected areas (WPA) in the study area (Kibale, Toro-Semiliki and Mount Rwenzori National parks) (Fig. 1) and Rwebitaba Zonal Agricultural Research Institute (in Kabarole district) have meteorological stations which provide climate data used in meteorological forecasts for the region. Weather stations are in government-aided establishments such as research stations and WPAs. Farmers in remote areas tend to find SF predictions to be inaccurate. Farming is a major livelihood source in the region. Access to SFs in local languages is due to the spread of FM radio and television stations. The region is inhabited by many ethnic groups, all of which use indigenous knowledge in their day-to-day lives. Data collection and analysis procedures Data was gathered from August to October 2015. A respondent survey using questionnaires was used to collect data on socio-economic factors, farm and institutional characteristics by trained research assistants. The respondent survey information was triangulated with data from focus group discussions (FGDs) and interviews. FGDs were used to get farmers’ views on the use of IFs and SFs. FGDs were held in Kyenjojo and Kabarole districts to get the farmers’ views on the use of IF and SF. A female FGD of 15 members and a male FGD of 16 members were carried out in Kyejonjo district. In Kabarole district, a female FGD of 15 members and male FGD of 17 members were conducted. The members were farmers who use IFs and SFs in their farming activities. To ensure that members of FGDs expressed their views freely, gender-segregated focus groups were used. The use of the mixed methods approach was for triangulation of the data obtained to enhance its validity and reliability. Data was analysed using Stata 16 statistical software. This study is nested in an earlier study (Nkuba et al. 2020a). A multi-stage stratified approach was used in the sampling of respondents. The first stage involved districts, the second stage counties, the third stage sub-counties and the smallest unit the household. The Uganda Bureau of Statistics disseminates the population data of households according to district, county, sub-county and household. The selection criteria were (i) farming systems, namely arable farming, pastoralism and agro-pastoralism and (ii) agro-ecological systems such as forested, lowlands, mountainous and wetlands. The statistically selected sample size of farmers was allocated to particular sub-counties in the selected districts using proportional allocation to size, where the size represents the number of households in the district. Based on a population of the study area of 102,496 households, according to the Uganda population census report of 2014, a sample size of 778 was selected with 95% confidence level and a margin of error of 3.5%. To allow for replacement in the sample of respondents who might drop out of the study, 19% of the statistically selected sample was included, giving a total study sample of 924. This was also to ensure a good sample size for sub-samples (for those who use both IFs and SFs, and IFs only). After data cleaning, 17 questionnaires were excluded due to incomplete responses. Of the 907 respondents, 580 were farmers, 270 pastoralists and 57 agro-pastoralists. This paper limits itself to the 580 farmers in the sample. Theoretical model The probit regression was used to analyse farmers’ use of forecasts. The probit model was used for examining the likelihood of a future climate event (for example rain onset and cessation) happening at a particular time using climate forecasts. There is no complete certainty regarding the occurrence of future climate outcomes that have been predicted. The predicted outcome may or may not happen. The predicted climate outcomes (dependent variables) in this study address use or non-use of IFs and SFs. IF and SF are specified as dichotomous outcomes with yes (use) and no (no-use), coded as 1 and 0 respectively. The probit model is a non-linear model that estimates with probabilistic maximum likelihood, often used for binary outcomes (Gujarati 2013). The regression results in this study are based on probit regression estimates. The probit model was estimated using Stata 16 software. There is potential for presence of self-selection in the use or non-use of forecasts. To overcome issues of selectivity convoluting the estimates, we have controlled for several conditioning factors that could be correlated with use of forecast and at the same time also influence outcomes of interest. We think this approach helps to minimise self-selection bias due to forecast use. We also tried to estimate models with and without forecasts to have a feel for the extent to which the estimates are being affected by selectivity. The estimates did not change when forecasts were excluded in the models. With these two approaches, we think that the possible threat from self-selection is not very serious in our analysis. We could use the instrumental variable approach because it was difficult to find strong excluded instruments that would satisfy all the instrument validity requirements. Empirical model The empirical model used in the analysis was specified as follows.1 Yij=β1Householdcharacteristics+β2Informationsources+β3Farmcharacteristics+β4Agro-ecologicalzone+β5SocialCapital+β6Percerptionofclimaterisk+β7Climaterisk+β8Wealth+ε For use of IF only2 Zij=β1Householdcharacteristics+β2Informationsources+β3Farmcharacteristics+β4Agro-ecologicalzone+β5SocialCapital+β6Percerptionofclimaterisk+β7Climaterisk+β8Wealth+ε For use of IF and SF where Yij(j = 1,2,3,4) representing the four models of using only IF and Zij(j = 1,2,3,4) representing the four models of using both IF and SF (Appendix Table 4). In this study, we control for several factors that could influence use of information as well as those factors that have been identified as affecting adoption of agricultural technologies in developing countries. Past empirical and theoretical literature has guided our theoretical choice of the factors that we include in our econometric models as explained below. Factors associated with farmers’ use of climate forecasts include education level, age, gender (Roncoli et al. 2002), livelihood choices (Vogel 2000; Ingrama et al. 2002; Patt and Gwata 2002; Ziervogel and Calder 2003; Crane et al. 2010; Haigh et al. 2015; Klemm and McPherson 2017), translation of SFs into local languages (Ingrama et al. 2002; Ziervogel and Downing 2004), source of forecasts for onset and cessation (Roncoli et al. 2002; Haigh et al. 2015), access to credit (Vogel 2000; Ingrama et al. 2002), access to non-farm enterprises (Ingrama et al. 2002; Ziervogel and Calder 2003; Crane et al. 2010; Klemm and McPherson 2017), access to agricultural extension (Vogel 2000; Ziervogel and Downing 2004; Coles and Scott 2009; Crane et al. 2010; Haigh et al. 2015; Klemm and McPherson 2017), access to improved crop varieties (Ingrama et al. 2002), climate risks (Ingrama et al. 2002), agro-ecological zone (Ingrama et al. 2002; Ziervogel 2004; Klemm and McPherson 2017), perception of climate risks (Vogel 2000; Ingrama et al. 2002; Ziervogel 2004; Haigh et al. 2015) and farm size (Vogel 2000; Coles and Scott 2009). Climate risks and perception of climate risks are important factors associated with farmers’ use of forecasts (Vogel 2000; Ingrama et al. 2002; Haigh et al. 2015). Farmers’ cognitive biases2 have an effect on their risk perceptions which influences climate forecasts use (Waldman et al. 2019). Unlike pastoralists, who practice herd mobility as an adaptive mechanism, farmers cannot migrate their crop fields. Changes in the onset and cessation of rains progress slowly towards their final manifestation in damage to crops (Slovic 2000a). The explanatory variables include Respondent characteristics (H): level of education, age, gender and farming experience; farm characteristics (F): type of crops grown and type of livestock kept; institutional characteristics (I): sources of IF, sources of SF, access to agricultural extension, credit access, improved crop access, non-farm access; agro-ecological system (A): forested, lowland, mountainous, wetland, mountainous and forested; social capital (S): membership of farmers’ organisation, farmer-to-farmer extension networks; perception of climate risks (P) drought increase, flood increase; climate risks (C): flood experience, drought experience; wealth (W): farm size. Variables and expected signs for use of IFs only and both IFs and SFs are shown in Appendix Table 5. Results Descriptive overview: use of forecasts Results show that almost half (49%) of the farmers used IFs only and half (50%) used both SFs and IFs (Appendix Table 6). Only one respondent used SF only. Over half (54%) of the respondents were male (Appendix Table 7). There were no significant differences in the use of either both IF and SF or IFS only between men and women respondents. Forecasts for rain onset and cessation were very crucial in farming. Considering the sub-sample of those who used IFs only, all the respondents used indigenous knowledge in predicting the onset of rains compared to 77% for the sub-sample of those who used both IFs and SFs. Most of the farmers (97%) reported that IF was reliable, compared to 43% for SF. Farmers who used both IFs and SFs reported that they used SFs to confirm the IFs based on what they had observed and learnt from elders. The most important crops grown were cereals and tubers (Appendix Table 3). Farmers’ organisations and farmer-to-farmer networks also played a crucial role in providing climate information (Appendix Table 3). Radio and non-government organisations were important SF dissemination channels (Appendix Table 8). Local FM stations provide SFs in local languages (Appendix Table 7), which has greatly improved access even to farmers with no formal education. Agricultural extension workers played a minimal role in the dissemination of SFs and information on rain season duration. Agricultural extension has been impacted by the involvement of Uganda Peoples Defence Force under Operation Wealth Creation, which plays a major role in the government distribution scheme3 for agricultural inputs such as improved seeds, seedlings and livestock during onset of rains. The results show that there were significant differences between farmers who used both IFs and SFs and those who used IFs only (Appendix Tables 3, 7, and 9 at 5% level of significance). Farmers from mountainous forested areas and bare mountainous areas were using IF only significantly more than both IFs and SFs (Appendix Table 7). Location and terrain appear to affect use of SFs, possibly because of differences in the level of infrastructural development between mountainous areas and lowlands. Farmers in lowlands were more likely to use SFs, while their counterparts in mountains were more likely to use IFs. This suggests constraints linked to quality of infrastructural development limiting access to SFs. Farmers growing cereals were using both IFs and SFs significantly less than IF only (Appendix Table 3). Cereals are very sensitive to water stress and are grown on a commercial scale in the study area. This increases the reliance on rain-fed agriculture in rural areas where there is limited use of irrigation, resulting in an increase in the use of SFs. Factors influencing farmers’ use of indigenous forecasts only The results show that factors positively and significantly associated with using IFs only among farmers were as follows: livelihood choices such as tubers and goat production, agro-ecology such as being resident in mountainous areas, reception of information from farmers’ organisations about the onset and cessation of rains access to government programme interventions on climate change adaptations and perceptions of climate variability and change as increases in floods and drought (Table 1). Factors negatively and significantly associated with using IFs only were non-farm access and livelihoods that depend on maize production.Table 1 Marginal effects of use of IFs only for arable farmers Variable Onset dy/dx Cessation dy/dx 5 day dy/dx Seasonal dy/dx Tubers 0.054(0.051) 0.214*(0.077) 0.201(0.079) 0.161**(0.074) Mountainous and forested 0.200(0.125) Mountainous 0.156(0.086) Farmers’ organisation as source of onset and cessation 0.264*(0.063) 0.306(0.082) 0.322*(0.095) 0.164(0.094) Credit access − 0.120(0.073) Non-farm access − 0.200*(0.078) Agricultural extension access − 0.170(0.095) Rainfall increase 0.108(0.083) Rainfall season change 0.120(0.079) Droughts increase 0.134(0.074) Floods increase 0.218(0.110) 0.244(0.116) 0.322(0.115) Access govt interventions on climate change adaptations 0.116(0.058) 0.176*(0.088) ln maize production − 0.017(0.010) − 0.025(0.015) − 0.036(0.016) − 0.022(0.015) ln local goat production 0.053(0.044) 0.073(0.046) 0.107(0.044) Log likelihood − 281.134 − 130.445 − 125.016 − 125.204 Pseudo R2 0.045 0.103 0.136 0.109 N 425 210 210 210 (Source: survey data 2015 overall title of the research project was ‘The Use of Indigenous and Scientific Forecasts in adaptation to climate variability and Change. A case study of Rwenzori region, Western Uganda’). Figures in parentheses are standard errors. *, and * denote that significance at the 1%, 5% and 10% levels respectively Government programme interventions on climate change adaptations include the provision of improved crop varieties and tree-planting materials to farmers, whose adoption depends on forecasts for the onset of rains. The agricultural inputs, such as seeds and seedlings, were delivered by National Agricultural Advisory Services, Operation Wealth Creation under the Office of the President and the Uganda Forestry Authority. Farmers had confidence in IFs and hence used it for predicting the onset of rains to effectively participate in government adaptation programmes. Cereals (especially maize), tubers and tree crops are sensitive to water stress, especially during the vegetative stage, and farmers’ confidence in IFs influenced their decisions on choice of crop enterprises and allocation of resources, depending on the onset and cessation forecasts. Farmers preferred to grow maize in the second rainy season, which is long enough and runs from August to November. Short-maturing cereals like sorghum and millet were preferred for the first rainy season, which is short and runs from March to May. A key informant reported that winds in the mountainous and forested areas on the windward side of Mount Rwenzori have high moisture content which easily reaches saturation point resulting in rainfall. Farmers in such areas rely a great deal on observing IF indicators like clouds and wind to get reliable forecasts. Factors influencing farmers’ farmers’ use of both indigenous forecasts and scientific forecasts The findings show that factors positively and significantly associated with using both IFs and SFs were reception of SFs in local language and English, attainment of higher education (diploma or advanced secondary school education), access to SFs through radio and TV, access to short-maturing crop varieties, availability of agricultural extension services, age and fellow farmer as source of SFs (Table 2).4 Factors negatively and significantly associated with using both IFs and SFs were livelihood that depend on vegetables, tubers and maize, being residents in mountainous forested areas, primary school education, Non-Governmental Organisation as source of onset and cessation forecasts and drought experience, and perceiving climate variability and change as seasonal rainfall change and drought increase (Table 2). Vegetable growers seemed not to have enough confidence in SFs as a basis for decision-making, probably because of vegetable crops’ water requirements. Command of English has a higher impact than local language on the use of SFs. Farmers who are comfortable with receiving SFs in English have fewer constraints than those who receive SFs in local languages. English may be more used by elite and well-to-do farmers. Farmers using local languages have less ability to access SFs compared to their counterparts who also use English.Table 2 Marginal effects of use of both IFs and SFs for arable farmers Variable Onset dy/dx Cessation dy/dx 5 day dy/dx Seasonal dy/dx Male 0.050(0.038) Vegetables − 0.107(0.051) − 0.131(0.053) − 0.134(0.044) Tubers − 0.127(0.045) Age 0.003(0.002) 0.004(0.002) Local language 0.464*(0.045) 0.402(0.058) 0.214(0.056) 0.201(0.047) English 0.577(0.063) 0.611(0.082) 0.367(0.115) 0.317(0.127) Mountainous and forested − 0.229(0.054) − 0.210(0.065) -0.089(0.047) Primary education − 0.135(0.053) -0.046(0.039) Higher education 0.375(0.163) 0.414*(0.144) Radio as source of scientific forecasts 0.249(0.090) 0.227(0.056) TV as source of scientific forecasts 0.113(0.090) 0.391(0.159) Fellow farmer as source of scientific forecasts 0.142(0.061) 0.152(0.066) 0.161*(0.060) NGO as source of onset and cessation forecasts − 0.339(0.040) − 0.264(0.052) − 0.208(0.047) Drought experience -0.098(0.057) − 0.126(0.071) Short mature crop access 0.156(0.063) 0.109(0.067) 0.094(0.053) Agricultural extension access 0.149(0.077) 0.180(0.089) Rainfall increase -0.058(0.041) Rainfall season change − 0.083(0.048) − 0.098(0.052) -0.053(0.040) Droughts increase − 0.127(0.054) − 0.151*(0.046) Unpredictable trains 0.051(0.041) ln maize production − 0.023(0.011) 0.017(0.010) -0.016(0.008) ln farm size 0.040(0.023) 0.027(0.026) 0.024(0.023) 0.019(0.019) ln likelihood − 270.210 − 194.653 − 195.332 -168.930 Pseudo R2 0.205 0.248 0.173 0.144 N 507 394 394 394 (Source: survey data 2015). Significant at 1%, significant at 5%, *significant at 10%. Figures in parentheses are standard errors There are few weather stations in the region (Fig. 1) whose rainfall data is used to make predictions, resulting in poor predictive accuracy and wide spatial variation for forecasts from UNMA. The rainfall data from the weather station in the mountainous area of Kabarole was used to make forecasts for the mountainous and forested area, quite far from the station.5 A key informant pointed out that there was apparently no weather station for rainfall data in the mountainous and forested area of Bundibugyo. This casts doubt on the reliability of SFs for seasonal forecasts for mountainous forested areas using extrapolated rainfall data from weather stations located in distant ecological zones. Forecasts were made in terms of regions and districts, yet farmers are interested in information at parish or village level. Receiving SFs in local languages made dissemination accessible to farmers, even to those with no formal schooling. Mountainous and forested areas have high precipitation in the form of mist and fog, which contributes positively to soil moisture availability. Climate risks like drought are negatively associated with the use SFs. It is probable that climate risks like droughts can influence farmers’ cognitive biases such as availability bias, anchoring effect and confirmation bias, which could impair farmers’ judgement in using SFs. Farmers can overestimate or underestimate the likelihood of droughts due to over-confidence in their recent memory of the severity and frequency of the occurrences. It is also possible to consider long dry spells as droughts. Climate risk perception such as drought increase and seasonal rainfall change was negatively associated with use of SFs. Climate risk perceptions can influence farmer cognitive biases such as framing effect, hindsight bias, anchoring effect and confirmation bias, which could impair farmers’ judgement in using SFs. The framing effect created by the limited coverage of dissemination mechanisms such as media about the effects of change in rain onset and cessation can lead farmers to underestimate the start and end of the rainfall season. Most media and meteorologists tend to focus more on rainfall distribution in terms of ‘above normal’ or ‘below normal’ in their dissemination of the seasonal climate forecast for the farming season, rather than duration of the rainy season. Farmers’ experiences can make them over-confident and consequently impair their judgement of frequency and severity of droughts. But there were also respondents who perceive that SFs are not credible, as exemplified by the following observation by a female participant in the FGD: ‘For me, I don’t think it’s truth because there is when they told us to prepare for heavy hail storms but they didn’t come’ (i.e., scientific forecasts are not reliable). There is also a lack of local specificity in SFs, rendering them less credible to the farmers. One FGD participant, for instance, remarked that ‘You can see that it [rain] has not happened here but when in another place it has’. Some doubt was also cast on the legitimacy of SFs, as indicated by one participant who declared that ‘They also just guess’. This indicates that farmers do not trust the processes that meteorologists use to make forecasts, and consequently, they distrust the forecasts as well. Farmers attached high credibility to IFs because of the long tradition of using and depending on them, as is evidenced by the following statement from a FGD participant: ‘from when we were born until we have grown to now it’s what we found our elders using. The elders before us this [IFs] is what they were using, their knowledge of long ago. For that we also grow up following it’. IFs knowledge is based on the long-term observation of nature and is passed on orally from one generation to another. There were 15 indicators used in IF identified in the study area (Nkuba et al. 2020b) implying that there is a wide variety of sources of forecasts, resulting in increased legitimacy. The indicators are biotic factors such as plants, insects and birds and abiotic factors such as wind and clouds. The abiotic factors are similar to what meteorologists use in their predictions. Farmers’ organisations and elderly farmers were social capital that provide forecast information within their specific local area. Religion also plays a role in the use of IFs; for example, many respondents said that the onset of rains corresponds to the day of Mother Mary, namely 18 August. One participant in the FGD reported that ‘Indigenous knowledge is a very important thing because that’s the knowledge God created us with. Because even what is taught us [scientific forecasts]… you would learn and by the time you reach home you have forgotten it. But indigenous knowledge is better. Because with scientific forecasts, someone will guess what’s not there’. Radio is the most widely used method of disseminating information about SFs. Radio stations broadcast in both English and local languages. TV is widely used by elite and well-to-do farmers. TV stations like National TV and Uganda Broadcasting Services are widely used as sources of forecasts by wealthier farmers. A key informant in the civil service stated that agricultural extension workers receive SFs from UNMA through the Ministry of Agriculture on a regular basis. There are factors that influence the use of both IFs and SFs, and IFs only, differently (Tables 1 and 2). For instance, access to improved crop varieties and agricultural extension services is positively associated with using both IFs and SFs and negatively associated with using IFs only. Access to agricultural extension is closely associated with SFs. Maize production was positively associated with use of both IFs and SFs but negatively associated with IFs for the 5-day forecasts. Short-range SFs such as 5-day forecasts provide reliable climate information (dry spells, or optimum planting and harvesting days) which is relevant to farmers’ involvement in water stress crops like maize. Being a resident in mountainous and forested areas, and perceiving climate variability and change as drought increase and seasonal rainfall change, was positively associated with using IFs only and negatively associated with using IFs and SFs. Mountainous and forested areas are positively associated with high precipitation due to orographic effects on cloud formation and condensation, contributing to less variability in rainfall. Having a higher level of education (secondary education and diploma) was positively associated with use of IFs and SFs. Increase in education increases uptake of SFs. However, primary education was negatively associated with use of IFs and SFs. Primary education probably causes the farmers to overestimate or underestimate due to over-confidence based on their recent past experiences. Reception of SFs from fellow farmers were positively associated with using IFs and SFs, while reception of onset and cessation of rain forecasts from farmers’ organisations was positively associated with using IFs only. Group decisions were more trusted in using IFs only. Confidence in using IFs only is built up through social gatherings of like-minded people. Information disseminated through farmer-to-farmer networks was more trusted in using SFs. This could arise from the confidence that the farmer-to-farmer networks built over time in using SFs with good outcomes. Discussion Results showed that significant factors associated with using both IFs and SFs were farm size, education, age, reception of scientific forecasts in local languages, agricultural extension access, short-mature crop access, farmer-to-farmer network and accessing forecasts through mass media. This study shows that IFs were used complementarily with SFs. On the other hand, significant factors associated with using IFs only were livelihood choices such as tuber and goat production, access to government interventions on climate change adaptations, agro-ecological zone and social capital. Research has shown that dissemination of SFs through local languages is the most preferred among rural households (Antwi-Agyei et al. 2021). Dissemination in local languages is usually facilitated by media such as local FM radio stations. The current study has also confirmed that radio and television play a crucial role in the dissemination of SFs. Radio has been reported to be an important dissemination mechanism for SFs in rural regions (Jost et al 2015a, b). This indicates that widening FM radio station broadcasting coverage in Uganda could provide an opportunity for increasing the utilisation of SFs in rural areas. The main shortcoming of using such media as radio and television for dissemination is that there is usually no possibility for feedback from farmers. This is perhaps one of the reasons that farmers seem to have greater confidence in forecasts from their fellow farmers: there is feedback and discussion of experiences and the implications of using such forecasts. National meteorological services need to employ dissemination mechanisms that provide feedback from farmers for improved effectiveness in the use of forecasts. The findings of the current study also show that the onset and cessation of rains forecasts from farmer-to-farmer extension networks had mixed results, with a positive association with use of IFs only and a negative association with use of both IFs and SFs. This shows that farmers had more confidence in IFs than in SFs because IFs have local specificity while SFs have broad spatial and temporal variations (Speranza et al. 2010). Farmers’ indigenous knowledge of climate forecasts is more trusted than forecasts by climate scientists (Kolawole et al. 2014; Roudier et al. 2014). The results of the current study have also shown that farmers have negative associations with use of forecasts from NGOs. The challenge is that NGO staff operating in rural areas may not be able to answer all the farmers’ questions about forecasts, which may engender a lack of trust in the forecasts among farmers (Ofoegbu and New 2021). This study shows that farmers involved in tuber production are more associated with use of IFs only. Farmers involved in tuber and maize production do not have confidence in SFs because of past experiences of poor spatial and temporal specificity and, therefore, in their eyes, of false forecasts (Ziervogel 2004). Our study shows mixed results regarding the association of education with climate forecast use. Post primary education was positively associated with use of SFs. Luseno et al. (2003) study in Kenya and Ethiopia showed that formal education improved confidence in and access to SFs. Similarly, a study done in Botswana showed that the more educated a farmer is, the more likely he/she will be to use SFs received through the media (Kolawole et al. 2014). However, primary education was negatively associated with the use of SFs. A study done is Zambia reported that one additional year in education reduced farmers’ perception of change in rain onset due to psychological factors (Waldman et al. 2019). This calls for further research on the effect of psychological factors on climate forecast use and a multidisciplinary approach for climate services research. Climate change perceptions have a mixed effect on the use of forecasts, with a positive association with use of IFs and a negative association with use of both IFs and SFs. Experienced farmers make adjustments in their livelihoods after experiencing a drought, or they make changes in their responses to the onset and cessation of rains because of the likelihood of a recurrence (Slovic et al. 2000b). Farmers experience high crop losses due the change in rainfall duration and severe droughts, thus increasing their confidence in using IFs only. The range of alternatives for farmers is limited, which increases the level of catastrophe to their livelihoods (Slovic et al. 2000c). This also influences the use of IFs only, in order to avoid or to minimise experiencing such risks as loss of seeds due to late onset of rains, or poor harvests due to early cessation of rainfall. Farmers’ assessment of risks, which influences their use of IFs, is related to climate risks that reduce household welfare due to crop yield losses (Slovic 2000, 2000a, 2000b, 2000c). However, the perception of unpredictable rains as climate change is associated with using both. This is consistent with Partey et al. (2018),who found that increased rainfall variability positively influences the use of climate information in Ghana. The seasonal climate and short-range forecasts from UNMA provide predictions of rainfall distribution which influence risk perceptions (UNMA 2017a), as illustrated by a participant in a FGD, who said that ‘Now like this season when they say the rains are going to be a lot, we become cautious of the floods’. This indicates that SFs are also positively associated with risk perceptions and climate change perceptions. The results from the FGDs show that SFs are associated with increased crop yields which is consistent with findings from studies done in Burkina Faso and Zimbabwe (Patt et al. 2005; Roncoli et al. 2008). Research shows that farmers who feel they have adaptive capacity for a particular climate risk (availability heuristic) tend to have lower the climate risk perception (Duinen et al. 2015) which can lead underestimating its effects (Tversky and Kahneman 1979). Farmers with drought experiences have hindsight bias leading to increased confidence in their degree of prediction of future droughts without using SFs. Drought increase is a dreaded risk whose calamities are known through loss of crop yields and change in rainfall season is a delayed risk with a slow manifestation of harm. The recent memory of climate risks like rainfall seasonal change and drought could influence the use of SFs due to hindsight bias and anchoring effect. The framing effect from various stakeholders and media can result in making rainfall season change look less dangerous than gradual onset of droughts, leading to bias in using SFs. Research has shown that maize farmers have biases related to rain onset (Waldman et al. 2017), which is in agreement with our findings. Research has shown that farmers’ climate risk perception and past experience can lead to poor decision-making related to climate forecast use, due to cognitive biases (Waldman et al. 2019). Farmers from mountainous forested areas, and mountainous areas, are positively associated with use of IFs only. Due to orographic factors, the wide spatial variations in rainfall in these agro-ecological zones cause farmers to have more confidence in IFs only due to IFs’ local specificity compared to SFs, with SFs’ low spatial coverage because of low meteorological station density, especially in the mountainous areas (Sultan et al. 2020). Furthermore, Leroux (2001) found that ‘with increasing altitude the nature of precipitation changes; raindrops become smaller, showers give way to more continuous rain, and thence to rain, drizzle, mist and fog’ (p. 50). Forests act as windbreakers which disturb surface circulation and have a greenhouse effect that reduces temperature variations (Leroux 2001). Relative humidity in forested areas is high and this consequently leads to forests having a strong influence on rainfall over wide areas (Leroux 2001). The use of both SFs and IFs can be enhanced by experts engaging in constant dialogue with end-users. Studies in Mozambique, Kenya, Ivory Coast, Senegal, Tanzania and Benin have shown that confidence in SFs greatly improved when national meteorological services received constructive feedback from stakeholders such as farmers (Ziervogel and Opere 2010; Lo and Dieng 2015). Research has also shown that the legitimacy, salience and credibility of SFs have been improved by the active engagement of farmers, resulting in farmers’ increased use of SFs (Bouroncle et al. 2019). Integration of SFs and IFs is critical for the effective utilisation of forecasts in rural livelihoods based on farming. Ziervogel and Opere (2010) have provided information on projects in Africa that had integrated to good effect meteorological and indigenous-based seasonal forecasts in the agricultural sector. Stakeholder engagement can be in form of farm rainfall data generation by farmers. Botswana’s Department of Meteorological services has installed rain-gauges for farmers who send daily rainfall data via telephone. Farmers involved in farm rainfall data generation get excited when rainfall data from their farms is read during the news bulletin on national television station. Botswana meteorologists conducted meteorology extension services during the agricultural shows and held feedback sessions with farmers involved in rainfall data generation before the COVID-19 pandemic. A pilot study on citizen science in Namibia and South Africa has shown that a farmer involved farm rainfall data generation had much trust in SF, constructive engagement with meteorologists, reduced vulnerability to climate change, improved crop yields, livestock sales and household incomes (Landman et al. 2020). The results of the study are applicable to humid tropical zones and may not be applicable to a semi-arid area. This calls for further research regarding factors associated with use of indigenous and scientific forecasts in semi-arid areas. Conclusion and recommendations This study has established that farming in the Rwenzori region is informed by the use of both IFs and SFs or IFs only. Factors that promote the use of SFs include post-primary school education, access to appropriate rural institutions and dissemination of forecasts through radio using local languages. However, primary education was negatively associated with use of SFs. This is attributed to farmers’ cognitive biases. This calls for further research on the effect of psychological factors on climate forecast use. Research on climate information use requires a transdisciplinary approach involving environmental psychologists, social scientists, agronomists and meteorologists. Farmer-to-farmer extension networks are positively associated with use of IFs only and negatively associated with the use of both IFs and SFs. This calls for active stakeholders’ engagement by national meteorological services. This will improve the salience, legitimacy and credibility of meteorological forecasts in rural areas. Climate risk perceptions and climate risks are negatively associated with the use of SFs. Addressing farmers’ cognitive biases associated with climate risks and climate risk perceptions by national meteorological systems actively engaging stakeholders in climate forecast dissemination mechanisms could improve uptake of SFs. Investments in ddissemination of weather forecasts should be maintained and or increased, as the study has shown that television and radio have a positive impact. SFs were found to be complementary to IFs. SFs reinforce IFs. This calls for co-production of climate information in order to promote an increased use of forecasts in rural areas. Indigenous forecasts will continue to play a major role in influencing land-use interventions among farmers in Uganda and sub-Saharan Africa. We do not contest the use of SFs among farmers because trust in SFs among farmers is likely to improve with an increase in the number and density of weather stations in rural areas. Investment in more weather stations (automatic weather stations and rain gauges) in farming areas is a key factor in obtaining more spatially specific and accurate SFs. This could result in the improved use of SFs which might lead to improved food security and reduce vulnerability to climate change. Governments in developing countries, the private sector and the global and regional development partners should support investments in weather stations and capacity building of national meteorological systems. These possibilities are of course influenced not only by the use of forecasts but also by a number of other equally important factors such as access to inputs, agricultural extension services and credit. A longitudinal study on the validity of indigenous forecasts should be explored. In places with bimodal rainfall distribution, data has to be collected for both seasons in a period of 3 years or more to get meaningful outcomes. For comparative purposes, similar research should be conducted in other agro-ecological zones such as semi-arid areas and temperate areas, which were not covered in this study. The study has also established that farmers use scientific and indigenous forecasts in making decisions under uncertainty. Some farmers receive SFs but do not use them, instead using IFs only. Some receive SFs and update them with IFs, resulting in using both indigenous and scientific forecasts. Supplementary Information Below is the link to the electronic supplementary material. ESM 1 (DOC 135 kb) 1 Burkina Faso, Senegal, Mali, Niger, Ghana, Kenya, Tanzania and Ethiopia. 2 Cognitive biases include hindsight bias, confirmation bias, framing effect, anchoring effect, availability bias and decision regret effect (Tversky and Kahneman 1974; Kahneman and Tversky 1982; Nicholls 1999). 3 Media reports have indicated farmers rejecting agricultural inputs because they were supplied without taking into account rain onset. 4 The independent variables in the two models (Tables 1 and 2) differ because of precision concerns. Variables with standard errors larger than the coefficients indicate a poor estimation (Gujarati 2013) and were therefore not included in the models. 5 The distance from Kabarole to Bundibugyo is 24 km. Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References ACCRA (2014) Planning for the future and adapting to climate change in Uganda. Lessons from ACCRA In A. C. C. R. Alliance (Ed.), (pp. 4). Kampala, Uganda Africa Climate Change Resilience Alliance Akter S, Krupnik TJ, Rossi F, Khanam F (2016) The influence of gender and product design on farmers’ preferences for weather-indexed crop insurance. Glob Environ Chang 38:217–229. 10.1016/j.gloenvcha.2016.03.010 Amare A, Simane B, Nyangaga J, Defisa A, Hamza D, et al (2019) Index-based livestock insurance to manage climate risks in Borena zone of southern Oromia, Ethiopia. Clim Risk Manag 25:100191. 10.1016/j.crm.2019.100191 Antwi-Agyei P, Dougill AJ, Abaidoo RC (2021) Opportunities and barriers for using climate information for building resilient agricultural systems in Sudan savannah agro-ecological zone of north-eastern Ghana. Clim Serv 22.10.1016/j.cliser.2021.100226 Bouroncle C Müller A Giraldo D Rios D Imbach P Girón E Portillo F Boni A van Etten J Ramirez-Villegas J A systematic approach to assess climate information products applied to agriculture and food security in Guatemala and Colombia Climate Services 2019 16 October 100137 10.1016/j.cliser.2019.100137 Cash DW, Borck JC, Patt AG (2006) Countering the loading-dock approach to linking science and decision making: comparative analysis of El Niño/Southern Oscillation (ENSO) forecasting systems. Sci Technol Human Values 31(4):465–494. 10.1177/0162243906287547 Cash DW, Clark WC, Alcock F, Dickson NM, Eckley N, et al (2003) Knowledge systems for sustainable development. Proc Natl Acad Sci 100(14):8086–8091. 10.1073/pnas.1231332100 Coles AR, Scott CA (2009) Vulnerability and adaptation to climate change and variability in semi‐arid rural southeastern Arizona, USA. Nat Resour Forum 33(4):297–309. https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1477-8947.2009.01253.x Crane TA Roncoli C Paz J Breuer N Broad K Ingram KT Hoogenboom G Forecast skill and farmers’ skills: seasonal climate forecasts and agricultural risk management in the southeastern United States Weather, Clim Soc 2010 2 1 44 59 10.1175/2009WCAS1006.1 Duinen R Filatova T Geurts P van der Veen A Empirical analysis of farmers’ drought risk perception: objective factors, personal circumstances, and social influence Risk Anal 2015 35 4 741 755 10.1111/risa.12299 25514996 Ebregt E Struik PC Abidin PE Odongo B Farmers' information on sweet potato production and millipede infestation in north-eastern Uganda. I. Associations between spatial and temporal crop diversity and the level of pest infestation NJAS - Wageningen J Life Sci 2004 52 1 47 68 10.1016/S1573-5214(04)80029-0 Ebregt E Struik PC Odongo B Abidin PE Piecemeal versus one-time harvesting of sweet potato in north-eastern Uganda with special reference to pest damage NJAS - Wageningen J Life Sci 2007 55 1 75 92 10.1016/S1573-5214(07)80005-4 El-Sharkawy MA, Cadavid L (2002) Response of cassava to prolonged water stress imposed at different stages of growth. Exp Agric 38(03):333–350. 10.1017/S001447970200306X Goddard L, Mason SJ, Zebiak SE, Ropelewski CF, Basher R, et al (2001) Current approaches to seasonal to interannual climate predictions. Int J Climatol 21(9):1111–1152. 10.1002/joc.636 Goddard L, Aitchellouche Y, Baethgen W, Dettinger M, Graham R, et al (2010) Providing seasonal-to-interannual climate information for risk management and decision-making. Procedia Environ Sci 1:81–101. 10.1016/j.proenv.2010.09.007 Gujarati DN Basic Econometrics. Boston 2013 United States of America McGraw-Hill companies Haigh T Takle E Andresen J Widhalm M Carlton JS Angel J Mapping the decision points and climate information use of agricultural producers across the US Corn Belt Clim Risk Manag 2015 7 20 30 10.1016/j.crm.2015.01.004 Hampson KJ, Chapota R, Emmanuel J, Tall A, Huggins-Rao S, et al (2014) Delivering climate services for farmers and pastoralists through interactive radio: scoping report for the GFCS Adaptation Programme in Africa. Retrieved from CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS)Copenhagen, Denmark. https://hdl.handle.net/10568/65728 Hansen JW Realizing the potential benefits of climate prediction to agriculture: Issues, approaches, challenges Agric Syst 2002 74 3 309 330 10.1016/S0308-521X(02)00043-4 Ingrama KT, Roncolia MC, Kirshen PH (2002) Opportunities and constraints for farmers of west Africa to use seasonal precipitation forecasts with Burkina Faso as a case study. Agric Syst 74:331–349. 10.1016/S0308-521X(02)00044-6 Isabirye P (2017) Factoring spatial variability of rainfall in the design of optimum rainguage network for Uganda. Thesis for Master of science for climate change and Disater Management. Busitema University, Uganda. https://ir.busitema.ac.ug/bitstream/handle/20.500.12283/1898/Isabirye%20-%20Abstract.pdf?sequence=2&isAllowed=y Jost C, Kyazze F, Naab J, Neelormi S, Kinyangi J, et al (2015a) Understanding gender dimensions of agriculture and climate change in smallholder farming communities. Clim Dev 8(3):133-144.10.1080/17565529.2015.1050978 Jost C, Kyazze F, Naab J, Neelormi S, Kinyangi J, et al (2015b) Understanding gender dimensions of agriculture and climate change in smallholder farming communities. Clim Dev 8(2):1-12.10.1080/17565529.2015.1050978 Kahneman D, Tversky A (1982) The psychology of preferences. Scientific American 246:160–73 Kalanda-Joshua M Ngongondo C Chipeta L Mpembeka F Integrating indigenous knowledge with conventional science: enhancing localised climate and weather forecasts in Nessa, Mulanje, Malawi Phys Chem Earth, Parts a/b/c 2011 36 14–15 996 1003 10.1016/j.pce.2011.08.001 Klemm T McPherson RA The development of seasonal climate forecasting for agricultural producers Agric for Meteorol 2017 232 384 399 10.1016/j.agrformet.2016.09.005 Kolawole OD, Wolski P, Ngwenya B, Mmopelwa G (2014) Ethno-meteorology and scientific weather forecasting: small farmers and scientists’ perspectives on climate variability in the Okavango Delta, Botswana. Clim Risk Manag 4:43–58. 10.1016/j.crm.2014.08.002 Landman WA Archer ERM Tadross MA Citizen science for the prediction of climate extremes in South Africa and Namibia Frontiers in Climate 2020 2 1 8 10.3389/fclim.2020.00005 Leroux M The meteorology and climate of tropical Africa 2001 UK Springer Science & Business Media, UK Liebenehm S, Waibel H (2014) Simultaneous estimation of risk and time preferences among small-scale cattle farmers in West Africa. Am J Agr Econ 96(5):1420–1438. 10.1093/ajae/aau056 Lo H, Dieng M (2015) Impact assessment of communicating seasonal climate forecasts in Kaffrine, Diourbel, Louga, Thies and Fatick (Niakhar) regions in Senegal: Final Report for CCAFS West Africa Regional Program. Retrieved from Copenhagen, Denmark. http://bit.ly/1ifHC3P Luseno WK Mcpeak JG Barrett CB Little PD Gebru G Assessing the value of climate forecast information for pastoralists: evidence from Southern Ethiopia and Northern Kenya World Dev 2003 31 9 1477 1494 10.1016/S0305-750X(03)00113-X Lybbert TJ Barrett CB Mcpeak JG Luseno WK Bayesian Herders: Updating of Rainfall Beliefs in Response to External forecasts World Dev 2007 35 3 480 497 10.1016/j.worlddev.2006.04.004 MacKerron D, Jefferies R (1986) The influence of early soil moisture stress on tuber numbers in potato. Potato Res 29(3):299–312. 10.1007/BF02359959 McNie EC (2007) Reconciling the supply of scientific information with user demands: an analysis of the problem and review of the literature. Environ Sci Policy 10(1):17–38. 10.1016/j.envsci.2006.10.004 Medany M Niang-Diop I Nyong T Tabo R Background paper on impacts, vulnerability and adaptation to climate change in Africa 2006 Ghana Paper presented at the UNFCCC Convention Mpandeli S, Maponya P (2013) The Use of Climate Forecasts Information by Farmers in Limpopo Province, South Africa. J Agric Sci 5(2):47. 10.5539/jas.v5n2p47 Muema E, Mburu J, Coulibaly J, Mutune J (2018) Determinants of access and utilisation of seasonal climate information services among smallholder farmers in Makueni County, Kenya. Heliyon 4(11):e00889. 10.1016/j.heliyon.2018.e00889 Mugalavai EM Kipkorir EC Raes D Rao MS Analysis of rainfall onset, cessation and length of growing season for western Kenya Agric for Meteorol 2008 148 6–7 1123 1135 10.1016/j.agrformet.2008.02.013 Nicholls N (1999) Cognitive illusions, heuristics, and climate prediction. Bull Am Meteor Soc 80(7):1385. 10.1175/1520-0477 Nkomwa EC, Kalanda JM, Ngongondo C, Monjerezi M, Chipungu F (2014) Assessing indigenous knowledge systems and climate change adaptation strategies in agriculture: a case study of Chagaka Village, Chikhwawa, Southern Malawi. Phys Chem Earth, Parts a/b/c 67:164–172. 10.1016/j.pce.2013.10.002 Nkuba M Chanda R Mmopelwa G Mangheni MN Lesolle D Kato E Do indigenous forecasts and scientific forecasts influence arable farmers’ and agro-pastoralists’ estimation of onset and cessation of rains? Empirical evidence from Rwenzori region, Western Uganda Agric Forest Meteorol 2019 278 107667 10.1016/j.agrformet.2019.107667 Nkuba M Chanda R Mmopelwa G Kato E Mangheni MN Lesolle D Influence of indigenous knowledge and scientific climate forecasts on arable farmers’ climate adaptation methods in the Rwenzori region, Western Uganda Environ Manag 2020 10.1007/s00267-020-01264-x Nkuba M Chanda R Mmopelwa G Mangheni MN Lesolle D Kato E Indigenous knowledge systems and indicators of rain: evidence from Rwenzori Region, Western Uganda Weather, Climate, and Society 2020 12 2 213 234 10.1175/wcas-d-19-0027.1 Nkuba M Chanda R Mmopelwa G Mangheni MN Lesolle D Adedoyin A Mujuni G Determinants of pastoralists’ use of indigenous knowledge and scientific forecasts in Rwenzoriregion, WesternUganda Clim Serv 2021 23 100242 10.1016/j.cliser.2021a.100242 Nkuba M, Chanda R, Mmopelwa G, Kato E, Mangheni MN, et al (2021b) Indigenous and scientific forecasts on climate change perceptions of arable farmers: Rwenzori region, Western Uganda. Chapter 82. pg 1685–1703. In W.Leal Filho, N.Ogugu, L.Adelake,D.Ayal and I.Da Silva (Eds), African Handbook of climate change adaptation. Vol 2. Climate Change, Technologies, and Resources Management. Springer Nature Switzerland. 10.1007/978-3-030-45106-6_113 Nkuba M, Chanda R, Mmopelwa G, Kato E, Mangheni MN, et al (2021c) Barriers to Climate change adaptation among Pastoralists: Rwenzori region, Western Uganda. Chapter 35. pg. 659–676. In W Leal Filho, N Ogugu, L Adelake,D Ayal and I Da Silva (Eds), African Handbook of climate change adaptation. Vol 2. Climate Change, Technologies, and Resources Management. Springer Nature Switzerland. 10.1007/978-3-030-45106-6_99 Ofoegbu C New M The role of farmers and organizational networks in climate information communication: the case of Ghana Int J Clim Change Strat Manag 2021 13 1 19 34 10.1108/IJCCSM-04-2020-0030 Orlove B Roncoli C Kabugo M Majugu A Indigenous climate knowledge in southern Uganda: the multiple components of a dynamic regional system Clim Change 2010 100 243 265 10.1007/s10584-009-9586-2 Oyekale AS Factors explaining farm households’ access to and utilization of extreme climate forecasts in Sub-Saharan Africa (SSA) Environ Econ 2015 6 1 91 103 Partey ST Dakorah AD Zougmoré RB Ouédraogo M Nyasimi M Nikoi GK Huyer S Gender and climate risk management: evidence of climate information use in Ghana Clim Change 2018 158 61 75 10.1007/s10584-018-2239-6 Patt A, Gwata C (2002) Effective seasonal climate forecast applications: examining constraints for subsistence farmers in Zimbabwe. Glob Environ Chang 12(3):185–195. 10.1016/S0959-3780(02)00013-4 Patt A Suarez P Gwata C Effects of seasonal climate forecasts and participatory workshops among subsistence farmers in Zimbabwe Proc Natl Acad Sci USA 2005 102 35 12623 12628 10.1073/pnas.0506125102 16116076 Roncoli C Ingram K Kirshen P Reading the rains: local knowledge and rainfall forecasting in Burkina Faso Soc Nat Resour 2002 15 5 409 427 10.1080/08941920252866774 Roncoli C, Jost C, Kirshen P, Sanon M, Ingram KT, et al (2008) From accessing to assessing forecasts: an end-to-end study of participatory climate forecast dissemination in Burkina Faso (West Africa). Clim Change 92(3-4):433-460.10.1007/s10584-008-9445-6 Roudier P Muller B d’Aquino P Roncoli C Soumaré MA Batté L Sultan B The role of climate forecasts in smallholder agriculture: lessons from participatory research in two communities in Senegal Clim Risk Manag 2014 2 42 55 10.1016/j.crm.2014.02.001 Slovic P Slovic P Perception of risk The perception of risk 2000 London, UK Earthscan publications 220 231 Slovic P, Fischhoff B, Lichtenstein S (2000a) Cognitive processes and societal risk taking The Perception of Risk. Earthscan, London, UK, pp 32–50 Slovic P, Fischhoff B, Lichtenstein S (2000b) Facts and fears: understanding perceived risk. In: Slovic P (ed) Perception of Risk. Earthscan, London, UK, pp 137–153 Slovic P, White G, Kunreuther H (2000c) Decision processes, rationality, and adjustments to natural hazards. In: Slovic P (ed) Perception of Risk. Earthscan, London, UK, pp 1–31 Snow JT, Biagini B, Benchwick G, George G, Hoedjes J, et al (2016) A new vision for weather and climate services in Africa. Retrieved from New York Speranza CI Kiteme B Ambenje P Urs W Makali S Indigenous knowledge related to climate variability and change: insights from droughts in semi-arid areas of former Makueni District, Kenya Clim Change 2010 100 295 315 10.1007/s10584-009-9713-0 Steduto P, Hsiao TC, Fereres E, and Raes D (2012) Crop yield response to water: FAO Roma Tversky A, Kahneman D (1974) Judgment under uncertainty: heuristics and biases. Science 185(4157):1124–1131. 10.1126/science.185.4157.1124 Tversky A, Kahneman D (1979) Prospect theory: an analysis of decision under risk. Econometrica 47(2):263–291. 10.2307/1914185 UNECA (2011) Assessment of Africa’s climatic records and recording networks including strategic for rescuing of climatic data. Retrieved from Adisa Baba, United Nations Economic Commission for Africa; African Climate Policy Centre UNMA (2017a) March to May 2017a seasonal rainfall outlook over Uganda Kampala Uganda; Uganda National Meteorogical Authority Uganda National Meteorogical Authority UNMA (2017b) September to December 2017b Seasonal rainfall outlook over Uganda Kampala, Uganda Uganda National Meteorogical Authority Uganda National Meteorogical Authority Vogel C Usable science: an assessment of long-term seasonal forecasts amongst farmers in rural areas of Sourth Africa S Afr Geogr J 2000 82 2 107 116 10.1080/03736245.2000.9713700 Waldman KB Blekking JP Attari SZ Evans TP Maize seed choice and perceptions of climate variability among smallholder farmers Glob Environ Chang 2017 47 April 51 63 10.1016/j.gloenvcha.2017.09.007 Waldman KB Vergopolan N Attari SZ Sheffield J Estes LD Caylor KK Evans TP Cognitive biases about climate variability in smallholder farming systems in Zambia Weather, Climate, and Society 2019 11 2 369 383 10.1175/WCAS-D-18-0050.1 Yooyongwech S Theerawitaya C Samphumphuang T Cha-um S Water-deficit tolerant identification in sweet potato genotypes (Ipomoea batatas (L.) Lam.) in vegetative developmental stage using multivariate physiological indices Sci Hortic 2013 162 242 251 10.1016/j.scienta.2013.07.041 Ziervogel G (2004) Targeting seasonal climate forecasts for integration into household level decisions: the case of smallholder farmers in Lesotho. Geogr J 170(1):6–21. 10.1111/j.0016-7398.2004.05002.x Ziervogel G Calder R Climate variability and rural livelihoods: assessing the impact of seasonal climate forecasts in Lesotho Area 2003 35 4 403 417 10.1111/j.0004-0894.2003.00190.x Ziervogel G, Downing TE (2004) Stakeholder networks: improving seasonal climate forecasts. Clim Change 65(1–2):73–101. 10.1023/B:CLIM.0000037492.18679.9e Ziervogel G, Opere A (2010) Integrating meteorological and indigenous knowledge-based seasonal climate forecasts for the agricultural sector. http://hdl.handle.net/10625/46185	36494529	PMC9734386	NO-CC CODE	2022-12-14 23:28:28	no	Br Dent J. 2022 Dec 9; 233(11):915	latin-1	Br Dent J	2,022	10.1038/s41415-022-5328-x	oa_other
==== Front Reg Environ Change Reg Environ Change Regional Environmental Change 1436-3798 1436-378X Springer Berlin Heidelberg Berlin/Heidelberg 1994 10.1007/s10113-022-01994-0 Original Article Factors associated with farmers’ use of indigenous and scientific climate forecasts in Rwenzori region, Western Uganda http://orcid.org/0000-0002-6434-9100 Nkuba Michael Robert [email protected] 1 Chanda Raban 1 Mmopelwa Gagoitseope 1 Kato Edward 2 Mangheni Margaret Najjingo 3 Lesolle David 1 Adedoyin Akintayo 4 Mujuni Godfrey 5 1 grid.7621.2 0000 0004 0635 5486 Department of Environmental Sciences, University of Botswana, 4775 Notwane Road, Private Bag 00704, Gaborone, Botswana 2 grid.419346.d 0000 0004 0480 4882 International Food Policy and Research Institute, Washington, D.C USA 3 grid.11194.3c 0000 0004 0620 0548 Department of Extension and Innovation Studies, College of Agricultural and Environmental Sciences, Makerere University, Kampala, Uganda 4 grid.7621.2 0000 0004 0635 5486 Department of Physics, University of Botswana, Gaborone, Botswana 5 grid.463702.4 Uganda National Meteorological Authority, Kampala, Uganda Communicated by Shuaib Lwasa 7 12 2022 2023 23 1 42 9 2021 20 10 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. Although scientific climate forecast (SF) distribution by national climate services has improved over time, farmers seem not to make good use of climate forecasts, a likely contributing factor to vulnerability to climate change. This study investigated factors associated with farmers’ use of SFs and indigenous forecasts (IFs) for agricultural use in the Rwenzori region, western Uganda. Household survey gathered data on demographic characteristics, climate information use and livelihood choices from 580 farmers. Data was analysed using the probit model. Results showed that significant factors associated with using both IFs and SFs were farm size, education, age, reception of scientific forecasts in local languages, agricultural extension access, short-mature crop access, farmer-to-farmer network and accessing forecasts through radio. This study shows that IFs were used complementarily with SFs. On the other hand, significant factors associated with using IFs only were livelihood choices such as tuber and goat production, access to government interventions on climate change adaptations, agro-ecological zone and social capital. Climate risks and climate risk perceptions negatively influenced the use of scientific forecasts. Co-production of climate information, capacity-building and active engagement of stakeholders in dissemination mechanisms can improve climate forecast use. Investments in more weather stations in various districts will therefore be a key factor in obtaining more accurate scientific forecasts and could lead to increased use of scientific climate forecasts. Governments in developing countries, the private sector, global and regional development partners should support investments in weather stations and capacity building of national meteorological systems. Supplementary Information The online version contains supplementary material available at 10.1007/s10113-022-01994-0. Keywords Scientific climate forecasts Indigenous forecasts Farmers Co-production Cognitive bias Uganda issue-copyright-statement© Springer-Verlag GmbH Germany, part of Springer Nature 2023 ==== Body pmcIntroduction Although climate services in the developing world have improved over time, scientific climate forecasts’ (SFs) use has remained low (Nkuba et al. 2021a). SFs are accessible from radio, television, mobile phones (Jost et al. 2015a, b), the internet, agricultural extension agents and farmer-to-farmer networks (Goddard et al. 2010). Development agencies involved in disaster risk reduction and in rural and agriculture development such as World Vision, Care International, Save the Children and Oxfam have also played a role in the dissemination of SFs (ACCRA 2014). SFs are communicated in local languages in Senegal, Malawi and Tanzania (Hampson et al. 2014; Lo and Dieng 2015). Using appropriate local languages to disseminate SFs improves access and use among farmers (Jost et al. 2015a, b). The main challenge in using SFs by stakeholders such as farmers is their probabilistic nature (Nicholls 1999): what has been predicted may or may not happen. Decision-making based on forecasts with shortcomings raises serious concerns among arable farmers (hereafter referred to as farmers) (Goddard et al. 2001). The limited predictive accuracy negatively influences use of SFs by farmers. Goddard et al. (2001) found that obstacles to using SFs include meteorologists’ failure to provide full information about the predictive accuracy or reliability of the forecasts, and their interpretation as deterministic forecasts. SFs lack the spatial and temporal specificity that farmers are most interested in. The SFs are disseminated in terms of regions or districts which have wide geographical coverage. Thus, predicted weather events may occur in some areas but not in all. Simply producing and disseminating climate forecasts do not make them valuable to farmers. However, Patt et al. (2005) reported that the use of SFs improved crop yields in Zimbabwe. This implies that there is improved farmers’ welfare through utility of SFs. Farmers’ limited use of SFs has been associated with lack of saliency, credibility, trust and legitimacy (Patt and Gwata 2002; Cash et al. 2003, 2006; McNie 2007). Furthermore, farmers have raised as major shortcomings of SFs poor spatial and temporal resolution with regard to failure to provide forecasts on onset and cessation of rains (Kalanda-Joshua et al. 2011; Nkomwa et al. 2014). One of the factors that has led to low use of SFs in Africa is low meteorological station density (Medany et al. 2006; UNECA 2011), which has resulted in poor spatial resolution with negative credibility and trust implications. The meteorological station density in Africa is eight times lower than the World Meteorological Organisation’s (WMO) minimum recommended level (Medany et al. 2006); the distribution of the meteorological stations in Africa does not match the existing agro-ecological systems. Meteorological station density is one of the cornerstones of index-based weather insurance (IBWI) (Akter et al. 2016). Increased investment in rural weather station infrastructure not only improves the use of SFs but also increases the uptake of IBWI (Amare et al. 2019) meaning lending institutions may not be in be position to provide much needed insurance and credit facilities to African farmers due to lack of climate information. To address the poor station density in Africa, the United Nations Development Programme (UNDP) has supported a number of African countries by providing them with automatic weather stations (Snow et al. 2016). Besides station density, other challenges of climate services in Africa include dysfunctional stations, or poorly maintained, or low data quality due to low national budget allocation to national meteorological services in many developing countries (UNECA 2011; Snow et al. 2016). The Uganda National Meteorological Authority (UNMA) provides biannual climate forecasts on the onset and cessation of rains, 10-day rainfall forecasts and rainfall distribution (UNMA 2017b). With support from UNDP, there has been increased investment in weather infrastructure in Uganda (Snow et al. 2016). Alongside the high access to SFs (Jost et al. 2015a, b), the Rwenzori region has a low meteorological station density (Fig. 1), with negative implications for the use of SFs. The meteorological authority collaborates effectively with other agencies involved in weather data generation such as agricultural research stations, wildlife protected areas, national universities and private tea and sugar plantations in rural areas (Snow et al. 2016). Only 5% of rain gauges in Uganda are functional and another 921 rain gauges are needed to attain an optimum level (Isabirye 2017).Fig. 1 Study area map There are multiple sources of indigenous forecasts (IFs) (Roncoli et al. 2002; Kalanda-Joshua et al. 2011; Kolawole et al. 2014; Nkomwa et al. 2014). Sources of IFs include farmer-to-farmer extension network, farmers’ organisation, elderly farmers and farmer’s own observations (Appendix Table 3). There are 10 (abiotic and biotic environmental) indicators that are observed to provide weather and climate information (Nkuba et al., 2020b). These indicators are relevant to farmers in their localities providing forecasts at high spatial resolution ( Roncoli et al., 2002; Orlove et al. 2010; Nkomwa et al. 2014). The temporal dimension of IF provides relevant information on onset and cessation of rains, which is very important in farmers’ adaptation to changes in rainfall seasons (Roncoli et al. 2002; Nkomwa et al. 2014). This suggests that IF has high credibility, trust and legitimacy among farmers. Farmers who use IFs only have expressed interest in using SFs after interaction with meteorologists (Orlove et al. 2010; Mpandeli and Maponya 2013). Climate variability negatively influences the use of IFs (Roncoli et al. 2002; Speranza et al. 2010). In light of the challenges of using IFs exclusively or SFs exclusively, some farmers have resorted to using a combination of SFs and IFs since SFs complements IFs (Ziervogel and Opere 2010; Roncoli et al. 2008; Mpandeli and Maponya 2013). Research has shown that some farmers begin with using IFs and then revise their forecast partially or completely after receiving an SF (Lybbert et al. 2007). There are farmers who revise their IFs after access to SFs and there are farmers who use IFs only and who do not revise their forecasts even after accessing SFs. There is high access to SFs (89%) in the study area. This is attributed to the proliferation of FM radio stations and the use of local languages. Climate forecasts are used in farmers’ decision-making related to when to plant, harvest and what crops to grow for a given cropping season. The climate forecasts are essential in climate change adaptation (Nkuba et al. 2020a). Farmers use climate forecasts to reduce their vulnerability to the impacts of climate risks such as droughts and floods (Hansen 2002). In bimodal rainfall regions of East Africa, onset determines which cereals are to be grown, with maize being grown during periods of early onsets, while sorghum and millet are grown during periods of late onsets (Mugalavai et al. 2008). Forecasts thus influence the choice of crop enterprises for coming farming seasons. Maize is very sensitive to water stress. For sweet potatoes (Ipomoea batatas), potatoes (Solanum tuberosum) and cassava (Manihot esculenta), water stress leads to poor tuber formation (El-Sharkawy and Cadavid 2002; MacKerron and Jefferies 1986; Yooyongwech et al. 2013), sweet potato weevil infestation (Ebregt et al. 2004, 2007), fruit cracking and blotchy ripening for vegetables (Steduto et al. 2012). The farmers’ use of forecasts for variations in the crop-water requirements greatly influences choice of crop. Some studies have investigated factors associated with climate forecast use. A study done in Northern Kenya and Southern Ethiopia (Luseno et al. 2003) found that location, education level and access to radio are significant factors in the use of SFs. Another study done in Botswana (Kolawole et al. 2014) reported that age and education level are associated with the use of climate forecasts. The two studies did not include factors such as the language in which SFs are received, livelihood choices, the agro-ecological zone or access to agricultural extension, which the current study takes into account. Access to forecasts seems to have gender dimensions, with men reportedly having more access due to their control of the radio in the rural households (Jost et al. 2015a, b). Jost et al. (2015a, b) found that women in Uganda preferred to receive forecasts from gatherings and community radios, while television was the preferred option for rural women in Bangladesh. A study in south-eastern Kenya (Muema et al. 2018) showed that age, gender, drought frequency, radio and access to improved varieties were factors associated with the use of SFs. A study in eight African countries1 in East and West Africa (Oyekale 2015) showed that education level, climate shocks, radio and gender were factors associated with the use of SFs. These studies did not look at factors associated with farmers’ use of IFs only and both IFs and SFs, hence the need for this study. The above literature review has highlighted factors associated with use of climate forecasts by farmers. Nevertheless, knowledge gap regarding the effect of crop type, agro-ecological zones and climate risk perceptions on the use of IFs only or both IFs and SFs in farming still exists. The objective of this paper is to investigate factors associated with farmers’ use of IFs only or both IFs and SFs in the Rwenzori region in western Uganda. We answer the question regarding what factors are associated with farmers’ use of IF singly or the combination of both IF and SF. This gap has not been adequately addressed in the international literature. The consensus globally, regionally and nationally is that climate change is occurring and its negative effects have been acknowledged. Access to climate and weather information therefore becomes imperative to the farming communities which experience the direct vulnerabilities from climate change. In this study, we investigate what can be done to increase use of SFs among rural farmers in Uganda, where there is also a tendency of farmers to use their own non-scientifically based indigenous forecasts (IFs). This study has highlighted to policymakers, climate scientists, climate change activists and government funding partners that farmers who exclusively use SF forecast are very few, with majority of farmers still depending on IFs which are inaccurate and arbitrary. This might have the consequence of farmers not being able to respond to climate change risks appropriately and timely if there is still over reliance on IF forecasts. This study identifies the factors that can be targeted by government and its stakeholders to turn around this situation and increase usage of SF forecasts. This study has also revealed that possibly it would be hard to promote use of SFs only in rural communities but rather a hybrid between SFs and IFs might lead to wider usage of SFs. Theoretical framework This study investigated decision-making under uncertainty. Some scholars emphasise the consequentialist approach to decision-making, implying that decision choices have repercussions. The farmers who are the recipients of SFs may or may not use them. National meteorological systems are the principal providers of SFs. The farmers receive SFs which are used to examine decisions under uncertainty. In this study, UNMA disseminates SFs to the farming community. Some of the farmers may not use SFs even after receiving them but use IFs only. There is a budget allocation by the Ugandan government for gathering and dissemination of SFs and early-warning information by funding UNMA. Farmers who use information from meteorological stations are using posterior beliefs which may be objective. Information from meteorology stations can facilitate farmers to change their knowledge (prior beliefs) about climate. Farmers may update their prior beliefs (indigenous knowledge) with information from meteorologists (posterior beliefs) and may follow the recommendations from the experts. Alternatively, the farmers may not believe in the accuracy of the information from the meteorologists, and hence, they may not update their prior beliefs, resulting in their not following the recommendations from the experts. Farmers receiving SFs may change or revise their prior beliefs based on indigenous knowledge, underscoring the importance of confidence in forecast information received (Luseno et al. 2003). The meteorological stations make use of historical rainfall data and other climate parameters to forecast what the rainfall pattern will be in the coming days and months (seasonal climate forecast). Meteorologists go further in recommending what actions the farmers ought to take. However, the final decision on what actions to take lies within the farmers. The value and use of climate forecasts by farmers are based on the range of actions and their capability to respond. This study seeks to fill the knowledge gaps regarding the factors associated with farmers’ decision-making under uncertainty based on the climate forecasts (SFs and IFs) they used. The paper contributes to the current debate on influence of climate information on decision-making under uncertainty. Materials and methods The study area The study was conducted in the Rwenzori region of western Uganda. In terms of climate, the region experiences bimodal rainfall, which contributes to having two cropping seasons with the first season running from February to May and the second season from August to December. The temperature ranges from 12 to 24 °C, annual rainfall ranges from 800 to 3000 mm, and elevation ranges from 500 to 5000 m above sea level. Most of the farmers have completed primary education (Nkuba et al. 2020a). The farmers’ ability to comprehend climate information can be enhanced through stakeholder engagement by Uganda National Meteorological Authority (Nkuba et al. 2019). The soils are mainly sandy loam and sand clay loam which support production of cereals, tubers and vegetables (Nkuba et al. 2020a). Tree planting for commercial crop production of coffee, cocoa, fruit trees and commercial woodlots is very vibrant in the study area (Nkuba et al. 2020a). Access to pastures supports climate change adaptation measures for livestock farmers such as livestock migration, herd mobility and livestock diversification (Nkuba et al. 2021c). The region was selected because it has several agro-ecological zones—mountainous, lowland, mountainous and forested, wetland and forested—in order to investigate the effect of agro-climate on climate forecast use. Wildlife protected areas (WPA) in the study area (Kibale, Toro-Semiliki and Mount Rwenzori National parks) (Fig. 1) and Rwebitaba Zonal Agricultural Research Institute (in Kabarole district) have meteorological stations which provide climate data used in meteorological forecasts for the region. Weather stations are in government-aided establishments such as research stations and WPAs. Farmers in remote areas tend to find SF predictions to be inaccurate. Farming is a major livelihood source in the region. Access to SFs in local languages is due to the spread of FM radio and television stations. The region is inhabited by many ethnic groups, all of which use indigenous knowledge in their day-to-day lives. Data collection and analysis procedures Data was gathered from August to October 2015. A respondent survey using questionnaires was used to collect data on socio-economic factors, farm and institutional characteristics by trained research assistants. The respondent survey information was triangulated with data from focus group discussions (FGDs) and interviews. FGDs were used to get farmers’ views on the use of IFs and SFs. FGDs were held in Kyenjojo and Kabarole districts to get the farmers’ views on the use of IF and SF. A female FGD of 15 members and a male FGD of 16 members were carried out in Kyejonjo district. In Kabarole district, a female FGD of 15 members and male FGD of 17 members were conducted. The members were farmers who use IFs and SFs in their farming activities. To ensure that members of FGDs expressed their views freely, gender-segregated focus groups were used. The use of the mixed methods approach was for triangulation of the data obtained to enhance its validity and reliability. Data was analysed using Stata 16 statistical software. This study is nested in an earlier study (Nkuba et al. 2020a). A multi-stage stratified approach was used in the sampling of respondents. The first stage involved districts, the second stage counties, the third stage sub-counties and the smallest unit the household. The Uganda Bureau of Statistics disseminates the population data of households according to district, county, sub-county and household. The selection criteria were (i) farming systems, namely arable farming, pastoralism and agro-pastoralism and (ii) agro-ecological systems such as forested, lowlands, mountainous and wetlands. The statistically selected sample size of farmers was allocated to particular sub-counties in the selected districts using proportional allocation to size, where the size represents the number of households in the district. Based on a population of the study area of 102,496 households, according to the Uganda population census report of 2014, a sample size of 778 was selected with 95% confidence level and a margin of error of 3.5%. To allow for replacement in the sample of respondents who might drop out of the study, 19% of the statistically selected sample was included, giving a total study sample of 924. This was also to ensure a good sample size for sub-samples (for those who use both IFs and SFs, and IFs only). After data cleaning, 17 questionnaires were excluded due to incomplete responses. Of the 907 respondents, 580 were farmers, 270 pastoralists and 57 agro-pastoralists. This paper limits itself to the 580 farmers in the sample. Theoretical model The probit regression was used to analyse farmers’ use of forecasts. The probit model was used for examining the likelihood of a future climate event (for example rain onset and cessation) happening at a particular time using climate forecasts. There is no complete certainty regarding the occurrence of future climate outcomes that have been predicted. The predicted outcome may or may not happen. The predicted climate outcomes (dependent variables) in this study address use or non-use of IFs and SFs. IF and SF are specified as dichotomous outcomes with yes (use) and no (no-use), coded as 1 and 0 respectively. The probit model is a non-linear model that estimates with probabilistic maximum likelihood, often used for binary outcomes (Gujarati 2013). The regression results in this study are based on probit regression estimates. The probit model was estimated using Stata 16 software. There is potential for presence of self-selection in the use or non-use of forecasts. To overcome issues of selectivity convoluting the estimates, we have controlled for several conditioning factors that could be correlated with use of forecast and at the same time also influence outcomes of interest. We think this approach helps to minimise self-selection bias due to forecast use. We also tried to estimate models with and without forecasts to have a feel for the extent to which the estimates are being affected by selectivity. The estimates did not change when forecasts were excluded in the models. With these two approaches, we think that the possible threat from self-selection is not very serious in our analysis. We could use the instrumental variable approach because it was difficult to find strong excluded instruments that would satisfy all the instrument validity requirements. Empirical model The empirical model used in the analysis was specified as follows.1 Yij=β1Householdcharacteristics+β2Informationsources+β3Farmcharacteristics+β4Agro-ecologicalzone+β5SocialCapital+β6Percerptionofclimaterisk+β7Climaterisk+β8Wealth+ε For use of IF only2 Zij=β1Householdcharacteristics+β2Informationsources+β3Farmcharacteristics+β4Agro-ecologicalzone+β5SocialCapital+β6Percerptionofclimaterisk+β7Climaterisk+β8Wealth+ε For use of IF and SF where Yij(j = 1,2,3,4) representing the four models of using only IF and Zij(j = 1,2,3,4) representing the four models of using both IF and SF (Appendix Table 4). In this study, we control for several factors that could influence use of information as well as those factors that have been identified as affecting adoption of agricultural technologies in developing countries. Past empirical and theoretical literature has guided our theoretical choice of the factors that we include in our econometric models as explained below. Factors associated with farmers’ use of climate forecasts include education level, age, gender (Roncoli et al. 2002), livelihood choices (Vogel 2000; Ingrama et al. 2002; Patt and Gwata 2002; Ziervogel and Calder 2003; Crane et al. 2010; Haigh et al. 2015; Klemm and McPherson 2017), translation of SFs into local languages (Ingrama et al. 2002; Ziervogel and Downing 2004), source of forecasts for onset and cessation (Roncoli et al. 2002; Haigh et al. 2015), access to credit (Vogel 2000; Ingrama et al. 2002), access to non-farm enterprises (Ingrama et al. 2002; Ziervogel and Calder 2003; Crane et al. 2010; Klemm and McPherson 2017), access to agricultural extension (Vogel 2000; Ziervogel and Downing 2004; Coles and Scott 2009; Crane et al. 2010; Haigh et al. 2015; Klemm and McPherson 2017), access to improved crop varieties (Ingrama et al. 2002), climate risks (Ingrama et al. 2002), agro-ecological zone (Ingrama et al. 2002; Ziervogel 2004; Klemm and McPherson 2017), perception of climate risks (Vogel 2000; Ingrama et al. 2002; Ziervogel 2004; Haigh et al. 2015) and farm size (Vogel 2000; Coles and Scott 2009). Climate risks and perception of climate risks are important factors associated with farmers’ use of forecasts (Vogel 2000; Ingrama et al. 2002; Haigh et al. 2015). Farmers’ cognitive biases2 have an effect on their risk perceptions which influences climate forecasts use (Waldman et al. 2019). Unlike pastoralists, who practice herd mobility as an adaptive mechanism, farmers cannot migrate their crop fields. Changes in the onset and cessation of rains progress slowly towards their final manifestation in damage to crops (Slovic 2000a). The explanatory variables include Respondent characteristics (H): level of education, age, gender and farming experience; farm characteristics (F): type of crops grown and type of livestock kept; institutional characteristics (I): sources of IF, sources of SF, access to agricultural extension, credit access, improved crop access, non-farm access; agro-ecological system (A): forested, lowland, mountainous, wetland, mountainous and forested; social capital (S): membership of farmers’ organisation, farmer-to-farmer extension networks; perception of climate risks (P) drought increase, flood increase; climate risks (C): flood experience, drought experience; wealth (W): farm size. Variables and expected signs for use of IFs only and both IFs and SFs are shown in Appendix Table 5. Results Descriptive overview: use of forecasts Results show that almost half (49%) of the farmers used IFs only and half (50%) used both SFs and IFs (Appendix Table 6). Only one respondent used SF only. Over half (54%) of the respondents were male (Appendix Table 7). There were no significant differences in the use of either both IF and SF or IFS only between men and women respondents. Forecasts for rain onset and cessation were very crucial in farming. Considering the sub-sample of those who used IFs only, all the respondents used indigenous knowledge in predicting the onset of rains compared to 77% for the sub-sample of those who used both IFs and SFs. Most of the farmers (97%) reported that IF was reliable, compared to 43% for SF. Farmers who used both IFs and SFs reported that they used SFs to confirm the IFs based on what they had observed and learnt from elders. The most important crops grown were cereals and tubers (Appendix Table 3). Farmers’ organisations and farmer-to-farmer networks also played a crucial role in providing climate information (Appendix Table 3). Radio and non-government organisations were important SF dissemination channels (Appendix Table 8). Local FM stations provide SFs in local languages (Appendix Table 7), which has greatly improved access even to farmers with no formal education. Agricultural extension workers played a minimal role in the dissemination of SFs and information on rain season duration. Agricultural extension has been impacted by the involvement of Uganda Peoples Defence Force under Operation Wealth Creation, which plays a major role in the government distribution scheme3 for agricultural inputs such as improved seeds, seedlings and livestock during onset of rains. The results show that there were significant differences between farmers who used both IFs and SFs and those who used IFs only (Appendix Tables 3, 7, and 9 at 5% level of significance). Farmers from mountainous forested areas and bare mountainous areas were using IF only significantly more than both IFs and SFs (Appendix Table 7). Location and terrain appear to affect use of SFs, possibly because of differences in the level of infrastructural development between mountainous areas and lowlands. Farmers in lowlands were more likely to use SFs, while their counterparts in mountains were more likely to use IFs. This suggests constraints linked to quality of infrastructural development limiting access to SFs. Farmers growing cereals were using both IFs and SFs significantly less than IF only (Appendix Table 3). Cereals are very sensitive to water stress and are grown on a commercial scale in the study area. This increases the reliance on rain-fed agriculture in rural areas where there is limited use of irrigation, resulting in an increase in the use of SFs. Factors influencing farmers’ use of indigenous forecasts only The results show that factors positively and significantly associated with using IFs only among farmers were as follows: livelihood choices such as tubers and goat production, agro-ecology such as being resident in mountainous areas, reception of information from farmers’ organisations about the onset and cessation of rains access to government programme interventions on climate change adaptations and perceptions of climate variability and change as increases in floods and drought (Table 1). Factors negatively and significantly associated with using IFs only were non-farm access and livelihoods that depend on maize production.Table 1 Marginal effects of use of IFs only for arable farmers Variable Onset dy/dx Cessation dy/dx 5 day dy/dx Seasonal dy/dx Tubers 0.054(0.051) 0.214*(0.077) 0.201(0.079) 0.161**(0.074) Mountainous and forested 0.200(0.125) Mountainous 0.156(0.086) Farmers’ organisation as source of onset and cessation 0.264*(0.063) 0.306(0.082) 0.322*(0.095) 0.164(0.094) Credit access − 0.120(0.073) Non-farm access − 0.200*(0.078) Agricultural extension access − 0.170(0.095) Rainfall increase 0.108(0.083) Rainfall season change 0.120(0.079) Droughts increase 0.134(0.074) Floods increase 0.218(0.110) 0.244(0.116) 0.322(0.115) Access govt interventions on climate change adaptations 0.116(0.058) 0.176*(0.088) ln maize production − 0.017(0.010) − 0.025(0.015) − 0.036(0.016) − 0.022(0.015) ln local goat production 0.053(0.044) 0.073(0.046) 0.107(0.044) Log likelihood − 281.134 − 130.445 − 125.016 − 125.204 Pseudo R2 0.045 0.103 0.136 0.109 N 425 210 210 210 (Source: survey data 2015 overall title of the research project was ‘The Use of Indigenous and Scientific Forecasts in adaptation to climate variability and Change. A case study of Rwenzori region, Western Uganda’). Figures in parentheses are standard errors. *, and * denote that significance at the 1%, 5% and 10% levels respectively Government programme interventions on climate change adaptations include the provision of improved crop varieties and tree-planting materials to farmers, whose adoption depends on forecasts for the onset of rains. The agricultural inputs, such as seeds and seedlings, were delivered by National Agricultural Advisory Services, Operation Wealth Creation under the Office of the President and the Uganda Forestry Authority. Farmers had confidence in IFs and hence used it for predicting the onset of rains to effectively participate in government adaptation programmes. Cereals (especially maize), tubers and tree crops are sensitive to water stress, especially during the vegetative stage, and farmers’ confidence in IFs influenced their decisions on choice of crop enterprises and allocation of resources, depending on the onset and cessation forecasts. Farmers preferred to grow maize in the second rainy season, which is long enough and runs from August to November. Short-maturing cereals like sorghum and millet were preferred for the first rainy season, which is short and runs from March to May. A key informant reported that winds in the mountainous and forested areas on the windward side of Mount Rwenzori have high moisture content which easily reaches saturation point resulting in rainfall. Farmers in such areas rely a great deal on observing IF indicators like clouds and wind to get reliable forecasts. Factors influencing farmers’ farmers’ use of both indigenous forecasts and scientific forecasts The findings show that factors positively and significantly associated with using both IFs and SFs were reception of SFs in local language and English, attainment of higher education (diploma or advanced secondary school education), access to SFs through radio and TV, access to short-maturing crop varieties, availability of agricultural extension services, age and fellow farmer as source of SFs (Table 2).4 Factors negatively and significantly associated with using both IFs and SFs were livelihood that depend on vegetables, tubers and maize, being residents in mountainous forested areas, primary school education, Non-Governmental Organisation as source of onset and cessation forecasts and drought experience, and perceiving climate variability and change as seasonal rainfall change and drought increase (Table 2). Vegetable growers seemed not to have enough confidence in SFs as a basis for decision-making, probably because of vegetable crops’ water requirements. Command of English has a higher impact than local language on the use of SFs. Farmers who are comfortable with receiving SFs in English have fewer constraints than those who receive SFs in local languages. English may be more used by elite and well-to-do farmers. Farmers using local languages have less ability to access SFs compared to their counterparts who also use English.Table 2 Marginal effects of use of both IFs and SFs for arable farmers Variable Onset dy/dx Cessation dy/dx 5 day dy/dx Seasonal dy/dx Male 0.050(0.038) Vegetables − 0.107(0.051) − 0.131(0.053) − 0.134(0.044) Tubers − 0.127(0.045) Age 0.003(0.002) 0.004(0.002) Local language 0.464*(0.045) 0.402(0.058) 0.214(0.056) 0.201(0.047) English 0.577(0.063) 0.611(0.082) 0.367(0.115) 0.317(0.127) Mountainous and forested − 0.229(0.054) − 0.210(0.065) -0.089(0.047) Primary education − 0.135(0.053) -0.046(0.039) Higher education 0.375(0.163) 0.414*(0.144) Radio as source of scientific forecasts 0.249(0.090) 0.227(0.056) TV as source of scientific forecasts 0.113(0.090) 0.391(0.159) Fellow farmer as source of scientific forecasts 0.142(0.061) 0.152(0.066) 0.161*(0.060) NGO as source of onset and cessation forecasts − 0.339(0.040) − 0.264(0.052) − 0.208(0.047) Drought experience -0.098(0.057) − 0.126(0.071) Short mature crop access 0.156(0.063) 0.109(0.067) 0.094(0.053) Agricultural extension access 0.149(0.077) 0.180(0.089) Rainfall increase -0.058(0.041) Rainfall season change − 0.083(0.048) − 0.098(0.052) -0.053(0.040) Droughts increase − 0.127(0.054) − 0.151*(0.046) Unpredictable trains 0.051(0.041) ln maize production − 0.023(0.011) 0.017(0.010) -0.016(0.008) ln farm size 0.040(0.023) 0.027(0.026) 0.024(0.023) 0.019(0.019) ln likelihood − 270.210 − 194.653 − 195.332 -168.930 Pseudo R2 0.205 0.248 0.173 0.144 N 507 394 394 394 (Source: survey data 2015). Significant at 1%, significant at 5%, *significant at 10%. Figures in parentheses are standard errors There are few weather stations in the region (Fig. 1) whose rainfall data is used to make predictions, resulting in poor predictive accuracy and wide spatial variation for forecasts from UNMA. The rainfall data from the weather station in the mountainous area of Kabarole was used to make forecasts for the mountainous and forested area, quite far from the station.5 A key informant pointed out that there was apparently no weather station for rainfall data in the mountainous and forested area of Bundibugyo. This casts doubt on the reliability of SFs for seasonal forecasts for mountainous forested areas using extrapolated rainfall data from weather stations located in distant ecological zones. Forecasts were made in terms of regions and districts, yet farmers are interested in information at parish or village level. Receiving SFs in local languages made dissemination accessible to farmers, even to those with no formal schooling. Mountainous and forested areas have high precipitation in the form of mist and fog, which contributes positively to soil moisture availability. Climate risks like drought are negatively associated with the use SFs. It is probable that climate risks like droughts can influence farmers’ cognitive biases such as availability bias, anchoring effect and confirmation bias, which could impair farmers’ judgement in using SFs. Farmers can overestimate or underestimate the likelihood of droughts due to over-confidence in their recent memory of the severity and frequency of the occurrences. It is also possible to consider long dry spells as droughts. Climate risk perception such as drought increase and seasonal rainfall change was negatively associated with use of SFs. Climate risk perceptions can influence farmer cognitive biases such as framing effect, hindsight bias, anchoring effect and confirmation bias, which could impair farmers’ judgement in using SFs. The framing effect created by the limited coverage of dissemination mechanisms such as media about the effects of change in rain onset and cessation can lead farmers to underestimate the start and end of the rainfall season. Most media and meteorologists tend to focus more on rainfall distribution in terms of ‘above normal’ or ‘below normal’ in their dissemination of the seasonal climate forecast for the farming season, rather than duration of the rainy season. Farmers’ experiences can make them over-confident and consequently impair their judgement of frequency and severity of droughts. But there were also respondents who perceive that SFs are not credible, as exemplified by the following observation by a female participant in the FGD: ‘For me, I don’t think it’s truth because there is when they told us to prepare for heavy hail storms but they didn’t come’ (i.e., scientific forecasts are not reliable). There is also a lack of local specificity in SFs, rendering them less credible to the farmers. One FGD participant, for instance, remarked that ‘You can see that it [rain] has not happened here but when in another place it has’. Some doubt was also cast on the legitimacy of SFs, as indicated by one participant who declared that ‘They also just guess’. This indicates that farmers do not trust the processes that meteorologists use to make forecasts, and consequently, they distrust the forecasts as well. Farmers attached high credibility to IFs because of the long tradition of using and depending on them, as is evidenced by the following statement from a FGD participant: ‘from when we were born until we have grown to now it’s what we found our elders using. The elders before us this [IFs] is what they were using, their knowledge of long ago. For that we also grow up following it’. IFs knowledge is based on the long-term observation of nature and is passed on orally from one generation to another. There were 15 indicators used in IF identified in the study area (Nkuba et al. 2020b) implying that there is a wide variety of sources of forecasts, resulting in increased legitimacy. The indicators are biotic factors such as plants, insects and birds and abiotic factors such as wind and clouds. The abiotic factors are similar to what meteorologists use in their predictions. Farmers’ organisations and elderly farmers were social capital that provide forecast information within their specific local area. Religion also plays a role in the use of IFs; for example, many respondents said that the onset of rains corresponds to the day of Mother Mary, namely 18 August. One participant in the FGD reported that ‘Indigenous knowledge is a very important thing because that’s the knowledge God created us with. Because even what is taught us [scientific forecasts]… you would learn and by the time you reach home you have forgotten it. But indigenous knowledge is better. Because with scientific forecasts, someone will guess what’s not there’. Radio is the most widely used method of disseminating information about SFs. Radio stations broadcast in both English and local languages. TV is widely used by elite and well-to-do farmers. TV stations like National TV and Uganda Broadcasting Services are widely used as sources of forecasts by wealthier farmers. A key informant in the civil service stated that agricultural extension workers receive SFs from UNMA through the Ministry of Agriculture on a regular basis. There are factors that influence the use of both IFs and SFs, and IFs only, differently (Tables 1 and 2). For instance, access to improved crop varieties and agricultural extension services is positively associated with using both IFs and SFs and negatively associated with using IFs only. Access to agricultural extension is closely associated with SFs. Maize production was positively associated with use of both IFs and SFs but negatively associated with IFs for the 5-day forecasts. Short-range SFs such as 5-day forecasts provide reliable climate information (dry spells, or optimum planting and harvesting days) which is relevant to farmers’ involvement in water stress crops like maize. Being a resident in mountainous and forested areas, and perceiving climate variability and change as drought increase and seasonal rainfall change, was positively associated with using IFs only and negatively associated with using IFs and SFs. Mountainous and forested areas are positively associated with high precipitation due to orographic effects on cloud formation and condensation, contributing to less variability in rainfall. Having a higher level of education (secondary education and diploma) was positively associated with use of IFs and SFs. Increase in education increases uptake of SFs. However, primary education was negatively associated with use of IFs and SFs. Primary education probably causes the farmers to overestimate or underestimate due to over-confidence based on their recent past experiences. Reception of SFs from fellow farmers were positively associated with using IFs and SFs, while reception of onset and cessation of rain forecasts from farmers’ organisations was positively associated with using IFs only. Group decisions were more trusted in using IFs only. Confidence in using IFs only is built up through social gatherings of like-minded people. Information disseminated through farmer-to-farmer networks was more trusted in using SFs. This could arise from the confidence that the farmer-to-farmer networks built over time in using SFs with good outcomes. Discussion Results showed that significant factors associated with using both IFs and SFs were farm size, education, age, reception of scientific forecasts in local languages, agricultural extension access, short-mature crop access, farmer-to-farmer network and accessing forecasts through mass media. This study shows that IFs were used complementarily with SFs. On the other hand, significant factors associated with using IFs only were livelihood choices such as tuber and goat production, access to government interventions on climate change adaptations, agro-ecological zone and social capital. Research has shown that dissemination of SFs through local languages is the most preferred among rural households (Antwi-Agyei et al. 2021). Dissemination in local languages is usually facilitated by media such as local FM radio stations. The current study has also confirmed that radio and television play a crucial role in the dissemination of SFs. Radio has been reported to be an important dissemination mechanism for SFs in rural regions (Jost et al 2015a, b). This indicates that widening FM radio station broadcasting coverage in Uganda could provide an opportunity for increasing the utilisation of SFs in rural areas. The main shortcoming of using such media as radio and television for dissemination is that there is usually no possibility for feedback from farmers. This is perhaps one of the reasons that farmers seem to have greater confidence in forecasts from their fellow farmers: there is feedback and discussion of experiences and the implications of using such forecasts. National meteorological services need to employ dissemination mechanisms that provide feedback from farmers for improved effectiveness in the use of forecasts. The findings of the current study also show that the onset and cessation of rains forecasts from farmer-to-farmer extension networks had mixed results, with a positive association with use of IFs only and a negative association with use of both IFs and SFs. This shows that farmers had more confidence in IFs than in SFs because IFs have local specificity while SFs have broad spatial and temporal variations (Speranza et al. 2010). Farmers’ indigenous knowledge of climate forecasts is more trusted than forecasts by climate scientists (Kolawole et al. 2014; Roudier et al. 2014). The results of the current study have also shown that farmers have negative associations with use of forecasts from NGOs. The challenge is that NGO staff operating in rural areas may not be able to answer all the farmers’ questions about forecasts, which may engender a lack of trust in the forecasts among farmers (Ofoegbu and New 2021). This study shows that farmers involved in tuber production are more associated with use of IFs only. Farmers involved in tuber and maize production do not have confidence in SFs because of past experiences of poor spatial and temporal specificity and, therefore, in their eyes, of false forecasts (Ziervogel 2004). Our study shows mixed results regarding the association of education with climate forecast use. Post primary education was positively associated with use of SFs. Luseno et al. (2003) study in Kenya and Ethiopia showed that formal education improved confidence in and access to SFs. Similarly, a study done in Botswana showed that the more educated a farmer is, the more likely he/she will be to use SFs received through the media (Kolawole et al. 2014). However, primary education was negatively associated with the use of SFs. A study done is Zambia reported that one additional year in education reduced farmers’ perception of change in rain onset due to psychological factors (Waldman et al. 2019). This calls for further research on the effect of psychological factors on climate forecast use and a multidisciplinary approach for climate services research. Climate change perceptions have a mixed effect on the use of forecasts, with a positive association with use of IFs and a negative association with use of both IFs and SFs. Experienced farmers make adjustments in their livelihoods after experiencing a drought, or they make changes in their responses to the onset and cessation of rains because of the likelihood of a recurrence (Slovic et al. 2000b). Farmers experience high crop losses due the change in rainfall duration and severe droughts, thus increasing their confidence in using IFs only. The range of alternatives for farmers is limited, which increases the level of catastrophe to their livelihoods (Slovic et al. 2000c). This also influences the use of IFs only, in order to avoid or to minimise experiencing such risks as loss of seeds due to late onset of rains, or poor harvests due to early cessation of rainfall. Farmers’ assessment of risks, which influences their use of IFs, is related to climate risks that reduce household welfare due to crop yield losses (Slovic 2000, 2000a, 2000b, 2000c). However, the perception of unpredictable rains as climate change is associated with using both. This is consistent with Partey et al. (2018),who found that increased rainfall variability positively influences the use of climate information in Ghana. The seasonal climate and short-range forecasts from UNMA provide predictions of rainfall distribution which influence risk perceptions (UNMA 2017a), as illustrated by a participant in a FGD, who said that ‘Now like this season when they say the rains are going to be a lot, we become cautious of the floods’. This indicates that SFs are also positively associated with risk perceptions and climate change perceptions. The results from the FGDs show that SFs are associated with increased crop yields which is consistent with findings from studies done in Burkina Faso and Zimbabwe (Patt et al. 2005; Roncoli et al. 2008). Research shows that farmers who feel they have adaptive capacity for a particular climate risk (availability heuristic) tend to have lower the climate risk perception (Duinen et al. 2015) which can lead underestimating its effects (Tversky and Kahneman 1979). Farmers with drought experiences have hindsight bias leading to increased confidence in their degree of prediction of future droughts without using SFs. Drought increase is a dreaded risk whose calamities are known through loss of crop yields and change in rainfall season is a delayed risk with a slow manifestation of harm. The recent memory of climate risks like rainfall seasonal change and drought could influence the use of SFs due to hindsight bias and anchoring effect. The framing effect from various stakeholders and media can result in making rainfall season change look less dangerous than gradual onset of droughts, leading to bias in using SFs. Research has shown that maize farmers have biases related to rain onset (Waldman et al. 2017), which is in agreement with our findings. Research has shown that farmers’ climate risk perception and past experience can lead to poor decision-making related to climate forecast use, due to cognitive biases (Waldman et al. 2019). Farmers from mountainous forested areas, and mountainous areas, are positively associated with use of IFs only. Due to orographic factors, the wide spatial variations in rainfall in these agro-ecological zones cause farmers to have more confidence in IFs only due to IFs’ local specificity compared to SFs, with SFs’ low spatial coverage because of low meteorological station density, especially in the mountainous areas (Sultan et al. 2020). Furthermore, Leroux (2001) found that ‘with increasing altitude the nature of precipitation changes; raindrops become smaller, showers give way to more continuous rain, and thence to rain, drizzle, mist and fog’ (p. 50). Forests act as windbreakers which disturb surface circulation and have a greenhouse effect that reduces temperature variations (Leroux 2001). Relative humidity in forested areas is high and this consequently leads to forests having a strong influence on rainfall over wide areas (Leroux 2001). The use of both SFs and IFs can be enhanced by experts engaging in constant dialogue with end-users. Studies in Mozambique, Kenya, Ivory Coast, Senegal, Tanzania and Benin have shown that confidence in SFs greatly improved when national meteorological services received constructive feedback from stakeholders such as farmers (Ziervogel and Opere 2010; Lo and Dieng 2015). Research has also shown that the legitimacy, salience and credibility of SFs have been improved by the active engagement of farmers, resulting in farmers’ increased use of SFs (Bouroncle et al. 2019). Integration of SFs and IFs is critical for the effective utilisation of forecasts in rural livelihoods based on farming. Ziervogel and Opere (2010) have provided information on projects in Africa that had integrated to good effect meteorological and indigenous-based seasonal forecasts in the agricultural sector. Stakeholder engagement can be in form of farm rainfall data generation by farmers. Botswana’s Department of Meteorological services has installed rain-gauges for farmers who send daily rainfall data via telephone. Farmers involved in farm rainfall data generation get excited when rainfall data from their farms is read during the news bulletin on national television station. Botswana meteorologists conducted meteorology extension services during the agricultural shows and held feedback sessions with farmers involved in rainfall data generation before the COVID-19 pandemic. A pilot study on citizen science in Namibia and South Africa has shown that a farmer involved farm rainfall data generation had much trust in SF, constructive engagement with meteorologists, reduced vulnerability to climate change, improved crop yields, livestock sales and household incomes (Landman et al. 2020). The results of the study are applicable to humid tropical zones and may not be applicable to a semi-arid area. This calls for further research regarding factors associated with use of indigenous and scientific forecasts in semi-arid areas. Conclusion and recommendations This study has established that farming in the Rwenzori region is informed by the use of both IFs and SFs or IFs only. Factors that promote the use of SFs include post-primary school education, access to appropriate rural institutions and dissemination of forecasts through radio using local languages. However, primary education was negatively associated with use of SFs. This is attributed to farmers’ cognitive biases. This calls for further research on the effect of psychological factors on climate forecast use. Research on climate information use requires a transdisciplinary approach involving environmental psychologists, social scientists, agronomists and meteorologists. Farmer-to-farmer extension networks are positively associated with use of IFs only and negatively associated with the use of both IFs and SFs. This calls for active stakeholders’ engagement by national meteorological services. This will improve the salience, legitimacy and credibility of meteorological forecasts in rural areas. Climate risk perceptions and climate risks are negatively associated with the use of SFs. Addressing farmers’ cognitive biases associated with climate risks and climate risk perceptions by national meteorological systems actively engaging stakeholders in climate forecast dissemination mechanisms could improve uptake of SFs. Investments in ddissemination of weather forecasts should be maintained and or increased, as the study has shown that television and radio have a positive impact. SFs were found to be complementary to IFs. SFs reinforce IFs. This calls for co-production of climate information in order to promote an increased use of forecasts in rural areas. Indigenous forecasts will continue to play a major role in influencing land-use interventions among farmers in Uganda and sub-Saharan Africa. We do not contest the use of SFs among farmers because trust in SFs among farmers is likely to improve with an increase in the number and density of weather stations in rural areas. Investment in more weather stations (automatic weather stations and rain gauges) in farming areas is a key factor in obtaining more spatially specific and accurate SFs. This could result in the improved use of SFs which might lead to improved food security and reduce vulnerability to climate change. Governments in developing countries, the private sector and the global and regional development partners should support investments in weather stations and capacity building of national meteorological systems. These possibilities are of course influenced not only by the use of forecasts but also by a number of other equally important factors such as access to inputs, agricultural extension services and credit. A longitudinal study on the validity of indigenous forecasts should be explored. In places with bimodal rainfall distribution, data has to be collected for both seasons in a period of 3 years or more to get meaningful outcomes. For comparative purposes, similar research should be conducted in other agro-ecological zones such as semi-arid areas and temperate areas, which were not covered in this study. The study has also established that farmers use scientific and indigenous forecasts in making decisions under uncertainty. Some farmers receive SFs but do not use them, instead using IFs only. Some receive SFs and update them with IFs, resulting in using both indigenous and scientific forecasts. Supplementary Information Below is the link to the electronic supplementary material. ESM 1 (DOC 135 kb) 1 Burkina Faso, Senegal, Mali, Niger, Ghana, Kenya, Tanzania and Ethiopia. 2 Cognitive biases include hindsight bias, confirmation bias, framing effect, anchoring effect, availability bias and decision regret effect (Tversky and Kahneman 1974; Kahneman and Tversky 1982; Nicholls 1999). 3 Media reports have indicated farmers rejecting agricultural inputs because they were supplied without taking into account rain onset. 4 The independent variables in the two models (Tables 1 and 2) differ because of precision concerns. Variables with standard errors larger than the coefficients indicate a poor estimation (Gujarati 2013) and were therefore not included in the models. 5 The distance from Kabarole to Bundibugyo is 24 km. Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References ACCRA (2014) Planning for the future and adapting to climate change in Uganda. Lessons from ACCRA In A. C. C. R. Alliance (Ed.), (pp. 4). Kampala, Uganda Africa Climate Change Resilience Alliance Akter S, Krupnik TJ, Rossi F, Khanam F (2016) The influence of gender and product design on farmers’ preferences for weather-indexed crop insurance. Glob Environ Chang 38:217–229. 10.1016/j.gloenvcha.2016.03.010 Amare A, Simane B, Nyangaga J, Defisa A, Hamza D, et al (2019) Index-based livestock insurance to manage climate risks in Borena zone of southern Oromia, Ethiopia. Clim Risk Manag 25:100191. 10.1016/j.crm.2019.100191 Antwi-Agyei P, Dougill AJ, Abaidoo RC (2021) Opportunities and barriers for using climate information for building resilient agricultural systems in Sudan savannah agro-ecological zone of north-eastern Ghana. Clim Serv 22.10.1016/j.cliser.2021.100226 Bouroncle C Müller A Giraldo D Rios D Imbach P Girón E Portillo F Boni A van Etten J Ramirez-Villegas J A systematic approach to assess climate information products applied to agriculture and food security in Guatemala and Colombia Climate Services 2019 16 October 100137 10.1016/j.cliser.2019.100137 Cash DW, Borck JC, Patt AG (2006) Countering the loading-dock approach to linking science and decision making: comparative analysis of El Niño/Southern Oscillation (ENSO) forecasting systems. Sci Technol Human Values 31(4):465–494. 10.1177/0162243906287547 Cash DW, Clark WC, Alcock F, Dickson NM, Eckley N, et al (2003) Knowledge systems for sustainable development. Proc Natl Acad Sci 100(14):8086–8091. 10.1073/pnas.1231332100 Coles AR, Scott CA (2009) Vulnerability and adaptation to climate change and variability in semi‐arid rural southeastern Arizona, USA. Nat Resour Forum 33(4):297–309. https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1477-8947.2009.01253.x Crane TA Roncoli C Paz J Breuer N Broad K Ingram KT Hoogenboom G Forecast skill and farmers’ skills: seasonal climate forecasts and agricultural risk management in the southeastern United States Weather, Clim Soc 2010 2 1 44 59 10.1175/2009WCAS1006.1 Duinen R Filatova T Geurts P van der Veen A Empirical analysis of farmers’ drought risk perception: objective factors, personal circumstances, and social influence Risk Anal 2015 35 4 741 755 10.1111/risa.12299 25514996 Ebregt E Struik PC Abidin PE Odongo B Farmers' information on sweet potato production and millipede infestation in north-eastern Uganda. I. Associations between spatial and temporal crop diversity and the level of pest infestation NJAS - Wageningen J Life Sci 2004 52 1 47 68 10.1016/S1573-5214(04)80029-0 Ebregt E Struik PC Odongo B Abidin PE Piecemeal versus one-time harvesting of sweet potato in north-eastern Uganda with special reference to pest damage NJAS - Wageningen J Life Sci 2007 55 1 75 92 10.1016/S1573-5214(07)80005-4 El-Sharkawy MA, Cadavid L (2002) Response of cassava to prolonged water stress imposed at different stages of growth. Exp Agric 38(03):333–350. 10.1017/S001447970200306X Goddard L, Mason SJ, Zebiak SE, Ropelewski CF, Basher R, et al (2001) Current approaches to seasonal to interannual climate predictions. Int J Climatol 21(9):1111–1152. 10.1002/joc.636 Goddard L, Aitchellouche Y, Baethgen W, Dettinger M, Graham R, et al (2010) Providing seasonal-to-interannual climate information for risk management and decision-making. Procedia Environ Sci 1:81–101. 10.1016/j.proenv.2010.09.007 Gujarati DN Basic Econometrics. Boston 2013 United States of America McGraw-Hill companies Haigh T Takle E Andresen J Widhalm M Carlton JS Angel J Mapping the decision points and climate information use of agricultural producers across the US Corn Belt Clim Risk Manag 2015 7 20 30 10.1016/j.crm.2015.01.004 Hampson KJ, Chapota R, Emmanuel J, Tall A, Huggins-Rao S, et al (2014) Delivering climate services for farmers and pastoralists through interactive radio: scoping report for the GFCS Adaptation Programme in Africa. Retrieved from CGIAR Research Program on Climate Change, Agriculture and Food Security (CCAFS)Copenhagen, Denmark. https://hdl.handle.net/10568/65728 Hansen JW Realizing the potential benefits of climate prediction to agriculture: Issues, approaches, challenges Agric Syst 2002 74 3 309 330 10.1016/S0308-521X(02)00043-4 Ingrama KT, Roncolia MC, Kirshen PH (2002) Opportunities and constraints for farmers of west Africa to use seasonal precipitation forecasts with Burkina Faso as a case study. Agric Syst 74:331–349. 10.1016/S0308-521X(02)00044-6 Isabirye P (2017) Factoring spatial variability of rainfall in the design of optimum rainguage network for Uganda. Thesis for Master of science for climate change and Disater Management. Busitema University, Uganda. https://ir.busitema.ac.ug/bitstream/handle/20.500.12283/1898/Isabirye%20-%20Abstract.pdf?sequence=2&isAllowed=y Jost C, Kyazze F, Naab J, Neelormi S, Kinyangi J, et al (2015a) Understanding gender dimensions of agriculture and climate change in smallholder farming communities. Clim Dev 8(3):133-144.10.1080/17565529.2015.1050978 Jost C, Kyazze F, Naab J, Neelormi S, Kinyangi J, et al (2015b) Understanding gender dimensions of agriculture and climate change in smallholder farming communities. Clim Dev 8(2):1-12.10.1080/17565529.2015.1050978 Kahneman D, Tversky A (1982) The psychology of preferences. Scientific American 246:160–73 Kalanda-Joshua M Ngongondo C Chipeta L Mpembeka F Integrating indigenous knowledge with conventional science: enhancing localised climate and weather forecasts in Nessa, Mulanje, Malawi Phys Chem Earth, Parts a/b/c 2011 36 14–15 996 1003 10.1016/j.pce.2011.08.001 Klemm T McPherson RA The development of seasonal climate forecasting for agricultural producers Agric for Meteorol 2017 232 384 399 10.1016/j.agrformet.2016.09.005 Kolawole OD, Wolski P, Ngwenya B, Mmopelwa G (2014) Ethno-meteorology and scientific weather forecasting: small farmers and scientists’ perspectives on climate variability in the Okavango Delta, Botswana. Clim Risk Manag 4:43–58. 10.1016/j.crm.2014.08.002 Landman WA Archer ERM Tadross MA Citizen science for the prediction of climate extremes in South Africa and Namibia Frontiers in Climate 2020 2 1 8 10.3389/fclim.2020.00005 Leroux M The meteorology and climate of tropical Africa 2001 UK Springer Science & Business Media, UK Liebenehm S, Waibel H (2014) Simultaneous estimation of risk and time preferences among small-scale cattle farmers in West Africa. Am J Agr Econ 96(5):1420–1438. 10.1093/ajae/aau056 Lo H, Dieng M (2015) Impact assessment of communicating seasonal climate forecasts in Kaffrine, Diourbel, Louga, Thies and Fatick (Niakhar) regions in Senegal: Final Report for CCAFS West Africa Regional Program. Retrieved from Copenhagen, Denmark. http://bit.ly/1ifHC3P Luseno WK Mcpeak JG Barrett CB Little PD Gebru G Assessing the value of climate forecast information for pastoralists: evidence from Southern Ethiopia and Northern Kenya World Dev 2003 31 9 1477 1494 10.1016/S0305-750X(03)00113-X Lybbert TJ Barrett CB Mcpeak JG Luseno WK Bayesian Herders: Updating of Rainfall Beliefs in Response to External forecasts World Dev 2007 35 3 480 497 10.1016/j.worlddev.2006.04.004 MacKerron D, Jefferies R (1986) The influence of early soil moisture stress on tuber numbers in potato. Potato Res 29(3):299–312. 10.1007/BF02359959 McNie EC (2007) Reconciling the supply of scientific information with user demands: an analysis of the problem and review of the literature. Environ Sci Policy 10(1):17–38. 10.1016/j.envsci.2006.10.004 Medany M Niang-Diop I Nyong T Tabo R Background paper on impacts, vulnerability and adaptation to climate change in Africa 2006 Ghana Paper presented at the UNFCCC Convention Mpandeli S, Maponya P (2013) The Use of Climate Forecasts Information by Farmers in Limpopo Province, South Africa. J Agric Sci 5(2):47. 10.5539/jas.v5n2p47 Muema E, Mburu J, Coulibaly J, Mutune J (2018) Determinants of access and utilisation of seasonal climate information services among smallholder farmers in Makueni County, Kenya. Heliyon 4(11):e00889. 10.1016/j.heliyon.2018.e00889 Mugalavai EM Kipkorir EC Raes D Rao MS Analysis of rainfall onset, cessation and length of growing season for western Kenya Agric for Meteorol 2008 148 6–7 1123 1135 10.1016/j.agrformet.2008.02.013 Nicholls N (1999) Cognitive illusions, heuristics, and climate prediction. Bull Am Meteor Soc 80(7):1385. 10.1175/1520-0477 Nkomwa EC, Kalanda JM, Ngongondo C, Monjerezi M, Chipungu F (2014) Assessing indigenous knowledge systems and climate change adaptation strategies in agriculture: a case study of Chagaka Village, Chikhwawa, Southern Malawi. Phys Chem Earth, Parts a/b/c 67:164–172. 10.1016/j.pce.2013.10.002 Nkuba M Chanda R Mmopelwa G Mangheni MN Lesolle D Kato E Do indigenous forecasts and scientific forecasts influence arable farmers’ and agro-pastoralists’ estimation of onset and cessation of rains? Empirical evidence from Rwenzori region, Western Uganda Agric Forest Meteorol 2019 278 107667 10.1016/j.agrformet.2019.107667 Nkuba M Chanda R Mmopelwa G Kato E Mangheni MN Lesolle D Influence of indigenous knowledge and scientific climate forecasts on arable farmers’ climate adaptation methods in the Rwenzori region, Western Uganda Environ Manag 2020 10.1007/s00267-020-01264-x Nkuba M Chanda R Mmopelwa G Mangheni MN Lesolle D Kato E Indigenous knowledge systems and indicators of rain: evidence from Rwenzori Region, Western Uganda Weather, Climate, and Society 2020 12 2 213 234 10.1175/wcas-d-19-0027.1 Nkuba M Chanda R Mmopelwa G Mangheni MN Lesolle D Adedoyin A Mujuni G Determinants of pastoralists’ use of indigenous knowledge and scientific forecasts in Rwenzoriregion, WesternUganda Clim Serv 2021 23 100242 10.1016/j.cliser.2021a.100242 Nkuba M, Chanda R, Mmopelwa G, Kato E, Mangheni MN, et al (2021b) Indigenous and scientific forecasts on climate change perceptions of arable farmers: Rwenzori region, Western Uganda. Chapter 82. pg 1685–1703. In W.Leal Filho, N.Ogugu, L.Adelake,D.Ayal and I.Da Silva (Eds), African Handbook of climate change adaptation. Vol 2. Climate Change, Technologies, and Resources Management. Springer Nature Switzerland. 10.1007/978-3-030-45106-6_113 Nkuba M, Chanda R, Mmopelwa G, Kato E, Mangheni MN, et al (2021c) Barriers to Climate change adaptation among Pastoralists: Rwenzori region, Western Uganda. Chapter 35. pg. 659–676. In W Leal Filho, N Ogugu, L Adelake,D Ayal and I Da Silva (Eds), African Handbook of climate change adaptation. Vol 2. Climate Change, Technologies, and Resources Management. Springer Nature Switzerland. 10.1007/978-3-030-45106-6_99 Ofoegbu C New M The role of farmers and organizational networks in climate information communication: the case of Ghana Int J Clim Change Strat Manag 2021 13 1 19 34 10.1108/IJCCSM-04-2020-0030 Orlove B Roncoli C Kabugo M Majugu A Indigenous climate knowledge in southern Uganda: the multiple components of a dynamic regional system Clim Change 2010 100 243 265 10.1007/s10584-009-9586-2 Oyekale AS Factors explaining farm households’ access to and utilization of extreme climate forecasts in Sub-Saharan Africa (SSA) Environ Econ 2015 6 1 91 103 Partey ST Dakorah AD Zougmoré RB Ouédraogo M Nyasimi M Nikoi GK Huyer S Gender and climate risk management: evidence of climate information use in Ghana Clim Change 2018 158 61 75 10.1007/s10584-018-2239-6 Patt A, Gwata C (2002) Effective seasonal climate forecast applications: examining constraints for subsistence farmers in Zimbabwe. Glob Environ Chang 12(3):185–195. 10.1016/S0959-3780(02)00013-4 Patt A Suarez P Gwata C Effects of seasonal climate forecasts and participatory workshops among subsistence farmers in Zimbabwe Proc Natl Acad Sci USA 2005 102 35 12623 12628 10.1073/pnas.0506125102 16116076 Roncoli C Ingram K Kirshen P Reading the rains: local knowledge and rainfall forecasting in Burkina Faso Soc Nat Resour 2002 15 5 409 427 10.1080/08941920252866774 Roncoli C, Jost C, Kirshen P, Sanon M, Ingram KT, et al (2008) From accessing to assessing forecasts: an end-to-end study of participatory climate forecast dissemination in Burkina Faso (West Africa). Clim Change 92(3-4):433-460.10.1007/s10584-008-9445-6 Roudier P Muller B d’Aquino P Roncoli C Soumaré MA Batté L Sultan B The role of climate forecasts in smallholder agriculture: lessons from participatory research in two communities in Senegal Clim Risk Manag 2014 2 42 55 10.1016/j.crm.2014.02.001 Slovic P Slovic P Perception of risk The perception of risk 2000 London, UK Earthscan publications 220 231 Slovic P, Fischhoff B, Lichtenstein S (2000a) Cognitive processes and societal risk taking The Perception of Risk. Earthscan, London, UK, pp 32–50 Slovic P, Fischhoff B, Lichtenstein S (2000b) Facts and fears: understanding perceived risk. In: Slovic P (ed) Perception of Risk. Earthscan, London, UK, pp 137–153 Slovic P, White G, Kunreuther H (2000c) Decision processes, rationality, and adjustments to natural hazards. In: Slovic P (ed) Perception of Risk. Earthscan, London, UK, pp 1–31 Snow JT, Biagini B, Benchwick G, George G, Hoedjes J, et al (2016) A new vision for weather and climate services in Africa. Retrieved from New York Speranza CI Kiteme B Ambenje P Urs W Makali S Indigenous knowledge related to climate variability and change: insights from droughts in semi-arid areas of former Makueni District, Kenya Clim Change 2010 100 295 315 10.1007/s10584-009-9713-0 Steduto P, Hsiao TC, Fereres E, and Raes D (2012) Crop yield response to water: FAO Roma Tversky A, Kahneman D (1974) Judgment under uncertainty: heuristics and biases. Science 185(4157):1124–1131. 10.1126/science.185.4157.1124 Tversky A, Kahneman D (1979) Prospect theory: an analysis of decision under risk. Econometrica 47(2):263–291. 10.2307/1914185 UNECA (2011) Assessment of Africa’s climatic records and recording networks including strategic for rescuing of climatic data. Retrieved from Adisa Baba, United Nations Economic Commission for Africa; African Climate Policy Centre UNMA (2017a) March to May 2017a seasonal rainfall outlook over Uganda Kampala Uganda; Uganda National Meteorogical Authority Uganda National Meteorogical Authority UNMA (2017b) September to December 2017b Seasonal rainfall outlook over Uganda Kampala, Uganda Uganda National Meteorogical Authority Uganda National Meteorogical Authority Vogel C Usable science: an assessment of long-term seasonal forecasts amongst farmers in rural areas of Sourth Africa S Afr Geogr J 2000 82 2 107 116 10.1080/03736245.2000.9713700 Waldman KB Blekking JP Attari SZ Evans TP Maize seed choice and perceptions of climate variability among smallholder farmers Glob Environ Chang 2017 47 April 51 63 10.1016/j.gloenvcha.2017.09.007 Waldman KB Vergopolan N Attari SZ Sheffield J Estes LD Caylor KK Evans TP Cognitive biases about climate variability in smallholder farming systems in Zambia Weather, Climate, and Society 2019 11 2 369 383 10.1175/WCAS-D-18-0050.1 Yooyongwech S Theerawitaya C Samphumphuang T Cha-um S Water-deficit tolerant identification in sweet potato genotypes (Ipomoea batatas (L.) Lam.) in vegetative developmental stage using multivariate physiological indices Sci Hortic 2013 162 242 251 10.1016/j.scienta.2013.07.041 Ziervogel G (2004) Targeting seasonal climate forecasts for integration into household level decisions: the case of smallholder farmers in Lesotho. Geogr J 170(1):6–21. 10.1111/j.0016-7398.2004.05002.x Ziervogel G Calder R Climate variability and rural livelihoods: assessing the impact of seasonal climate forecasts in Lesotho Area 2003 35 4 403 417 10.1111/j.0004-0894.2003.00190.x Ziervogel G, Downing TE (2004) Stakeholder networks: improving seasonal climate forecasts. Clim Change 65(1–2):73–101. 10.1023/B:CLIM.0000037492.18679.9e Ziervogel G, Opere A (2010) Integrating meteorological and indigenous knowledge-based seasonal climate forecasts for the agricultural sector. http://hdl.handle.net/10625/46185	36494539	PMC9734387	NO-CC CODE	2022-12-14 23:28:28	no	Br Dent J. 2022 Dec 9; 233(11):945	latin-1	Br Dent J	2,022	10.1038/s41415-022-5338-8	oa_other
==== Front Anal Bioanal Chem Anal Bioanal Chem Analytical and Bioanalytical Chemistry 1618-2642 1618-2650 Springer Berlin Heidelberg Berlin/Heidelberg 36477907 4471 10.1007/s00216-022-04471-z Editorial Electrochemical biosensors — driving personalized medicine Lobo-Castañón Maria Jesús [email protected] 12María Jesús Lobo-Castañón holds a PhD in Chemistry and leads the Electroanalysis research group at the University of Oviedo in Spain. Since 2017, she has been Full Professor in Analytical Chemistry at the said university. Her research interests focus on the development of electrochemical sensors for clinical diagnosis and food analysis, using different molecular recognition elements, such as enzymes, DNA, and aptamers. Campuzano Susana [email protected] 3Susana Campuzano is Full Professor at the Analytical Chemistry Department of the Chemistry Faculty of the Universidad Complutense de Madrid (Spain) and Head of the “Electroanalysis and Electrochemical (Bio)Sensors” (GEBE-UCM) research group. Her areas of interest include the development of affinity-based electrochemical bioplatforms with potential for multiplexed and/or multi-omics determinations in clinical and food safety. 1 grid.10863.3c 0000 0001 2164 6351 Departamento de Química Física Y Analítica, Universidad de Oviedo, 33006 Oviedo, Spain 2 grid.511562.4 Instituto de Investigación Sanitaria del Principado de Asturias, 33011 Oviedo, Spain 3 grid.4795.f 0000 0001 2157 7667 Analytical Chemistry Department, Chemistry Faculty, University Complutense of Madrid, 28040 Madrid, Spain 8 12 2022 12 30 11 2022 © 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. ==== Body pmcPersonalization is everything nowadays; we can customize consumer products (T-shirts, phone cases, gifts, etc.) and get personalized recommendations on streaming services for movies, music, and so on. It is therefore not surprising that medicine and nutrition are fully taking root in this domain. We are privileged witnesses that the clinical management of patients is drastically changing from an approach based on general clinicopathological profiles to a personalized one that considers molecular alterations at the individual level, which is known as personalized medicine (PM). In addition to being a right for patients, PM can make a decisive contribution to making our lives longer and better and can play a key role in engaging us in our self-care. For society, it means a more rational use of resources and significant savings in healthcare costs and unnecessary effort and suffering. PM is becoming increasingly important in multiple diseases, such as oncological and neurodegenerative diseases, the most important example of which is Alzheimer’s disease, which involve many genetic disorders and unique molecular profiles. Although PM is having a growing impact on research and healthcare, mainly in oncology patients, where it is contributing to higher survival rates, its implementation in egalitarian healthcare has a long road ahead. This path will be conditioned, among other things, by the development and application of new technologies that allow the identification, validation, and determination of new biomarkers at multiplexed and/or multiomics levels in a decentralized, affordable and sustainable manner, a field in which electrochemical biosensing technology has much to say and offer. Having first-hand knowledge of what cutting-edge electrochemical biosensing can offer in PM, we approached the preparation of this topical collection with tremendous enthusiasm. Our aim was to put even more focus on this subject of the highest and most timely relevance, through unique contributions from renowned researchers who share our concerns in this area. All of them lead research groups responsible for pioneering contributions, and from the moment we contacted them, they were compelled to participate in this project by providing unique contributions. This is how this topical collection was born, which takes advantage of the opportunity and diffusion channels provided by the journal Analytical & Bioanalytical Chemistry, making visible some representative breakthroughs of the last few years in the field of electrochemical biosensing research to meet the needs in personalized healthcare. The collection comprises five review articles and three research articles (2 selected by the Editorial team of this journal as Paper in Forefront). These outstanding contributions focus on state-of-the-art bio-devices that have demonstrated pioneering and decisive capabilities for the determination of clinical biomarkers of different nature (proteins, point mutations, exosomes, metal ions, and antimicrobial compounds). They are applied to personalized medicine of both known diseases (cancerous, neurological, inflammatory, infectious, etc.) and those, like COVID-19, which may arise unexpectedly, turning our lives upside down. Designs implemented using nanomaterials, different bioreceptors, less conventional electrode substrates (paper, flexible, wearable), and electrochemical techniques (photoelectrochemistry) are included. The versatility, rapid adaptation, and the groundbreaking advances that electrochemical biosensing approaches are constantly demonstrating to drive personalized medicine make us think that this topical collection is just a foretaste of the wonderful contributions we will witness in the near future. Much remains to be done on this subject, which, due to its relevance both at a scientific and social level, does not stop feeding back and is more awake than ever. We expect that the papers in this collection will serve as inspiration for further work to address new bioelectrochemical technologies that contribute to making personalized medicine a reality. At this point, we must admit that we are flattered to be involved in this topical collection, and to belong to the Editorial Board of Analytical & Bioanalytical Chemistry. We are tremendously grateful to many people and for much. Firstly, to all the important scientists who have contributed (Dr. Jahir Orozco from Colombia, Dr. Mustafa Sezgintürk from Turkey, Dr. Cecilia Cristea from Romania; Dr. Martin Bartosik from the Czech Republic; Dr. Fabiana Arduini and Dr. Ilaria Palchetti from Italy, Dr. Damion K. Corrigan from the United Kingdom; and Dr. Alfredo de la Escosura-Muñiz from Spain); secondly to other great scientists who have participated in this TC judging these contributions as reviewers. Thirdly to the Editorial team of Analytical & Bioanalytical Chemistry and particularly to Maite Menes for her kind support with the management of this topical collection. Finally, to our dear María Cruz Moreno-Bondi, who proposed us as Guest Editors and to whom we would love to dedicate it. We hope and trust that this topical collection will be to your liking and satisfy your expectations. Published in the topical collection Electrochemical Biosensors – Driving Personalized Medicine with guest editors Susana Campuzano Ruiz and Maria Jesus Lobo-Castañón. Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.	36477907	PMC9734388	NO-CC CODE	2022-12-14 23:28:28	no	Anal Bioanal Chem. 2022 Dec 8;:1-2	utf-8	Anal Bioanal Chem	2,022	10.1007/s00216-022-04471-z	oa_other
==== Front Chin J Integr Med Chin J Integr Med Chinese Journal of Integrative Medicine 1672-0415 1993-0402 Springer Nature Singapore Singapore 36484921 3627 10.1007/s11655-022-3627-3 Original Article Therapeutic Efficacy of Shexiang Baoxin Pill Combined with Exercise in Patients with Heart Failure with Preserved Ejection Fraction: A Single-Center, Double-Blind, Randomized Controlled Trial Liu Si-pei 12 Zhou Jian-guan 3 Jin Yan 2 Guo Yan 4 Zhou Shi-wei 3 Lin Mei-lan 12 Zhang Jun 3 Wang Xiao-nv 3 Guan Xia-fei 3 Wang Lei [email protected] 12 1 grid.410745.3 0000 0004 1765 1045 Nanjing Drum Tower Hospital, Clinical College of Nanjing University of Chinese Medicine, Nanjing, 210008 China 2 grid.410745.3 0000 0004 1765 1045 Department of Rehabilitation, College of Acupuncture and Moxibustion and Massage Health Preservation and Rehabilitation, Nanjing University of Chinese Medicine, Nanjing, 210023 China 3 Division of Cardiovascular Rehabilitation, Ruian Hospital of Traditional Chinese Medicine, Ruian, Zhejiang Province, 325299 China 4 grid.410318.f 0000 0004 0632 3409 Xiyuan Hospital, China Academy of Chinese Medical Sciences, Beijing, 100091 China 9 12 2022 19 21 9 2022 © The Chinese Journal of Integrated Traditional and Western Medicine 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. Objective To evaluate the therapeutic efficacy of Shexiang Baoxin Pill combined with exercise in heart failure patients with preserved ejection fraction (HFpEF). Methods Sixty patients with HFpEF were randomly divided into group A (n=20), receiving Shexiang Baoxin Pill combined with home-based exercise training based on conventional drugs for 12 weeks; group B (n=20), receiving conventional drugs combined with home-based exercise training for 12 weeks; and group C (n=20), receiving conventional drug treatment only. Peak oxygen uptake (peakVO2), anaerobic threshold (AT), 6-min walking test (6MWT), Pittsburgh Sleep Quality Index (PSQI), and SF-36 questionnaire (SF-36) results before and after treatment were compared among groups. Results After the 12-week intervention, patients in group C showed significant declines in peakVO2, AT, 6MWT, PSQI, and SF-36 compared with pre-treatment (P<0.01), while groups A and B both showed significant improvements in peakVO2, AT, 6MWT, PSQI, and SF-36 results compared with pre-treatment (P<0.01). Compared with group C, patients in groups A and B showed significant improvements in peakVO2, AT, 6MWT, PSQI, and SF-36 (P<0.01). In addition, patients in group A showed more significant improvements in physical function, role-physical, vitality, and mental health scores on the SF-36 questionnaire, and PSQI scores than those in group B (P<0.01). Conclusions Exercise training improved exercise tolerance, sleep quality and quality of life (QoL) in patients with HFpEF. Notably, Shexiang Baoxin Pill played an active role in sleep quality and QoL of patients with HFpEF. (The trial was registered in the Chinese Clinical Trial Registry (No. ChiCTR2100054322)) Keywords heart failure preserved ejection fraction exercise training Shexiang Baoxin Pill ==== Body pmcAcknowledgements We thank the patients and staff of the Ruian Hospital Division of Cardiovascular Rehabilitation Department for their valuable contributions to the study. Supported by the Chinese Association of Integrative Medicine-Hutchison Research Fund (No. HMP2005002P) Conflict of Interest The authors declare no conflicts of interest. Author Contributions Wang L, Guan XF and Guo Y were equally contributed to the conception of the manuscript; Liu SP, Zhou JG and Jin Y were contributed to the writing and revision of the manuscript; Liu SP, Zhou JG, Zhou SW, Lin ML, Zang J and Wang XN participated in the patients’ recruitment and data collection; Liu SP, Jin Y and Wang L analyzed the data; Wang L and Guan XF were contributed to review and revision of the manuscript; All authors have agreed with the content and approve of the manuscript for submission. ==== Refs References 1. Zhuang CC Luo XF Wang QY Wang WJ Sun RM Zhang XF The effect of exercise training and physiotherapy on diastolic function, exercise capacity and quality of life in patients with heart failure with preserved ejection fraction: a systematic review and meta-analysis Kardiol Pol 2021 7910 1107 1115 10.33963/KP.a2021.0101 2. Zakeri R Cowie MR Heart failure with preserved ejection fraction: controversies, challenges and future directions Heart 2018 1045 377 384 10.1136/heartjnl-2016-310790 3. Naing P Forrester D Kangaharan N Muthumala A Myint SM Playford D Heart failure with preserved ejection fraction: a growing global epidemic Aust J Gen Pract 2019 487 465 471 10.31128/AJGP-03-19-4873 4. Oktay AA Shah SJ Diagnosis and management of heart failure with preserved ejection fraction: 10 key lessons Curr Cardiol Rev 2015 111 42 52 5. Tucker WJ Nelson MD Beaudry RI Halle M Sarma S Kitzman DW Impact of exercise training on peak oxygen uptake and its determinants in heart failure with preserved ejection fraction Cardiac Fail Rev 2016 22 95 101 6. Lee KS Lennie TA Heo S Song EK Moser DK Prognostic importance of sleep quality in patients with heart failure Am J Crit Care 2016 256 516 525 10.4037/ajcc2016219 7. Fu TC Huang SC Hsu CC Wang CH Wang JS Cardiac rehabilitation in patients with heart failure Acta Cardiol Sin 2014 305 353 359 8. Haykowsky MJ Daniel KM Bhella PS Sarma S Kitzman DW Heart failure: exercise-based cardiac rehabilitation: who, when, and how intense? Can J Cardiol 2016 3210 S382 S387 10.1016/j.cjca.2016.06.001 9. Piepoli MF Hoes AW Agewall S Albus C Brotons C Catapano AL 2016 European Guidelines on cardiovascular disease prevention in clinical practice Rev Espan de Cardiol (English) 2016 6910 939 10. Fukuta H Goto T Wakami K Kamiya T Ohte N Effects of exercise training on cardiac function, exercise capacity, and quality of life in heart failure with preserved ejection fraction: a meta-analysis of randomized controlled trials Heart Fail Rev 2019 244 535 547 10.1007/s10741-019-09774-5 11. Fukuta H Goto T Wakami K Ohte N Effects of drug and exercise intervention on functional capacity and quality of life in heart failure with preserved ejection fraction: a meta-analysis of randomized controlled trials Eur J Prev Cardiol 2016 231 78 85 10.1177/2047487314564729 12. Wang JP Yang R Zhang FL Jia CX Wang PP Liu JJ The effect of Chinese herbal medicine on quality of life and exercise tolerance in heart failure with preserved ejection fraction: a systematic review and meta-analysis of randomized controlled trials Front Physiol 2018 9 15 10.3389/fphys.2018.01420 29410630 13. Fang HY Zeng HW Lin LM Chen X Shen XN Fu P A network-based method for mechanistic investigation of Shexiang Baoxin Pill’s treatment of cardiovascular diseases Sci Rep 2017 7 11 28127060 14. Lu L Sun XD Chen C Qin YT Guo XM Shexiang Baoxin Pill, Derived from the traditional Chinese medicine, provides protective roles against cardiovascular diseases Front Pharmacol 2018 9 10 10.3389/fphar.2018.01161 29434546 15. Fukuta H Effects of exercise training on cardiac function in heart failure with preserved ejection fraction Cardiac Fail Rev 2020 6 e27 10.15420/cfr.2020.17 16. Leggio M Fusco A Loreti C Limongelli G Bendini MG Mazza A Effects of exercise training in heart failure with preserved ejection fraction: an updated systematic literature review Heart Fail Rev 2020 255 703 711 10.1007/s10741-019-09841-x 17. Schmidt C Moreira-Goncalves D Santos M Leite-Moreira A Oliveira J Physical activity and exercise training in heart failure with preserved ejection fraction: gathering evidence from clinical and pre-clinical studies Heart Fail Rev 2022 272 573 586 10.1007/s10741-020-09973-5 18. Moher D Hopewell S Schulz KF Montori V Gotzsche PC Devereaux PJ CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials Int J Surg 2012 101 28 55 10.1016/j.ijsu.2011.10.001 19. Turri-Silva N Vale-Lira A Verboven K Quaglioti Durigan JL Hansen D Cipriano G High-intensity interval training versus progressive high-intensity circuit resistance training on endothelial function and cardiorespiratory fitness in heart failure: a preliminary randomized controlled trial PLoS One 2021 1610 25 20. Liao YH Yang JF Zhang J Cheng X Xie MX Li XL An expert consensus for diagnosis and treatment of diastolic heart failure J Clin Cardiol (Chin) 2020 3601 1 10 21. Wang LM Shen YQ Chinese expert consensus on exercise rehabilitation in chronic stable heart failure Chin Circ J (Chin) 2014 z2 113 119 22. Salzano A De Luca M Israr MZ Crisci G Eltayeb M Debiec R Exercise intolerance in heart failure with preserved ejection fraction Heart Fail Clin 2021 173 397 413 10.1016/j.hfc.2021.03.004 23. Parati G Lombardi C Castagna F Mattaliano P Filardi PP Agostoni P Heart failure and sleep disorders Nat Rev Cardiol 2016 137 389 403 10.1038/nrcardio.2016.71 24. Awotidebe TO Adeyeye VO Adedoyin RA Ogunyemi SA Oke KI Ativie RN Assessment of functional capacity and sleep quality of patients with chronic heart failure Hong Kong Physiother J 2017 36 17 24 10.1016/j.hkpj.2016.10.001 30931035 25. Suna JM Mudge A Stewart I Marquart L O’Rourke P Scott A The effect of a supervised exercise training programme on sleep quality in recently discharged heart failure patients Eur J Cardiovasc Nurs 2015 143 198 205 10.1177/1474515114522563 26. Esnaasharieh F Dehghan M Shahrbabaki PM The relationship between sleep quality and physical activity among patients with heart failure: a cross-sectional study BMC Sports Sci Med Rehabil 2022 141 8 27. Hassanpour Dehkordi A Khaledi Far A Effect of exercise training on the quality of life and echocardiography parameter of systolic function in patients with chronic heart failure: a randomized trial Asian J Sport Med 2015 61 e22643 28. Dallas K Dinas PC Chryssanthopoulos C Dallas G Maridaki M Koutsilieris M The effects of exercise on VO2peak, quality of life and hospitalization in heart failure patients: a systematic review with meta-analyses Eur J Sport Sci 2021 219 1337 1350 10.1080/17461391.2020.1846081 29. Wang ZX Liu ZC Gao SW Wang BH Research progress on the mechanism of Shexiang Baoxin Pill J Emerg Trad Chin Med (Chin) 2020 2907 1309 1312 30. Wu ZX Jang ZT Luo J Zhou SB Xiao T Li F Effect of Shexiang Baoxin Pill on myocardial remodeling and gap junction proteins Cx43 and Cx45 in heart failure rats Chin Trad Patent Med (Chin) 2015 3706 1329 1332 31. Wu BW Li J Jin B Li HY Shi HM Luo XP Mechanism of promoting angiogenesis by Shexiang Baoxin Pill Chin Trad Patent Med (Chin) 2018 4006 1384 1388 32. Zhang BL Wang YL Jing XM Advances of sleep disorder in heart failure patients J North Sichuan Med Coll (Chin) 2017 3203 471 474 33. Zhou XD Shi DD Zhang ZJ Antidepressant and anxiolytic effects of the proprietary Chinese medicine Shexiang Baoxin Pill in mice with chronic unpredictable mild stress J Food Drug Anal 2019 271 221 230 10.1016/j.jfda.2018.08.001 34. Li YZ Ou YL Zhang YN Jin C Gong S Yu GL Observation on the efficacy of Shexiang Baoxin Pill on coronary heart disease with angina pectoris and mild to moderate anxiety and depression Modern J Integ Trad Chin West Med (Chin) 2022 3108 1062 1066 35. Lu YS Liu M Efficacy of escitalopram combined with Shexiang Baoxin Pill in chronic heart failure complicated by depression Hubei J Trad Chin Med (Chin) 2015 3711 49 50 36. Wang KB Effect of Shexiang Baoxin Pill on chronic heart failure Cardiovasc Dis J Integr Chin West Med (Chin) 2015 323 85 86 37. Yao HL Xu MW Wu Q Effect of Shexiang Baoxin Pill combined with sacubitril valsartan sodium Tablets on cardiac function and SF-36 scores in chronic heart failure patients J North Pharm (Chin) 2020 1710 27 28 38. Zhu YH Clinical observation of Shexiang Baoxin Pill in the treatment of chronic heart failure Med J Chin People Health (Chin) 2014 2607 86 87 39. Zhang KJ Effect of exercise rehabilitation combined with Shexiang Baoxin Pill on cardiac function and daily living ability of chronic heart failure patients with coronary heart disease Chin J Convalesc Med (Chin) 2019 2807 722 723	36484921	PMC9734389	NO-CC CODE	2022-12-14 23:28:28	no	Chin J Integr Med. 2022 Dec 9;:1-9	utf-8	Chin J Integr Med	2,022	10.1007/s11655-022-3627-3	oa_other
==== Front Sugar Tech Sugar Tech Sugar Tech 0972-1525 0974-0740 Springer India New Delhi 1223 10.1007/s12355-022-01223-2 Research Article Optimization of Extraction Parameters and Characterization of Tunisian Date Extract: A Scientific Approach Toward Their Utilization http://orcid.org/0000-0002-6764-6688 Messadi Nesrine [email protected] 1 http://orcid.org/0000-0002-0587-4773 Mechmeche Manel 1 http://orcid.org/0000-0002-4244-4174 Setti Khaoula 1 Tizemmour Zoulikha 1 Hamdi Moktar 2 http://orcid.org/0000-0001-6515-0235 Kachouri Faten 1 1 grid.419508.1 0000 0001 2295 3249 Laboratory of Innovation and Valorization for Sustainable Food Industry, Superior School of Food Industry at Tunis (ESIAT), University of Carthage, 58, Street Alain Savary, 1003 Tunis, Tunisia 2 grid.419508.1 0000 0001 2295 3249 Laboratory Microbial Ecology and Technology (LETMI), National Institute of Applied Sciences and Technology (INSAT), University of Carthage, BP, 676, 1080 Tunis, Tunisia 8 12 2022 113 24 5 2022 5 11 2022 © The Author(s), under exclusive licence to Society for Sugar Research & Promotion 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 response surface methodology (RSM) was used in order to select the extraction conditions of extract from Kentichi date powder; a by-product of the date-processing process. Powder/solvent ratio, extraction temperature, and extraction time all had an impact on sugar yield, and these model factors have quadratic effects influencing sugar yield. Optimal extraction was obtained with 300 g/L powder/solvent ratio, 32.7 °C extraction temperature, and 2.1 h extraction time. Under these conditions, Kentichi date powder's (KDP) sugar yield was 77.1%, which was close to the predicted value of the model (80.50%). The results of Kentichi date powder extract (KDPE) showed that the total sugar content is 160.09 g/L. However, the protein content is 10.31 g/L with a majority of the essential amino acids (essentially glutamic acid (28.39 mg/L) and aspartic acid (9.65 mg/L)). The determination of antioxidant activity of KDPE showed a high activity (DPPH IC50 = 4.8 mg/mL, ABTS IC50 = 3 mg/mL, FRAP = 4.70 μmol AAE/mL and, TAA = 18.04 μmol Fe(II)/mL). The results show also that the freeze-drying technique has a lot of potential for producing powder from KDPE with many desirable properties. The findings indicate that KDPE with a high nutritional value could be used as a component for the formulation of functional foods. Supplementary Information The online version contains supplementary material available at 10.1007/s12355-022-01223-2. Keywords Kentichi date powder extract Response surface methodology Antioxidant activity Freeze-drying ==== Body pmcIntroduction Nowadays, the COVID-19 pandemic has put humanity in jeopardy all over the world. Therefore, people must strengthen their immune systems in order to combat the virus (Galanakis et al. 2020). A diet rich in fruits and vegetables can help to boost the immune system (Moreb et al. 2021). Many studies have shown that fruits and vegetables are rich in fiber and phytochemicals and can prevent or reduce the risk of diseases including cardiovascular disease, diabetes, obesity, certain types of cancer, inflammation, stroke, and septic shock (Schauder et al. 2020). Date palm (Phoenix dactylifera L.) belongs to the Arecaceae family (Angiosperms, monocots) which comprises 200 genera and more than 2 500 species (Al-Alawi et al. 2017). According to the literature, low-quality dates are used to feed animals and are often processed into date paste and date syrup, which are both widely used in the food industry by-products (Najjar et al. 2020). These by-products have health and nutritional benefits, including an immunostimulating effect on the reticuloendothelial system and strong antioxidant activity (Zerizer et al. 2014). Several researchers showed that dates are rich in phenolic antioxidants, but their value varies depending on the variety of dates, agronomic, and environmental conditions (Alam et al. 2021). Furthermore, dates contain nutrients such as protein (essential amino acids), fiber, fat, minerals, and vitamins (Ibrahim et al. 2020). In addition, the most essential components of date fruits are carbohydrates including soluble sugars (glucose, fructose, and sucrose) and dietary fiber (cellulose, hemicelluloses, pectin, and fructans) (Kamal-Eldin et al. 2020). The date is an important nutrient and energy source due to its high carbohydrate content (Siddiqi et al. 2020). Thus, fresh dates contain about 157 cal per 100 g, whereas dry dates contain more than 300 cal per 100 g (Aljaloud et al. 2020). Dates have the potential to contain a variety of bioactive phytochemicals, and dates have simple phenolic acids (gallic acid, vanillic acid, syringic acid), carotenoids (lutein, β-carotene), flavonoids and their derivatives (catechin, epicatechin, quercetin, apigenin), phytosterols (cholesterol, campesterol, -sitosterol), phenylpropanoids (caffeic acid, 5-O-caffeoylshikimic acid, ferulic acid) and anthocyanins (Zihad et al. 2021). Due to these antioxidant components, dietary recommendations recommend increasing the consumption of antioxidant-rich foods. Date sugar can be one of those foods with extremely high antioxidant levels (Phillips et al. 2009). However, it became important to extract sugar from dates. Different unconventional extraction sugar procedures have been utilized in order to maximize the extraction yield of bioactive macromolecules from various plant sections (Nadar et al. 2018). Thus, water extraction usually requires a combination with other innovative techniques, such as ultrasonic-assisted extraction (UAE), microwave-assisted extraction (MAE), pressurized water extraction (PWE), and enzyme-assisted extraction (EAE), which facilitate polysaccharide dissolution via biodegradation or mechanical disintegration of plant cell walls (Chen et al. 2019). Nevertheless, the extraction, isolation, and fractionation techniques change depending on the eventual purpose. For food application, aqueous extraction provides numerous advantages because water is not only economical and environmentally friendly, but it is also non-flammable and nontoxic, allowing for clean processing and pollution prevention (Filly et al. 2016). Therefore, the aqueous treatment approach for extracting sugar from Kentichi date powder appears to be a productive and promising method for the increased recovery of fruit sugars at relatively low temperatures and without the utilization of any toxic solvents. Response Surface Methodology (RSM) is a method for developing and optimizing experimental data that are based on statistical and mathematical concepts. This modeling approach is beneficial when there are numerous input parameters that interact with one another and the interaction influences the system's output known as "response" (Peng et al. 2020). RSM utilizes an experimental design such as the central composite design (CCD) to fit a model using the least squares method, and the diagnostic testing tests offered by analysis of variance (ANOVA) are then used to show whether the suggested model is adequate. The response surface plots could be used to evaluate the surfaces and find the optimal values (Leili et al. 2020). The RSM technique has been utilized in numerous areas and has demonstrated its efficacy as a numerical method by comparing it to experimental studies and other numerical research (Farouk et al. 2022). However, no scientific research is published to study the impacts of extraction time and temperature on sugar extract from date powder using simple technology, as well as its functional characteristics and antioxidant capacity. The aims of this study were to maximize sugar yield of Tunisian date powder, employing RSM (a central composite rotary design), to characterize the optimized extract from Tunisian date powder and to assess its antioxidant capacity. An essay on preserving the extract through a freeze-drying method is also evaluated in order to increase its shelf life while keeping its nutritive value. Material and Methods Plant Material The dates used in this work are Tunisian dates of "Kentichi" variety purchased at the central market (Tunis) in the Tamar stage (fully mature). The dates were cleaned, dried at 50 °C for 15 min, and ground to a powder with a mean particle size of 500 µm. The powder obtained was stored in airtight bottles in the refrigerator at 4 °C until later use (Yefsah-Idres et al. 2019). Kentichi Date Powder Extract The Kentichi date powder (KDP) was extracted with selected combinations of independent factors such as powder/solvent ratio, extraction time, and extraction temperature. To separate the extract from the insoluble residue, centrifugation at 6000 rpm for 10 min was performed to remove particles. The sugar yield was determined by the method of Dubois et al. (1956). Sugar ratio of the Kentichi date powder (KDPE) was expressed as the ratio of the extracted sugar in the total sugar content of the KDPE (Arrutia et al. 2020). The ratio of sugar extraction yield (%) was calculated as following Eq. 1:1 Sugarextractionyield%=MseMsp×100 where Mse is the concentration of sugar in the extract (g/100 g); Msp is the concentration of sugar in the pulp (g/100 g). Optimized extract from Kentichi date powder (KDPE) was then employed for composition and antioxidant analyses. Experimental Design A centered and rotating compound plan (CCRD) with three variables and five levels was created. To obtain a second-order polynomial model that describes the sugar yield (Kentichi date powder extract (KDPE)) Y (dependent variable) as a function with three independent variables, the ratio between solvent and date powder (X1) from 100 to 500 g/L, extraction temperature (X2) from 20 to 60 °C, and extraction time (X3) from 0 to 8 h (Table 1). Table S1 shows the coded and uncoded variables used in the design of surface response methodology.Table 1 Independent variables and their levels for optimizing the extraction of sugar from Kentichi date powder Independent variables Symbols Coded factor levels Coded Uncoded − 1.68 − 1 0 1 1.68 Powder/solvent ratio (g/L) X1 x1 100 200 300 400 500 Extraction temperature (°C) X2 x2 20 30 40 50 60 Extraction time (h) X3 x3 0 2 4 6 8 Second-order polynomial model, employed to describe the dependent variables Y as a function of the independent variables Xi, is presented in following Eq. 2:2 Y=β0+∑i=13βiXi+∑i=13βiiXi2+∑∑i<j=13βijXj+ε where Xi and Xj are the input factors that influence the response Y, while β0,βi,βii, and βij (i ≠ j) were constant coefficients regression of the model, the coefficients β, which must be determined in the second-order model, are obtained by the least-squares method, and ε is the error. Data analyses were performed using Expert Design Software, version 7. The ANOVA procedure was used to perform variance analysis. The mean values were regarded to be significantly different at p ≤ 0.05. Freeze-Drying KDPE was placed on the stainless steel trays of the freeze-dryer and frozen at − 20 °C for 24 h before freeze-drying. The trays were then transmitted to the freeze-dryer, and the vacuum turned on, allowing some of the pulp's free water to be sublimed off. This procedure was carried out for 72 h at 0.0041 mbar chamber pressure. The temperature of the condenser was − 65 ± 1 °C. The freeze-dried powder (FDP) was stored in an airtight dark glass container until the analysis stage. Analytical Methods The pH and titratable acidity were measured for KDPE according to the standard analytical method (AOAC 1990). The total sugars content was determined for KDPE and FDP using the sulfuric acid and phenol method (Dubois et al. 1956). The DNS method was employed for the determination of reducing sugars content of KDPE (Miller 1959), with glucose as a standard for total and reducing sugars. The sucrose content of KDPE was determined with the formula established by Chafi et al. (2015) using following Eq. 3:3 Sucrose%=total sugars%-reducing sugars% The protein assay of KDPE and FDP is carried out by the method of Lowry (1951), using bovine serum albumin as a standard. All experiments were performed in triplicate. Free Sugars Analysis Fructose, glucose, sucrose, maltose, and lactose contents were analyzed employing HPLC technique. KDPE was filtered using a 0.45 µm filter. In equal volumes, the analytical standards and sample were injected separately (5 µL). HPLC analysis was carried out employing liquid chromatography in conjunction with a diode-array detector (DAD). Separation was performed with Shimadzu SHIMPACK VP-ODS 4.6 × 150 mm, 5 µm RP column. As the mobile phase for the chromatographic analysis, a mixture of methanol and distilled water (10:90 v/v) was used. A 5 µL volume was injected at a rate of 1.2 mL/min with a run time was 20 min. The temperature of the column was kept at 30 °C throughout the experiment (Alghamdi et al. 2020). Amino Acid Analysis High-performance liquid chromatography with post-column fluorescence derivatization (HPLC-FLD) was used to determine the concentration of free amino acids in KDPE, employing a C18 column with a particle dimension of 5 μm (250 × 4.6 mm) (Agilent 1200 series) and a flow rate of 1 mL/min. The excitation wavelength (λEx) of 340 nm and the emission wavelength (λEm) of 440 nm were used in the fluorescence detection setup. The mobile phase of a binary gradient included solvent A: ACN, methanol, water (45:45:10), and solvent B: Na2HPO4 2.75 g/L (pH 6.5), which were created according to Mechmeche et al. (2016). The amino acids were identified and quantified using a standard amino acid mixture (Sigma Chemical). The amino acid content was expressed as milligrams of amino acids per liter of KDPE. Phytochemical Composition Extract The contents of active molecules were determined according to the protocol of Bigrali et al. (2008) with some modifications by stirring 100 mL of (KDPE and FDP) with 300 mL of methanol using a magnetic stirrer, for 24 h. The mixture was filtered using filter paper. The filtrate was centrifuged at 4000 rpm for 10 min. The supernatant was evaporated by rotary vacuum evaporator at 40 °C. The concentrated extracts are stored at 4 °C in dark glass bottles until use. Total Phenolic Content Total phenolic content (TP) was estimated by the colorimetric method using the Folin–Ciocalteu reagent (Waterhouse 2002). 100 µL of each extract is added to 500 µL of Folin reagent (diluted ten times with distilled water) and 1 mL of distilled water. After incubation for one minute, 1.5 mL of Na2 CO3 (20%) is added and mixed. Absorbance was measured after 2 h of incubation in the dark at 760 nm. The results are expressed in milligram gallic acid equivalents (GAE) per milliliter. Total Flavonoid Content Total flavonoid content (TF) was determined by the colorimetric method using the reagent of aluminum chloride and quercetin as standard according to the protocol described by Biglari et al. (2008). 1.5 mL of each extract was mixed with an equal volume of a 2% (AlCl3, 6H2O) solution. After incubation for 10 min at room temperature, the absorbance was measured at 367 nm. The results are expressed as milligram quercetin equivalents (QE) per milliliter. Condensed Tannins Content The method of Laouini et al. (2018) is employed to determine the condensed tannins content (CT), using catechin as a standard. A volume of extract 0.5 mL was mixed with 3 mL of the mixture of vanillin and methanol (4%), and 1.5 mL of hydrochloric acid is added and mixed thoroughly. The resulting mixture was allowed to stand for 15 min at 20 °C. The absorbance of each was measured at 500 nm. The results are expressed in milligram catechin equivalents (CE) per milliliter. Antioxidant Activity Antioxidant activity was determined by measuring the capability to scavenge different radicals by DPPH and ABTS method for KDPE and FDP, reducing power (FRAP), and phosphomolybdate method (TAA) for KDPE. DPPH Radical Scavenging Activity The potential to scavenge 2,2-diphenyl-2-picrylhydrazyl free radicals (DPPH) was determined by the Jan et al. (2013) method. 0.5 mL of each extract at different concentrations is added to 0.5 mL of the freshly prepared DPPH· solution (2 mg/50 mL of methanol). The mixture is stored in the dark for 15 min. The absorbance was measured at 517 nm against a blank which is methanol. The results are expressed as a percentage of inhibition; radical scavenging activity (%) was determined using following Eq. 4:4 DPPH scavenging activity%Acontrol-AsampleAcontrol×100 where Acontrol is the absorbance of DPPH· in methanol instead of samples. The antioxidant capacity of the various extracts was calculated graphically from the percentage of inhibition relative to the concentration of the extracts IC50, which corresponds to the concentration required to reduce 50% of the DPPH· radical. ABTS Radical Scavenging Activity The capacity to scavenge free radicals (2,2'-Zinobis- (3 ethylbenzothiazoline)-6-sulfonic acid ammonium salt) radical ABTS was determined by the method of Lien et al. (1999). The preparation of ABTS·+ radical is carried out according to the method described by Jan et al. (2013). An aqueous solution of ABTS at a concentration of 7 mM prepared in ultrapure water is mixed with a solution of potassium persulfate at a concentration of 2.45 mM. The mixture is stored in the dark and at room temperature for 12 h to 16 h. ABTS·+ solution was diluted to have an absorbance of 0.750 ± 0.025 at 734 nm in ethanol. Once the concentrations were diluted, 1 mL aliquots of the extracts were mixed with 1 mL of the ABTS·+ solution after incubation for 6 min, the absorbance was measured at 734 nm against a blank which is ethanol. The results are expressed as a percentage of inhibition; radical scavenging activity (%) was calculated using following Eq. 5:5 ABTS;scavenging;activity%=Acontrol-AsampleAcontrol×100 where Acontrol is the absorbance of ABTS·+ in ethanol instead of samples. The IC50, which corresponds to the concentration required to eliminate 50% of the ABTS·+ radical, was determined graphically from the percentage of inhibition relative to the concentration of the extracts. Antioxidant Activity by the FRAP Method (Ferric Reducing Antioxidant Power) The reducing power was performed by the FRAP method described by Oyaizu (1986). The objective of this method is to measure the potential to reduce the ferric iron (Fe3+) present in the ferrous iron complex (Fe2+). Using iron sulfate (FeSO4) as a standard, 1 mL of the extract was mixed with 2.5 mL of a 1% potassium ferrocyanide [(K3Fe (CN6)] solution and 2.5 mL of phosphate buffer (0.2 M; pH 6.6). Then, the mixture is incubated at 50 °C for 20 min. To stop the reaction, 2.5 mL of 10% trichloroacetic acid is added. After centrifugation at 6504 rpm/10 min, 2.5 mL of supernatant is added to 2.5 mL of distilled water and 0.5 mL of 0.1% iron chloride (FeCl3). The absorbance of the reaction medium is measured at 700 nm. The results are expressed as micromoles of ferrous iron (Fe II) equivalents per milliliter. Antioxidant Activity by the Phosphomolybdate Method Total antioxidant activity (TAA) of sample in the phosphomolybdenum method was based on the reduction in molybdenum Mo (VI) to molybdenum Mo (V) in the presence of antioxidant compounds and therefore the formation of a green phosphate/complex Mo (V) at acidic pH (Laloo and Sahu 2011), using ascorbic acid as a standard. 0.3 mL of extract was mixed with 3 mL of reagent solution (0.6 M sulfuric acid, 28 mM sodium phosphate, and 4 mM ammonium molybdate). This mixture is incubated in a water bath at 95 °C for 90 min then cooled at room temperature for 6 min. Absorbance was measured at 695 nm. Results are expressed as micromole of ascorbic acid equivalents per milliliter. Results and Discussion Kentichi Date Powder Extract Response surface methodology (RSM) is a combination of mathematical and statistical methods for modeling and evaluating problems. The goal of this technique is to optimize the response surface, which is affected by various process parameters (Aslan and Cebeci 2007). Predicted Model and Statistical Analysis Extraction optimization for KDP extract is based on various combinations of variables, as described in the experimental design. The effect of three independent variables: powder/solvent ratio (X1), extraction temperature (X2), and extraction time (X3), as well as their interactions, on yield sugar extraction in KDP (Y), was studied. Analysis of variance (ANOVA) was performed to study the relevance of the proposed models and to identify the most important factors. The p value and F test are used to confirm the significance of each coefficient and to determine the interaction of each parameter for the sample. In the present study, the p value of the model is 0.0083 for sugars extraction from KDP; however, the coefficient of determination and the Adjusted coefficient of determination are (R2 = 0.824) and (R2-Adj = 0.665) for KDP, respectively. They approved that the model was statistically significant (p < 0.05). The result implied that it was suitable for this experiment (Table 2). Another researcher has reported an R2 coefficient of determination ranging from 94.26 to 98.66% (Appiah-Nkansah et al. 2016) for dried bagasse. The F values of the model indicate that they are not significant concerning the pure error as shown in Table 2. These results identify the accuracy and general availability of the polynomial model. The quadratic polynomial regression model proposed for sugars extraction was calculated using following Eq. 6:Table 2 Analysis of variance (ANOVA) for the quadratic polynomial model of extraction of sugar from Kentichi date powder Source Degree of freedom Sum of square Mean square F value p value Prob > F Model 9 4985.92 553.99 5.21 0.0083 A-(Powder/solvent) Ratio 1 211.09 211.09 1.98 0.1892 B-Extraction temperature 1 8.50 8.51 0.08 0.7831 C-Extraction time 1 6.05 6.05 0.06 0.8163 AB 1 93.71 93.71 0.88 0.3700 AC 1 3.84 3.84 0.04 0.8532 BC 1 0.82 0.82 0.01 0.9318 A2 1 538.62 538.62 5.06 0.0481 B2 1 2992.31 2992.31 28.14 0.0003 C2 1 1888.31 1888.31 17.76 0.0018 Residual 10 1063.41 106.34 Lack of Fit 5 1063.41 212.68 Pure error 5 0 0 Cor total 19 6049.33 Standard deviation 10.31 R2 0.82 Mean 58.62 R2-adj 0.66 Coefficient of variation 17.58 Predicted R2 − 0.35 Press 8205.00 Adequate precision 5.77 6 Y=80.45-3.93X1-0.78X2-0.66X3+3.42X12-0.69X13+0.32X23-6.11X12-14.40X22-11.44X32 Indeed, the quadratic factors presented in the model of sugars extraction from Kentichi date powder (powder/solvent ratio, extraction temperature, and extraction time) (Table 2) are significantly different (p ≤ 0.05). Study of Single-Factor Experimental Analysis The effects of powder/solvent ratio on sugar yield of KDP are represented in Fig. 1a. The other extraction variables were set as follows: extraction time of 2 h, extraction temperature at 30 °C. According to the findings, sugar yield increased from 100 g/L to 300 g/L, where it maximized at 68.88% and then, decreased from 300 to 500 g/L. The diffusion of the solvent into cells was frequently enhanced by the liquid/material ratio, which also facilitated polysaccharide desorption from the cells (Chen et al. 2017).Fig. 1 Effects of different extraction parameters on sugar yield (a: extraction powder/solvent; b: extraction time; c: extraction temperature) The effects of extraction time on the extraction yield of sugar are illustrated in Fig. 1b. Extraction time was from 0 to 8 h, while other extraction variables were set as follows: extraction temperature at 30 °C, and powder/solvent ratio of 300 g/L. Results demonstrated that maximum sugar yield (71.1%) was reached during a 2 h extraction interval, beyond 2 h, the sugar yield decreased, suggesting that the prolonged extraction process caused polysaccharide degradation (Prakash Maran et al. 2013). The use of aqueous extraction with the RSM method to extract sugar from date palm by-products has not been used in any other research. Iwassa et al. (2019) showed that the yields of soluble sugars increased gradually in the subcritical water extraction of asparagus by-products from 10 to 90 min and that these yields were stable for 120 min. In Fig. 1c, we observed that the sugar production varied significantly from 20 °C to 30 °C. An increase in temperature favored the extraction of sugar. At this point, the following conditions were set: powder/solvent ratio, 300 g/L; and extraction time, 2 h. The extreme increase in temperature led to a lower extraction recovery (Setyaningsih et al. 2022). Study of Independent Variables Interactions The three-dimensional (3D) response surface and the two-dimensional (2D) contour projection of the regression equation are graphical representations that aid in the determination of the relationship between two variables and the identification of optimal experimental conditions when the third variable is set to zero levels (Li et al. 2020). Elliptical contour plots show a significant interaction between variables, whereas circular contour plots indicate non-significant interactions between variables (Sahu et al. 2020). The effect of the powder/solvent ratio and extraction temperature on the sugar yield of KDP extract was determined, setting the extraction time to zero (Fig. 2). The sugar yield increased with increasing powder/solvent ratio, up to 300 g/L, and increasing extraction temperature, up to 30 °C. Beyond a powder/solvent ratio of 300 g/L or extraction temperature of 30 °C, there was a decrease in the sugar yield (Fig. 2a). In general, the increase in powder/solvent ratio facilitated sugar desorption from the cells. However, the reduction in sugar yield was caused by partial decomposition of sugar at high temperatures. The contour plots also had a circular shape, indicating that the interaction between powder/solvent ratio and extraction temperature is not significant (Fig. 2b). Additionally, setting the extraction temperature to zero, the effect of the powder/solvent ratio and extraction time on the sugar yield of KDP extract was studied (Fig. 2c). With the extraction time increased from 0 to 2 h or the powder/solvent ratio increased from 100 to 300 g/L, the yield increased at first and then, decreased when these two variables kept increasing thereafter. The contour plots showed circular shapes which justify that the effect of interaction between powder/solvent ratio and extraction time on sugar yield is not significant (Fig. 2d). In addition, Fig. 2e also shows the effect of temperature and extraction time on sugar yield when the powder/solvent ratio is set to zero. From these observations, we discovered that sugar output increased with temperature and time. Beyond an extraction time of 2 h or extraction temperature of 30 °C, there was no further increase in the sugar yield. The contour plots revealed a circular outline, indicating that there is no interaction between these two parameters and sugar yield (Fig. 2f).Fig. 2 Response surface and contour plots for sugar yield from Kentichi date powder showing the effects of powder/solvent ratio and extraction temperature (a, b), the effects of powder/solvent ratio, and extraction time (c, d), and the effects of extraction temperature and extraction time (e, f) Moreover, the negative values for the quadratic term of powder/solvent ratio, temperature, and time (Eq. 6) indicate that the extraction conduction at high levels of this variable, i.e., to values over 300 g/L, 30 °C, and 2 h, respectively, can decrease the sugar yield. Model Validity To validate the established model, the optimized conditions were tested experimentally under the following conditions: powder/solvent ratio is 300 g/L, extraction temperature is 32.7 °C, and extraction time is 2.1 h. The fitted equation predicted a yield value of 80.50%. To confirm the model prediction, optimal extraction conditions were applied. An average value of about 77.1% of sugar content was obtained by three independent real experiments. There is not a significant difference (p > 0.05) between the experimental and theoretical values obtained from the model, which confirmed that the response model was adequate for optimization. Extract Composition KDP presents a concentration of 67.91 g/100 g total sugar, with a 77.1% extraction yield. As a result, KDPE is a cost-effective way to produce sugar. Furthermore, dry dates are one of the most abundant and inexpensive international fruits (Ghnimi et al. 2017). The pH and titratable acidity values are 5.97–2.22% in KDPE, respectively (Table 3). The pH result is similar to this El-Nagga and Abd El-Tawab (2012) studied date syrup extraction by different methods. However, the acidity result is higher than our results. Likewise, the level of total sugars in KDPE is 160.09 g/L, of which 3.04 g/L are reducing sugars, and 157.05 g/L are sucrose (Table 3). The presence of reducing sugar in the waste explained the low content. These results are very far from those reported by El-Nagga and Abd El-Tawab (2012) studied a different variety of dates.Table 3 Physicochemical properties of Kentichi date powder extract Parameters KDPE pH 5.97 ± 0.03 Titratable acidity (%) 2.22 ± 0.01 Total sugars (g/L) 160.09 ± 0.29 Reducing sugars (g/L) 3.04 ± 0.03 Sucrose (g/L) 157.05 ± 0.32 Free sugars1 (%) Fructose 13.74 Glucose 14.26 Sucrose 72 Maltose – Lactose – Protein (g/L) 10.31 ± 0.07 Amino acids1 (mg/L KDPE) Aspartic acid 9.65 Glutamic acid 28.59 Serine + Histidine + Glutamine 4.68 Glycine + Threonine + Arginine 6.27 Alanine 6.29 Tyrosine 5.03 Phenylalanine 1.81 Isoleucine 2.83 Leucine 0.96 Valine + Methionine 4.24 Total 70.35 TP (mg GAE2/mL) 3.13 ± 3.47 TF (mg QE2/mL) 0.25 ± 1.10 CT (mg CE2/mL) 0.16 ± 0.40 1Contents of free sugars and amino acids were calculated by HPLC analyses 2Reference compound: GAE gallic acid equivalents, QE quercetin equivalents, CE catechin equivalents Free sugars composition was evaluated by HPLC. One of the important components of KDPE is that sucrose is present in higher quantities than glucose and fructose. Approximately, about 72% is sucrose, 14.26% glucose, and 13.74% fructose; on the other hand, maltose and lactose are no longer present in KDPE (Table 3). These results are comparable to those obtained by Djaoud et al. (2020) analyzed the free sugars in syrup from a secondary date variety. Generally, soft date cultivars' fruits are dominated by inverted sugars (glucose and fructose) and constitute little or no sucrose, whereas dry date cultivars' fruits may contain a high proportion of sucrose (Ghnimi et al. 2017). Sugars, in addition to their primary role of sweetness, also play other functions in the food industry, such as preservation, fermentation, color, flavor, texture, solubility, hygroscopicity, crystallinity, and viscosity (Zaitoun et al. 2018). In addition, the protein content is 10.31 g/L in KDPE. This result is much lower than a date fiber isolate given by Ben Yahmed et al. (2020). The amino acid contents were determined using HPLC and were also employed in the current study to evaluate the nutritional quality of protein in KDPE. Furthermore, the free amino acid content attends 70.35 mg/L essentially glutamic acid (28.39 mg/L) and aspartic acid (9.65 mg/L), but leucine (0.95 mg/L) has a lower content (Table 3). As a result, Kumar et al. (2021) reported that the optimal assimilation of proteins from various plant origins can provide enough necessary amino acids to meet human health requirements. The KDPE has low free amino acids content when compared to other protein isolates such as Moringa oleifera seed (Aderinola et al. 2018). Proteins seem to be well for their role in the physical structure of processed foods, aiding in the formation of a different number of gels, emulsions, and foams (Allen Foegeding 2015). The biochemical components of dates are affected by culture conditions including growth zone and produce period (fully mature stage), and it differs considerably between cultivars (Ben Yahmed et al. 2020), as well as different extraction methods (El-Nagga and Abd El–Tawab 2012). KDPE contained 3.13 mg GAE/mL of total phenolic content (Table 3), indicating that it was a valuable source of phenolic antioxidants. These results are higher than those given by El-Nagga and Abd El-Tawab (2012). Then, the results showed that KDPE has total flavonoid contents that are 0.25 mg QE/mL. However, the condensed tannins content obtained is 0.16 mg CE/mL in KPDE (Table 3). KDPE contained a wide range of bioactive components capable of inhibiting the effect of reactive oxygen species implicated in human diseases such as cardiovascular disease and cancer. As a result, essential biological macromolecules may be protected from oxidation (Benmeddour et al. 2012). Date varieties, geographic origin, fruit storage time, extraction conditions such as solvent used, plant material/solvent ratio, and extraction time are all factors that influence phenolic compounds (Masmoudi-Allouche et al. 2016). Antioxidant Activity The antioxidant activity of KDPE was determined using (DPPH, ABTS, FRAP, and TAA methods) with two aqueous and methanolic extracts, as shown in Fig. 3. The results indicate a significant difference between the two aqueous and methanolic extracts (p ≤ 0.05).Fig. 3 Antioxidant activity by DPPH, ABTS (a, b), and TAA, and FRAP (c, d) methods with two aqueous () and methanolic () extracts for Kentichi date powder extract (KDPE) The KDPE aqueous extract was endowed with anti-free radical activity, an IC50 concentration (DPPH) by about 4.8 mg/mL (Fig. 3a). On the other hand, the IC50 value for the KDPE methanolic extract was found to be 14 mg/mL (Fig. 3b). Then, the ABTS test of KDPE aqueous extracts gives results whose IC50 concentration is 3 mg/mL (Fig. 3a). However, the IC50 value of methanolic extracts is about 7 mg/mL (Fig. 3b). Antioxidant activity assayed by FRAP showed that the KDPE aqueous extract is 4.7 µmol Fe(II)E/mL (Fig. 3c). On the other hand, the methanolic extract gives a low level of antioxidant activity (3.37 µmol Fe(II)E/mL) (Fig. 3d). However, the TAA of KDPE is 18.04 and 15.86 µmol AAE/mL for the aqueous and methanolic extracts, respectively. It should be noted that aqueous extracts have a higher antioxidant capacity than methanolic extracts (Fig. 3), showing that antioxidants in the KDPE are mostly polar. As a result, extract can be thought a low-cost source of active molecules with natural antioxidants, implying that the extract's improved antioxidant activity was mostly attributable to the active components of KDP that are readily soluble in water. Additionally, the migration of the active component to the food simulant is affected by the polarity of the migrant and simulating media. Also, the sum of glucose, fructose, sucrose, and fructans is termed “water soluble carbohydrate” (Al-Sheikh Ahmed et al. 2020). Phillips et al. (2009) confirmed this result by comparing the antioxidant activity content among natural sweeteners (fruit sugars, e.g., date sugar) versus refined sugar and found that date sugar has the highest antioxidant activity of the natural sweeteners studied. Faraji and Lindsay (2004) demonstrated that antioxidant activity for fructose, sucrose, raffinose, sorbitol, or mannitol was confirmed when integrated at 16% of the aqueous phase in model emulsions of fish oil in water. The date-fruit syrup waste extract (DSWE) contained larger hydrophilic phenolic molecules that could rapidly migrate to the aqueous phase, resulting in a high TPC release profile in the water medium (Saleh 2011). As a result, Rangaraj et al. (2021) showed that the antioxidant activity of films containing DSWE was higher in the water medium than in the 95% ethanol medium. On the other hand, Hu et al. (2016) proved that the antioxidant activities observed for complex carbohydrates correlate with the presence of phenolic and/or protein components. KDPE contains a concentration of aspartic acid and glutamic acid, both of which have high antioxidant properties due to the presence of additional electrons that can be produced when free radicals interact with them (He et al. 2013). Effect of Freeze-Drying Process on Physicochemical Properties of Kentichi Date Powder Extract Fruits and their extracts have a limited shelf life because of their high water content (Silva-Espinoza et al. 2020). The pulp drying for the production of fruit powders allows can be preserved for a long time, enabling it to be employed in the production of instant beverages and other industrial applications (Cordeiro 2020). Freeze-drying is the process of producing high-quality products by sublimating a frozen sample at reduced pressure while avoiding high temperatures. The preservation of taste, flavor, and thermo-sensitive compounds with biological activity are among the advantages of freeze-drying (Uscanga et al. 2021). Table 4 shows the total sugar content, protein content, and antioxidant activity of the lyophilized powder of KDPE.Table 4 Physicochemical properties of freeze-dried powder of Kentichi date powder extract Parameters FDP Total sugars (mg/g DM1) 742.8 ± 0.31 Protein (mg/g DM1) 88.8 ± 0.08 DPPH IC50 (mg/g) 3.2 ± 0.13 ABTS IC50 (mg/g) 2.8 ± 0.18 1Dry matter The FDP results show that the total sugars and protein content are 742.8–88.8 mg/g dry weight, respectively; in other words, the increase after drying is 41.84 and 158.6%, respectively. ABTS and DPPH tests showed that FDP has an increase of 6.67–33.34%, with IC50 values for ABTS and DPPH of 2.8–3.2 mg/g, respectively. A similar drying trend was also shown by Assefa and Keum (2016), and Shonte et al. (2020), who found an increase in antioxidant activity and protein content of freeze-dried yuzu and stinging nettle powders. Due to water removal, the relative increases in mineral content, total acidity, carbohydrates, and total sugars in the dried fruits were significantly higher than in the fresh samples (Radojcin et al. 2021). On the other hand, proteins in most foods retain their nutritional value and digestibility when dried (Guiné 2018). Numerous researchers have indicated the benefits of freeze-drying over other drying methods in terms of chemical and nutritional properties (Radojcin et al. 2021). Conclusion In this study, aqueous extraction was used to extract sugar from Kentichi date powder. In order to determine the optimal treatment conditions, the Box–Behnken (BBK) design for response surface methodology (RSM) was utilized. It was demonstrated that optimal values of powder/solvent ratio, extraction temperature, and extraction time were 300 g/L, 32.7 °C, and 2.1 h, respectively. The sugar yield reached 77.1% under the above-optimized conditions. The Kentichi date powder extract (KDPE) findings revealed a total sugar content of 160.09 g/L. Aspartic acid (9.65 mg/L) and glutamic acid (28.39 mg/L) make up the majority of the 10.31 g/L of protein. KDPE showed exhibited excellent antioxidant activity in a dose-dependent manner in various in vitro models such as DPPH radical scavenging activity, ABTS + radical scavenging activity, FRAP method, and Total antioxidant activity TAA. The freeze-drying of extract appears to be a viable option for preserving nutritional quality for a long time. Overall, in terms of extraction from fruit, aqueous extraction had the benefits of efficiency, low cost, and being environmentally friendly. However, the product of KDPE is a matrix that could be used as a basic ingredient in the formulation of functional foods. Therefore, this research makes an important addition to the literature, considering the lack of reports on the extraction of the date palm by-product using a simple technique. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 15 kb) Authors Contribution NM wrote the main manuscript text of the thesis; MM contributed to statistic analysis; KS contributed to protein analysis; ZT contributed to antioxidant activity analysis; MH reviewed the manuscript; FK supervised the work. All authors reviewed the manuscript. Funding The authors have no relevant financial or non-financial interests to disclose. Declarations Conflict of interest The authors have no conflict of interest. Ethical Approval 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 Aderinola TA Fagbemi TN Enujiugha VN Alashi AM Aluko RE Amino acid composition and antioxidant properties of Moringa oleifera seed protein isolate and enzymatic hydrolysates Journal of Heliyon 2018 4 10 e00877 10.1016/j.heliyon.2018.e00877 30386828 Al-Alawi RA Al-Mashiqri JH Al-Nadabi JSM Al-Shihi IB Baqi Y Date palm tree (Phoenix dactylifera L.): Natural products and therapeutic options Journal of Frontiers in Plant Science. 2017 8 845 10.3389/fpls.2017.00845 28588600 Alam MA Alhebsi MSR Ghnimi S Kamal-Eldin A Inability of total antioxidant activity assays to accurately assess the phenolic compounds of date palm fruit (Phoenix dactylifera L.) Journal of Nutrition & Food Sciences 2021 22 32 40 Alghamdi BA Alshumrani ES Bin Saeed MS Rawas GM Alharthi NT Baeshen MN Mohammed Helmi N Alam MZ Suhail M Analysis of sugar composition and pesticides using HPLC and GC–MS techniques in honey samples collected from Saudi Arabian markets Saudi Journal of Biological Sciences. 2020 27 12 3720 3726 10.1016/j.sjbs.2020.08.018 33304183 Aljaloud S Colleran HL Ibrahim SA Nutritional value of date fruits and potential use in nutritional bars for athletes Journal of Food and Nutrition Sciences 2020 11 463 480 10.4236/fns.2020.116034 Allen Foegeding E Food protein functionality—A new model Journal of Food Science 2015 80 12 C2670 C2677 10.1111/1750-3841.13116 26513109 Al-Sheikh Ahmed S Zhang J Farhan H Zhang Y Yu Z Islam S Chen J Cricelli S Foreman A Van den Ende W Ma W Dell B Diurnal changes in water soluble carbohydrate components in leaves and sucrose associated TaSUT1 gene expression during grain development in wheat International Journal of Molecular Sciences 2020 21 21 8276 10.3390/ijms21218276 33167324 AOAC. 1990. Official methods of analysis of AOAC international. Washington, MD: AOAC International. Appiah-Nkansah NB Zhang K Rooneyc W Wang D Model study on extraction of fermentable sugars and nonstructural carbohydrate from sweet sorghum using diffusion process Journal of Industrial Crops and Products 2016 83 654 662 10.1016/j.indcrop.2015.12.056 Arrutia F Adam M Calvo-Carrascal MA Mao Y Binner E Development of a continuous-flow system for microwave-assisted extraction of pectin-derived oligosaccharides from food waste Journal of Chemical Engineering 2020 395 125056 10.1016/j.cej.2020.125056 Aslan N Cebeci Y Application of Box-Behnken design and response surface methodology for modeling of some Turkish coals Journal of Fuel 2007 86 1–2 90 97 10.1016/j.fuel.2006.06.010 Assefa AD Keum YS Effect of extraction solvent and various drying methods on polyphenol content and antioxidant activities of yuzu (Citrus junos Sieb ex Tanaka) Food Measure 2016 11 2 576 585 10.1007/s11694-016-9425-x Ben Yahmed N Dauptain K Lajnef I Carrere H Trably E Smaali I New sustainable bioconversion concept of date byproducts (Phoenix dactylifera L.) to biohydrogen, biogas and date-syrup International Journal of Hydrogen Energy 2020 46 1 297 305 10.1016/j.ijhydene.2020.09.203 Benmeddour Z Mehinagi E Meurlay DL Louaileche H Phenolic composition and antioxidant capacities of ten Algerian date (Phoenix dactylifera L.) cultivars: A comparative study Journal of Functional Foods 2012 5 1 346 354 10.1016/j.jff.2012.11.005 Biglari F Abbas FM Alkarkhi FMA Easa A Antioxidant activity and phenolic content of various date palm (Phoenix dactylifera) fruits from Iran Journal of Food Chemistry 2008 107 4 1636 1641 10.1016/j.foodchem.2007.10.033 Chafi A Benabbes R Bouakka M Hakkou A Kouddane N Berrichi A Pomological study of dates of some date palm varieties cultivated in Figuig oasis Journal of Materials and Environmental Science 2015 6 5 1266 1275 Chen C Zhang B Huang Q Fu X Liu R Microwave-assisted extraction of polysaccharides from Moringa oleifera Lam. leaves: Characterization and hypoglycemic activity Industrial Crops and Products 2017 100 1 11 10.1016/j.indcrop.2017.01.042 Chen G Fang C Ran C Tan Y Yu Q Kan J Comparison of different extraction methods for polysaccharides from bamboo shoots (Chimonobambusa quadrangularis) processing by-products International Journal of Biological Macromolecules 2019 130 903 914 10.1016/j.ijbiomac.2019.03.038 30849468 Cordeiro, C.A.M. 2020. Tecnologia de Alimentos. Brasil (2st edn., pp. 206). Djaoud K Arkoub-Djermoune L Remini H Sait S Tazarourte M Hadjal S Romero A Madani K Boulekbache-Makhlouf L Syrup from common date variety (Phoenix dactylifera l): Optimization of sugars extraction and their quantification by high performance liquid chromatography Journal of Current Nutrition & Food Science 2020 16 4 530 542 10.2174/1573401315666190115160950 Dubois M Gilles KA Hamilton JK Rebers PA Smith F Colorimetric method for determination of sugars and related substances Journal of Analytical Chemistry 1956 28 3 350 356 10.1021/ac60111a017 El-Nagga EA Y.A. Abd El–Tawab. Compositional characteristics of date syrup extracted by different methods in some fermented dairy products Journal of Annals of Agricultural Science 2012 57 1 29 36 10.1016/j.aoas.2012.03.007 Faraji H Lindsay RC Characterization of the antioxidant activity of sugars and polyhydric alcohols in fish oil emulsions Journal of Agricultural and Food Chemistry 2004 52 23 7164 7171 10.1021/jf035291k 15537333 Farouk WM Abdullah AS Mohammed SA Alawee WH Omara ZM Essa FA Modeling and optimization of working conditions of pyramid solar still with different nanoparticles using response surface methodology Case Studies in Thermal Engineering 2022 33 101984 10.1016/j.csite.2022.101984 Filly A Fabiano-Tixier AS Louis C Fernandez X Chemat F Water as a green solvent combined with different techniques for extraction of essential oil from lavender flowers Comptes Rendus Chimie 2016 19 6 707 717 10.1016/j.crci.2016.01.018 Galanakis CM Aldawoud TMS Rizou M Rowan NJ Ibrahim SA Food ingredients and active compounds against the coronavirus disease (COVID-19) pandemic: A comprehensive review Journal of Foods 2020 9 11 1701 10.3390/foods9111701 33233560 Ghnimi S Umer S Azharul K Kamal-Eldin A Date fruit (Phoenix dactylifera L.): An underutilized food seeking industrial valorization Journal of Nutrition & Food Sciences 2017 6 1 10 Guiné R The drying of foods and its effect on the physical-chemical, sensorial and nutritional properties International Journal of Food Engineering 2018 4 93 100 He R Girgih AT Malomo SA Ju X Aluko RE Antioxidant activities of enzymatic rapeseed protein hydrolysates and the membrane ultrafiltration fractions Journal of Functional Foods 2013 5 1 219 227 10.1016/j.jff.2012.10.008 Hu S Yin J Nie S Wang J Phillips GO Xie M Cui SW In vitro evaluation of the antioxidant activities of carbohydrates Journal of Bioactive Carbohydrates and Dietary Fibre 2016 7 2 19 27 10.1016/j.bcdf.2016.04.001 Ibrahim SA Ayad AA Williams LL Ayivi DR Gyawali R Krastanov A Aljaloud SO Date fruit: A review of the chemical and nutritional compounds, functional effects and food application in nutrition bars for athletes Journal of Food Science and Technology 2020 56 4 1503 1513 10.1111/ijfs.14783 Iwassa IJ Ribeiro MAS Meurer ED Cardozo-Filho L Bolanho BC Silva C Effect of subcritical water processing on the extraction of compounds composition and functional properties of asparagus by-product Journal of Food Process Engineering 2019 42 4 e13060 10.1111/jfpe.13060 Jan S Khan MR Rashid U Bokhari J Assessment of antioxidant potential, total phenolics and flavonoids of different solvent fractions of monotheca buxifolia fruit Journal of Public Health and Research Perspectives 2013 4 5 246 254 10.1016/j.phrp.2013.09.003 24298440 Kamal-Eldin A George N Sobti B AlRashidi N Ghnimi S Ali AA Andersson AAM Andersson R Antony A Hamed F Dietary fiber components, microstructure, and texture of date fruits (Phoenix dactylifera, L.) Scientific Reports 2020 10 1 21767 10.1038/s41598-020-78713-4 33303871 Kumar M Tomar M Potkule J Reetu S Punia J Dhakane-Lad S Singh S Dhumal PC Pradhan B Bhushan T Anitha O Alajil A Alhariri A. Amarowicz Kennedy JF Functional characterization of plant-based protein to determine its quality for food applications Journal of Food Hydrocolloids 2021 123 106986 10.1016/j.foodhyd.2021.106986 Laloo D Sahu AN Antioxidant activities of three Indian commercially available Nagakesar: An in vitro study Journal of Chemical and Pharmaceutical Research 2011 3 1 277 283 Laouini SE Kelef A Ouahrani MR Free radicals scavenging activity and phytochemical composition of astermisia (Herba-Alba) extract growth in Algeria Journal of Fundamental and Applied Sciences 2018 10 1 268 10.4314/jfas.v10i1.20 Leili M Khorram NS Godini K Azarian G Rahim Moussavi R Peykhoshian A Application of central composite design (CCD) for optimization of cephalexin antibiotic removal using electro -oxidation process Journal of Molecular Liquids 2020 313 1 113556 10.1016/j.molliq.2020.113556 Li S Qin X Zhang G Xu Y Lv Y Optimization of content of components over activated carbon catalyst on CO2 reforming of methane using multi-response surface methodology International Journal of Hydrogen Energy 2020 45 16 9695 9709 10.1016/j.ijhydene.2020.01.226 Lien EJ Ren S Bui HH Wang R Quantitative structure-activity relationship analysis of phenolic antioxidants Journal of Free Radical Biology and Medicine 1999 26 3–4 285 294 10.1016/S0891-5849(98)00190-7 9895218 Lowry OH Protein measurement with the folin phenol reagent Journal of Biological Chemistry 1951 193 1 265 275 10.1016/S0021-9258(19)52451-6 14907713 Masmoudi-Allouche F Touati S Mnafgui K Gharsallah N El Feki A Allouche N Phytochemical profile, antioxidant, antibacterial, antidiabetic and anti-obesity activities of fruits and pits from date palm (Phoenix dactylifera L.) grown in south of Tunisia Journal of Pharmacognosy and Phytochemistry. 2016 5 3 15 22 Mechmeche M Kachouri F Chouabi M Ksontini H Setti K Hamdi M Optimization of extraction parameters of protein isolate from tomato seed using response surface methodology Journal of Food Analytical Methods 2016 10 3 809 819 10.1007/s12161-016-0644-x Miller GL Use of dinitrosaiicyiic acid reagent for determination of reducing sugar Journal of Analytical Chemistry 1959 31 3 426 428 10.1021/ac60147a030 Moreb NA Albandary A Jaiswal S Jaiswa AK Fruits and vegetables in the management of underlying conditions for COVID-19 high-risk groups Journal of Foods 2021 10 2 389 10.3390/foods10020389 33578926 Nadar SS Rao P Rathod VK Enzyme assisted extraction of biomolecules as an approach to novel extraction technology: A review Food Research International 2018 108 309 330 10.1016/j.foodres.2018.03.006 29735063 Najjar Z Stathopoulos C Chockchaisawasdee S Utilization of date by-products in the food industry Emirates Journal of Food and Agriculture 2020 32 11 808 815 10.9755/ejfa.2020.v32.i11.2192 Oyaizu M Antioxidative activities of products of browning reaction prepared from glucosamine Japanese Journal of Nutrition and Dietetics 1986 44 6 307 315 Peng Y Khaled U Al-Rashed AAAA Meer R Goodarzi M Sarafraz MM Potential application of response surface methodology (RSM) for the prediction and optimization of thermal conductivity of aqueous CuO (II) nanofluid: A statistical approach and experimental validation Physica 2020 554 124353 10.1016/j.physa.2020.124353 Phillips KM Carlsen MH Blomhoff R Total antioxidant content of alternatives to refined sugar American Dietetic Association 2009 109 1 64 71 10.1016/j.jada.2008.10.014 Prakash Maran J Sivakumar V Thirugnanasambandham K Sridhar R Optimization of microwave assisted extraction of pectin from orange peel Carbohydrate Polymers 2013 97 2 703 709 10.1016/j.carbpol.2013.05.052 23911504 Radojcin M Pavkov I Kovacevic DB Putnik P Wiktor A Stamenkovic Z Kešelj K Gere A Effect of selected drying methods and emerging drying intensification technologies on the quality of dried fruit: A review Journal of Processes 2021 9 1 132 10.3390/pr9010132 Rangaraj VM Rambabu K Banat F Mittal V Effect of date fruit waste extract as an antioxidant additive on the properties of active gelatin films Food Chemistry 2021 355 129631 10.1016/j.foodchem.2021.129631 33799252 Sahu R Meghavarnam AK Janakiraman S Response surface methodology: An effective optimization strategy for enhanced production of nitrile hydratase (NHase) by Rhodococcus rhodochrous (RS-6) Journal of Heliyon 2020 6 10 e05111 10.1016/j.heliyon.2020.e05111 33088939 Saleh EA Tawfik MS Abu-Tarboush HM Phenolic Contents and antioxidant activity of various date palm (Phoenix dactylifera L.) fruits from Saudi Arabia Food and Nutrition Sciences 2011 2 10 1134 1141 10.4236/fns.2011.210152 Schauder S Thomsen MR Nayga RM Jr Agent-based modeling insights into the optimal distribution of the fresh fruit and vegetable program Journal of Preventive Medicine Reports 2020 20 101173 10.1016/j.pmedr.2020.101173 32923315 Setyaningsih W Putro AW Fathimah RN Kurnia KA Darmawan N Yulianto B Jiwanti PK Carrera CA Palma M A microwave-based extraction method for the determination of sugar and polyols: Application to the characterization of regular and peaberry coffees Arabian Journal of Chemistry 2022 15 103660 10.1016/j.arabjc.2021.103660 Shonte TT Duodu KG Kock HL Effect of drying methods on chemical composition and antioxidant activity of underutilized stinging nettle leaves Journal of Heliyon 2020 6 5 e03938 10.1016/j.heliyon.2020.e03938 32478185 Siddiqi SA Rahman S Khan MM Rafiq S Inayat A Khurram MS Seerangurayar T Jamil F Potential of dates (Phoenix dactylifera L.) as natural antioxidant source and functional food for healthy diet Science of the Total Environment 2020 748 141234 10.1016/j.scitotenv.2020.141234 32798862 Silva-Espinoza MA Ayed C Foster T Camacho MM Martínez-Navarrete N The Impact of freeze-drying conditions on the physico-chemical properties and bioactive compounds of a freeze-dried orange puree Journal of Foods 2020 9 1 32 10.3390/foods9010032 Uscanga MA Salvador A Camacho MM Martínez-Navarrete N Impact of freeze-drying shelf temperature on the bioactive compounds, physical properties and sensory evaluation of a product based on orange juice International Journal of Food Science & Technology 2021 56 10 5409 5416 10.1111/ijfs.15086 Waterhouse AL Determination of total phenolics Current Protocols in Food Analytical Chemistry. 2002 6 1 I1 1 Yefsah-Idres A Benrima A Hammouchi K Bennazoug Y Essai de valorisation de la datte Mech-degla par sa substitution au sucre blanc dans la formulation d’un biscuit Revue Agrobiologia 2019 9 2 1543 1559 Zaitoun M Ghanem M Harphoush S Sugars: Types and their functional properties in food and human health International Journal of Public Health Research 2018 6 4 93 99 Zerizer S Houssem-Eddine K Kabouche Z Immunostimulatory activity of Phoenix dactylifera International Journal of Pharmacy and Pharmaceutical Sciences 2014 3 73 76 Zihad SMNK Uddin SJ Sifat N Lovely F Rouf R Shilpi JA Göransson U Antioxidant properties and phenolic profiling by UPLC-QTOF-MS of Ajwah, Safawy and Sukkari cultivars of date palm Biochemistry and Biophysics Reports 2021 25 100909 10.1016/j.bbrep.2021.100909 33521336	0	PMC9734394	NO-CC CODE	2022-12-14 23:28:28	no	Sugar Tech. 2022 Dec 8;:1-13	utf-8	Sugar Tech	2,022	10.1007/s12355-022-01223-2	oa_other
==== Front Eur Child Adolesc Psychiatry Eur Child Adolesc Psychiatry European Child & Adolescent Psychiatry 1018-8827 1435-165X Springer Berlin Heidelberg Berlin/Heidelberg 36484855 2121 10.1007/s00787-022-02121-4 Original Contribution Monthly correlates of longitudinal child mental health during the COVID-19 pandemic according to children and caregivers http://orcid.org/0000-0002-3265-0898 Rappaport Lance M. [email protected] 1 Mactavish Alexandra 1 Mastronardi Carli 1 Babb Kimberley A. 1 Menna Rosanne 1 Amstadter Ananda B. 2 Battaglia Marco 34 1 grid.267455.7 0000 0004 1936 9596 Department of Psychology, University of Windsor, 401 Sunset Ave, Windsor, ON N9B3P4 Canada 2 grid.224260.0 0000 0004 0458 8737 Virginia Institute for Psychiatric and Behavioral Genetics, Virginia Commonwealth University, Richmond, VA USA 3 grid.155956.b 0000 0000 8793 5925 Child, Youth and Emerging Adults Programme, Centre for Addiction and Mental Health, Toronto, ON Canada 4 grid.17063.33 0000 0001 2157 2938 Department of Psychiatry, University of Toronto, Toronto, ON Canada 9 12 2022 112 29 7 2022 5 12 2022 © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany 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. Multiple reviews identify the broad, pervasive initial impact of the global COVID-19 pandemic on the mental health of children, who may be particularly vulnerable to long-term psychiatric sequelae of the ongoing pandemic. However, limited longitudinal research examines persistence of, or change in, children’s distress or psychiatric symptomatology. From June 2020 through December 2021, we enrolled two cohorts of families of children aged 8–13 from Southwestern Ontario into a staggered baseline, longitudinal design that leveraged multi-informant report (N = 317 families). In each family, one child and one parent/guardian completed a baseline assessment, 6 monthly follow-up assessments, and one final follow-up assessment 9 months post-baseline. At each assessment, the child and parent/guardian completed the CoRonavIruS health Impact Survey and measures of child anxiety, depressive, irritability, and posttraumatic stress syndromes. Children’s mental health, indexed by the severity of multiple syndromes, fluctuated over the study period. Elevated local monthly COVID-19 prevalence, hospitalization, and death rates were associated with monthly elevations in children’s reported worry about contracting COVID-19 and stress related to stay-at-home orders. In turn, both elevated monthly worry about contracting COVID-19 and stress related to stay-at-home orders were associated with monthly elevations in child- and parent-/guardian-report of children’s emotional distress and psychiatric syndromes. This study illustrates the importance of, and informs the potential design of, longitudinal research to track the mental health of children, who may be particularly vulnerable to broad psychosocial sequelae of health crises such as the COVID-19 pandemic. Supplementary Information The online version contains supplementary material available at 10.1007/s00787-022-02121-4. Keywords COVID-19 pandemic Longitudinal studies Child Anxiety Depressive symptoms http://dx.doi.org/10.13039/501100000226 Ontario Ministry of Health and Long-Term Care 702 Rappaport Lance M. WE-Spark Health Institute ==== Body pmcIntroduction Beyond the physical health implications of the SARS-CoV-2 virus, the ongoing COVID-19 pandemic may pose a long-term risk for the mental health of children and youth worldwide. Multiple reviews summarize initial global research on children’s mental health during the onset of the pandemic (e.g., through Summer 2020). Briefly, they document that a substantial minority of children and youth in both community and clinical samples reported elevated severity of a variety of psychiatric syndromes [e.g., 1, 2]. However, there is an urgent need to understand not only the acute, initial impact of the pandemic, but also the psychosocial implications of its chronic course. Longitudinal research is needed to clarify the psychosocial sequelae that commonly follow global disasters, especially to safeguard the mental health of children, who are particularly vulnerable [3]. Longitudinal research has begun to document the sustained psychosocial impact of the ongoing COVID-19 pandemic, such as higher rates of depressive and anxiety disorders [e.g., 4]. However, results are mixed [e.g., 5], even across countries within the same study [6]. Variability in empirical findings may result from the timing of longitudinal assessments (e.g., during months marked by high COVID-19 severity). Consequently, in one longitudinal study of youth, mood concerns increased at the outset of the pandemic, decreased over summer 2020, and increased in the fall [7]. Similarly, American health registry data illustrates substantial variability over months of each year in the rate of pediatric emergency department visits for mental health conditions [8]. Additionally, most prior research assessed child mental health through January 2021 at the latest [9]. Ongoing longitudinal research with more frequent (e.g., monthly) assessment is needed to track change in, or sustained impact on, children’s mental health throughout the ongoing pandemic to guide clinical assessment and inform the development and provision of mental health resources. Prior research also inferred the impact of constructs of interest to public policy (e.g., attending school virtually) based on local policy in place when assessments occurred. Higher temporal resolution (e.g., monthly) assessment over a longer time span facilitates within-person examination of how variation in public policy and children’s experiences are associated with monthly variation in child distress. Moreover, it is important to assess the impact of the pandemic on each child rather than rely solely on environmental data from public databases [e.g., 10]. The present study This study sought to characterize the sustained psychosocial impact of the COVID-19 pandemic on children within three aims: (i) evaluate change in child distress and psychiatric symptomatology, specifically internalizing symptomatology, over time; (ii) evaluate the magnitude and type of sustained distress and internalizing symptomatology over time; (iii) identify time-varying correlates of acute child distress and internalizing symptomatology to clarify contexts in which children demonstrate acutely worsened distress. Practically, this study enrolled two cohorts of families of children aged eight to thirteen years using a staggered baseline design to maximize coverage of the ongoing pandemic by a planned series of baseline assessment and seven follow-up assessments over nine months. Specifically, this study examined whether monthly fluctuations in child distress and internalizing symptomatology were associated with monthly fluctuations in (i) the local monthly case prevalence per 100,000 persons; (ii) the local monthly hospitalization rate; (iii) the local monthly death rate; (iv) child attendance at school virtually or in person; (v) child worry that they or a loved one would contract COVID-19 or that contracting COVID-19 would adversely affect their mental or physical health; and (vi) stress related to stay-at-home orders or the cancelation of events to curtail the spread of the SARS-CoV-2 virus. Methods Participants Families of a child aged eight to thirteen years were recruited between June and July 2020 or November 2020 and January 2021 from Southwestern Ontario via local school boards, news, and social media. From each family, one child and one parent/guardian participated for the duration of the study, which was completed entirely online. Study inclusion criteria required that both children and adults be sufficiently proficient in English to provide informed consent and complete study questionnaires and have regular internet access. There were no explicit exclusion criteria. Based on preliminary power analyses, this study was designed to recruit 330 families to provide 80% statistical power to detect change in emotional distress over time. For a separate aim outside the scope of this paper, we recruited additional families to augment power to examine baseline psychosocial predictors of sustained emotional distress. Design and procedure Following consent and assent, parents/guardians then completed a baseline assessment of children’s demographic information; the broad impact of the COVID-19 pandemic on the child and family; and the child’s emotional distress including symptoms of common internalizing (i.e., irritability, depressive, posttraumatic stress, and anxiety) syndromes. Children then completed the same measures to self-report on their emotional distress including symptoms of internalizing syndromes. Following baseline assessment, families completed 6 monthly assessments wherein the same parent/guardian and child reported on the broad impact of the COVID-19 pandemic on the child over the past two weeks and on the child’s distress on the same measures from baseline assessment. Finally, each child and parent/guardian completed one final longitudinal assessment of the same measures nine months after baseline assessment. Each family was compensated CAD $12 for time spent completing the baseline assessment; CAD $9 for each follow-up assessment. This study was cleared by the University of Windsor Research Ethics Board (#20–123). Following recruitment, parents/guardians provided informed consent; children provided assent. Measures COVID-19 impact At each assessment, each child and adult completed the CoRonavIruS health Impact Survey [CRISIS; 11] to assess diverse impacts of the COVID-19 pandemic on children’s daily lives (e.g., emotional distress). Baseline assessment queried the three months prior to the pandemic and the past two weeks. Each follow-up assessment queried only the past two weeks. Only data regarding the past two weeks were used from the baseline assessment. Baseline parent/guardian assessment included demographic items designed to harmonize data collected across international research [12]. In October 2020, we added one question to assess whether each child attended school in person or virtually in response to provincial policy that allowed each family to choose the mode of school attendance. Due to the staggered baseline, longitudinal design, enough participants provided data between July 2020 and October 2021 to compute monthly interitem reliability for each study measure (see Figs. 1, 2 and 3 and Figures S1–S3). Child- (ωpolychoric = 0.88–0.96; αpolychoric = 0.87–0.94) and parent-/guardian-report (ωpolychoric = 0.85–1.06; αpolychoric = 0.86–0.95) of child emotional distress evidenced high interitem reliability at each month. From the CRISIS, two scales, child worry about contracting COVID-19 (ωpolychoric = 0.81–0.94; αpolychoric = 0.84–0.92) and stress related to stay-at-home orders or the cancelation of events (ωpolychoric = 0.74–0.87; αpolychoric = 0.77–0.90) were extracted based on conceptual similarity and interitem reliability (see Supplement). During baseline assessment, children (M = 2.86, SD = 0.90) and parents/guardians (M = 3.09, SD = 0.85) each completed one item to report on the child’s current physical health from 0 (“Poor”) to 4 (“Excellent”).Fig. 1 Change in child-reported distress (A), depressive symptomatology (B), and irritability (C). The x-axis reflects, at each month, the number of children who provided data, mean, standard deviation, and number and percent of children who exceeded any available clinical threshold Fig. 2 Change in child-reported panic/somatic (A), social anxiety (B), and separation anxiety (C) syndrome severity. The x-axis reflects, at each month, the number of children who provided data, mean, standard deviation, and number and percent of children who exceeded any available clinical threshold Fig. 3 Change in child-report of posttraumatic stress symptomatology (A), generalized anxiety symptomatology (B), worry about contracting COVID-19 (C), and stress related to stay-at-home orders or the cancelation of important events (C). For plots (A) and (B), the x-axis reflects, at each month, the number of children who provided data, the mean, standard deviation, and number and percent of children who exceeded any available clinical threshold. For plot (C), blue depicts worry about contracting COVID-19; red depicts stress related to stay-at-home orders or the cancelation of important events. The x-axis reflects the number of children who provided data; number and percent of children who reported attending school virtually; monthly mean and standard deviation of worry about contracting COVID-19; and monthly mean and standard deviation of stress related to stay-at-home orders or the cancelation of important events Psychiatric symptomatology At each assessment, each child and parent/guardian reported on the child’s irritability and depressive, posttraumatic stress, and anxiety syndrome severity over the preceding two-week period. Each measure was chosen for validation in epidemiological assessment of psychopathology per child- and parent-/guardian-report in community samples including disorder-specific thresholds (see Supplement). Each family completed the Screen for Child Anxiety Related Emotional Disorders [13] to assess the dimensional severity of four anxiety (i.e., panic/somatic anxiety, generalized anxiety, social anxiety, separation anxiety) syndromes. School-related anxiety was omitted given the intermittent provincial closure of schools during the study period. In this study, interitem reliability was high at each month for child- (ωpolychoric = 0.85–1.13; αpolychoric = 0.88–0.96) and parent-/guardian-report (ωpolychoric = 0.86–1.17; αpolychoric = 0.87–0.95) on all 4 subscales. Each child and parent/guardian reported on each child’s depressive symptom severity on the Short Mood and Feelings Questionnaire [SMFQ; 14] and irritability on the Affective Reactivity Index [ARI; 15]. In this study, interitem reliability was high at each month for the SMFQ based on child- (ωpolychoric = 0.93–1.10; αpolychoric = 0.95–0.99) and parent-/guardian-report (ωpolychoric = 0.92–1.02; αpolychoric = 0.94–0.98) and ARI based on child- (ωpolychoric = 0.81–1.03; αpolychoric = 0.89–0.97) and parent-/guardian-report (ωpolychoric = 0.83–0.98; αpolychoric = 0.91–0.96). Finally, each child and parent/guardian reported the severity of each child’s posttraumatic stress symptoms on the Child PTSD Symptoms Scale for DSM-5 [16]. Interitem reliability was high at each month based on child- (ωpolychoric = 0.98–1.07; αpolychoric = 0.96–0.99) and parent-/guardian-report (ωpolychoric = 0.95–1.04; αpolychoric = 0.93–0.99). Data analytic plan Multilevel, mixed effects models estimated average change in child distress and psychiatric symptoms (i.e., irritability and symptoms of depressive, anxiety, and posttraumatic stress syndromes) from June 2020 through December 2021. Putative time-varying correlates (e.g., worry about contracting COVID-19) were added to each model to evaluate correlates of children’s monthly distress. Models included random effects for each statistically significant fixed effect to simultaneously estimate interindividual heterogeneity. Data on the monthly mean local prevalence, hospitalization, and death rates were drawn from Public Health Ontario for each family’s geographic region [17]. Analyses were conducted in a stepwise fashion. As participants were enrolled in two cohorts, all analyses included an effect-coded variable to adjust for any difference between cohorts. Similarly, all analyses accounted for monotonic, linear change from baseline assessment to the last survey administered to each family to adjust for a potential initial elevation bias relative to later responses [18]. Initial analyses examined the association of child distress and psychopathology with COVID-19 epidemiological data. Next, increasing polynomial fixed then random effects were added to model change in child- and parent-/guardian-reported distress and psychopathology over time (see Tables S1–S4). Time was coded as the number of months from March 2020. Next, putative time-varying correlates were added to each model. Each correlate was person-centered to examine intraindividual associations with distress and psychopathology while adjusting for any interindividual associations [19]. Each correlate was examined separately (see Tables S5–S10). Lastly, all putative time-varying covariates were added to final models to evaluate their unique association with monthly variation in child distress and psychopathology. Additional analyses examined the association of local COVID-19 prevalence, hospitalization, and death rates with child worry about contracting COVID-19 and stress related to stay-at-home orders or the cancelation of important events to examine potential indirect associations of COVID-19 epidemiological data with child emotional distress and psychopathology. Models were fit with full information maximum likelihood estimation using the lavaan [20], ggplot2 [21], and nlme packages [22] in R version 4.1.2 [23]. We present raw p-values but note whether p-values would be retained following false discovery rate adjustment [24]. Analyses were checked for normality of residuals; all primary results discussed below were robust to removal of potential outliers based on Cook’s D. Data and analytic code are available upon specific request to the corresponding author. Results Participants In total, 377 families completed baseline assessment. Software obstacles (e.g., captcha) blocked most potential non-human respondents, though 60 responses resulted from non-human respondents who were subsequently removed from analyses. Data from 317 families were retained for analysis. Of these families, children ranged in age from eight to thirteen (M = 10.83, SD = 1.48); parents/guardians ranged in age from 21 to 58 (M = 41.26, SD = 5.85). Most parents/guardians identified as female [287 (90.45%)]; children’s sex was balanced between female [164 (51.74%)] and male [153 (48.26%)]. Parent-/guardian-report of children’s ancestry approximated the demographic composition of Southwestern Ontario [see Table S11; 25] with slightly greater representation of children with ancestry from England, Ireland, Scotland or Wales. Most adult respondents were parents [282 (88.96%)]; 29 (9.15%) were grandparents and five were siblings or an unrelated guardian (1.58%). In 237 (74.76%) families, at least one parent/guardian completed a four-year postsecondary degree. In 124 (39.12%) families, someone in the home was an essential worker, 26 (20.97%) of whom were first responders or healthcare providers. Almost all families lived in the Windsor-Essex region [297 (93.69%)]; 12 (3.79%) families were from Middlesex-London and seven (2.21%) were from Sarnia-Lambton or Chatham-Kent. 160 (50.47%) families endorsed receiving government assistance prior to the pandemic. However, to facilitate comparison with data from other countries, this study operationalized government assistance according to the CRISIS [12] to include social assistance programs widely available in Canada and Ontario including the Ontario Child Benefit provided to low- and moderate-income families [26]. Descriptive data Descriptive data for each month of the study are reported in Figs. 1, 2 and 3 and S1–S3 (see Supplement). The severity and prevalence of child distress and psychiatric syndromes fluctuated over time. According to normative data for each scale, fluctuations were small yet statistically robust (see Supplemental Tables S1–S4). However, within this convenience sample, a substantial percent of children remained above putative clinical thresholds on a broad set of internalizing syndromes over time. Of the 317 families whose data were retained for analysis, 250 (78.86%) completed at least 33% of the scheduled one-month follow-up assessment; 232 (73.19%) the two-month follow-up assessment; 212 (66.88%) the three-month follow-up assessment; 204 (64.35%) the four-month follow-up assessment; 199 (62.78%) the five-month follow-up assessment; 190 (59.94%) the six-month follow-up assessment; and 183 (57.73%) completed the scheduled nine-month follow-up assessment. Sixteen families (5.05%) withdrew from the study; otherwise, individual reasons why families chose not to complete a given follow-up assessment are unknown. Following a liberal alpha threshold of 0.10 to adjust for multiple testing, missing data at the final, 9-month follow-up was associated only with worse baseline child physical health as self-reported (p = 0.033) or observed by a parent/guardian (p = 0.036). Otherwise, there was no statistically significant evidence that data missing at the final 9-month follow-up assessment was associated with demographic information reported at baseline assessment by the child or parent/guardian (ps > 0.12; see Supplement). Child and parent/guardian baseline report of children’s physical health were moderately to strongly correlated, r = 0.59, 95% CI [0.51, 0.67], p < 0.01. Therefore, subsequent analyses adjusted for child overall physical health at baseline computed as the average of child- and parent/guardian-report at baseline. Association of monthly COVID-19 prevalence, hospitalization rate, and death rate with child mental health Elevated local monthly COVID-19 prevalence, b = 0.01, 95% CI (0.007, 0.014), p = 4.31 × 10–10, rate of hospitalization, b = 0.27, 95% CI (0.19, 0.35), p = 2.53 × 10–10, and death rate due to COVID-19, b = 0.26, 95% CI (0.17, 0.35), p = 1.04 × 10–8 were associated with children’s elevated monthly reported worry about themselves or a loved one contracting COVID-19. Similarly, as stay-at-home orders increased to address acute increases in both local and provincial prevalence, hospitalization rates, and death rates, children reported elevated stress related to stay-at-home orders or the cancelation of important events on months characterized by elevated local monthly COVID-19 prevalence, b = 0.005, 95% CI (0.002, 0.008), p = 0.002, hospitalization rate, b = 0.15, 95% CI (0.07, 0.23), p = 0.0004, and death rate, b = 0.14, 95% CI (0.05, 0.22), p = 0.001. Regarding indices of child mental health, parents/guardians reported elevated child distress on months characterized by elevated COVID-19 prevalence, b = 0.003, 95% CI (0.001, 0.005), p = 0.004, hospitalization rate, b = 0.09, 95% CI (0.04, 0.14), p = 0.0009, and death rate, b = 0.09, 95% CI (0.03, 0.14), p = 0.003. Monthly variation in COVID-19 prevalence was weakly associated with children’s self-reported monthly distress, b = 0.002, 95% CI (0, 0.005), p = 0.046, while monthly variation in COVID-19 prevalence, b = 0.01, 95% CI (0.001, 0.019), p = 0.023, and hospitalization rate, b = 0.28, 95% CI (0.06, 0.51), p = 0.014, were associated with children’s self-reported separation anxiety symptoms. There was otherwise limited evidence of a direct association of children’s psychopathology with COVID-19 prevalence, hospitalization rate, or death rate (ps > 0.055). Change in child mental health over time Child- and parent-/guardian-report of child distress; irritability; and symptoms of depressive, panic, and posttraumatic stress syndromes decreased initially but varied over time from June 2020 through December 2021 (see Figs. 1, 2 and 3 and Figures S1–S3). Anxiety symptoms, specifically symptoms of generalized anxiety disorder, social anxiety disorder, and separation anxiety disorder increased over time after accounting for the documented tendency for initial participant responses to be elevated [see Tables S1–S2; 18], though parent-/guardian-reported separation anxiety disorder severity decreased initially (see Table S4). Correlates of monthly child mental health Final analyses considered all putative time-varying correlates of child emotional distress and psychopathology concurrently along with adjustment for covariates based on methodology and systematic change over time. Preliminary analyses examined the association of each time-varying correlate with child emotional distress and psychiatric syndromes without adjustment for any covariates or systematic change in child distress or psychopathology over time (see Table S5–S10). Monthly elevations in child-reported fear that their health or the health of a loved one would be harmed by COVID-19 were associated with elevated monthly child emotional distress and broad psychiatric symptoms of irritability, depressive, posttraumatic stress, and generalized anxiety syndromes based on child-report (see Table 1) and parent-/guardian-report (see Table 2) as well as elevated monthly child-reported symptoms of panic or somatization even after adjustment for multiple testing as described above. Similarly, monthly elevations in child-reported stress related to stay-at-home orders or the cancelation of important events were associated with elevated monthly child emotional distress and symptoms of depressive, posttraumatic stress, and generalized anxiety syndromes based on child-report (see Table 1) or parent-/guardian-report as well as symptoms of irritability and separation anxiety disorder based on parent-/guardian-report (see Table 2). However, in final analyses, monthly variation in children’s attendance at school virtually was associated only with parent-/guardian-report of elevated child monthly emotional distress and depressive symptoms (see Table 2).Table 1 Correlates of child-report of distress and psychiatric symptomatology—b [p value] (95% CI) Distress Depression Irritability PTSD GAD Panic/somatization Social anxiety Separation anxiety Number of People 251 251 251 250 250 251 250 249 Number of assessments 1058 1068 1068 1065 1062 1068 1065 1059 Fixed effects Intercept 1.24 [0.122] (− 0.32, 2.81) 0.74 [0.898] (− 10.52, 12.00) 6.56 [0.056] (− 0.13, 13.26) − 5.79 [0.666] (− 31.94, 20.37) 1.53 [0.319] (− 1.46, 4.52) 5.32 [0.317] (− 5.04, 15.67) 4.16** [0.005] (1.31, 7.01) − 0.12 [0.912] (− 2.30, 2.06) Virtual school attendance (within-person) 0.02 [0.550] (− 0.04, 0.08) 0.49* [0.038] (0.03, 0.95) − 0.02 [0.883] (− 0.30, 0.25) 0.58 [0.290] (− 0.48, 1.64) 0.07 [0.669] (− 0.26, 0.41) 0.25 [0.240] (− 0.17, 0.68) 0.10 [0.553] (− 0.23, 0.43) 0.27 [0.056] (− 0.005, 0.55) Virtual school attendance (between-person) 0.09 [0.397] (− 0.12, 0.29) − 0.32 [0.731] (− 2.13, 1.49) − 0.30 [0.545] (− 1.26, 0.66) 2.63 [0.230] (− 1.65, 6.90) − 1.03 [0.213] (− 2.66, 0.59) − 0.32 [0.668] (− 1.78, 1.14) − 0.26 [0.743] (− 1.79, 1.28) 0.17 [0.778] (− 0.99, 1.32) Worry about contracting COVID-19 (within-person) 0.13 + [2.585e−05] (0.07, 0.18) 0.56** [0.004] (0.18, 0.94) 0.26* [0.031] (0.03, 0.50) 1.43 [0.001] (0.57, 2.29) 0.54 [0.002] (0.20, 0.88) 0.50 [0.003] (0.18, 0.83) 0.04 [0.762] (− 0.21, 0.29) 0.18 [0.094] (− 0.03, 0.39) Worry about contracting COVID-19 (between-person) 0.27 + [7.200e−07] (0.17, 0.37) 2.24 + [3.720e−06] (1.32, 3.17) 0.84* [0.0008] (0.36, 1.33) 7.18 + [0] (4.99, 9.37) 2.62 + [0] (1.79, 3.45) 2.52 + [0] 3.28) 1.99 + [1.120e−06] (1.21, 2.77) 1.13* [0.0002] (0.56, 1.71) Stress related to y-at- home orders (within-person) 0.23 + [0] (0.17, 0.28) 0.67 [0.004] (0.21, 1.12) 0.20 [0.051] (0.0001, 0.41) 1.47 [0.002] (0.56, 2.39) 0.53 [0.003] (0.18, 0.88) 0.20 [0.216] (− 0.11, 0.50) 0.23 [0.075] (− 0.02, 0.49) 0.16 [0.134] (− 0.05, 0.38) Stress related to stay-at- home orders (between-person) 0.33 + [0] (0.24, 0.42) 1.12 [0.007] (0.31, 1.93) 0.82* [0.0002] (0.40, 1.25) 1.77 [0.072] (− 0.14, 3.68) 0.83* [0.024] (0.11, 1.55) − 0.30 [0.366] (− 0.96, 0.35) − 0.51 [0.141] (− 1.20, 0.17) 0.70 [0.007] (0.19, 1.21) Each model is adjusted for recruitment cohort; survey in the longitudinal design; monthly local COVID-19 prevalence; and change over time as indicated by preliminary analyses; and random effects of each statistically significant within-person fixed effect term. See Table S12 for the full model PTSD posttraumatic stress disorder, GAD generalized anxiety disorder + p < 0.0001 p < 0.001 p < 0.01 p < 0.05 Table 2 Correlates of parent-/guardian-report of child distress and psychiatric symptomatology—b [p value] (95% CI) Distress Depression Irritability PTSD GAD Panic/somatization Social anxiety Separation anxiety Number of people 251 251 251 251 251 251 251 251 Number of assessments 1073 2077 1077 1077 1072 1077 1073 1073 Fixed effects Intercept − 0.22 [0.771] (− 1.72, 1.27) − 3.69 [0.490] (− 14.11, 6.73) 0.93 [0.770] (− 5.28, 7.15) 22.62* [0.049] (0.27, 44.96) 4.27 [0.006] (1.25, 7.29) 5.99 [0.175] (–2.62, 14.59) 5.48 + [6.754e–05] (2.81, 8.15) 4.02 [0.003] (1.40, 6.65) Virtual school attendance (within-person) 0.18 + [2.00e−08] (0.12, 0.25) 0.54* [0.012] (0.12, 0.96) 0.18 [0.161] (− 0.07, 0.44) 0.42 [0.368] (– 0.48, 1.32) 0.06 [0.687] (– 0.24, 0.36) – 0.19 [0.278] (– 0.54, 0.15) 0.08 [0.556] (– 0.19, 0.35) 0.12 [0.331] (– 0.12, 0.35) Virtual school attendance (between-person) 0.07 [0.498] (− 0.13, 0.27) 0.44 [0.606] (− 1.22, 2.10) − 1.12* [0.036] (− 2.16, − 0.08) 1.70 [0.337] (– 1.76, 5.17) – 0.64 [0.457] (– 2.31, 1.04) – 0.49 [0.506] (– 1.93, 0.95) – 0.41 [0.59] (– 1.89, 1.08) – 0.08 [0.893] (– 1.24, 1.08) Worry about contracting COVID-19 (Within-Person) 0.06** [0.010] (0.01, 0.10) 0.400* [0.013] (0.09, 0.70) 0.27 [0.006] (0.08, 0.46) 1.11 [0.002] (0.40, 1.82) 0.38 [0.004] (0.13, 0.64) 0.25 [0.052] (0, 0.50) 0.16 [0.126] (– 0.04, 0.37) 0.113 [0.218] (– 0.07, 0.29) Worry about contracting COVID-19 (between-person) 0.14 [0.006] (0.04, 0.25) 1.08* [0.013] (0.23, 1.94) 0.52 [0.054] (− 0.006, 1.05) 5.30 + [2.000e–08] (3.52, 7.07) 1.55* [0.0005] (0.69, 2.40) 1.87 + [9.200e–07] (1.14, 2.59) 1.59 + [5.930e–05] (0.83, 2.35) 1.01* [0.0008] (0.43, 1.59) Stress related to stay-at-home orders (within-person) 0.11 + [1.126e−05] (0.06, 0.16) 0.53 [0.008] (0.14, 0.91) 0.28 [0.009] (0.07, 0.49) 1.36*** [0.0009] (0.56, 2.15) 0.31* [0.025] (0.04, 0.58) 0.24 [0.068] (– 0.02, 0.49) 0.07 [0.517] (– 0.14, 0.27) 0.28 [0.007] (0.08, 0.48) Stress related to stay-at-home orders (between-person) 0.31 + [0] (0.22, 0.40) 1.57 + [5.009e−05] (0.83, 2.31) 0.73 [0.002] (0.27, 1.19) 1.42 [0.075] (– 0.13, 2.97) 0.66 [0.087] (– 0.09, 1.40) – 0.17 [0.606] (– 0.80, 0.47) – 0.42 [0.214] (– 1.09, 0.24) 0.58* [0.026] (0.07, 1.09) Each model is adjusted for recruitment cohort; survey in the longitudinal design; monthly local COVID-19 prevalence; change over time as indicated by preliminary analyses; and random effects of each statistically significant within-person fixed effect. See Table S13 for the full model PTSD posttraumatic stress disorder, GAD generalized anxiety disorder + p < 0.0001 *p < 0.001 p < 0.01 *p < 0.05 Discussion Longitudinal data here from June 2020 through December 2021 document a broad impact of the COVID-19 pandemic on children’s emotional distress and internalizing symptomatology. Child- and caregiver-report of children’s symptomatology also indicate a substantial minority of children who consistently exceed clinical thresholds indicative of a possible internalizing disorder (see Figs. 1, 2 and 3 and Figures S1–S3). Evidence of a broad impact over time that affects a substantial minority of children and youth extends data on the initial emotional and psychiatric impact of the COVID-19 pandemic to children and youth worldwide [1, 2] including in this study [27]. The pattern of change through January 2021 begins to harmonize results of otherwise seemingly discordant longitudinal investigations [e.g., 5, 9], which may result from the timing of longitudinal assessments. Through monthly assessment of children’s emotional distress, this study illustrates variability in children’s distress and psychiatric symptomatology over time. This identifies potential correlates of variability in the rate of American pediatric emergency department visits for mental health conditions [8]. Monthly assessment of children’s emotional distress and mental health here also facilitated examination of time-varying correlates to characterize months when children reported elevated distress or psychopathology. Per child- and caregiver-report, child emotional distress and psychopathology were associated with local COVID-19 prevalence, hospitalization, and death rates indirectly via increased child worry that they or a loved one would contract COVID-19 or through stress related to stay-at-home orders or the cancelation of important events. The lack of evident associations with social anxiety or separation anxiety demonstrates specificity given the limited relevance of both syndromes with worry about contracting COVID-19 or stress related to stay-at-home orders or the cancelation of important events. We also note limited evidence that monthly variation in children’s attendance in school virtually was associated with change in children’s mental health. However, parent-/guardian-report of children’s mental health may identify an association of virtual school attendance with elevated monthly emotional distress and depressive symptoms less evident in children’s report of their own distress. The associations of child distress and mental health with worry about contracting COVID-19 and stress related to stay-at-home orders or the cancelation of important events were highlighted in another recent longitudinal investigation [10] and emphasize the delicate balance required of public policy. These results underscore the importance of sustained, widespread efforts to (i) track local COVID-19 and related epidemiology; (ii) comprehensively monitor child and adolescent mental health; and (iii) communicate with families, children, and the public about all measures taken to enhance safety in public spaces and rationale when implementing or removing non-pharmaceutical interventions. We also note that, consistent with public health strategies employed in 2020 and 2021, the assessment of stress related to stay-at-home orders or the cancelation of important events used here emphasized disruptions to daily activities or interactions with family and friends (see Supplement). The evident empirical associations of worry about contracting COVID-19 and stress related to stay-at-home orders or the cancelation of important events with children’s mental health highlight the importance, to children’s mental health, of public health strategies that reduce risk while facilitating social interaction (e.g., widespread mask adoption, vaccination). Results of this study should be considered in light of several limitations including potential type 2 error; uncertain generalizability to other populations given the convenience sampling strategy employed; and lack of a control sample either prior to the pandemic or concurrent (e.g., from another country). However, although a relatively modest sample size without assessment prior to the pandemic, this study provided higher temporal resolution, granular assessment of child distress and psychopathology over the course of the COVID-19 pandemic. The modest sample size also highlights the prominent effect size of evident associations of worry about contracting COVID-19 and stress related to stay-at-home orders or the cancelation of important events with children’s acute, monthly fluctuations in myriad psychiatric outcomes. Future research is needed to compare data from the present study with that collected from other countries and regions, such as countries who implemented different public health strategies [e.g., 1]. In particular, given evidence of health disparities in the impact of the COVID-19 pandemic [e.g., 28], future research is critically needed to evaluate the impact of the pandemic on marginalized communities and in developing countries. Finally, this study examined internalizing syndromes. Future research is needed to explore the broad impact of the ongoing COVID-19 pandemic on externalizing psychopathology in children and youth including the implications of monthly variations in worry about contracting COVID-19 and stress related to stay-at-home orders or the cancelation of important events. Conclusion Critically, rather than provide inferences to be generalized globally, this study illustrates the importance of local efforts to track the broad, potentially sustained, psychosocial impact of the COVID-19 pandemic on children’s mental health throughout the ongoing pandemic and its recovery. Moreover, this study highlights key methodological features for tracking efforts including broad assessment of psychopathology, the integration of multi-informant reports [e.g., 29], and frequent assessment. This study also indicates that a substantial minority of children may demonstrate a sustained psychosocial impact throughout the COVID-19 pandemic. Otherwise, child- and caregiver-report of children’s well-being evidence variability in emotional distress and broad internalizing psychopathology over time. Finally, this study demonstrates that local COVID-19 severity (e.g., prevalence, hospitalizations) was associated with acute, monthly fluctuations in both children’s worry that they or a loved one might contract COVID-19 and stress related to stay-at-home orders or the cancelation of important events, which were associated with acute elevations in emotional distress and broad psychopathology according to both child- and caregiver-report. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 2091 KB) Acknowledgements The authors thank Erienne Cookson, University of Windsor, who assisted with data collection, and Dr. Patrick E. Shrout, Ph.D., New York University, and Dr. Brent I. Rappaport, Ph.D., Washington University in St. Louis, for feedback in data analysis and manuscript preparation. Author contributions L.R.: Conceptualization, Methodology, Software, Formal analysis, Investigation, Resources, Data Curation, Writing – Original Draft, Visualization, Supervision, Project administration, Funding acquisition. A.M.: Conceptualization, Validation, Data Curation, Writing – Original Draft. C.M.: Methodology, Investigation, Writing – Original Draft. K.B.: Conceptualization, Funding acquisition, Writing – Review & Editing. R.M.: Conceptualization, Funding acquisition, Writing – Review & Editing. A.A.: Conceptualization, Funding acquisition, Writing – Review & Editing. M.B.: Conceptualization, Validation, Funding acquisition, Writing – Review & Editing. Funding This study was funded by the Government of Ontario Ministry of Health and Long-Term Care COVID-19 Rapid Response Fund and the WE-Spark Health Institute awarded to LMR. The views expressed in the publication are the views of the authors and do not necessarily reflect those of the Province of Ontario. Data availability Data, materials, and code are available by request to the corresponding author. Declarations Conflict of interest The authors have no relevant financial or non-financial interests to disclose. ==== Refs References 1. Racine N McArthur BA Cooke JE Global prevalence of depressive and anxiety symptoms in children and adolescents during COVID-19: a meta-analysis JAMA Pediatr 2021 10.1001/jamapediatrics.2021.2482 2. Viner R Russell S Saulle R School closures during social lockdown and mental health, health behaviors, and well-being among children and adolescents during the first COVID-19 wave: a systematic review JAMA Pediatr 2022 10.1001/jamapediatrics.2021.5840 3. Early Childhood Development Action Network (2020) A joint statement on early childhood development and COVID-19: a call for coordinated action to protect and support all young children and their caregivers. Retrieved from https://www.mcusercontent.com/8103bc6125ed66e0964ae244d/files/462ed6c4–97cd-4bce-9a58–8f0efa8d17f2/Call_To_Action.pdf. Accessed 13 Sep 2022 4. Loades ME Chatburn E Higson-Sweeney N Rapid systematic review: the impact of social isolation and loneliness on the mental health of children and adolescents in the context of COVID-19 J Am Acad Child Adolesc Psychiatry 2020 59 1218 1239 10.1016/j.jaac.2020.05.009 32504808 5. Hawes MT Szenczy AK Olino TM Trajectories of depression, anxiety and pandemic experiences; a longitudinal study of youth in New York during the spring-summer of 2020 Psychiatry Res 2021 298 113778 10.1016/j.psychres.2021.113778 33550176 6. Orgilés M Francisco R Delvecchio E Psychological symptoms in Italian, Spanish and Portuguese youth during the COVID-19 health crisis: a longitudinal study Child Psychiatry Hum Dev 2021 10.1007/s10578-021-01211-9 7. Hawke LD Szatmari P Cleverley K Youth in a pandemic: a longitudinal examination of youth mental health and substance use concerns during COVID-19 BMJ Open 2021 11 e049209 10.1136/bmjopen-2021-049209 8. Radhakrishnan L Leeb RT Bitsko RH Pediatric emergency department visits associated with mental health conditions before and during the COVID-19 pandemic—United States, January 2019–January 2022 MMWR Morb Mortal Wkly Rep 2022 71 319 10.15585/mmwr.mm7108e2 35202358 9. Ravens-Sieberer U Kaman A Erhart M Quality of life and mental health in children and adolescents during the first year of the COVID-19 pandemic: results of a two-wave nationwide population-based study Eur Child Adolesc Psychiatry 2021 10.1007/s00787-021-01889-1 10. Weissman DG Rodman AM Rosen ML Contributions of emotion regulation and brain structure and function to adolescent internalizing problems and stress vulnerability during the COVID-19 Pandemic: a longitudinal study Biol Psychiatry Glob Open Sci 2021 1 272 282 10.1016/j.bpsgos.2021.06.001 34901918 11. Merikangas K, Milham M, Stringaris A (2021) The CoRonavIruS health impact survey (CRISIS) V0.3. nimh-comppsych 12. Nikolaidis A Paksarian D Alexander L The coronavirus health and impact survey (CRISIS) reveals reproducible correlates of pandemic-related mood states across the Atlantic Sci Rep 2021 11 8139 10.1038/s41598-021-87270-3 33854103 13. Birmaher B Khetarpal S Brent D The screen for child anxiety related emotional disorders (SCARED): scale construction and psychometric characteristics J Am Acad Child Adolesc Psychiatry 1997 36 545 553 10.1097/00004583-199704000-00018 9100430 14. Angold A Costello EJ Messer SC Pickles A Development of a short questionnaire for use in epidemiological studies of depression in children and adolescents Int J Methods Psychiatr Res 1995 5 237 249 15. Stringaris A Goodman R Ferdinando S The affective reactivity index: a concise irritability scale for clinical and research settings J Child Psychol Psychiatry 2012 53 1109 1117 10.1111/j.1469-7610.2012.02561.x 22574736 16. Foa EB Asnaani A Zang Y Psychometrics of the Child PTSD symptom scale for DSM-5 for trauma-exposed children and adolescents J Clin Child Adolesc Psychol 2018 47 38 46 10.1080/15374416.2017.1350962 28820616 17. Public Health Ontario Ontario COVID-19 data tool 2022 Ontario Ontario Agency for Health Protection and Promotion 18. Shrout PE Stadler G Lane SP Initial elevation bias in subjective reports Proc Natl Acad Sci 2018 115 E15 E23 10.1073/pnas.1712277115 29255039 19. Curran PJ Bauer DJ The disaggregation of within-person and between-person effects in longitudinal models of change Annu Rev Psychol 2011 62 583 619 10.1146/annurev.psych.093008.100356 19575624 20. Rosseel Y lavaan: an R package for structural equation modeling J Stat Softw 2012 48 1 10.18637/jss.v048.i02 21. Wickham H ggplot2: elegant graphics for data analysis 2009 New York Springer 22. Pinheiro J, Bates D, DebRoy S, et al (2016) nlme: linear and nonlinear mixed effects models 23. R Core Team (2020) R: a language and environment for statistical computing 24. Benjamini Y Hochberg Y Controlling the false discovery rate: a practical and powerful approach to multiple testing J R Stat Soc Ser B Methodol 1995 57 289 300 10.1111/j.2517-6161.1995.tb02031.x 25. Statistics Canada (2020) Data tables, 2016 Census 26. Government of Ontario Ministry of Children, Community and Social Services (2021) Ontario child benefit. In: ontario.ca. http://www.ontario.ca/page/ontario-child-benefit. Accessed 14 Feb 2022 27. Mactavish A, Mastronardi C, Menna R, Babb KA, Battaglia M, Amstadter AB, Rappaport LM (2021) Children’s mental health in southwestern Ontario during summer 2020 of the COVID-19 pandemic. J Can Acad Child Adolesc Psychiatry 30:177–190 28. Webb Hooper M Nápoles AM Pérez-Stable EJ COVID-19 and racial/ethnic disparities JAMA 2020 323 2466 10.1001/jama.2020.8598 32391864 29. De Los RA Augenstein TM Wang M The validity of the multi-informant approach to assessing child and adolescent mental health Psychol Bull 2015 141 858 900 10.1037/a0038498 25915035	36484855	PMC9734395	NO-CC CODE	2022-12-14 23:28:28	no	Eur Child Adolesc Psychiatry. 2022 Dec 9;:1-12	utf-8	Eur Child Adolesc Psychiatry	2,022	10.1007/s00787-022-02121-4	oa_other
==== Front Nat Rev Nephrol Nat Rev Nephrol Nature Reviews. Nephrology 1759-5061 1759-507X Nature Publishing Group UK London 36474115 664 10.1038/s41581-022-00664-y Year in Review Progress towards solving the donor organ shortage http://orcid.org/0000-0002-8919-2445 Anderson Douglas J. 12 Locke Jayme E. [email protected] 12 1 grid.265892.2 0000000106344187 Comprehensive Transplant Institute, University of Alabama at Birmingham, Birmingham, AL USA 2 grid.265892.2 0000000106344187 Department of Surgery, University of Alabama at Birmingham, Birmingham, AL USA 6 12 2022 12 © 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. Kidney transplantation is the best therapy for kidney failure, but is limited by donor organ availability and the risks associated with immunosuppression. Studies in 2022 provided encouraging data about the outcomes of COVID-19 among transplant recipients, the effects of changes to organ allocation policy in the US and progress in xenotransplantation, raising hope that the organ shortage can be solved. Key advances Organ transplantation is not an independent risk factor for death from COVID-19 when data are adequately adjusted for other comorbidities3. The performance of US Organ Procurement Organizations is not driven by the organ acceptance and utilization rates of local transplant centres5; policy changes could increase organ recovery, allocation and utilization. Successful transplantation of genetically modified pig kidneys into recently deceased humans is encouraging for the potential future of clinical xenotransplantation7,8. Subject terms Nephrology Kidney diseases Organ transplantation ==== Body pmcKidney transplantation is the preferred treatment for kidney failure, with transplant recipients enjoying increases in longevity and quality of life. However, the Achilles’ heel of transplantation remains the shortage of available organs. As the supply of these life-saving organs is limited, only a minority of patients can realize the benefits. The past few years have offered additional challenges as the COVID-19 pandemic wreaked havoc on global health care systems. However, key studies published in 2022 have provided reasons for optimism regarding the future of kidney transplantation (Fig. 1).Fig. 1 Major findings in kidney transplantation in 2022. Key studies reported in 2022 provided reasons for optimism regarding the impact of COVID-19 on kidney transplant recipients3 (a), the potential for US policy changes to increase rates of organ transplantation5 (b) and the future of kidney xenotransplantation7,8 (c). “the Achilles’ heel of transplantation remains the shortage of available organs” COVID-19 has dominated the attention of the healthcare community for nearly 3 years, and no area of medicine or patient population has been unaffected. COVID-19 is of particular concern for immunocompromised patients, and early studies showed increased mortality in this group1,2. The risk to transplant recipients was less clear, but was suspected to be similarly increased. In 2022, a meta-analysis of 31 observational studies from the pre-vaccine era reported that receipt of a solid organ transplant was not associated with increased mortality up to 30 days after infection with SARS-COV-2, after adjustment for comorbidities3. Notably, organ transplant recipients did have increased 30-day COVID-19 mortality when data from studies without adjustment were included in the analysis. A subgroup analysis of the adjusted outcome data showed that kidney transplant recipients were not at increased risk of death from COVID-19 compared with the general population. Transplant recipients require monitoring of immunosuppression and therefore have frequent interactions with the healthcare system, so they might have benefited from early detection of infection and escalation of COVID-19 care. Importantly, the meta-analysis also showed that transplant recipients were at more risk of ICU admission and acute kidney injury (AKI) owing to COVID-19 than the general population. This increased risk of AKI is notable given that most of the transplant recipients in the pooled analysis were kidney recipients. The data did not enable the authors to assess whether the risk of rejection or organ loss was due to COVID-19 itself or to reduction of immunosuppression in response to SARS-COV-2 infection. As most of the studies included in the meta-analysis were from the first wave of the pandemic, it is unclear whether the findings would extend to the subsequent waves of the Delta and Omicron variants, which had differing effects across regions owing to the timing of surges and social determinants of health. The effects of the later variants would also be confounded by increasing vaccine availability, use of novel therapeutics such as tixagevimab–cilgavimab and increasing rates of natural immunity owing to previous SARS-COV-2 infection. In 2021, at the height of the pandemic, major policy changes were enacted in the US with the goal of increasing organ transplantation and decreasing geographic variability in transplant rates. Rather than the previous local–regional–national allocation system based on donor service areas (DSAs) and United Network for Organ Sharing (UNOS) regions, kidneys from deceased donors are now allocated first within a geographical circle with a radius of 250 nautical miles, then shared nationally if not used within that circle. Moreover, the Center for Medicare and Medicaid Service (CMS) revised their outcome measures for Organ Procurement Organizations (OPOs) with the aim of incentivizing them to be more aggressive in pursuing organ donation and recovery. The overall result of these changes is something to celebrate — 2021 saw more kidneys transplanted in the US than in any previous year, and 2022 is on track to exceed this record4. However, much work is still to be done. Organ discard rates are increasing and the simultaneous nature of the policy changes, enacted during a pandemic, makes accurate assessment of their effects challenging. Using data from before the 2021 change in donor kidney allocation, Doby et al.5 demonstrated that low-performing OPOs had low performance across all donor groups without consistent trends in kidney utilization or discard rates within the DSAs that they serve. This finding discredits a criticism of the CMS revision of outcome measures — that OPO performance is driven by local transplant centre organ acceptance practices. This criticism should be further abated by the 2021 changes in allocation policy, which effectively unlink OPOs from centres within their DSAs. Doby et al.5 clearly demonstrate that organ recovery, allocation and utilization are inextricably linked. As further policy changes are considered in the US, including a proposed move to a continuous distribution model free of any geographic boundaries, care must be taken to consider how such changes would affect each component of this triad. The challenges of kidney allocation and utilization are driven by the shortage of available organs for transplantation. Decades of advocacy work to increase donation of kidneys from both deceased and living donors have been fruitful, but these successes do not outweigh the increasing prevalence of kidney failure. Indeed, annual additions to waiting lists exceed the number of kidney transplants performed each year and, until recently, the number of patients waiting for a transplant increased annually6. Ending the organ shortage and achieving durable immune tolerance have long been the ‘holy grails’ of transplantation. The past year saw a major step towards a possible solution. Two groups, working independently, reported the successful transplantation of a genetically modified pig kidney into a recently deceased person7,8. In both studies, the transplanted kidney produced urine and did not undergo hyperacute rejection. This success builds on decades of preclinical research in nonhuman primates aimed at advancing xenotransplantation as a solution to the organ shortage, and represents a major step towards a future clinical trial. In preparation for such a trial, Porrett et al.7 also demonstrated the feasibility of performing pre-transplantation crossmatching across the species barrier. A negative pre-transplant crossmatch test is considered standard-of-care for kidney allotransplantation, so this proof-of-concept for xenotransplantation is another key milestone towards the clinic. The work of these two groups in 2022 was further bolstered by the announcement of a successful xenotransplant of a genetically modified pig heart into a human9. Together, these studies strongly suggest that clinical xenotransplantation is on the horizon. “clinical xenotransplantation is on the horizon” As we emerge from the COVID-19 pandemic, the goal remains to offer the benefits of kidney transplantation to as many patients as possible. Although the challenges of COVID-19 might finally be abating, kidney transplantation is still hindered by a lack of available organs. We must strive to make the most of the available donor organs via effective policies that address all phases of the transplant system (recovery, allocation and utilization) and to push forward with preclinical research in the hope that xenotransplantation or another advance will end the organ shortage. Competing interests D.J.A. and J.D.L. receive grant funding from United Therapeutics and its subsidiaries Lung Biotechnology and Revivicor. D.J.A. and J.D.L. receive grant funding from Hansa Biopharma. J.D.L. is a consultant for Sanofi. ==== Refs References 1. Baek MS Lee M-T Kim W-Y Choi JC Jung S-Y COVID-19-related outcomes in immunocompromised patients: a nationwide study in Korea PLoS One 2021 16 e0257641 10.1371/journal.pone.0257641 34597325 2. Suárez-García I In-hospital mortality among immunosuppressed patients with COVID-19: analysis from a national cohort in Spain PLoS One 2021 16 e0255524 10.1371/journal.pone.0255524 34343222 3. Gatti M Clinical outcome in solid organ transplant recipients affected by COVID-19 compared to general population: a systematic review and meta-analysis Clin. Microbiol. Infect. 2022 28 1057 1065 10.1016/j.cmi.2022.02.039 35289294 4. Organ Procurement and Transplantation Network data (OPTN, accessed 27 October 2022); https://optn.transplant.hrsa.gov/data/ 5. Doby BL Examining utilization of kidneys as a function of procurement performance Am. J. Transplant. 2022 22 1614 1623 10.1111/ajt.16985 35118830 6. Lentine KL OPTN/SRTR 2020 Annual Data Report: Kidney Am. J. Transplant. 2022 22 Suppl. 2 21 136 10.1111/ajt.16982 35266618 7. Porrett PM First clinical-grade porcine kidney xenotransplant using a human decedent model Am. J. Transplant. 2022 22 1037 1053 10.1111/ajt.16930 35049121 8. Montgomery RA Results of two cases of pig-to-human kidney xenotransplantation N. Engl. J. Med. 2022 386 1889 1898 10.1056/NEJMoa2120238 35584156 9. Griffith BP Genetically modified porcine-to-human cardiac xenotransplantation N. Engl. J. Med. 2022 387 35 44 10.1056/NEJMoa2201422 35731912	36474115	PMC9734397	NO-CC CODE	2022-12-14 23:28:28	no	Nat Rev Nephrol. 2022 Dec 6;:1-2	utf-8	Nat Rev Nephrol	2,022	10.1038/s41581-022-00664-y	oa_other
==== Front AIDS Behav AIDS Behav AIDS and Behavior 1090-7165 1573-3254 Springer US New York 36472685 3942 10.1007/s10461-022-03942-9 Original Paper “I want the doctors to know that I am as bright as a candle”: Experiences with and Hopes for Doctor Interactions Among Malaysian Key Populations and People Living with HIV Earnshaw Valerie A. [email protected] 1 Cox Jon 2 Wong Pui Li 3 Saifi Rumana 34 Walters Suzan 5 Azwa Iskandar 3 Omar Sharifah Faridah Syed 3 Collier Zachary K. 6 Hassan Asfarina Amir 4 Lim Sin How 7 Wickersham Jeffrey 8 Haddad Marwan S. 9 Kamarulzaman Adeeba 34 1 grid.33489.35 0000 0001 0454 4791 Department of Human Development and Family Sciences, University of Delaware, 111 Alison Hall West, Newark, DE 19716 USA 2 grid.33489.35 0000 0001 0454 4791 Department of Art and Design, University of Delaware, Newark, DE USA 3 grid.10347.31 0000 0001 2308 5949 Faculty of Medicine, Universiti Malaya, Kuala Lumpur, Malaysia 4 grid.10347.31 0000 0001 2308 5949 Centre of Excellence for Research in AIDS, Universiti Malaya, Kuala Lumpur, Malaysia 5 grid.137628.9 0000 0004 1936 8753 School of Global Public Health, New York University, New York, NY USA 6 grid.33489.35 0000 0001 0454 4791 School of Education, University of Delaware, Newark, DE USA 7 grid.10347.31 0000 0001 2308 5949 Department of Social and Preventive Medicine, Universiti Malaya, Kuala Lumpur, Malaysia 8 grid.47100.32 0000000419368710 School of Medicine, Yale University, New Haven, CT USA 9 grid.428181.6 Center for Key Populations, Community Health Center, Inc, New Britain, CT USA 6 12 2022 110 21 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. Stigma in healthcare settings is a pernicious barrier to HIV prevention and treatment in contexts with strong HIV-related structural stigma. Previous work has documented substantial stigma towards key populations and people living with HIV (PLWH) among Malaysian doctors. The perspectives of Malaysian key populations and PLWH, however, remain understudied. In 2021, 34 Malaysian participants representing key populations and PLWH engaged in a photovoice study designed to qualitatively explore their experiences with and hopes for doctor interactions. Many participants reported stigma from their doctors, perceiving that doctors view them as not normal, sinful, misguided, and incapable. Several emphasized that they wear figurative masks to conceal aspects of themselves from doctors. Yet, many also remain hopeful for constructive relationships with doctors. They want their doctors to know that they are bright, capable, kind, and valuable. Interventions are needed to address stigma among doctors working in contexts with strong structural stigma. Resumen El estigma en los ambientes de atención médica es una barrera perniciosa en la prevención y el tratamiento del VIH. Investigaciones anteriores han documentado un estigma sustancial hacia los grupos de población clave y las personas que viven con el VIH (PLWH por sus siglas en inglés) entre los médicos de Malasia. Sin embargo, las perspectivas de los grupos de población clave y las PLWH en Malasia siguen sin estudiarse. En 2021, 34 participantes que representaban los grupos de población clave y PLWH en Malasia participaron en un estudio de fotovoz diseñado para explorar cualitativamente sus experiencias y esperanzas en las interacciones con los médicos. Muchos participantes describieron el estigma de sus médicos, percibiendo que los médicos los ven como no normales, pecaminosos, equivocados e incapaces. Varios enfatizaron que usan máscaras figurativas para ocultar aspectos de ellos mismos a los médicos. Sin embargo, muchos también mantienen la esperanza de tener relaciones constructivas con los médicos. Quieren que sus médicos sepan que son inteligentes, capaces, amables y valiosos. Se necesitan intervenciones para abordar el estigma estructural entre los médicos que trabajan en la prevención y el tratamiento del VIH. Keywords HIV Key Populations Photovoice Stigma http://dx.doi.org/10.13039/100000025 National Institute of Mental Health R34MH124390 http://dx.doi.org/10.13039/100000026 National Institute on Drug Abuse K01DA053159 Walters Suzan ==== Body pmcIntroduction HIV-related stigma in healthcare settings is a barrier to every step of the HIV care cascade, playing a fundamental role in HIV disparities globally [1]. Stigma is a social process supported by power that differentiates people based on social statuses and leads to devaluation [2]. Stigma is a multilevel phenomenon. At the structural level, it is manifested as societal-level conditions, cultural norms, and institutional policies [3, 4]. Among healthcare providers, it is manifested as prejudice (i.e., negative feelings), stereotypes (i.e., group-based beliefs) and discrimination (i.e., unfair or unjust treatment) [5]. Key populations [i.e., populations most likely to be exposed to HIV including men who have sex with men (MSM), transgender women (TGW), people who inject drugs (PWID), and female sex workers (FSW)] as well as people living with HIV (PLWH1) who anticipate and experience more stigma from healthcare providers have worse HIV prevention and treatment outcomes [1, 8–10]. Stigma is a particularly pernicious barrier to HIV prevention and treatment in places where strong structural stigma persists towards key populations and PLWH [11–16]. As examples, indicators of HIV-related structural stigma are associated with increased HIV risk among African Americans and people who inject drugs in the United States [13, 14], MSM in Europe [17], and MSM and TGW in Asia [18]. Malaysia is an example of a country where strong HIV-related structural stigma persists [19–26]. Previous work has identified substantial prejudice towards, stereotypes about, and intentions to discriminate against key populations and PLWH among Malaysian healthcare providers [27–29]. Yet, the perspectives of Malaysian key populations and PLWH on their healthcare experiences remain understudied. The current study uses photovoice to qualitatively explore Malaysian key populations’ and PLWH’s experiences with healthcare providers to inform intervention efforts to address stigma in Malaysian and other similar healthcare settings with strong structural stigma, ultimately reducing HIV disparities in these contexts. HIV Disparities and Structural Stigma in Malaysia Malaysia is representative of many socio-cultural contexts wherein the HIV epidemic is concentrated among members of key populations, many of whom are severely stigmatized and do not know their HIV status, have not received HIV prevention services, and/or are not receiving antiretroviral therapy. In 2020, rates of HIV among Malaysian key populations were estimated to be 15.8 to 54.0 times those of the general population [30]. Yet, members of key populations were, at best, half as likely as members of the general population to know their HIV status [30]. Many members of key populations were not reached by HIV prevention programs, with as little as 1.4% of PWID receiving prevention programming [30]. Moreover, antiretroviral therapy coverage for key populations was below the UNAIDS goal of 90% (ranging 34.6–62.6% where data are available) [30]. These HIV disparities occur within a social environment characterized by substantial stigma towards key populations. Like many countries worldwide, stigma towards key populations in Malaysia has deep historical roots and is currently structurally sanctioned [19, 24–26]. The British Empire spread legal codes to its colonies, including Malaysia, that criminalized same-sex sexual practices with punishments including fines and lengthy imprisonment. As of 2020, Malaysia was one of 67 countries that continues to criminalize consensual same-sex conduct [19]. Secular and Shariah laws criminalize the gender expression of transgender individuals, and Shariah law prohibits gender-affirming surgery among Muslims [20, 26]. Additionally, although government officials have recently announced plans to decriminalize drug use to better address addiction, Malaysia has harsh penalties for drug possession (e.g., death penalty) [21, 24]. Laws prohibit activities related to sex work, and FSW report being charged for sex-work related crimes on the basis of carrying condoms [22–24]. These laws contribute to a social environment characterized by profound stigma towards key populations and PLWH, including within healthcare settings. Stigma in Malaysian Healthcare Settings Previous work has documented substantial prejudice towards, stereotypes about, and intentions to discriminate against members of key populations and PLWH among practicing doctors and medical students in Malaysia [27–29]. For example, medical students report explicit negative feelings towards MSM and PWID, and ambivalence towards PLWH [27]. Medical students who are Muslim and Malay tend to endorse greater prejudice towards key populations than those of other religions and ethnicities [27, 31]. Moreover, doctors and medical students who endorse greater prejudice toward and fear of key populations and PLWH also report greater intentions to discriminate against these patients [28, 29, 31]. This work additionally suggests that prejudice towards key populations and PLWH is correlated such that individuals who endorse more prejudice towards one key population or PLWH group also endorse more prejudice towards other groups [28]. Less work has been conducted to explore the perceptions and experiences of key populations and PLWH in Malaysian healthcare settings. The current study explores Malaysian key populations’ and PLWH’s experiences with doctor interactions. It focuses on meta-perceptions, which are thoughts about how others see the self [32]. Meta-perceptions play a powerful role in shaping the extent to which individuals anticipate stigma from others. Key populations and PLWH who believe that healthcare providers feel prejudice towards and believe stereotypes about them may anticipate stigma from those providers. In turn, key populations who anticipate stigma from healthcare providers are less likely to access HIV testing and pre-exposure prophylaxis (PrEP), and PLWH who anticipate stigma from healthcare providers are less likely to access HIV care [1, 8–10]. The current study additionally explores Malaysian key populations’ and PLWH’s hopes for provider interactions, with a focus on what they would like providers to know about them. Greater understanding of key populations’ and PLWH’s experiences with and hopes for doctor interactions can inform interventions to reduce stigma in healthcare settings and close HIV disparities in Malaysia. These findings may also be useful in other environments characterized by substantial stigma. Methods Procedure Members of key populations and PLWH in Malaysia engaged in an online photovoice project in the fall of 2021. Photovoice, a qualitative method involving photography and story-telling [33], was chosen as a research method because it yields rich data and allows for creative expression among participants. Moreover, online, asynchronous photovoice methods offer high levels of confidentiality for participants who do not have to speak with members of the study team or travel to a data collection site. The photovoice project was facilitated by a website, with content available in both Bahasa Malay and English. The first page of the website included a video that introduced participants to the photovoice project. The doctor in the video noted that doctors providing HIV prevention and treatment services may struggle to understand the lives of their patients who have used drugs, sexual and gender minorities, sex workers, and PLWH. The doctor explained that the goal of the study is to help doctors connect with and better understand the experiences of their patients so that they can provide better HIV-related care. After watching the introductory video, participants viewed an informed consent form and indicated their agreement to participate. After providing consent, participants could watch a series of instructional videos about photography. These videos were created for the current study and featured a professional photographer and professor of photography. Lastly, participants were introduced to the photovoice prompts. These prompts included: (1) what makes you happy? (2) what makes you sad? (3) what is important to you? (4) what challenges do you face? (5) how do doctors see you? and (6) what do you want doctors to know about you? Each prompt was accompanied by a video featuring doctors explaining why knowing the answer to the question would help them provide better care, as well as an example of a photovoice response. To respond to the prompt, participants were invited to submit a photograph and caption via the website. A member of the research team was available to answer participant questions about the study. All participants responded to all of the study prompts, and received monetary compensation for their time. All procedures were approved by institutional review boards at the Universities of Malaya and Delaware. For more information about the study website and procedures, please see Earnshaw et al. [34]. Participants Participants were recruited in partnership with local community-based organizations that serve key populations and PLWH. Digital flyers in Bahasa Malay and English were shared via social media and WhatsApp. The study was advertised as “Photovoices in Healthcare: Connecting through Photography” and participants were told that they would be asked to answer questions about their experiences with doctors. Individuals were eligible to participate if they were 18 years or older, had access to a camera phone with internet connection, and belonged to a key population group or were living with HIV. The recruitment materials directed individuals to contact the study research assistant, who screened individuals for study eligibility via WhatsApp. In total, 34 individuals participated including 12 (35.3%) MSM, 8 (23.5%) PLWH, 7 (20.6%) TGW, 7 (20.6%) FSW, and 6 (17.6%) PWID. Five (14.7%) individuals identified as MSM and PLWH, and one individual (2.9%) identified as TGW and PLWH. Half completed the project in Malay and half completed it in English. Over half of participants identifying as MSM (66.7%), TGW (71.4%), and PLWH (62.5%) responded in English whereas over half of participants identifying as FSW (71.4%) and PWID (83.3%) responded in Malay. To best protect participants’ identities and encourage trust in the project, socio-demographic data that could be used to identify participants (e.g., age, gender, ethnicity, religion) were not collected. Analysis The current analysis focuses on participant responses to the prompts “how do doctors see you?” and “what do you want doctors to know about you?”. Rapid Qualitative Inquiry methods were employed to analyze the data [35]. Malay captions were first translated to English by two members of the study team. We then took an iterative and team-based approach to data analyses. Initial findings based on photographs and captions were summarized and examples of findings were identified by three members of the study team, including two Malaysian and one U.S. team members. Themes were compiled as they emerged from the data. Next, findings were discussed with members of the full team, which included Malaysian and U.S. researchers and clinicians. This discussion focused on reaching consensus about findings and included identifying new questions about the data. Findings were then re-summarized by the three members of the team who led the initial summary and then re-discussed by the full team, and conclusions were refined. As a trustworthiness check, results were then shared with members of the project’s Scientific and Community Advisory Board belonging to key population groups to verify the team’s conclusions. Results “How do doctors see you?”: Participants’ Meta-Perceptions In response to the question “how do doctors see you?” participants reported that doctors’ perceptions of them ranged from positive to negative. Several participants noted that doctors view them positively and want the best for them. A PWID participant responding in English whose photograph featured objects and a setting that promotes physical and emotional wellbeing (i.e., sunscreen, apple, stuffed animal, glasses, and pool) reported that doctors perceive them to be resilient and encourage them to be healthy. They stated:The doctors see me as someone who is brave and strong being able to go through life despite having problems with my health, but they also see that I need to give a lot more attention and love on myself so that I can stay as healthy as possible. Other than that they also said to avoid thinking about things that upset me cause it really effect my physical. Similarly, a MSM participant responding in English whose photograph featured yellow flowers bathed in sunlight noted that doctors view them as a bright flower who is capable of taking care of themselves. Other participants noted that doctors’ views of them were kind and professional (e.g., “very satisfactory, kind and professional” [TGW participant responding in Malay]) as well as understanding and non-judgmental. As shown in Fig. 1 A, a FSW participant responding in Malay whose photograph featured a beautiful, expensive dress on a dress form noted that their doctor treats them like a human being. Fig. 1 How do doctors see you? In contrast, many participants characterized doctors’ perceptions of them as stigmatizing. A PWID participant responding in Malay noted that “there are a number of unprofessional doctors, too much stigma, double standards” and a MSM participant responding in English reported that “they judge and stereotype me without treating (me) as an individual.” As shown in Fig. 1B, another MSM participant responding in English reported that they feel that doctors see them as not normal and shocking, including during HIV testing visits. Their photograph featured feet with painted toenails, suggestive of gender transgression. A TGW participant responding in English whose photograph featured a branch on the ground noted:Even though, I’m a transwomen but my ID card identify myself as man so most of healthcare workers see me as man. This is not what I want or identify. I just hope I could be the branch from the picture which is genderless. Another TGW participant responding in English noted that “doctors see me as someone who committed sin, and I should repent and be as I should be (a man).” Participants reported that these stigmatizing perceptions resulted in doctors judging them as incapable of taking care of themselves. A MSM participant responding in English captioned their photograph featuring a withered purple flower: “Small. Misguided. Helpless. Even when I do have the medical expertise, it’s not enough to convince them that I am capable of understanding my body and my limits. To them I am withered flower in need of caring.” Participants additionally questioned whether doctors care about them. One PLWH participant responding in Malay whose photograph featured HIV medications stated: “I always doubt whether a doctor was really sincere in treating me or just fulfilling his duty.” Participants also responded to the question by emphasizing that they conceal parts of themselves, or wear a figurative mask, in healthcare settings. Some participants reported that doctors are aware of the concealment, and may even see their true self. A FSW participant responding in English noted that doctors see them as “Look good and bold outside but afraid to open up. I keep it low key and my doctor see that in me.” Their photograph featured the outer petals of a bright pink flower, the interior of which is hidden from view. A MSM participant responding in Malay whose photograph included pushpins with bright, round plastic heads noted that they are like the pins: “the first side is round where I will face (doctors) and share what I go through voluntarily and transparently, and the next side is sharp where I try to hide from them.” Other participants suggested that doctors are either unaware that they are concealing or do not see the real them. As shown in Fig. 1 C, a MSM and PLWH participant responding in English noted that they mostly hide who they are from others due to distrust, sharing a calm and “sugarcoated” exterior with doctors. Their photograph featured a calm landscape of trees that is zoomed out, emphasizing distance. “What do you want doctors to know about you?”: Participants’ Hopes for Doctors Participant responses to the question “what do you want doctors to know about you?” emphasized their humanity and desire to be seen and treated equally to other patients. Many participants responded that they wanted to be known by doctors as unique, kind, or valuable in some way. As shown in Fig. 2 A, a MSM participant responding in English whose photograph featured a candle noted that they wanted their doctors to know that they are intelligent, capable, and bright. Similarly, a FSW participant responding in English whose photograph featured a doll collection shared that “despite the challenges and trials I go through every day, I am a cheerful and childlike person.” Some participants emphasized their value or worth. A PWID participant responding in Malay whose photograph featured a leafy plant noted that “I want the doctors to know that I want to be like a leaf, even though it is unattractive, there are still some benefits where it can shade us, and when it falls, it is still useful as fertilizer.” Other participants emphasized that they are normal, or like other patients. A PLWH participant responding in Malay whose photograph featured a group of goats noted that “I want the doctor to see me the same as any other patient without any difference in terms of the type of disease or otherwise.” Fig. 2 What do you want doctors to know about you? Several participants wanted doctors to see and treat them without stigma or discrimination. As shown in Fig. 2B, a TGW participant responding in English whose photograph featured horses in a forest noted that she wants to be seen equally to other patients, without stigma or discrimination. Other participants called on doctors to treat them without judgment. For example, a PWID participant responding in Malay whose photograph featured a white cake artfully topped with strawberries noted that they wanted doctors to “easily and calmly accept what is presented without seeing and judging the individual that prepares it.” Another PWID participant responding in Malay whose photograph featured a hand holding a vertical bar, reminiscent of a jail cell, stated “I hope the doctors see me as someone who dares to come out of the existing cocoon. Don’t get caught up in the insults that are offensive and demeaning.” A MSM participant responding in English whose photograph featured a single, bright purple flower emphasized that they wanted to be treated as an individual based on their life circumstances rather than their risk group membership. They stated: “I want the doctors to know that I may be bisexual but my orientation doesn’t make me a high risk behaviour client. Treat me as an individual based on my story/history without prejudice to my sexuality.” Participants also wanted doctors to see beyond the figurative masks they wore in healthcare settings and recognize the vulnerability or pain underneath. They addressed their needs related to HIV, including challenges surrounding HIV risk, fear of HIV testing, and concerns related to living with HIV. A FSW participant responding in Malay whose photograph featured the sky seen through a broken roof stated: “Support. I want doctors to know about me that I am often a place of refuge for others but that I also need protection. I sometimes fail to protect myself from my negative and unhealthy environment.” A TGW participant responding in English whose photograph featured HIV and STI tests noted: “I am very apprehensive about doing the HIV screening. Malaysian in general are still lack of information regarding HIV/AIDS which lead to Stigma and Discrimination.” A MSM and PLWH participant responding in Malay whose photograph featured pelicans with their wings spread defensively stated thatI want the doctor to know about me that I am strong but there are times when I get scared when I face the limitation of this disease where I need the support of medical knowledge in terms of facts and moral support. Finally, many participants appealed to doctors for help and support. Some requested help to be healthy. A FSW participant responding in Malay whose photograph featured a rice paddy landscape noted “As healthy as paddy is to produce rice, please serve us so that we are healthy.” Others called for guidance. A MSM and PLWH participant responding in English whose photograph featured the reflection of a sun setting over the ocean noted:In between light and shadow there’s a reflection of who we are. Guide us so that when light and shadow meet, only ourselves remain. Guide us on a better path so that one day we can stand on our own or even help people like us. Several participants asked for emotional support. As shown in Fig. 2 C, a PWID participant responding in Malay whose photograph featured two hands holding each other noted that they need moral support from doctors to develop self-esteem. As shown in Fig. 2D, one MSM participant responding in English whose photograph featured the back of a person reaching towards the sky and wearing angel wings noted that they wanted freedom to be themselves around a doctor who will tend to their wings rather than clip them. Discussion Results of this study provide insight into the experiences of key populations and PLWH in Malaysian healthcare settings. Like many other countries with persistent HIV disparities, Malaysia is a context characterized by pronounced HIV-related structural stigma. Previous work has found that Malaysian doctors hold prejudice towards, stereotypes about, and intend to discriminate against key populations and PLWH [27–29, 31]. The current study confirms that key populations and PLWH perceive stigma from their doctors. As examples, participants perceive that their doctors view them as not normal, sinful, small, misguided, helpless, incapable, and withered. Several emphasized that they wear a figurative mask in healthcare settings, concealing and protecting aspects of themselves from their doctors. Their photographs exude distance (e.g., zoomed out landscape) and defensiveness (e.g., sharp pushpins). These perceptions extend to HIV prevention and treatment care visits, with key populations perceiving that their doctors view them as shocking during HIV testing visits and PLWH questioning whether their HIV doctors sincerely care about them. These findings are consistent with research from other global settings with strong HIV-related structural stigma (e.g., countries that criminalize same-sex practices), wherein key populations and PLWH report experiencing stigma in healthcare settings [36–38] and disparities in HIV services persist [39, 40]. Participants also reported positive perceptions and experiences, many of which are compatible with the values of patient-centered care [41]. In patient-centered relationships, patients are understood by doctors as people in the context of their social environments and are respected, informed, and listened to [41]. Consistent with these values, participants noted that their doctors see them as strong, brave, and capable. They emphasized that their doctors are non-judgmental and understanding. Some of the participants’ photographs referencing their doctors featured objects and settings that promote health behaviors and wellbeing (e.g., sunscreen, apple). Results of the current study additionally suggest that key populations and PLWH who have had negative experiences in healthcare settings may remain open to and hopeful for positive and patient-centered relationships with their doctors. As examples, participants reported that they want their doctors to know that they are intelligent, capable, bright, cheerful, kind, giving, unique, normal, and valuable. The desire to feel human and valuable stands in sharp contrast to the dehumanization and devaluation inherent to stigma processes [2]. Several participants reported wanting to take off their masks within healthcare settings to share their vulnerability and pain with their doctors. They also desired support and guidance from their doctors, or to be cared for. Participants’ photographs evoked cooperation (e.g., holding hands) and freedom (e.g., wings). Ensuring that patients are empowered to be involved in their care is an additional key component of patient-centered care [41]. There were many similarities and some key differences in responses among participants identifying with various key population groups and responding in Malay versus English. The main themes identified in this study, including perceived stigma from doctors and hopes for positive relationships with doctors, were observed among participants belonging to all key population and language groups. Yet, there were also several key nuances in their responses. For example, although participants belonging to most key population groups described masking in healthcare settings, TGW participants did not. TGW in Malaysia may be less able to conceal their gender identity in healthcare settings if the sex on their ID cards does not match their gender expression. Additionally, several MSM participants’ responses emphasized their individuality and several FSW participants’ responses called attention to harmful social environments. These findings underscore the importance of incorporating an intersectional stigma lens when researching and intervening in stigma experienced by key populations and PLWH [42]. This lens encourages researchers and interventionists to seek to understand and address both shared and unique experiences of stigma among communities affected by HIV. Strengths, Limitations and Future Directions This program of research is focused on addressing HIV-related stigma as a barrier to HIV prevention and treatment in Malaysia. This study therefore included participants from multiple key population groups, including MSM, TGW, FSW, and PWID, as well as PLWH. Previous work suggests that Malaysian doctors endorse stigma towards all of these groups [27]. As noted above, results of the current study suggest that participants of all key population groups and PLWH perceive stigma from and hope for positive interactions with doctors. Yet, there were some key differences in the responses of individuals belonging to different key population groups. More work employing diverse methods and larger sample sizes is needed to understand variations in experiences with, perceptions of, and hopes for doctor interactions between individuals belonging to different key population, language, ethnic, gender, and other groups. Such work can inform efforts to tailor intervention strategies to the unique needs of individual patients. This study employed photovoice, which yielded rich qualitative data and provided some insights that may not have emerged through other methods (e.g., images of landscapes exuded distance or of sharp pushpins that exuded defensiveness). Yet, other methods, such as quantitative surveys collected from larger samples and with representative sampling strategies, may be useful for characterizing the extent to which key populations and PLWH experience negative interactions in healthcare settings or understanding the outcomes of these negative interactions. Moreover, mixed-methods that blend photovoice with quantitative methods may enrich analyses by enabling interpretation of photovoice results alongside quantitative indicators of engagement in healthcare settings and experiences with doctors. This photovoice project was conducted online due to the COVID-19 pandemic. Online, asynchronous methods may have promoted participation among individuals concerned with disclosure given that they did not need to come to a specific location for a study engaging key populations or PLWH. Yet, online, asynchronous methods made some aspects of traditional photovoice projects difficult, such as group discussion and participatory analysis of photographs and captions [33, 43, 44]. Future work should seek ways to incorporate participatory analysis (e.g., via online discussion) to ensure that findings reflect participants’ experiences. Strategies for incorporating participatory analysis in future online, asynchronous photovoice projects are further discussed in Earnshaw et al. [34] This study was conducted in Malaysia, a country with persistent HIV disparities and strong structural stigma towards key populations. Research in such contexts is critical for developing interventions to support some of the most vulnerable key populations and PLWH globally. Yet, generalizability of findings may be limited to similar socio-cultural settings. More work is needed to understand the extent to which the current findings generalize to other settings, including those with strong structural stigma towards key populations and PLWH such as the United States and Europe [13, 14, 16]. Conclusion Key populations and PLWH in Malaysia report that many of their doctors hold negative, stigmatizing perceptions of them. As a result, some wear protective masks in healthcare settings to conceal their lives and struggles. Yet, key populations and PLWH also report wanting to take off their masks, to be vulnerable with their doctors, and hope for the opportunity to receive support and care from their doctors. It is critical to build safe and welcoming healthcare environments where key populations and PLWH can be open about themselves and receive support and care from all doctors. To achieve this, multi-level interventions are needed to address stigma in healthcare settings in Malaysia and similar contexts [1, 45]. Such interventions should implement evidence-based stigma intervention tools at the institutional level (e.g., change policies, redress systems, restructure facilities) and individual healthcare provider level (e.g., enhance knowledge, build skills, facilitate contact) [1]. Photovoice projects hold promise for reducing stigma at the individual level by facilitating extended contact between members of key populations and PLWH with healthcare providers. For example, results of a randomized controlled trial demonstrated that healthcare providers who attended a performance featuring photovoice presentations from individuals in recovery from substance use disorders endorsed fewer stereotypes, less prejudice, and less support for discrimination than providers in the control condition [46]. Future research may seek to identify best practices for leveraging photovoice for stigma reduction. Finally, social change efforts are crucial for eradicating stigma at the structural level and achieving health equity for key populations and PLWH globally. Acknowledgements The authors would like to thank all of the study participants and the Malaysian AIDS Council for their partnership, as well as Jordan Silberman for his partnership on website development and Frederick Altice for his collaboration on this program of research. Authors’ contributions All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Valerie Earnshaw, Jon Cox, Wong Pui Li, Rumana Saifi, Suzan Walters, and Asfarina Binti Amir Hassan. All authors read, revised, and approved the final manuscript. Funding This work was supported by the National Institute of Mental Health (R34MH124390) and National Institute on Drug Abuse (K01DA053159). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Availability of data and material Data may be available from the corresponding author upon request. Declarations 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. The study received institutional review board approval from the University of Delaware (1588354) and University of Malaya Medical Centre. Consent to participate All participants provided informed consent. Consent for publication In addition to consenting to the study, participants consented to have their photographs shared via publication. Conflict of interest The authors declare no conflicts of interest. 1 We recognize that preferred terminology within the HIV field evolves. As recommended by the current UNAIDS Terminology Guidelines [6] and National Institute of Allergy and Infectious Diseases HIV Language Guide [7], we use the term people living with HIV (PLWH) in this manuscript. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Nyblade L Stockton MA Giger K Bond V Ekstrand ML Lean RM Stigma in health facilities: why it matters and how we can change it BMC Med 2019 17 25 1 15 30651111 2. Major B Dovidio JF Link BG Calabrese SK Major B Dovidio JF Link BG Stigma and its implications for health: introduction and overview The Oxford Handbook of Stigma, discrimination, and Health 2018 New York Oxford University Press 3 29 3. Hatzenbuehler ML Link BG Introduction to the special issue on structural stigma and health Soc Sci Med 2014 103 103 1 6 10.1016/j.socscimed.2013.12.017 24445152 4. Hatzenbuehler ML Major B Dovidio JF Link BG Structural stigma and health The Oxford Handbook of Stigma, discrimination, and Health 2018 New York Oxford University Press 105 21 5. Earnshaw VA Chaudoir SR From conceptualizing to measuring HIV stigma: a review of HIV stigma mechanism measures AIDS Behav 2009 13 6 1160 77 10.1007/s10461-009-9593-3 19636699 6. United Nations Programme HIV/AIDS. UNAIDS terminology guidelines 2015. Geneva, Switzerland: UNAIDS. Available at: https://www.unaids.org/sites/default/files/media_asset/2015_terminology_guidelines_en.pdf. Accessed July 26, 2022. 7. National Institute of Allergy and Infectious Diseases. NIAID HIV Language Guide. Available at: https://www.hptn.org/sites/default/files/inline-files/NIAID%20HIV%20Language%20Guide%20-%20March%202020.pdf. Accessed 26 Jul 2022. 8. Nyblade L Stangl A Weiss E Ashburn K Combating HIV stigma in health care settings What works? J Int AIDS Soc 2009 12 1 15 10.1186/1758-2652-12-15 19660113 9. Earnshaw VA Bogart LM Courtney I Zanoni H Bangsberg DR Orrell C Exploring treatment needs and expectations for people living with HIV in South Africa: a qualitative study AIDS Behav 2018 22 2543 52 10.1007/s10461-018-2101-x 29619585 10. Eaton LA Driffin DD Kegler C Smith H Conway-Washington C White D The role of stigma and medical mistrust in the routine health care engagement of black men who have sex with men Am J Public Health 2015 105 2 e75 82 10.2105/AJPH.2014.302322 11. Altman D Aggleton P Williams M Kong T Reddy V Harrad D Men who have sex with men: Stigma and discrimination Lancet 2012 380 439 45 10.1016/S0140-6736(12)60920-9 22819652 12. Beyrer C Pushback The current wave of anti-homosexuality laws and impacts on health PLoS Med 2014 11 6 e1001658 10.1371/journal.pmed.1001658 24959671 13. Kerr J Jackson T Stigma, sexual risks, and the war on drugs: examining drug policy and HIV/AIDS inequities among African Americans using the Drug War HIV/AIDS Inequities Model Int J Drug Policy 2016 37 31 41 10.1016/j.drugpo.2016.07.007 27565526 14. Flath N Tobin K King K Lee A Latkin C Enduring consequences from the war on drugs: how policing practices impact HIV risk among people who inject drugs in Baltimore city Subs Use Misuse 2017 52 1003 10 10.1080/10826084.2016.1268630 15. Lyons CE, Schwartz SR, Murray SM, Shannon K, Diouf D, Mothopeng T, et al. The role of sex work laws and stigmas in increasing HIV risks among sex workers. Nat Commun. 2020;11(1). 16. Poteat T Reisner SL Radix A HIV epidemics among transgender women Curr Opin HIV AIDS 2014 9 2 168 73 10.1097/COH.0000000000000030 24322537 17. Pachankis JE Hatzenbuehler ML Berg RC Mirandola M Marcus U Schmidt AJ Anti-LGBT and anti-immigrant structural stigma: an intersectional analysis of sexual minority men’s HIV risk when migrating to or within Europe J Acquir Immune Defic Syndr 2018 76 4 356 66 10.1097/QAI.0000000000001519 18. Anderson JE Kanters S Lack of sexual minorities’ rights as a barrier to HIV prevention among men who have sex with men and transgender women in Asia: a systematic review LGBT Health 2014 2 16 26 10.1089/lgbt.2014.0024 26790014 19. Mendos LR, Lelis RC, de la Pena EL, Savalev I, Tan D. State-sponsored homophobia 2020: Global legislation update [Internet]. 2020 p. 331. Available at: https://ilga.org/downloads/ILGA_World_State_Sponsored_Homophobia_report_global_legislation_overview_update_December_2020.pdf. Accessed July 27, 2022. 20. Human Rights Watch. “I’m scared to be a woman:” Human rights abuses against transgender people in Malaysia. 2014. Available at: https://www.hrw.org/sites/default/files/reports/malaysia0914_ForUpload.pdf. Accessed July 27, 2022. 21. Anti Death Penalty Asia Network. Reform in limbo: Report on the death penalty in Malaysia. 2021. Available at: https://adpan.org/wp-content/uploads/2021/06/ADPAN-Malaysia-Report-Reform-in-Limbo.pdf. Accessed July 27, 2022. 22. Global Network of Sex Work Projects. How sex work laws are implemented on the ground and their impact on sex workers: Malaysia case study. Available at: https://www.nswp.org/resource/nswp-briefing-papers/case-studies-how-sex-work-laws-are-implemented-the-ground-and-their. Accessed 27 July 2022. 23. Suppiah P Kaur S Transnational sex workers in Malaysia: methodological challenges in data collection J Mod Lang 2018 28 118 36 24. Syariah Criminal Offences Enactment. 1997. Available at: http://www2.esyariah.gov.my/esyariah/mal/portalv1/enakmen2011/Eng_enactment_Ori_lib.nsf/f831ccddd195843f48256fc600141e84/3f7f4673a90e5f2c48257679000ddf32?OpenDocument. Accessed 7 July 2022. 25. Barmania S, Reiss MJ. Islam and health policies related to HIV prevention in Malaysia. Springer International Publishing; 2018. 26. Barmania S Aljunid SM Transgender women in Malaysia, in the context of HIV and Islam: a qualitative study of stakeholders’ perceptions BMC Int Health Hum Rights 2017 17 1 10 10.1186/s12914-017-0138-y 28052765 27. Jin H Earnshaw VA Wickersham JA Kamarulzaman A Desai MM John J An assessment of health-care students’ attitudes toward patients with or at high risk for HIV: implications for education and cultural competency AIDS Care 2014 26 10 1223 8 10.1080/09540121.2014.894616 24625279 28. Earnshaw VA Jin H Wickersham J Kamarulzaman A John J Altice FL Exploring intentions to discriminate against patients living with HIV/AIDS among future healthcare providers in Malaysia Trop Med Int Health 2014 19 6 672 9 10.1111/tmi.12306 24666546 29. Tee YC Earnshaw VA Altice FL Jin H Kamarulzaman A Wickersham JA Evaluating physicians’ intention to discriminate against patients living with HIV in Malaysia AIDS Behav 2019 23 4 1039 47 10.1007/s10461-018-2362-4 30560483 30. UNAIDS Country Factsheets. Malaysia. Available at: https://www.unaids.org/en/regionscountries/countries/malaysia. Accessed July 26, 2022. 31. Vijay A Earnshaw VA Tee YC Pillai V White Hughto JM Clark K Factors associated with medical doctors’ intentions to discriminate against transgender patients in Kuala Lumpur, Malaysia LGBT Health 2018 5 1 61 8 10.1089/lgbt.2017.0092 29227183 32. Kenny DA. Interpersonal perception: a Social Relations Analysis. Guilford Press; 1994. 33. Wang C Burris MA Photovoice: Concept, methodology, and use for participatory needs assessment Health Educ Behav 1997 24 369 87 10.1177/109019819702400309 9158980 34. Earnshaw VA, Cox J, Wong PL, Saifi R, Walters SM, Azwa I, et al. Acceptability and feasibility of online, asynchronous photovoice with key populations and people living with HIV. AIDS Behav. In press. 35. Beebe J Rapid qualitative Inquiry: a Field Guide to Team-Based Assessment 2014 2 Lanham Rowman & Littlefield 36. Schwartz SR Nowak RG Orazulike I Keshinro B Ake J Kennedy S The immediate effect of the same-sex marriage prohibition act on stigma, discrimination, and engagement on HIV prevention and treatment services in men who have sex with men in Nigeria: analysis of prospective data from the TRUST cohort Lancet HIV 2015 2 7 e299 306 10.1016/S2352-3018(15)00078-8 26125047 37. Fay H Baral SD Trapence G Motimedi F Umar E Iipinge S Stigma, health care access, and HIV knowledge among men who have sex with men in Malawi, Namibia, and Botswana AIDS Behav 2011 15 6 1088 97 10.1007/s10461-010-9861-2 21153432 38. Wiginton JM Murray SM Poku O Augustinavicius J Jackman KMP Kane J Disclosure of same-sex practices and experiences of healthcare stigma among cisgender men who have sex with men in five sub-saharan african countries BMC Public Health 2021 21 1 2206 10.1186/s12889-021-12151-3 34861835 39. Arreola S Santos GM Beck J Sundararaj M Wilson PA Hebert P Sexual stigma, criminalization, investment, and access to HIV services among men who have sex with men worldwide AIDS Behav 2015 19 2 227 34 10.1007/s10461-014-0869-x 25086670 40. Hagopian A Rao D Katz A Sanford S Barnhart S Anti-homosexual legislation and HIV-related stigma in african nations: what has been the role of PEPFAR? Glob Health Action 2017 10 1 1306391 10.1080/16549716.2017.1306391 28580879 41. Epstein RM Street RL The values and value of patient-centered care Ann Fam Med 2011 9 2 100 3 10.1370/afm.1239 21403134 42. Berger MT Workable sisterhood the political journey of stigmatized women with HIV/AIDS 2004 Princeton Princeton University Press 43. Teti M Koegler E Conserve DF Handler L Bedford M A scoping review of photovoice research among people with HIV J Assoc Nurses AIDS Care 2018 29 504 27 10.1016/j.jana.2018.02.010 29576252 44. Golden T Reframing photovoice: building on the method to develop more equitable and responsive research practices Qual Health Res 2020 30 960 72 10.1177/1049732320905564 32081060 45. Rao D Elshafei A Nguyen M Hatzenbuehler ML Frey S Go VF A systematic review of multi-level stigma interventions: state of the science and future directions BMC Med 2019 17 41 1 11 30651111 46. Flanagan EH Buck T Gamble A Hunter C Sewell I Davidson L “Recovery Speaks”: a photovoice intervention to reduce stigma among primary care providers Psychiatr Serv 2016 67 5 566 9 10.1176/appi.ps.201500049 26766754	36472685	PMC9734400	NO-CC CODE	2022-12-14 23:28:28	no	AIDS Behav. 2022 Dec 6;:1-10	utf-8	AIDS Behav	2,022	10.1007/s10461-022-03942-9	oa_other
==== Front J Relig Health J Relig Health Journal of Religion and Health 0022-4197 1573-6571 Springer US New York 36469230 1695 10.1007/s10943-022-01695-2 Original Paper Spiritual Well-Being and Care Burden in Caregivers of Patients with Breast Cancer in Turkey http://orcid.org/0000-0002-4848-0993 Türkben Polat Hilal [email protected] 1 http://orcid.org/0000-0002-7278-2094 Kiyak Sibel [email protected] 2 1 grid.411124.3 0000 0004 1769 6008 Department of Fundamentals of Nursing, Seydişehir Kamil Akkanat Faculty of Health Sciences, Necmettin Erbakan University, Konya, Turkey 2 grid.411124.3 0000 0004 1769 6008 Department of Obstetrics and Gynecology Nursing, Seydişehir Kamil Akkanat Faculty of Health Sciences, Necmettin Erbakan University, Konya, Turkey 5 12 2022 114 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. This research was carried out to identify the relationship between the spiritual well-being and caregiver burden in caregivers of patients with breast cancer. The study was conducted with family caregivers of patients with breast cancer who presented to the oncology clinic of a university hospital for treatment. The study sample included a total of 138 family caregivers who met the criteria for participation and agreed to participate in the study. The data were collected using a participant information form, caregiver burden scale and three-dimensional spiritual well-being scale. The caregivers have a moderate level of caregiver burden and their spiritual well-being was quite high. The caregiver burden of female caregivers was found to be significantly higher than that of male caregivers (p = 0.040). There is a negatively significant relationship between caregiver burden and spiritual well-being (p = 0.000, r = − 0.357). The caregiver burden is significantly higher among the 24-h caregivers compared to that among the 3-h and 4–6-h caregivers (p = 0.003). The spiritual well-being of the caregivers who provide care between 3 h and 4–6 h a day was significantly higher than that of those who provide 24-h care (p = 0.001). Increasing spiritual well-being may help to reduce caregiver burden in caregivers of those with breast cancer. Keywords Breast cancer Spiritual well-being Caregiver burden Family caregiver ==== Body pmcIntroduction Breast cancer in women is one of the most common cancers all over the world. With 2,261,419 new cases in 2020, the incidence rate of breast cancer is 11.7% (World Health Organization, 2020). Breast cancer can be treated using methods such as surgery, chemotherapy, radiotherapy and immunotherapy. The side effects of these treatments increase the care needs of patients (Baider, 2014). Caring for a patient with breast cancer can often strain family caregivers psychologically, socially and financially, leading to negative consequences called caregiver burden (CB). CB is defined as ‘‘a multidimensional response to the physical, psychological, social and financial stress factors associated with the caregiving experience’’ (Zarit et al., 1980). Breast cancer patients are now treated on an outpatient basis without the need for hospitalisation. Care needs are often provided by voluntary family caregivers (Schulz et al, 2020). Caregivers help patients with their daily work, provide medication and manage symptoms (Sun et al, 2019). In this process, caregivers' CB and needs are often unnoticed, and the treatment process is often shaped by the needs of the patient (care recipient) (Jite et al., 2021). Caregivers often receive neither adequate education nor social support (Adelman et al., 2014). Studies on CB have reported a moderate CB (Jite et al, 2021; Lee et al., 2018; Vahidi et al., 2016; Yusuf et al., 2011; Zuo et al., 2020). However, some studies report advanced CB (Garcia et al., 2020). The CB of those who provide care for patients with cancer is affected by many factors. One of them is spiritual well-being (SWB). SWB is an expression of spirituality and a measure of a person's spiritual health (Spatuzzi et al., 2019). Spiritual well-being is a feeling of one's contentment defined as “the affirmation of life in a relationship with God, self, community and environment that nurtures and celebrates wholeness.” (National Interfaith Coalition on Aging, 1975). Various studies show that people who are spiritually strong feel more positive about their roles and communicate better when caring for their patients (Newberry et al., 2013; Sankhe et al., 2017; Spazutti et al., 2019; Tan et al., 2015; Vigna et al., 2020). The International Council of Nurses has included spirituality in nursing codes. ‘Spiritual distress’ and ‘risk for spiritual distress’ diagnoses are included in the nursing diagnosis list of the North American Nursing Diagnosis Association (Wilkinson & Barcus, 2016). Nurses are responsible for identifying CBs that may affect the health of caregivers and SWBs that are known to be associated with CBs as well as for providing care. SWB can be used as a cost-effective and effective intervention to reduce the CB of caregivers of patients with breast cancer. In line with this information, our research was carried out to define the relationship between the SWB of caregivers of patients with breast cancer and their CB. Study Questions What is the level of CB of caregivers of patients with breast cancer? What is the level of SWB of caregivers of patients with breast cancer? Is there a relationship between the SWB of caregivers of patients with breast cancer and their CB? Method Study Design It is a descriptive and correlational study. Study Setting and Sample The study was conducted in the inpatient unit of a university hospital oncology clinic and in the outpatient chemotherapy unit. The study conducted with family caregivers of patients with breast cancer who applied to a university hospital oncology clinic for treatment. A total of 138 family caregivers who met the criteria for participation in the study and agreed to participate in the study constituted the study sample. Study Power Gpower 3.1.9.2 was used in calculating the sample size (Faul et al., 2009). The mean care load score of caregivers of cancer patients in the study by Bahrami et al. (2014) was used for calculating the sample size. With a 95% confidence (1 − α), 95% test power (1 − β), d = 0.333 effect size and two-way t-test, the sample size for this study was determined as 133 people. The study was completed with 138 people. The posthoc power of the research is 0.957. Inclusion Criteria for Participants in the Study Being ≥ 18 years of age, being a family member responsible for the care of a patient diagnosed with breast cancer, being referred to by the caregivers themselves or the patient as a primary care provider, not having any health problems, not having a cancer diagnosis or neurological cognitive disorder, being able to communicate in Turkish, and being a caregiver for ≥ 3 months. Exclusion Criteria for Participants Wanting to withdraw from the research for any reason while the research is in progress and being < 18 years of age. Data Collection Technique and Tools Data were collected using the participant information form, caregiver burden scale and spiritual well-being scale. Participant Information Form It is a form prepared by the researchers, consisting of a total of 16 questions related to the age, socio-demographic status and employment status of the participants (Spatuzzi et al., 2019; Vigna et al., 2020). Caregiver Burden Scale (CBS) It is a scale used to evaluate the CB of those who care for a person or elderly in need of care (Zarit et al., 1980). ‘‘The Turkish validity and reliability of the scale was assessed by İnci and Erdem. The reliability coefficient of the scale is between 0.87 and 0.99. The scale consists of 22 statements. It is a 5-point Likert type scale (never, rarely, sometimes, frequently, and nearly always). A minimum score of 0 and a maximum score of 88 points can be obtained on the scale. In this study, scale scores were evaluated as no/little burden (0–20 points), moderate burden (21–40 points), severe burden (41–60 points) and very severe burden (61–88 points)’’ (İnci & Erdem, 2010). The Cronbach’s α value of this study is 0.92. Three-Factor Spiritual Well-Being Scale (SWBS) The validity and reliability of this scale was assessed by Ekşi and Kardaş (Ekşi & Kardaş, 2017; Kardaş, 2019). Three-factor Spiritual Well-being Scale in Turkish was given in Table 1 (Ekşi and Kardaş 2017; Kardaş 2019), spiritual and social dimensions of a person's life and to determine the quality of this process. It is a 5-point Likert type scale. The transcendency subscale consists of items 1, 4, 5, 8, 9, 12, 13, 16, 17, 20, 21, 24, 25, 27 and 29; the harmony with nature subscale consists of items 2, 6, 10, 14, 18, 22 and 28 and the anomie subscale consists of items 3, 7, 11, 15, 19, 23 and 26. When a total score is desired, reverse scoring is applied for the items in the anomie subscale. The Cronbach's α value of the scale was determined as 0.763’’ (Ekşi & Kardaş, 2017; Kardaş, 2019). The Cronbach’ alpha value of this study is 0.81. Table 1 Three-factor spiritual well-being scale Bana hiç uygun değil Bana uygun değil Bana biraz uygun Bana oldukça uygun Bana tamamen uygun 1 İlahi bir güce bağlı olmak bana güven verir (1) (2) (3) (4) (5) 2 Doğaya saygı duyulması gerektiğini düşünürüm (1) (2) (3) (4) (5) 3 Hayata dair bir hoşnutsuzluk duygusu hissederim (1) (2) (3) (4) (5) 4 Bir problemle karşılaştığımda Allah’ın yardımını hissederim (1) (2) (3) (4) (5) 5 Allah’ın gizli ve açık tüm duygu ve düşüncelerimi bildiğine inanırım (1) (2) (3) (4) (5) 6 Bütün canlıların saygıyı hak ettiğini düşünürüm (1) (2) (3) (4) (5) 7 Hayatımda büyük bir boşluk var (1) (2) (3) (4) (5) 8 Günlük hayatta Allah’ın kudretine şahit olurum (1) (2) (3) (4) (5) 9 Allah’ın beni sevdiğine ve önemsediğine inanırım (1) (2) (3) (4) (5) 10 Yeryüzündeki tüm canlılara iyi davranırım (1) (2) (3) (4) (5) 11 Hayattan zevk almam (1) (2) (3) (4) (5) 12 Hayatımın her anında Allah’ın varlığını hissederim (1) (2) (3) (4) (5) 13 Daha güçlü bir varlığa sığınma duygusu beni rahatlatır (1) (2) (3) (4) (5) 14 Kendimi doğanın bir parçası olarak görürüm (1) (2) (3) (4) (5) 15 Hayatımın amacını halen bulabilmiş değilim (1) (2) (3) (4) (5) 16 Yaşadığım her olayda bir hayır olduğuna inanırım (1) (2) (3) (4) (5) 17 İnancım, nasıl bir hayat süreceğime dair bana yol gösterir (1) (2) (3) (4) (5) 18 Yeryüzündeki bütün canlıların hakları benim için önemlidir (1) (2) (3) (4) (5) 19 Sorunlarımı çözmeye nereden başlayacağımı bilemem (1) (2) (3) (4) (5) 20 Yalnız kaldığımda Allah’ı ve yarattıklarını düşünürüm (tefekkür ederim) (1) (2) (3) (4) (5) 21 İnanç ve değerlerim, zorluklar karşısında dayanabilme gücümü arttırır (1) (2) (3) (4) (5) 22 Doğayla uyum içinde yaşarım (1) (2) (3) (4) (5) 23 Zorluklar yaşadığımda bunalmış hissederim (1) (2) (3) (4) (5) 24 İnancım, yaşadığım sıkıntılarda dahi olumlu tarafların olabileceğini görmemi sağlar (1) (2) (3) (4) (5) 25 Hayatta hiçbir şey sebepsiz değildir (1) (2) (3) (4) (5) 26 Hayatın beni mutsuz eden olaylardan ibaret olduğunu düşünürüm (1) (2) (3) (4) (5) 27 Her şeyin elimde olmadığını bilmek üzüldüğüm olaylar karşısında bir teselli kaynağıdır (1) (2) (3) (4) (5) 28 Yeryüzündeki her doğal varlığın eşsiz olduğuna inanırım (1) (2) (3) (4) (5) 29 Dünya hayatının geçici olduğuna inanmak beni hırslarımdan arındırır (1) (2) (3) (4) (5) Data Collection The data of the study were collected between August 2021 and October 2021. The data were collected through face-to-face interviews with caregivers of patients with breast cancer. The data collection took approximately 10–15 min. Due to the ongoing coronavirus disease-19 pandemic, the interviewers and interviewees wore masks and maintained social distance during the interview. Ethical Principles Ethics committee approval was obtained before starting the research (App. Date/ No: 07.07.2021/ 2021–12-63). Institutional permission was obtained from the hospital (E-14567952–900-71,409). Oral and written consents of the participants were obtained. The research was carried out according to the principles of the Declaration of Helsinki. Data Analysis Data analysis was performed using Statistical Package for Social Science 22.0 package program. Descriptive data were evaluated using percentile, mean, standard deviation, minimum and maximum values. The conformity of the data to the normal distribution according to the groups was determined by the Kolmogorov–Smirnov/Shapiro–Wilk test. The one-way analysis of variance, independent samples t-test and post-hoc Tukey’s test were used for the analysis of the data. The relationship between the total scale scores was determined using Pearson’s correlation analysis. P < 0.05 was accepted as the level of statistical significance. Results The descriptive characteristics of the patients are given in Table 2.Table 2 The descriptive characteristics of the patients (N = 138) N % Marital status Married 125 90.6 Single 13 9.4 Education İlliterate 15 10.9 Literate 7 5.1 Primary school 82 59.4 High school 17 12.3 University 17 12.3 Disease stage 1 13 9.4 2 16 11.6 3 39 28.3 4 70 50.7 X¯ ± SD Min Max The average age/year 53.75 ± 11.91 28 87 Disease duration/month 28.49 ± 36.15 3 220 The total SWBS scores and CBS scores are given in Table 3 based on the descriptive characteristics of the caregivers. The CBS of female caregivers were significantly higher than those of male caregivers (p = 0.040). A statistically significant difference was observed between the duration of care and SWB (p = 0.001) and CB (p = 0.003) scales mean scores. According to the post-hoc Tukey’s test, caregivers who provide 24-h care had a significantly higher CBS score than the score of those who provide care for 3 h (p = 0.044) and 4–6 h (p = 0.012). According to the post-hoc Tukey’s test results, the total SWB scores of caregivers who provide care between 3 hours (p = 0.014) and 4–6 hours (p = 0.006) a day were significantly higher than those who provide care for 24 hours a day. A statistically significant difference was observed between the SWBS scores of those who received education about the disease and those who did not (p = 0.007). The SWBS scores of those who did not receive education were significantly higher (Table 3).Table 3 The total SWBS and CBS scores based on the descriptive characteristics of the caregivers CBS SWBS n % X¯ ± SD Test and p value X¯ ± SD Test and p value Sex Female 50 36.2 36.12 ± 16.83 t = − 2.075 119.33 ± 9.37 t = 1.194 Male 88 63.8 29.93 ± 16.85 p = 0.040 121.39 ± 9.99 p = 0.234 Marital status Married 117 84.8 32.17 ± 16.69 t = -0.005 121.11 ± 9.18 t = 1.323 Single 21 15.2 32.19 ± 19.36 p = 0.996 118.05 ± 12.62 p = 0.188 Education Primary school 60 43.5 32.70 ± 16.29 F = 0.259 120.82 ± 9.12 F = 0.584 High school 37 26.8 33.09 ± 21.57 p = 0.772 119.26 ± 11.16 p = 0.559 University 41 29.7 30.58 ± 13.45 121.64 ± 9.53 Income Good 22 15.9 28.74 ± 14.95 F = 0.618 123.52 ± 11.33 F = 1.48 Middle 100 72.5 32.56 ± 16.70 p = 0.541 120.40 ± 9.38 p = 0.231 Bad 16 11.6 34.51 ± 21.79 118.21 ± 9.79 Duration of care hour/a day 24 houra 81 58.7 36.38 ± 17.92 F = 4.89 117.90 ± 9.58 F = 6.18 3 hourb 18 13.0 25.09 ± 15.63 p = 0.003 125.32 ± 6.82 p = 0.001 4–6 hourc 18 13.0 23.10 ± 10.87 a > b, c 125.99 ± 10.30 b, c > a 7–12 hourd 21 15.2 29.81 ± 14.14 122.64 ± 8.93 Relationship Spouse 72 52.2 31.40 ± 16.88 F = 0.256 121.20 ± 9.69 F = 0.265 Children 50 36.2 33.56 ± 17.03 p = 0.775 120.20 ± 10.43 p = 0.768 Other Relatives 16 11.6 31.32 ± 18.59 119.53 ± 8.52 Education about the disease Yes 65 47.1 33.65 ± 17.23 t = 0.961 118.29 ± 9.87 t = − 2.726 No 73 52.9 30.86 ± 16.88 p = 0.338 122.74 ± 9.30 p = 0.007 Average age of caregiver X¯ ± SD Min Max 46.09 ± 13.00 18 72 Abbreviation: CB Caregiver burden, SWB Spiritual well-being F one-way analysis of variance t: independent samples t-test, Statistically significant values (p < .05) are shown in bold Caregivers' total CB scale, total SWB and subscale scores are given in Table 4. The caregivers had a moderate CB, whereas their SWB was quite high.Table 4 Caregivers' total CBS, total SWB and subscale scores X¯ ± SD Min–Max SWBS total 120.64 ± 9.79 89.07–142.03 Transcendence 67.20 ± 4.19 41.13–70.33 Harmony with nature 28.71 ± 2.49 13.71–30.71 Anomie 17.41 ± 6.10 6.43–32.71 CBS 32.17 ± 17.04 0–70.18 Abbreviation: CBS Caregiver burden scale, SWB Spiritual well-being scale Table 5 demonstrates the negatively significant relationship between CB and SWB (p = 0.000, r = -0.357). There was no significant relationship of disease duration, disease stage, caregiver’s age and SWB with CB (Table 5).Table 5 The relationship between SWBS and CBS total scores and some variables SWBS CBS r p R p CBS − 0.357 0.000 – – Disease duration 0.034 0.688 0.044 0.606 Disease stage − 0.158 0.065 0.160 0.060 Caregiver age 0.007 0.936 − 0.032 0.709 Statistically significant values (p < .05) are shown in bold Abbreviation: CBS Caregiver burden scale, SWBS Spiritual well-being scale, r: Pearson’s correlation Participants reported no/little burden (27%), moderate burden (36%), severe burden (30%) and very severe burden (7%). Discussion This research aimed to identify the CB and SWB of family caregivers of patients with breast cancer and the relationship between them. As breast cancer is the most common cancer, the research was conducted with the caregivers of these patients. A total of 50.7% of the patients of the caregivers participating in this study had stage 4 breast cancer. Due to the increased care needs of advanced cancer patients as a result of treatments such as chemotherapy and radiotherapy, many family caregivers have a high CB (Roij et al., 2021). In this study, the CB of female caregivers was significantly higher than that of male caregivers. According to the literature, women are at a higher risk of having a high CB than men (Han et al., 2013; Jansen et al., 2018). This may be due to the fact that women, whose responsibilities are already excessive in daily life, are also the caregivers of an individual with cancer. As the responsibilities of a caregiver increase, the care process becomes tiring, dependent and long. According to our research results, the CB of those who provide 24-h care is significantly higher than those who care for 3 h and 4–6 h. Similarly, caregivers who provide care for more than 6 h a day have the highest CB (Zuo et al., 2020). Providing intensive and uninterrupted care leads to an increase in the CB (Vigna et al., 2020). For this reason, caregivers should be supported socially, psychologically and economically as the duration of care increases. Caregivers should be supported with practical support, better case management and greater recognition of the role of caregivers (Heath et al., 2018). In this study, the SWB of those who did not receive education about the disease was significantly higher than that of those who received education. In Turkey, patients diagnosed with cancer and their caregiver family members are informed about the stage of the disease, its treatment, side-effects of the treatment and emergency situations. The educational contents are prepared in a standard manner by the healthcare professionals of the hospital education unit. However, each individual's educational needs are different. The education should be tailored to the educational needs of the patients and caregivers. The caregivers who participated in this study had a moderate CB. Studies on CB reported moderate CB (Jite et al., 2021; Zuo et al., 2020) as well as severe CB (Garcia et al., 2020). Excessive CB may cause the care receiver to receive inappropriate and unsafe care (Lafferty et al., 2016). Increased CB is associated with depression and anxiety in patients (Dionne et al., 2016). At the same time, CB negatively affects the patient's quality of life (An et al., 2019). The needs of caregivers may change at different times during the care process (Treanor, 2020). Basically, caregivers with a high level of CB are at risk for a continued increase in CB in the following years (Jansen et al., 2021). In the present study, 36% of the participants had moderate care burden. Similarly Asadi et al. reported 39.5% moderate burden (Asadi et al., 2019). SWB is a protective factor against psychological and physiological diseases. (Delgado-Guay et al., 2014). SWB is an important factor in coping with the difficulties that caregivers experience during the care of patients with cancer. SWB is a concept that includes both religion and spirituality (Akkuş et al., 2022). More religion or spirituality was associated with lower depressive symptoms and less personality disorder (Power & McKinney, 2014), less post-traumatic stress and perceived stress (Arévalo et al., 2008) also this may affect caregiver burden. In this study, the SWB of the caregivers was found to be quite high. SWB is an important factor that can affect CB and the physical health of caregivers (Spatuzzi et al., 2019). Individuals with high spirituality feel less CB (Vigna et al., 2020). Some studies show that spirituality and religious beliefs reduce distress in caregivers (Hosseini et al., 2016; Koenig, 2015) and also this may be used as coping strategies for stressful situations (Torabi Chafjiri et al., 2017). The most important result of this research is that as the SWB of caregivers increases, their CB decreases. Similarly, there are studies reporting that SWB is negatively related to CB, and that spirituality can be used as an effective and low-cost intervention to reduce the CB (Rafati et al., 2020). Caregivers with low levels of spirituality are at higher risk of CB, anxiety and stress (Newberry et al., 2013). In this study, the total SWB scores of the caregivers who provide care between 3 h and 4–6 h a day are significantly higher than the scores of those who care for 24 h a day. This finding also supports the negative relationship between SWB and CB. Due to the increase in the elderly population and in the number of chronic and fatal diseases, the number of individuals who need care and the number of caregivers increase every year. The CB of caregivers of women with breast cancer, one of the most common types of cancer, also increases in this process. Caregivers should be supported in terms of treatment, care and financial issues to eliminate the negative effects of CB on both the caregiver and the patient. Limitations The first limitation of the study is that the data were collected only from a single hospital. Therefore, the results of this study should be cautiously generalized to other settings. It is recommended to conduct longitudinal design studies. Many factors affecting the concept of spiritual well-being, such as psychosocial characteristics of caregivers, perceptions of social support, and attachment patterns, were not considered. Conclusion and Recommendations According to our results, caregivers have a moderate level of CB and their SWB is quite high. There is a negative relationship between CB and SWB. Nurses, whose main role is to provide care, have important duties in reducing the CB of family caregivers. Follow-up of patients with breast cancer who are cared for at home should be provided by primary health care services. Family physicians and nurses should be in frequent contact with caregivers to reduce the CB and develop coping strategies. Caregivers of patients with breast cancer should be provided effective health counselling, psychosocial care and moral support following the diagnosis of cancer. Since female caregivers have a high care burden, they should be supported more and protected from burnout. Caregivers with longer daily care periods have a higher CB and lower SWB. It is recommended that caregivers with long daily caregiving periods are supported by social services in their care activities. Spirituality can be an important factor influencing the role of a caregiver. As spirituality has different meanings and roles in different cultures and religious beliefs, its relationship with CB may vary. Therefore, conducting more studies in different countries is recommended. Funding The authors received no financial support for the research, authorship and/or publication of this article. This research received no specific grant from any funding agency in the public, commercial, or not for-profit sectors. Data Availability The data that support the findings of this study are available on request from the corresponding author. Declarations Conflict of interest The author declares they have no potential conflict of interest. The authors report no actual or potential conflicts of interest. Ethical Approval All procedures in the study performed in accordance with the ethical standards of the 1964 Helsinki Declaration and its later amendments. Informed Consent Was obtained from all participants who were included in the study. This study was verbal presented in 10th International Medicine and Health Sciences Researchers Congress, 27–28 August 2022 Ankara /Turkey. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Adelman RD Tmanova LL Delgado D Dion S Lachs MS Caregiver burden: A clinical review JAMA 2014 311 10 1052 1060 10.1001/jama.2014.304 24618967 Akkuş Y Karacan Y Ünlü K Deniz M Parlak A The effect of anxiety and spiritual well-being on the care burden of caregivers of cancer patients during the COVID-19 pandemic Supportive Care in Cancer : Official Journal of the Multinational Association of Supportive Care in Cancer 2022 30 2 1863 1872 10.1007/s00520-021-06611-0 34613473 An Y Fu G Yuan G Quality of life in patients with breast cancer: the influence of family caregiver's burden and the mediation of patient's anxiety and depression The Journal of Nervous and Mental Disease 2019 207 11 921 926 10.1097/NMD.0000000000001040 31517713 Arévalo S Prado G Amaro H Spirituality, sense of coherence, and coping responses in women receiving treatment for alcohol and drug addiction Evaluation and Program Planning 2008 31 1 113 123 10.1016/j.evalprogplan.2007.05.009 17825910 Asadi P Fereidooni-Moghadam M Dashtbozorgi B Masoudi R Relationship between care burden and religious beliefs among family caregivers of mentally ill patients Journal of Religion and Health 2019 58 4 1125 1134 10.1007/s10943-018-0660-9 29936675 Bahrami M Farzi S The effect of a supportive educational program based on COPE model on caring burden and quality of life in family caregivers of women with breast cancer Iranian Journal of Nursing and Midwifery Research 2014 19 2 119 126 24834079 Baider L Surbone A Universality of aging: family caregivers for elderly cancer patients Frontiers in Psychology 2014 5 744 10.3389/fpsyg.2014.00744 25076927 Delgado-Guay MO Spirituality and religiosity in supportive and palliative care Current Opinion in Supportive and Palliative Care 2014 8 3 308 313 10.1097/SPC.0000000000000079 25029394 Dionne-Odom JN Hull JG Martin MY Lyons KD Prescott AT Tosteson T Li Z Akyar I Raju D Bakitas MA Associations between advanced cancer patients' survival and family caregiver presence and burden Cancer Medicine 2016 5 5 853 862 10.1002/cam4.653 26860217 Ekşi H Kardaş S Spiritual well-being: scale development and validation Spiritual Psychology and Counseling 2017 2 73 88 10.12738/spc.2017.1.0022 Faul F Erdfelder E Buchner A Lang AG Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses Behavior Research Methods 2009 41 4 1149 1160 10.3758/BRM.41.4.1149 19897823 García-Torres F Jabłoński MJ Gómez Solís Á Jaén-Moreno MJ Gálvez-Lara M Moriana JA Moreno-Díaz MJ Aranda E Caregiver burden domains and their relationship with anxiety and depression in the first six months of cancer diagnosis International Journal of Environmental Research and Public Health 2020 17 11 4101 10.3390/ijerph17114101 32526823 Han Y Yuan J Luo Z Zhao J Wu J Liu R Lopez V Determinants of hopelessness and depression among Chinese hospitalized esophageal cancer patients and their family caregivers Psycho-Oncology 2013 22 11 2529 2536 10.1002/pon.3315 23703786 Heath A Carey LB Chong S Helping carers care: an exploratory study of factors impacting informal family carers and their use of aged care services Journal of Religion and Health 2018 57 3 1146 1167 10.1007/s10943-018-0593-3 29524071 Hosseini S Chaurasia A Cooke M Oremus M Effect of religious involvement on cognition from a life-course perspective: protocol for a systematic review and meta-analysis British Medical Journal Open 2016 6 9 e011301 10.1136/bmjopen-2016-011301 İnci F Erdem M Bakım verme yükü ölçeği’nin Türkçe'ye uyarlanması geçerlilik ve güvenilirliği Anadolu Hemşirelik Ve Sağlık Bilimleri Dergisi 2008 11 4 85 95 Jansen L Dauphin S De Burghgraeve T Schoenmakers B Buntinx F van den Akker M Caregiver burden: an increasing problem related to an aging cancer population Journal of Health Psychology 2021 26 11 1833 1849 10.1177/1359105319893019 31814462 Jansen L Dauphin S van den Akker M De Burghgraeve T Schoenmakers B Buntinx F Prevalence and predictors of psychosocial problems in informal caregivers of older cancer survivors-a systematic review: still major gaps in current research European Journal of Cancer Care 2018 27 6 e12899 10.1111/ecc.12899 30168877 Jite IE Adetunji AA Folasire AM Akinyemi JO Bello S Caregiver burden and associated factors amongst carers of women with advanced breast cancer attending a radiation oncology clinic in Nigeria African Journal of Primary Health Care & Family Medicine 2021 13 1 e1 e8 10.4102/phcfm.v13i1.2812 Kardaş S Erratum: correcting the name of the spiritual well-being scale as the three-factor spiritual well-being scale Spiritual Psychology and Counseling 2019 4 85 85 10.12738/spc.2019.4.1.0068 Koenig HG Religion, spirituality, and health: a review and update Advances in Mind-Body Medicine 2015 29 3 19 26 26026153 Lafferty A Fealy G Downes C Drennan J The prevalence of potentially abusive behaviours in family caregiving: findings from a national survey of family carers of older people Age and Ageing 2016 45 5 703 707 10.1093/ageing/afw085 27225014 Lee YH Liao YC Shun SC Lin KC Liao WY Chang PH Jhang SY Yu CJ Yang PC Hsieh PY Lai YH Trajectories of caregiver burden and related factors in family caregivers of patients with lung cancer Psycho-Oncology 2018 27 6 1493 1500 10.1002/pon.4678 29476636 National Interfaith Coalition on Aging Spiritual wellbeing: a definition 1975 National Interfaith Coalition on Aging Newberry AG Choi CW Donovan HS Schulz R Bender C Given B Sherwood P Exploring spirituality in family caregivers of patients with primary malignant brain tumors across the disease trajectory Oncology Nursing Forum 2013 40 3 E119 E125 10.1188/13.ONF.E119-E125 23615145 Power L McKinney C The effects of religiosity on psychopathology in emerging adults: intrinsic versus extrinsic religiosity Journal of Religion and Health 2014 53 5 1529 1538 10.1007/s10943-013-9744-8 23771804 Rafati F Mashayekhi F Dastyar N Caregiver burden and spiritual well-being in caregivers of hemodialysis patients Journal of Religion and Health 2020 59 6 3084 3096 10.1007/s10943-019-00939-y 31650381 Sankhe A Dalal K Agarwal V Sarve P Spiritual care therapy on quality of life in cancer patients and their caregivers: a prospective non-randomized Single-cohort study Journal of Religion and Health 2017 56 2 725 731 10.1007/s10943-016-0324-6 27853982 Schulz R Beach SR Czaja SJ Martire LM Monin JK Family caregiving for older adults Annual Review of Psychology 2020 71 635 659 10.1146/annurev-psych-010419-050754 Spatuzzi R Giulietti MV Ricciuti M Merico F Fabbietti P Raucci L Bilancia D Cormio C Vespa A Exploring the associations between spiritual well-being, burden, and quality of life in family caregivers of cancer patients Palliative & Supportive Care 2019 17 3 294 299 10.1017/S1478951518000160 29749317 Sun V Raz DJ Kim JY Caring for the informal cancer caregiver Current Opinion in Supportive and Palliative Care 2019 13 3 238 242 10.1097/SPC.0000000000000438 31157656 Tan JY Lim HA Kuek NM Kua EH Mahendran R Caring for the caregiver while caring for the patient: exploring the dyadic relationship between patient spirituality and caregiver quality of life Supportive Care in Cancer : Official Journal of the Multinational Association of Supportive Care in Cancer 2015 23 12 3403 3406 10.1007/s00520-015-2920-5 26335405 TorabiChafjiri R Navabi N Shamsalinia A Ghaffari F The relationship between the spiritual attitude of the family caregivers of older patients with stroke and their burden Clinical Interventions in Aging 2017 12 453 458 10.2147/CIA.S121285 28280318 Treanor CJ Psychosocial support interventions for cancer caregivers: reducing caregiver burden Current Opinion in Supportive and Palliative Care 2020 14 3 247 262 10.1097/SPC.0000000000000508 32769618 Vahidi M Mahdavi N Asghari E Ebrahimi H EivaziZiaei J Hosseinzadeh M NamdarAreshtanab H Kermani IA Other side of breast cancer: factors associated with caregiver burden Asian Nursing Research 2016 10 3 201 206 10.1016/j.anr.2016.06.002 27692249 Van Roij J Brom L Sommeijer D van de Poll-Franse L Raijmakers N Self-care, resilience, and caregiver burden in relatives of patients with advanced cancer: results from the eQuiPe study Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer 2021 29 12 7975 7984 10.1007/s00520-021-06365-9 34215933 Vigna PM de Castro I Fumis R Spirituality alleviates the burden on family members caring for patients receiving palliative care exclusively BMC Palliative Care 2020 19 1 77 10.1186/s12904-020-00585-2 32493301 Wilkinson JM Barcus L Nursing Diagnosis Handbook (Translate: Kapucu S., Akyar I., Korkmaz F) 2016 11 Pearson World Health Organization. (2020, December). The global cancer observatory reports, International agency for research on cancer number of new cases in 2020, both sexes all ages. Retrieved from Nov 1, 2021 https://gco.iarc.fr/today/data/factsheets/cancers/20-Breast-fact-sheet.pdf. Yusuf AJ Adamu A Nuhu FT Caregiver burden among poor caregivers of patients with cancer in an urban African setting Psycho-Oncology 2011 20 8 902 905 10.1002/pon.1814 20680930 Zarit SH Reever KE Bach-Peterson J Relatives of the impaired elderly: correlates of feelings of burden The Gerontologist 1980 20 649 655 10.1093/geront/20.6.649 7203086 Zuo Y Luo BR Peng WT Liu XR He YL Zhang JJ Informal caregiver burden and influencing factors in gynaecological oncology patients hospitalized for chemotherapy: a cross-sectional study The Journal of International Medical Research 2020 48 11 300060520974927 10.1177/0300060520974927 33259238	36469230	PMC9734401	NO-CC CODE	2022-12-14 23:28:28	no	J Relig Health. 2022 Dec 5;:1-14	utf-8	J Relig Health	2,022	10.1007/s10943-022-01695-2	oa_other
==== Front Rev Manag Sci Review of Managerial Science 1863-6683 1863-6691 Springer Berlin Heidelberg Berlin/Heidelberg 605 10.1007/s11846-022-00605-w Original Paper Impact of board of directors on insolvency risk: which role of the corruption control? Evidence from OECD banks Sallemi Marwa 1 Ben Hamad Salah 2 http://orcid.org/0000-0003-1032-3965 Ould Daoud Ellili Nejla [email protected] 3 1 Higher Institute of Management of Gabes, Gabès, Tunisia 2 University of Economics and Management of Tunis, Tunis, Tunisia 3 grid.444459.c 0000 0004 1762 9315 College of Business, Abu Dhabi University, Abu Dhabi, UAE 3 12 2022 138 12 2 2022 7 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 examines the relationship between corporate governance and banking stability by considering the moderating role of corruption controls. This study applies the Generalized Moments Method using a sample of panel data collected from 74 banks in 10 Organization for Economic Co-operation and Development countries during the period 2006–2016. The empirical results reveal that banking governance is positively associated with banking stability as measured by the Z-score. Additionally, the findings indicate that effective corruption control significantly moderates the power of the board of directors in boosting banking stability. This study discerns the fundamental role of the board of directors as the main corporate governance mechanism and internal player in the stability of banking institutions. Keywords Board of directors Corruption control index (CCI) Corporate governance Z-score OECD banks JEL Classification C33 G20 G30 ==== Body pmcIntroduction In recent years, banking institutions have experienced severe global crises. These disruptions correlate with recurrent periods of financial stress. Some institutions have gone bankrupt, while others have suffered from financial turmoil due to economic and financial decline in their countries. This was essentially the result of a poor corporate governance system (Choi 2000). In normal instances, bank functions are considered as a solid network heading to its objectives. However, disturbances caused by fluctuations in the monetary sphere threaten the banking environment’s stability. Djebali and Zaghdoudi (2020) consider banking stability as “the degree of resistance of the bank to protect itself against financial shocks and adverse events”. Moreover, it guarantees continuity of bank solvency in bad economic situations. Banks’ insolvency risks must be reduced to maintain an acceptable level of stability. The drawbacks of the traditional financial system appeared during the international crisis that shook the world in 2007 (Trabelsi 2011; Fakhfekh et al. 2016). This crisis has contributed to a decline in financial institutions and the collapse of several economies worldwide. This has led to many questions about banks’ ability to defend themselves against economic disturbances and shocks as well as about adequate solutions to financial failures (Bourkhis and Nabi 2013; Rosman et al. 2014). Since then, the banking profession has become more complex. The main concern of banking institutions is to maintain the continuity of their economic and financial growth. Consequently, new measures have been suggested to guarantee profitable management of banking institutions (Becht et al. 2011). Interestingly, banking governance has received particular attention, and governments worldwide have offered reforms to the compensation structures of bank executives in order to preserve the stability of their financial systems. Some researchers studied the poor performance of financial institutions, in particular, following the crisis which affected the subprime mortgage sector in the United States in 2007and depicted various short comings in the system of banking governance. The Organization for Economic Cooperation and Development (OECD) attributed the financial crisis to the weaknesses and shortcomings of corporate governance arrangements (Kirkpatrick 2009). Thus, better bank supervision is deemed necessary to effectively manage risks (Acharya and Richardson 2009; Demirgüç-Kunt et al. 2018). The benefits of sound governance frameworks are well recognized (Claessens and Yurtoglu 2013; Boubacar Diallo 2017), particularly in terms of improving performance and efficiency, allowing better access to financing, lower cost of capital, and more favor to all stakeholders. Similarly, weak banking governance can increase risk by affecting the quality of banking assets and causing financial volatility, which is often associated with a lack of transparency and fragility in terms of risk taking. All of these factors complicate the mission of supervisors from legal and procedural points of view (Mehran et al. 2011). Strong banking regulations are implemented to avoid complex financial intermediation activities such as securitization, which encourages excessive risk-taking. Furthermore, risk specifically increases systemic hazards, and subsequently leads banks to stability problems. Therefore, better governance measures are meant to fix the level of risk taking (Gorton and Rosen 1995; Dinç 2005). Thus, the different banking risks to which the bank is exposed, namely financial risks (Barth et al. 1998; Campbell 2007ab), operational risks, and accidental risks (Greuning and Brajovic-Bratanovic 2004) have been investigated. The problems of banking governance include the attention given to national legislative systems, adoption of codes of good practice, development of good governance models, identification of the board of directors, and role of management (Lombardi 2012). In addition, the principles of the OECD (2015) stipulate that corporate governance is aligned with the strategic choices of the institution, paying particular attention to the effective monitoring of management by the board of directors and the board's responsibility towards the firm and shareholders. The OECD defines the governance framework as “promoting transparent and fair markets and the efficient allocation of resources” while paying particular attention to the role played by stakeholders in control and monitoring. Additionally, the boards of directors of banking institutions play a key role in controlling and managing risks in both internal and external governance measures in the financial market. Banking governance considers the distribution of rights and liabilities between various parties in the bank, including shareholders, leaders, board of directors, and other stakeholders. It also specifies the rules and procedures for decision-making. Process. Therefore, a bank’s governance structure determines which participants have power to combat corruption, which is a coherent theme in terms of banking stability. In recent years, increasing attention has been paid to how banking institutions interact with their stakeholders. Corruption is linked to irregularity in the decision-making process in exchange for improper incentives or advantages. The control of corruption is at the heart of development policies. Recent studies justify the effectiveness of anti-corruption measures from an innovation perspective (Anokhin and Schulze 2009; Dang and Yang 2016; Xu and Yano 2017). This study explores the possible links between the board of directors and banking stability by considering the moderating role of corruption control. To the best of our knowledge, no study has examined the impact of the board of directors on a bank’s stability by considering the role played by corruption control; therefore, this study sheds light on this topic. OECD banks were the subject of this study for two reasons. First, OECD banks are interested in examining the impact of corruption control on improving their stability. Second, OECD countries were selected based on their tight regulations and high resilience to the financial crisis of 2007–2008. As shown in Fig. 1, developed countries (Canada, Finland, France, Ireland, and Switzerland) have a higher level of corruption control than emerging countries (Bahrain, Jordan, Qatar, Turkey, and United Arab Emirates). Finland has the highest control of corruption among developed countries, and the United Arab Emirates has the highest control of corruption among emerging countries.Fig. 1 Control of Corruption in the 10 OECD Countries in 2016. Source: Worldwide Governance Indicators. http://info.worldbank.org/governance/wgi/Home/Reports The research questions were as follows. What is the impact of board governance on the insolvency risk and banking stability? To what extent does corruption control moderate the relationship between board of directors and banking stability? This study makes two significant contributions to the literature: First, we explored the explanatory factors, particularly the attributes of the board of directors, that contribute to banking stability. Second, we examine the relationship between bank stability and the degree of corruption control. Empirically, we contribute to the literature by observing that many studies have examined the direct relationship between banking governance and banking stability (Dong et al. 2017; Ghosh 2018; Raouf and Ahmed 2020), but relatively little attention has been paid to the interaction effect of a country's corruption control in the banking sector. Our contribution relates to the integration of CI as a moderating variable in the study of the relationship between the board of directors and banking stability. Thus, we evaluate whether corruption control strengthens or weakens managerial relationships, which can open up new perspectives. This study uses a sample of 74 banks in 10 OECD countries from 2006 to 2016. The empirical results show the importance of controlling corruption, which significantly moderates the behavior of board members in improving banking stability. The remainder of this paper is organized as follows. Section 2 surveys the literature and develops hypotheses. Section 3 describes the research design, including the sample, model, and measurement of the variables. Section 4 presents the empirical results. Finally, Sect. 5 provides the conclusions and recommendations for future research. Theoretical framework and development of hypotheses Conceptual and theoretical framework The financial literature includes theories on banking stability and regulation. These theories consist of agency (Jensen and Meckling 1976), and political regulation theories (La Porta et al. 1998). According to agency theory, the corporate governance mechanisms are used to control managers and subsequently increase their efficiency. However, a managerial entrenchment strategy weakens the power of these mechanisms to manage the firm properly and benefit shareholders (Alexandre and Paquerot 2000). In addition, executive incentives are generally considered effective tools to motivate and align managers’ interests with those of shareholders (Berle and Means 1932; Holmstrom 1979; Grossman and Hart 1983; Murphy 1999; Bebchuk and Fried 2004; Bebchuk et al. 2010). The theory of political regulation states that politicians do not necessarily seek to maximize social well-being, but their own well-being (La Porta et al. 1998). Politicians can compel banks to lend funds to companies for political purposes (Chen et al. 2015). Conversely, large banks can abide by the decisions of politicians so that the rules of supervision of the financial system fit their own interests. Thus, banking supervision by public authorities can negatively affect bank efficiency and credit allocation stability. In addition to the abovementioned theories, Giammarino et al. (1993) model is inspired by agency theory and proposes the creation of effective oversight mechanisms that solve conflicts of interest and reduce the costs arising from agency conflicts in the process of banking governance. This model relates capital ratio and risk-taking through asymmetric information. Indeed, the regulator has no information on the quality of a bank’s assets or ideas regarding its choice to change its risk profile. The model proposed by Giammarino et al. (1993) considers the agency relationships among depositors, banks, and regulators, which represents the probability of default and gives great importance to the regulatory financial system based on deposit insurance and a precise level of equity (Vilanova 2006). The effect of board of directors on banking stability In all countries worldwide, the main objective of banking institutions is to establish a strong board of directors system that aims to improve corporate governance by setting standards for the size, composition, structure, and responsibilities of board members. Following previous studies examining the impact of the board of directors on banking stability (Karkowska and Acedański 2020; Fernandes et al. 2021; Gilani et al. 2021; Marie et al. 2021), this study considers the attributes of the board of directors, including board size, board independence, board diversity, board duality, and number of board meetings. Board size In corporate governance literature, there is no agreement on the impact of board size on board efficiency. Daily et al. (1999) and Kiel and Nicholson (2003) confirm that a large number of directors might be necessary in large financial institutions to increase the pool of expertise and available resources as well as the potential for establishing contacts with various customers and depositors. In addition, Bouwman (2011) indicated that a large number of directors not only strengthened the board of directors, but also provided essential resources such as professional networks, domain knowledge, skills, and experience. However, Jensen (1993) and Hermalin and Weisbach (2003) state that a high number of directors decreases the efficiency and control of risk management because of the higher agency costs related to problems encountered by directors and difficulties in coordination and communication. Empirically, Ladipo and Nestor (2009) indicate that the best-performing European banks have smaller boards. In addition, Pathan and Faff (2013) reveal that American banks with smaller boards have superior financial performance. More recently, Karkowska and Acedański (2020) confirm a negative association between board size and banking stability. Hence, our first hypothesis is as follows. H1.a There is a negative association between board size and bank stability. Board of directors composition Board independence This study considers board independence another essential characteristic that can significantly impact bank efficiency and risk. Several studies confirm an inverse relationship between board independence and banking stability (Linck et al. 2008; Minton et al. 2011). In addition, Erkens et al. (2012) confirm that board independence within banks reduces stock market returns. Similarly, Pathan and Faff (2013) note that board independence decreases banking performance. Therefore, the presence of independent directors on the board does not always guarantee better performance. Independent directors should use their skills better and apply their expertise and knowledge in their decision-making processes (Adams and Ferreira 2007). In contrast, several theoretical reasons explain why greater board independence can be beneficial and effective at the board level (Fama and Jensen 1983). These arguments highlight the important role of independent directors in protecting their reputation in the market, which should make them more effective at monitoring and disciplining managers, reducing opportunistic costs, and protecting shareholders’ interests. Other studies confirm that board independence has a positive effect on banking performance (Agrawal and Knoeber 1996; Skully 2002; Park and Shin 2004; Pathan et al. 2007). More recently, Ghosh (2018) shows that greater board independence is essential not only to further enhance profitability but also to reduce the insolvency risk within banking institutions. More recently, Fernandes et al. (2021), Gilani et al. (2021), and Trinh et al. (2021) confirm that board independence is positively associated with banking stability. Thus, our second hypothesis predicts a positive relationship between the presence of independent directors on a board and banking stability. H1.b There is a positive association between board independence and banking stability. Women on board The greater presence of women on boards affects corporate governance dynamics in several ways. Several studies have shown that a higher proportion of female directors negatively affects performance (Adams and Ferreira 2007; Berger et al. 2014) whereas others have confirmed the opposite effect (Carter et al. 2003; Erhardt et al. 2003; Owen and Temesvary 2018). Adams and Ferreira (2007) explain the negative impact of excessive monitoring or lack of experience. Berger et al. (2014) confirm that the presence of women on boards increases banks’ risk-taking decisions in Germany, although the economic impact is marginal. However, the evidence of its impact on bank performance and risk is far from simple. Kanter (1977) suggests that the performance benefits are only achieved when the proportion of women in the board reaches the “critical mass” that will allow women to form coalitions within the board, help each other’s and influence the culture of the board. Several empirical studies confirm a positive relationship between a higher proportion of female directors and accounting performance (Carter et al. 2003; Erhardt et al. 2003; Gulamhussen and Santa 2015). Owen and Temesvary (2018) show that women’s participation has a positive effect as soon as a minimum threshold of gender diversity is reached. However, this positive effect is only seen in better-capitalized banks. Regarding gender differences in attitudes towards risk, organizational psychology and economics documents show that women tend to be more risk averse than men (Palvia et al. 2020). More recently, Marie et al. (2021) reveal that board diversity has a positive and significant impact on Egyptian bank stability. Based on these arguments, we formulate the following hypotheses: H1.c There is a positive association between the presence of women on a board of directors and banking stability. The board of directors Board duality Another important feature of the board is the dual appointment of the CEO and the chairperson. Two opposing arguments exist in the literature regarding the potential effects of board duality on bank operations and performance. Organizational theorists (Anderson and Anthony 1986) argue that combined leadership structures on top of the hierarchy can reduce information costs and improve the stability and performance of banking institutions. Empirical studies in the banking sector have provided mixed results. Agency theorists consider board duality to weaken board supervisory powers and increase governance costs (Lipton and Lorsch 1992; Jensen 1993; Lasfer 2006). Additionally, duality allows leaders to take advantage of their power to achieve personal interests (Dey et al. 2011). Hence, the third hypothesis is as follows. H1.d There is a negative association between the duality and banking stability. Number of board meetings This study tests whether the frequency of meetings has performance advantages in terms of efficiency and banking risk taking. There is no consensus in the empirical literature regarding the impact types. Andres and Vallelado (2008) reveal a non-significant relationship between board meetings and banking performance. Conger et al. (1998) confirmed a positive relationship between the number of board meetings and internal governance, which indirectly facilitates performance through reduced agency costs and lower risk-taking behavior. Adams and Mehran (2003) and Grove et al. (2011) added that compared to non-financial companies, banks require more frequent board meetings because of their business complexity. Liang et al. (2013) reveal that the frequency of such meetings improves both the performance of banks and the quality of their assets, leading to an interesting reduction in banking risk. Based on this argument, we propose the following hypotheses: H1.e There is a positive association between the number of meetings and banking stability. The moderating effect of corruption control on the relationship between board of directors and banking stability In the theoretical literature, the alignment hypothesis states that blockholders can impose increased oversight on management and use their power to push managers to make appropriate decisions that will increase shareholder wealth and firm value (Jensen and Meckling 1976; Shleifer and Vishny 1986). According to this hypothesis, having blockholders in an ownership structure could leave managers with fewer discretionary opportunities to engage in credit-corruption. The actions, interactions, and decision-making processes of the board of directors within the framework of a sustainable corporate governance system aim to prevent corruption in companies (Elkington 2002; Lozano 2012; Allais et al. 2017; Fuente et al. 2017; Adnan et al. 2018; Schrippe and Duarte Ribeiro 2019). The board of directors is responsible for the adoption and implementation of the corruption prevention code within the company as well as the definition of the sustainable strategic objectives of the corporation. Thus, the sustainability of banking institutions is the main example of overall business sustainability which aims to avoid negative externalities in the environment (Elkington 2002), and subsequently leads to better banking and economic stability. A bank’s sustainability is based on activities that positively contribute to sustainability (i.e., ecological, social, economic, political, and territorial) by the board of directors and management of the institution (Schrippe and Duarte Ribeiro 2019). From this perspective, corruption requires firms to implement sustainable strategies to prevent it (Atangana Ondoa 2014). Corruption is a fundamental obstacle to economic and social development and hinders sustainable economic development, particularly in developing countries (Spyromitros and Panagiotidis 2022). For banks, corruption increases costs and presents excessive risks, which reduce stability. Undoubtedly, banking institutions need strict anti-corruption measures to protect their reputations and stakeholders’ interests (World Bank 2020). In banking governance codes, ethical behavior plays a crucial role in preventing corruption and is mainly seen as an effect of anthropological, ethical, and moral attitudes. According to Beck et al. (2005), the corruption of financial intermediaries mitigates the efficient allocation of capital for small businesses, forcing them to miss profitable investment opportunities and subsequently reduce their economic growth. La Porta et al. (1998) confirm that corruption negatively affects bank loans in countries with poor investor-protection laws. In countries with strong investor protection laws, banks have the assurance of getting their money back, even if debtors are in payment default. The transparency and strength of the legal system serve to reduce the level of corruption, which in turn leads to better banking efficiency. Even in the case of borrower bankruptcy, the bank is protected by law. It can recover funds or take possession of business. Corruption reduces protection and paralyzes banking systems. In many countries, efforts to reduce or eliminate corruption have failed (Persson et al. 2013; Ajim Uddin et al. 2020). Thus, anti-corruption measures are advantageous and can reduce the risk of bank loss for banks (Demirgüç-kunt and Detragiache 2000; Beck, et al. 2005). These measures are effective in helping companies move from seeking political relationships to strengthening innovation (Anokhin and Schulze 2009; Dang and Yang 2016; Xu and Yano 2017). The governance system within banks and national anti-corruption laws are the two main pillars operating in complex scenarios. Sustainable banking governance aims to establish an efficient board of directors to facilitate decision making and ensure banking stability. In addition, national legislation in each country aims to prevent corruption among financial and non-financial companies and reduce negative impacts on the environment. Thus, a relevant question arises regarding the role of the board of directors in the application of national law and anti-corruption measures to ensure a decrease in insolvency risk and improve banking stability by integrating the corruption control index. Following these arguments, different hypotheses are developed: H2 The relationship between the attributes of board of directors and banking stability is affected by the degree of corruption control. H2.a The relationship between board size and banking stability is affected by the degree of corruption control. H2.b The relationship between board independence and banking stability is affected by the degree of corruption control. H2.c The relationship between the presence of women on board of directors and banking stability is affected by the degree of corruption control. H2.d The relationship between the duality of the chairperson of the board and the CEO, and banking stability is affected by the degree of corruption control. H2.e The relationship between the number of meetings and risk of insolvency is affected by the degree of corruption. Research methodology Sample and data This study uses a sample of 74 commercial banks operating in ten OECD countries (Canada, Switzerland, France, Ireland, Finland, Turkey, Qatar, Bahrain, Jordan, and the UAE) from 2006 to 2016. The year 2006 marks the beginning of the subprime mortgage crisis (Amadeo 2021) whereas 2016 marks the beginning of the recovery from the financial crisis of 2007–2008 (McKinsey Global Institute 2018). Data were collected from DataStream and included information on banking governance mechanisms (measured by the attributes of the board of directors) and the control variables. Macroeconomic variables were extracted from the World Bank’s website. Variables The dependent variable is the Z-score, which measures financial stability. Table 1 presents all independent variables, along with their respective notations, measures, expected signs, and sources of data.Table 1 Definition of independent variables Variables Notation Measure Expected sign Source of data Board Size Bsize Number of directors in the board (−) DataStream Board independence Bindp The percentage of the independent directors in the board (+) DataStream Women on board Bfemale The percentage of the female directors in the board (+) DataStream Board duality Bduality It takes the value of 1 if the chairman of the board is the CEO, 0 otherwise (−) DataStream Number of meetings Bmeetings The number of board’s meetings (+) DataStream Size of the bank Size The size is measured by the logarithm of the total assets (−) DataStream Loans to Asset LTA This variable measures the liquidity degree of the bank. It’s calculated by dividing total liabilities on total Assets (−) DataStream Non-performing loans NPL Measures the quality of bank credit risk management. It’s calculated by dividing the amount of non-performing loans on total loans (−) DataStream Gross Domestic Product GDP This variable measures the gross domestic product of the country (+) World Bank Inflation rate Inf This variable measures the degree of price increase for goods and services (−) World Bank Corruption Control Index CCI It is an index measuring the control of corruption (+) World Bank Methodology Impact of board of directors on Banking Stability The impact of the board of directors on banking stability is tested using the following general model:1 Zit=β0+β1Bsizeit+β2Bindpit+β3Bfemaleit+β4Bdualit+β5Bmeetingsit+β6Sizit+β7LTAit+β8NPLit+β9GDPit+β10Infit+εit Several banking risk measures have been used in the literature. In this study, banking risk is proxied by the Z-score, a synthetic indicator considered a measure of bank insolvency risk. In addition, this measure is widely used to assess bank financial health (Schaeck and Wolfe 2006; Worrell et al. 2007). This indicator is commonly mentioned in the literature (Boyd and Runkle 1993; Lepetit et al. 2008; Laeven and Levine 2009; Houston et al. 2010; Demirgüç-Kunt and Huizinga 2010). According to Houston et al. (2010), this risk-taking measure is linked mainly to creditors’ rights and shared information, which subsequently promotes economic growth (Fig. 2). A higher Z-score indicates a decrease in insolvency risk, suggesting that the bank is more stable. Banking stability reflects its ability to remain solvent under difficult economic conditions by using its capital and reserve accounts. In other words, it is a bank’s ability to resist and protect itself from adverse events and macroeconomic shock. The Z-score was calculated according to Čihák and Hesse (2010), as follows:Zit=ROAit+EitTAitσROAit Fig. 2 The effect of board of directors’ attributes on banking stability where ROA is the return on assets, E/TA is the ratio of equity, and σ ROA is the standard deviation of ROA. The moderating role of the corruption control index The objective of this section is to study the moderating effect of CI on the relationship between the board of directors and banking stability. To test Hypothesis (H2), CCI was used to moderate the relationship between the attributes of the board of directors and bank stability (Fig. 3). Specifically, the following model was developed to better understand whether the level of corruption in a country affects the relationship between board governance and banking stability.2 Zit=β0+β1CCIit+β2Bsizeit+β3Bsize×CCIit+β4Sizeit+β5LTAit+β6NPLit+β7GDPit+β8Infit+εit. 3 Zit=β0+β1CCIit+β2Bindpit+β3Bindp∗CCIit+β4Sizeit+β5LTAit+β6NPLit+β7GDPit+β8Infit+εit, 4 Zit=β0+β1CCIit+β2Bfemaleit+β3Bfemale∗CCIit+β4Sizeit+β5LTAit+β6NPLit+β7GDPit+β8Infit+εit, 5 Zit=β0+β1CCIit+β2Bdualit+β3Bdual∗CCIit+β4Sizeit+β5LTAit+β6NPLit+β7GDPit+β8Infit+εit. 6 Zit=β0+β1CCIit+β2Bmeetingsit+β3Bmeetings∗CCIit+β4Sizeit+β5LTAit+β6NPLit+β7GDPit+β8Infit+εit Fig. 3 The effect of corruption on the board of directors-bank stability relationship Empirical results Descriptive statistics Table 2 presents descriptive statistics for the variables included in this study. The average of Z-score of all the banks included in the sample (the dependent variable) for the period between 2006 and 2016 was 66.70. As for independent variables, the mean (standard deviation) of board size was 15.71 (4.02), the average of board independence (standard deviation) was 45.30% (14.18%). The average (standard deviation) of women on board was 34.39% (20.28%), and almost 35% of board members were female directors. The average (standard deviation) of board duality is 0.52 (0.5). The number of meetings on the board of directors was 12.35, with a minimum of three meetings and a maximum of 47 meetings per year.Table 2 Descriptive statistics Variables Mean STD Min Max Z-score 66.70 28.89 − 27.99 99.00 Bsize 15.71 4.02 6.00 28.00 Bindp 45.30 14.18 1.02 86.12 Bfemale 34.39 20.28 0.00 71.94 Bdual 0.51 0.50 0.00 1.00 Bmeetings 12.35 5.92 3.00 47.00 Size 17.84 2.18 11.59 21.69 LTA 64.38 19.57 9.5 123.59 NPL 5.39 7.33 0.05 67.80 GDP 2.74 3.25 − 8.26 7.60 Inf 1.08 0.92 − 1.14 2.81 CCI 0.91 0.67 − .019 2.50 With regard to control variables, the average size of the banks was 17.84 with a minimum of 11.60 and a maximum of 21.69. The average bank liquidity "LTA" was around 64.39%, its minimum was 9.5%, while its maximum was high, reaching 123.59%. This can be explained by banks’ excessive liabilities included in the sample. The values of the bank’s credit risk “NPL” were ranging between 0.05% and 67.8% with an average of 5.39%, which shows that most banks had a low rate of non-performing loans. With regard to the moderating variable, the average corruption control index was 0.91, with a minimum of − 0.19 and a maximum of 2.5. This index was collected from the World Bank and is commonly used in literature (Kaufmann et al. 2010). It must be between − 2.5 and 2.5, and a higher value of CCI indicates less corruption. Overall, most banks in our sample have good corruption controls. Table 3 presents the annual mean values of all the variables over the years included in the analysis. The mean of Z-score index was 73.40% in 2006, which decreased to 63.30% in 2016. This finding can be explained by the negative impact of the financial crisis on banks’ stability. As for the board of directors variables, the mean board size was 16.10 which 2006 and remained almost stable throughout the study period. The means of board independence and duality increased from 2006 to 2016, while the means of board diversity and the number of board meetings decreased from 2006 to 2016. The means of size, loans to assets, GDP, and the corruption control index decreased from 2006 to 2016, while the means of non-performing loans and inflation increased from 2006 to 2016.Table 3 The mean of all the variables over the years Variables 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 Z-score 73.40 73.66 73.03 65.67 64.27 64.33 63.93 63.97 60.13 61.87 63.30 Bsize 16.10 15.39 15.72 15.64 15.45 15.71 15.82 15.59 15.90 15.95 15.94 Bindp 42.70 42.18 41.16 41.48 46.55 47.90 46.71 50.46 50.33 46.85 45.01 Bfemale 37.97 37.20 36.84 31.01 31.69 30.46 36.73 35.87 34.20 33.95 32.23 Bdual 0.48 0.44 0.52 0.58 0.57 0.56 0.55 0.54 0.54 0.53 0.53 Bmeetings 12.60 12.58 12.86 12.36 12.51 12.37 12.79 13.10 12.09 11.57 11.21 Size 19.90 19.34 19.39 18.44 17.45 17.54 15.57 14.60 14.62 13.56 15.55 LTA 71.09 71.87 70.74 69.19 68.43 60.88 58.05 57.90 57.45 59.58 59.65 NPL 1.40 1.27 1.77 3.24 3.55 4.05 8.66 8.54 13.24 7.04 6.64 GDP 5.32 7.24 8.50 1.71 0.96 0.61 1.06 0.85 0.38 1.94 1.71 Inf − 0.43 − 0.99 0.47 2.22 0.11 2.26 2.69 2.66 2.44 0.21 0.20 CCI 0.95 0.94 0.96 0.97 0.89 0.88 0.85 0.80 0.81 0.81 0.88 Table 4 presents the mean values of the variables included in this study per country. Developed countries (Finland, Canada, Switzerland, Ireland, and France) have higher Z-scores than emerging countries (the UAE, Qatar, Jordan, Bahrain, and Turley). More particularly, Finland had the highest Z-score (91.98%), while Turkey had the lowest Z-score (32.93%). As for the board of directors variables, the highest mean of board size is in Canada (18.87), the highest mean of board independence is in France (58.74%), with a slight difference compared to Finland (58.51%), the highest mean of board diversity is in Finland (57.37%), the highest mean of board duality is in Bahrain (0.85), and the highest mean of the number of board meetings is in Ireland (14.66), with a slight difference compared to Finland (14.54). These results indicate that developed countries have stronger boards of directors that are associated with higher Z-scores.Table 4 The mean of all the variables per country Variables Finland Canada Switzerland Ireland France UAE Qatar Jordan Bahrain Turkey Z-score 91.98 91.42 87.81 88.79 82.86 45.96 48.88 36.82 37.94 32.93 Bsize 18.16 18.87 14.75 17.54 16.56 15.75 15.67 14.22 13.55 12.79 Bindp 58.51 49.44 54.95 53.45 58.74 47.02 39.99 35.62 23.92 27.13 Bfemale 57.37 47.49 43.88 45.45 45.40 33.26 17.61 14.43 13.23 24.78 Bdual 0 0.32 0.40 0.36 0.4 0.75 0.83 0.57 0.85 0.64 Bmeetings 14.54 13.45 13.98 14.66 13.74 11.14 10.09 10.74 9.61 9.76 Size 23.72 37.61 24.87 28.98 20.42 3.81 7.24 6.94 8.70 10.56 LTA 87.18 88.73 77.87 75.60 74.35 53.17 47.81 52.80 38.39 44.04 NPL 2.12 2.01 3.13 3.54 4.82 6.15 8.26 8.44 7.63 8.13 GDP 7.41 7.43 6.42 6.15 5.60 0.23 − 2.87 0.64 − 2.07 − 1.56 Inf − 1.05 − 0.92 − 1.12 0.85 − 1.03 2.71 3.66 2.33 1.86 3.64 CCI 1.19 1.30 1.11 1.29 1.48 1.06 0.21 0.48 0.94 0.22 Similarly, the highest means of size, loans to total assets, GDP, and corruption control index are in developed countries, while the highest means of non-performing loans and inflation are in emerging countries. Table 5 presents the correlation matrix and shows that all the correlation coefficients are less than 0.7 in absolute value (Kervin 2010) which confirms the absence of multicollinearity between the independent variables. Table 5 Correlation Matrix Zscore Bsize Bindp Bfemale Bdual Bmeetings Size LTA NPL GDP Inf CCI Zscore 1.00 Bsize 0.06 1.00 Bindp − 0.10* 0.01 1.00 Bfemale − 0.02 0.007 − 0.07 1.00 Bdual − 0.15* − 0.04 0.11* 0.01 1.00 Bmeetings 0.04 0.11* − 0.04 − 0.10* − 0.01 1.00 Size 0.61* 0.03 − 0.05 − 0.03 − 0.06* 0.02 1.00 LTA 0.006 0.07* − 0.04 0.01 − 0.02 0.04 − 0.23* 1.00 NPL − 0.24* − 0.10* 0.13* − 0.006 0.17* 0.07* − 0.37* 0.18* 1.00 GDP 0.14* 0.03 0.03 − 0.07* − 0.13* 0.02 0.15* − 0.07* − 0.03 1.00 Inf − 0.17* − 0.05 0.004 0.03 0.02 − 0.03 − 0.24* − 0.11* − 0.01 0.13* 1.00 CCI 0.39 * 0.03 0.03 − 0.04 0.09* 0.06 0.46* − 0.04 0.03 0.14* − 0.17* 1.00 Significant at 5% In addition, Table 6 indicates that the variance inflation factors (VIF) are lower than four, which is the threshold suggested by Benavent and Evrard (2002). This further confirms the results of the correlation matrix and absence of multicollinearity.Table 6 Variable Inflation Factors (VIF) Variables VIF 1/VIF Size 1.43 0.698 CCI 1.39 0.720 LTA 1.17 0.851 GDP 1.10 0.913 Inf 1.09 0.916 Bdual 1.07 0.936 Bmeetings 1.07 0.936 Bindp 1.04 0.963 Bfemale 1.03 0.966 NPL 1.03 0.970 Mean VIF 1.13 Empirical results Impact of board of directors on bank stability In this study, a two-step Generalized Method of Moments (GMM) system was estimated. This estimation was suggested in the Monte Carlo studies conducted by Blundell and Bond (1998) and Blundell and Bond (2000). The Sargan test was applied as part of the estimation to determine the validity of the instruments. Tables 7 and 8 illustrate the Sargan test results, which confirm the validity of the instruments. The results show that the lagged variable is significant at 1% level, indicating that it is difficult to quickly change the degree of insolvency risk within banks, and the decisions made during a year affect risk-taking in the future. In other words, granting credit in year N affects the banking risk level until the loan is paid fully. Consequently, this finding confirms that banking institutions’ insolvency risk is affected by positions generated by previous risk-taking decisions.Table 7 Board of directors’ attributes and banking stability Dependent variable: Z-score Independent variables Coefficient Z-Statistic Z-score L1 0.89 17.01 Constant − 4.70* − 3.31 Bsize − 0.87* 16.20 Bindp 0.06* 8.27 Bfemale 0.005 0.66 Bdual 0.16 0.27 Bmeetings − 0.29* − 8.69 Size − 0.67* − 6.54 LTA − 0.02* 3.08 NPL − 0.004* − 3.33 GDP 0.03 − 0.55 Inf − 0.02 1.19 Hausman 0.34 Durbin–Watson 0.56 AR(1)-test (p value) − 3.22* (0.001) AR (2)-test (p value) Sargan test (p value) − 2.001 (0.45) 58.44 (0.28) Significant at a level of 10%, significant at a level of 5%, * significant at a level of 1%level Table 8 Results of the integration of corruption control as a moderating effect on the governance banking relationship—banking stability Arellano-Bond dynamic panel-data estimation Dependent variable: Z-score Independent variables Model (2) Model (3) Model (4) Model (5) Model (6) Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Z-Score L 1 0.77* 19.04 0.77* 20.56 0.76* 23.89 0.76* 20.46 0.77* 22.93 Constant 8.89* 7.11 10.21* 5.33 15.16* 10.54 15.10* 10.86 11.95* 9.15 CCI 13.3* 16.08 9.13* 13.49 4.75* 10.91 6.07* 12.70 8.03* 11.73 Bsize − 0.45* 10.13 BsizeCCI 0.52* − 14.67 Bindp − 0.09*** 9.09 BindpCCI 0.08 − 8.57 Bfemale − 0.01 2.08 BfemleCCI 0.17** 2.20 Bdual 0.69* 0.70 BdualCCI − 1.26* − 2.44 Bmeetings − 0.21*** 2.95 BmeetingsCCI 0.21 − 5.72 Size − 1.23* − 16.21 − 1.21* − 13.66 − 1.27* − 15.85 − 1.24* − 15.29 − 1.18* − 13.77 LTA − 0.04* 4.40 − 0.05* 6.11 − 0.04* 5.68 − 0.05* 5.79 − 0.05* 6.87 NPL − 0.61* − 36.86 − 0.61* − 42.89 − 0.60* − 40.01 − 0.61* − 40.28 − 0.62* − 38.02 GDP 0.21* 4.91 0.22 4.56 0.22* 4.33 0.23* 4.72 0.24* 5.48 Inflation − 0.01 − 1.42 − 0.01 − 1.49 − 0.002 − 0.36 0.23 4.72 − 0.01* − 1.73 Hausman 0.05 0.04 0.38 0.28 0.14 Durbin–Watson 0.50 0.50 0.50 0.50 0.51 AR (1) test (p value) − 2.80* (0.005) − 2.81* (0.004) − 2.78* (0.005) − 2.81* (0.004) − 2.81*** (0.004) AR (2) test (p value) − 1.15 (0.24) − 1.17 (0.23) − 1.19 (0.23) − 1.18 (0.23) − 1.18 (0.23) Sargan test (p value) 62.59 (0.27) 61.54 (0.29) 62.95 (0.26) 62.07 (0.28) 61.37 (0.20) * Significant at a level of 10%, Significant at a level of 5%, * Significant at a level of 1% In addition, the empirical results reveal that board size (Bsize) negatively affects banking stability which confirms H1.a. This result corroborates those of Jensen (1993), Hermalin and Weisbach (2003), and Dong et al. (2017) and shows that the relationship between board size and banking stability is negative. This finding is explained by the high number of directors on the board, which decreases the efficiency and control of risk management because of the higher agency costs. Thus, a large board size is associated with lower stability in banking institutions (Pathan and Faff 2013). These empirical results do not confirm the findings of Daily et al. (1999), Kiel and Nicholson (2003), and Elyasiani and Zhang (2015), who demonstrate a positive relationship between board size and banking stability, because large boards have a high field of expertise, a high level of available resources, and a high potential to establish contacts with various customers and depositors. Board independence (Bindp) positively affects banking stability; therefore, H1.b is also validated. This result confirms Ghosh (2018), who confirms that independent directors decrease the asymmetry of information within a company, subsequently leading to an increase in banking stability. As for the presence of female directors on the board (Bfemale), Table 7 shows that the coefficient of the number of women on board is positive and significant at 5%, which validates H1.c. This result supports those of Kanter (1977), Erhardt et al. (2003), and Temesvary et al. (2018) and reveals that the presence of women on boards has a positive effect on banking stability and that female directors are able to form coalitions and positively influence the company’s culture. Adams and Ferreira (2007) suggest the opposite results because of the lack of experience of women in corporate management. In addition, the results indicate that board duality does not affect the risk of insolvency; therefore, H1.d is rejected. According to Aebi et al. (2012) and Berger et al. (2014), duality in the board of directors (Bdual) does not lead to an increase in opportunistic managerial behavior and, subsequently, does not deteriorate or improve stability within banks. This result contradicts Dong et al. (2017) findings, whose empirical results indicate that board duality is associated with a decrease in bank profitability and an increase in insolvency risk. The second characteristic of the board of directors is the number of meetings (Bmeetings). It was noticed that the association of this variable with stability was significantly negative at a level of 1%; hence, Hypothesis H1.e was rejected. According to Vafeas (1999) and Liang et al. (2013), the frequency of board meetings is negatively linked to performance, and therefore, stability. However, this result contradicts the findings of Conger et al. (1998), who confirm a positive relationship between the number of board meetings and banking stability because a high number of meetings mitigates information asymmetry and increases transparency by improving performance and reducing bank insolvency risk. Regarding the control variables, bank size (Size), liquidity (LTA), and non-performing loans (NPLs) negatively affect banking stability. Thus, large banks are riskier than small ones, based on the risk related to weighted assets, liquidity, and volatility of returns (Demsetz and Strahan 1997; Khan et al. 2017). Liquidity risk (measured by an increase in the LTA ratio) negatively affects banking stability. These results are consistent with those reported by Bourke (1989), Molyneux and Thornton (1992), Barth et al. (1998), Beck et al. (2003), Kosmidou et al. (2005) and Khan et al. (2017). According to these researchers, a high level of liquidity reduces a bank's vulnerability to default and therefore increases its financial stability. For non-performing loans, they have a negative and significant impact on Z-score. This result indicates that an increase in the credit risk reduces financial stability. As a result, a high level of credit risk leads to the deterioration of assets and, subsequently, higher destabilization and lower stability (Iyer and Puri 2012; Louhichi and Boujelbène 2016). The moderating role of the corruption control Table 8 illustrates the estimation results for the control corruption index’s impact on the relationship between governance and banking stability. The results show that the interaction term BsizeCCI was significantly positive (0.52, p < 0.01). This result confirmed that H2. Thus, the moderating effect positively affects the relationship between the board size and bank stability. In the presence of high corruption control, a high number of directors improves the field of expertise, increases the potential to establish contact with various clients and depositors, and reduces banks’ insolvency risk of the bank (Daily et al. 1999; Kiel and Nicholson 2003). In other words, a high degree of corruption control in banks leads to better surveillance of board members’ decisions (Elkington 2002; Lozano 2012; Allais et al. 2017; Fuente et al. 2017; Adnan et al. 2018). The interaction term BindpCCI was significantly positive (0.08, p < 0.01). This result confirmed that H2. The moderating effect of the CCI positively affects the relationship between the independence of the board and banking stability. Controlling corruption strengthens market discipline and influences the behavior of independent directors to be firmer and stricter in protecting their reputation in the market (Fama and Jensen 1983). This promotes control of the governing body and reduces costs due to agency disputes, which positively influences the solvency and stability of the bank. Indeed, the presence of independent directors on the board limits managers’ opportunistic behavior, which leads to a good agency relationship between shareholders and managers. In such a context of efficient banking governance, members of the board of directors are obligated to increase their efforts and collaborate, leading to a reduction in insolvency risk by achieving performance sustainability and banking stability (Lombardi et al. 2019). Similarly, regarding the composition of the board, Table 8 shows that the interaction term BfemaleCCI is significantly positive (0.17, p < 0.01). This result is consistent with H2, confirming the moderating effect of CCI on the relationship between the presence of women on boards and the risk of bank insolvency. This result suggests a good corruption control in the banking sector. It strengthens the position of women since they do not belong to informal social networks dominated by men and also because of their risk-averse behavior, which consequently increases their effect on decision-making and improves banking stability (Gulamhussen and Santa 2015). Thus, female directors have a positive influence on accounting and market performance, and thereafter, on banking stability. These results suggest the predominance of the advantages associated with the participation of female directors on boards, which are linked to tighter control (Watson et al. 1993), conservatism in monitoring strategies and risks (Wiersema and Bantel 1992) and the quality and solidity of monitoring strategies and risks (Forbes and Milliken 1999; McInerney-Lacombe et al. 2008). Corruption control moderates the effect of the board of directors on insolvency risk and banking stability. Hypothesis H2 is supported by the moderating effect of CCI on the relationship between board duality and banking stability. The results show that the interaction term BdualCCI is negative (− 1.26) and significant at the 5% level. This indicates that good corruption control can reduce board duality, because it leads to the absence of supervision impartiality, thus achieving banking stability (Rachdi and El Gaied 2009). CCI is a form of internal and external auditing that reduces managerial misconduct. In the case of duality, managers’ misconduct is associated with a decrease in bank profitability and increased insolvency risk (Dong et al. 2017). According to agency theory, duality allows managers to leverage their power for their own benefit to the detriment of the company (Dey et al. 2011; Dong et al. 2017). In addition, other researchers indicate that the government ownership of banks facilitates the financing of politically desirable projects that optimize the private well-being of politicians instead of maximizing social welfare (La Porta et al. 2002; Sapienza 2004; Dinç 2005; Khwaja and Mian 2005). Therefore, H2 was validated. Hypothesis H2 considers the moderating effect of CCI on the relationship between the number of meetings held by a board of directors and banking stability. The results show that the interaction term BmeetingsCCI is positive (0.21) and significant at 1%. This finding validates H2 and suggests that a high corruption control level facilitates bank transparency. In this case, a high number of meetings helps directors make the right decisions in the bank, which improves performance and reduces risk. Conger et al. (1998) suggested a positive relationship between the number of board meetings and a company's internal oversight, which results in better performance due to reduced agency costs and lower risk. Similarly, Marie et al. (2021) showed that the frequency of board meetings improves bank performance, leading to better banking stability. Our analysis provides evidence that corruption control plays an important moderating role in strengthening the positive impact of the attributes of the board of directors on financial stability. Banks with larger and dual boards are under stronger pressure and restrictions to avoid risky strategic decisions. Any changes in banks’ corporate governance regulations are expected to have a significant impact on their financial stability. Impact of board of directors on banking stability: developed versus emerging countries Table 9 illustrates the impact of the corruption control index on the relationship between the boards of directors and banking stability in developed countries (Finland, Canada, Switzerland, Ireland, and France). The results indicate a positive and significant interaction term, BsizeCCI (0.35, p < 0.01). This result validates (H2) which states that the Corruption Control Index (CCI) positively moderates the relationship between board size and bank stability. In other words, the presence of many directors in a healthy and non-corrupt environment increases the scope of their expertise, expands the corporate network with various clients and depositors, enhances the efficiency of the decision-making process, and increases bank stability.Table 9 Impact of board of directors on banking stability in Developed countries: Arellano-Bond dynamic panel-data estimation Dependent variable: Z-score Independent variables Model (2) Model (3) Model (4) Model (5) Model (6) Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Z-Score L 1 0.71* 17.43 0.71* 9.06 0.71* 16.79 0.68* 10.85 0.69* 11.90 Constant 16.59 2.19 14.03 1.21 16.01 2.18 11.35 1.48 21.51* 2.83 CCI 10.49 0.68 23.49* 4.79 21.18* 5.35 18.01* 3.04 2.98 0.27 Bsize 0.31 0.22 BsizeCCI 0.35 0.06 Bindp 0.47* 5.56 BindpCCI 0.61 4.38 Bfemale 0.39* 3.84 BfemleCCI 0.25 3.51 Bdual 5.30 0.53 BdualCCI − 7.16** − 1.05 Bmeetings − 1.14* − 0.90 BmeetingsCCI 0.62** 0.81 Size 2.40 0.89 0.07 0.03 1.70* 0.32 2.07 0.84 1.7 0.74 LTA 0.23* 2.18 0.08 0.83 0.22 2.46 0.22 2.22 0.26* 4.14 NPL − 1.24* − 6.19 − 1.51* − 5.39 − 1.31* − 4.76 − 1.76* − 6.54 − 1.70* − 6.92 GDP 21.86 2.19 18.46 1.21 21.17 2.17 14.87 1.47 28.23* 2.81 Inflation − 1.95* − 2.97 − 1.64 − 2.56 − 1.41 2.31 − 1.53* − 2.89 − 1.36 − 2.35 Hausman 0.05 0.03 0.01 0.04 0.03 Durbin–Watson 0.58 0.56 0.86 0.56 0.63 AR (1) test (p value) − 2.60* (0.009) − 2.61*(0.009) − 2.75(0.005) − 2.46(0.01) − 2.48(− 0.01) AR (2) test (p value) − 1.37(0.16) − 1.3 (0.18) − 1.64(0.09) − 1.31(0.18) − 1.37 (0.17) Sargan test (p value) 23.31 (0.99) 19.41(1.00) 23.46(0.99) 23.30 (0.99) 20.79(0.99) * Significant at a level of 10%, Significant at a level of 5%, * Significant at a level of 1% Regarding the composition of the board, the BindpCCI interaction term was positive and significant (0.61, p < 0.01). These results corroborate those of Fernandes et al. (2021), Gilani et al. (2021), and Trinh et al. (2021). In addition, the effective control of corruption in banking institutions increases the position and power of women on the board (BfemleCCI), which promotes and enhances banking stability (0.25, p < 0.01) (Marie et al. 2021). The BdualCCI interaction term is negative (− 7.16) and significant at 1%. The moderating effect of CCI on the relationship between the number of board meetings (BmeetingsCCI) and banking stability is positive (0.62) and significant at 1%. The results for the developed countries are almost the same as those for the overall sample. Table 10 presents the impact of corruption control on the relationship between the board of directors and banking stability in emerging countries (the UAE, Qatar, Jordan, Bahrain, and Turkey). The results deviate slightly from those of developed countries. In fact, the interaction term BsizeCCI has no effect on the relationship between the board of directors and banking stability (0.03, p = 0.32). This result rejects H2. Regarding the composition of the board of directors, CCI was found to positively moderate the relationship between independent directors (BindpCCI) on the board and banking stability (0.03, p < 0.01). Similarly, the interaction term BfemaleCCI is positively and significantly correlated with banking stability (0.07, p < 0.01). The relationship between board duality (BdualCCI) and banking stability was negatively moderated by CCI (− 3.21, p < 0.1), whereas the impact of the number of board meetings (BmeetingsCCI) on banking stability was positively moderated by CCI (0.29, p < 0.01).Table 10 Impact of board of directors on banking stability in Emerging countries: Arellano-Bond dynamic panel-data estimation Dependent variable: Z-score Independent variables Model (2) Model (3) Model (4) Model (5) Model (6) Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Z-Score L 1 0.90 17.61 0.91* 11.08 0.90* 13.19 0.89* 82.03 0.89* 14.79 Constant 5.34* 5.10 8.44* 12.93 7.48* 15.88 8.23* 10.73 0.18 0.14 CCI − 2.31 − 1.26 4.34* 5.97 5.50* 11.82 5.38*** 4.42 1.22 0.69 Bsize 0.05 1.15 BsizeCCI 0.03 0.32 Bindp − 0.04** − 8.55 BindpCCI 0.03 3.00 Bfemale − 0.05* − 8.86 BfemleCCI 0.07 6.80 Bdual − 2.67* − 8.45 BdualCCI − 3.21 − 2.23 Bmeetings 0.43*** 7.36 BmeetingsCCI 0.29 3.16 Size 0.18* 2.83 0.22*** 3.46 0.03 0.53 0.17* 1.75 0.04 0.02 LTA 0.02 2.02 0.002 2.52 0.01* 1.74 0.02* 2.96 0.01 0.07 NPL − 0.10* − 9.13 − 0.09* − 7.04 − 0.12* − 10.73 − 0.10* − 8.06 − 0.11* − 8.93 GDP 0.19* 8.88 0.20 9.81 0.19* 10.26 0.21* 17.64 0.17* 7.11 Inflation − 0.03* − 6.51 − 0.03* − 6.16 − 0.02* − 5.00 − 0.03* − 8.65 − 0.04* − 11.39 Hausman 0.00 0.04 0.01 0.00 0.0009 Durbin–Watson 0.61 0.64 0.61 0.62 0.62 AR (1) test (p value) − 2.47(0.013) − 2.51(0.012) − 2.46 (0.013) − 2.45(0.013) − 2.48 (0.01) AR (2) test (p value) 1.60 (0.10) 1.68* (0.09) 1.52 (0.12) 1.62 (0.10) 1.60 (0.10) Sargan test (p value) 35.95 (0.96) 36.48 (0.95) 37.68 (0.94) 34.91 (0.97) 37.04 (0.95) * Significant at a level of 10%, Significant at a level of 5%, * Significant at a level of 1% Comparing the results for developed and emerging countries, the difference is generally pronounced at the interaction term level. In particular, the significance of the coefficient is lower for emerging countries than it is for developed countries. This indicates that the degree of significance of the overall sample is essentially explained by the results for developed countries, which have stricter regulations and force the implementation of stronger corporate governance to achieve banking stability. Impact of board of directors on banking stability: pre and post-crisis Table 11 demonstrates the moderating effect of corruption control on the relationship between the attributes of the board of directors and banking stability during the pre-crisis period between 2006 and 2008. The interaction terms BsizeCCI (0.97) and BindpCCI (0.35) are positive and significant at 1%. This could be explained by the fact that in a highly regulated environment, additional and independent directors strengthen corporate governance and subsequently enhance banking stability. As For the board diversity, the term BfemaleCCI is positive (0.31) and significant at 5%. This result confirms H2 and the moderating effect of CCI on the relationship between the presence of women on boards and bank stability. The coefficient of BdualCCI is negative (− 14.54) and significant at the 5%, whereas the BmeetingsCCI interaction coefficient is positive (0.15) and significant at the 5%. These results are similar to those obtained for the entire analysis period.Table 11 Impact of board of directors on banking stability: Pre- crisis Arellano-Bond dynamic panel-data estimation Dependent variable: Z-score Independent Variables Model (2) Model (3) Model (4) Model (5) Model (6) Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Z-Score L 1 0.66 3.14 0.64* 4.06 0.61* 3.33 0.64* 3.32 0.63* 3.74 Constant 1.90* 0.06 33.51* 1.08 21.90 0.75 32.05* 1.16 20.54* 0.61 CCI 16.08 0.53 3.96* 0.23 5.73* 0.54 0.12* 0.01 11.35 0.68 Bsize 1.07 0.73 BsizeCCI 0.97 0.43 Bindp 0.15 0.76 BindpCCI 0.35 1.04 Bfemale − 0.09 − 1.22 BfemleCCI 0.31* 1.16 Bdual 8.96* 1.90 BdualCCI − 14.54* − 0.78 Bmeetings 0.22* 0.43 BmeetingsCCI 0.15* 0.24 Size 1.33 0.56 0.18* 0.09 0.47* 0.22 0.31 0.15 0.20* 0.10 LTA 0.18 1.36 0.2 1.02 0.24 1.22 0.25 1.34 0.22 1.09 NPL − 0.52 − 0.85 − 0.41* − 0.61 − 0.44* − 0.58 − 0.45* − 0.61 − 0.40* − 0.58 GDP 0.07* 0.17 0.27** 0.85 0.24* 0.66 0.05* 0.02 0.19 0.57 Inflation − 0.02* − 0.22 − 0.04* − 0.84 − 0.04* -0.33 − 0.06 − 0.55 − 0.06 − 0.54 Hausman 0.002 0.0009 0.0001 0.004 0.0002 Durbin–Watson 1.58 1.63 1.79 1.63 1.66 AR (1) test (p value) − 0.85* (0.03) − 0.05*(0.95) 0.09 (0.07) 0.02 (0.97) 0.001* (0.09) AR (2) test (p value) 1.50 (0.11) 1.58 (0.32) − 1.62 (0.41) 1.49 (0.13) 1.37 (0.41) Sargan test (p value) 2.77 (0.59) 0.48 (0.92) 0.98 (0.80) 0.70 (0.87) 0.58 (0.89) Significant at a level of 10%, * Significant at a level of 5%, *** Significant at a level of 1% Table 12 presents the moderating effect of corruption control on the relationship between the attributes of the board of directors and banking stability during the post-crisis period between 2009 and 2016. The results indicate that the moderating effect of corruption control on the relationship between the board of directors and banking stability is weaker than that in the pre-crisis period. Only the impacts of board size and the number of board meetings are moderated by corruption control. More precisely, the interaction terms of BsizeCCI and BmeetingsCCI were positive (0.29, 0.48) and significant at the 5% and 1% levels, respectively. This could be explained by the crisis effect, which hindered and degraded the quality of banking governance, thus deteriorating banking stability during this period (Díaz and Huang 2017).Table 12 Impact of board of directors on banking stability: Post-crisis Arellano-Bond dynamic panel-data estimation Dependent variable: Z-score Independent variables Model (2) Model (3) Model (4) Model (5) Model (6) Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Coeff. Z-Statistic Z-Score L 1 0.65* 38.46 0.65* 31.54 0.66* 34.66 0.66* 38.16 0.67* 38.49 Constant 13.63* 2.93 12.79* 2.88 12.03* 3.20 11.31* 2.76 9.52* 2.61 CCI 5.84** 2.18 5.61* 1.84 1.73 1.10 1.13 0.61 6.72** 2.51 Bsize 0.16 0.08 BsizeCCI 0.29* 2.11 Bindp 0.09 1.28 BindpCCI 0.14 2.13 Bfemale 0.02 1.36 BfemleCCI 0.02 1.17 Bdual 3.07 1.03 BdualCCI − 0.63 − 0.31 Bmeetings 0.24 1.37 BmeetingsCCI 0.48* 3.72 Size 0.37 1.29 0.35 1.13 0.27 0.87 0.28 0.93 0.17 0.54 LTA 0.01 1.01 0.01 0.78 0.01 0.66 0.01 0.87 0.01 0.76 NPL − 0.63* − 6.13 − 0.59* − 5.32 − 0.62* − 5.61 − 0.64* − 6.78 − 0.67* − 7.28 GDP 0.05 0.66 0.06 0.72 0.03 0.38 0.03 0.42 0.05 0.68 Inflation − 0.11* − 2.84 − 0.14* − 3.53 − 0.11* − 2.95 − 0.11* 3.07 − 0.11* − 2.91 Hausman 0.001 0.04 0.008 0.05 0.06 Durbin–Watson 0.65 0.64 0.66 0.65 0.67 AR (1) test (p value) − 2.52 (0.01) − 2.50 (0.01) − 2.52 (0.01) − 2.52 (0.01) − 2.53 (0.01) AR (2) test (p value) − 0.79 (0.42) − 0.78 (0.43) − 0.81 (0.41) − 0.80 (0.42) − 0.78 (0.43) Sargan test (p value) 29.42 (0.79) 27.08 (0.63) 28.33 (0.84) 28.01 (0.35) 28.78 (0.32) * Significant at a level of 10%, Significant at a level of 5%, * Significant at a level of 1% Bank governance should be reformed to reduce the negative impact of financial crises on stability. A major challenge for policymakers is determining an optimal financial balance that can maximize the benefits of reforms while minimizing the challenges associated with their implementation (Ghosh 2018). Conclusion This study examines the impact of corporate governance on bank stability by considering the moderating role of corruption controls. Empirical analysis was conducted using a sample of banking establishments in OECD countries between 2006 and 2016. The empirical results combine CCI, which indirectly addresses the relationship between the board of directors and the insolvency risk of banks. These results provide a new way to diagnose and analyze the relationship between the governance of the board of directors and banking stability. Governments play a fundamental role in controlling corruption, which, in turn, influences the relationship between the board of directors and banking stability. Thus, reforming governance in the banking sector seems delicate. Moreover, it is necessary to ensure the good practice of controlling corruption in different countries, which influences banks to guarantee more solid financial systems based on good governance to improve solvency and banking stability. This study has several implications. The practical implication would help financial regulators in the different countries included in the analysis to determine the appropriate framework of effective board of directors’ rules and guidance and implement effective anti-corruption measures by their respective banking institutions. This will reduce insolvency risk, maintain banking stability, and strengthen banks’ resilience in the face of different crises, such as the COVID-19 pandemic. This would ensure sustainable economic development. The strategic implication is related to the significant role played by corruption in a bank’s stability and financial decision-making process, which leads to a competitive advantage. As a result, anti-corruption practices should be recognized as strategic decisions related to a bank’s stability. This study had two limitations. First, this study only considered the board of directors’ variables. To explain banking stability better, other mechanisms of banking governance could be considered, such as the audit committee and ownership structure. Second, a further possible extension of this research could adopt a comparative approach between the countries used in the sample, considering the difference between the banking systems and the regulatory quality of these countries. This could have different effects on the relationship between banking governance and bank stability and would inform appropriate corrective actions to be taken by each country. Exploration of these newly suggested axes will be the subject of future research. Funding This research was not funded by any funding agency. Declarations Conflict of interest 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 Acharya VV Richardson M Causes of the financial crisis Crit Rev 2009 21 2–3 195 210 10.1080/08913810902952903 Adams RB Ferreira D A theory of friendly boards J Finance 2007 62 1 217 250 10.1111/j.1540-6261.2007.01206.x Adams R Mehran H Is Corporate Governance Different for Bank Holding Companies? Econ Policy Rev 2003 9 123 142 Adnan SM Hay D Van Staden CJ The influence of culture and corporate governance on corporate social responsibility disclosure: a cross country analysis J Clean Prod 2018 198 820 832 10.1016/j.jclepro.2018.07.057 Aebi V Sabato G Schmid M Risk management, corporate governance, and bank performance in the financial crisis J Bank Finance 2012 36 3213 3226 10.1016/j.jbankfin.2011.10.020 Agrawal A Knoeber CR Firm performance and mechanisms to control agency problems between managers and shareholders J Financ Quant Anal 1996 31 3 377 397 10.2307/2331397 Alexandre H Paquerot M Efficacité des structures de contrôle et enracinement des dirigeants Finance Contrôle Stratégie 2000 3 2 5 29 Allais R Roucoules L Reyes T Governance maturity grid: a transition method for integrating sustainability into companies J Clean Prod 2017 140 213 226 10.1016/j.jclepro.2016.02.069 Amadeo K (2021) Subprime mortgage crisis, its timeline and effect. The Balance. December, 31. https://www.thebalance.com/subprime-mortgage-crisis-effect-and-timeline-3305745 Anderson CA Anthony RN The new corporate director 1986 New York Wiley Publication Anokhin S Schulze WS Entrepreneurship, innovation, and corruption J Bus Ventur 2009 24 5 465 476 10.1016/j.jbusvent.2008.06.001 Atangana Ondoa H The determinants of corporate corruption in Cameroon Econ Bull 2014 34 2 938 950 Barth JR, Caprio G, Levine R (1998) Financial regulation and performance: cross- country evidence. World Bank Policy Research Working Paper (2037) Bebchuk L Fried J Pay without performance 2004 Cambridge, MA Harvard University Press Bebchuk LA Cohen A Spamann H The wages of failure: executive compensation at Bear Stearns and Lehman 2000–2008 Yale J Regul 2010 27 257 282 Becht M Bolton P Röell A Why bank governance is different Oxf Rev Econ Policy 2011 27 3 437 463 10.1093/oxrep/grr024 Beck T Demirgüç-Kunt A Levine R Bank concentration and crises 2003 Washington The World Bank Beck T Demirgüç-Kunt A Levine R SMEs, growth, and poverty: cross- country evidence J Econ Growth 2005 10 3 199 229 10.1007/s10887-005-3533-5 Berger AN Kick T Schaeck K Executive board composition and bank risk taking J Corp Finance 2014 28 48 65 10.1016/j.jcorpfin.2013.11.006 Berger AN, Bouwman CHS (2014). Bank Liquidity Creation, Monetary Policy, and Financial Crises Working Paper. Berle AA Means GC The modern corporation and private property 1932 New Brunswick, NJ Transaction Blundell R Bond S Initial conditions and moment restrictions in dynamic panel data models J Econom 1998 87 1 115 143 10.1016/S0304-4076(98)00009-8 Blundell R Bond S GMM estimation with persistent panel data: an application to production functions Econom Rev 2000 19 3 321 340 10.1080/07474930008800475 Bourke P Concentration and other determinants of bank profitability in Europe, North America and Australia J Bank Finance 1989 13 1 65 79 10.1016/0378-4266(89)90020-4 Bourkhis K Nabi MS Islamic and conventional banks' soundness during the 2007–2008 financial crisis Rev Financ Econ 2013 22 2 68 77 10.1016/j.rfe.2013.01.001 Bouwman CH Corporate governance propagation through overlapping directors Rev Financ Stud 2011 24 7 2358 2394 10.1093/rfs/hhr034 Boyd JH Runkle DE Size and performance of banking firms: testing the predictions of theory J Monet Econ 1993 31 1 47 67 10.1016/0304-3932(93)90016-9 Campbell JL Why would corporations behave in socially responsible ways? An institutional theory of corporate social responsibility Acad Manag Rev 2007 32 3 946 967 10.5465/amr.2007.25275684 Campbell JY Bank insolvency and the problem of non-performing loans J Bank Regul 2007 9 1 25 45 10.1057/palgrave.jbr.2350057 Carter DA Simkins BJ Simpson WG Corporate governance, board diversity and firm value Financ Rev 2003 38 1 33 53 10.1111/1540-6288.00034 Chen M Jeon BN Wang R Wu J Corruption and bank risk-taking: evidence from emerging economies Emerg Markets Rev 2015 24 122 148 10.1016/j.ememar.2015.05.009 Choi J Asian financial crisis: financial, structural, and international dimensions Int Finance Rev 2000 1 3 4 10.1016/S1569-3767(00)01003-7 Čihák MM Hesse H Islamic banks and financial stability: an empirical analysis J Financ Serv Res 2010 38 2–3 95 113 10.1007/s10693-010-0089-0 Claessens S Yurtoglu BB Corporate governance in emerging markets: a survey Emerg Markets Rev 2013 15 1 33 10.1016/j.ememar.2012.03.002 Conger J Finegold D Lawler III EE CEO appraisals: holding corporate leadership accountable Organ Dyn 1998 27 1 7 20 10.1016/S0090-2616(98)90037-7 Daily CM Certo ST Dalton DR A decade of corporate women: some progress in the board room, none in the executive suite Strateg Manag J 1999 20 1 93 100 10.1002/(SICI)1097-0266(199901)20:1<93::AID-SMJ18>3.0.CO;2-7 Dang L Yang R Anti-corruption, marketization and firm behaviors: evidence from firm innovation in China Econ Polit Stud 2016 4 1 39 61 10.1080/20954816.2016.1152093 De Andres P Vallelado E Corporate governance in banking: the role of the board of directors J Bank Finance 2008 32 12 2570 2580 10.1016/j.jbankfin.2008.05.008 Demirgüç-Kunt A Detragiache E Monitoring banking sector fragility: a multivariate logit approach World Bank Econ Rev 2000 14 2 287 307 10.1093/wber/14.2.287 Demirgüç-Kunt A Huizinga H Bank activity and funding strategies: The impact on risk and returns J Financ Econ 2010 98 3 626 650 10.1016/j.jfineco.2010.06.004 Demirgüç-Kunt A Klapper L Singer D Ansar S Hess J The Global Findex Database 2017: measuring financial inclusion and the fintech revolution 2018 Washington The World Bank Demsetz RS Strahan PE Diversification, size, and risk at bank holding companies J Money Credit Bank 1997 29 3 300 313 10.2307/2953695 Dey A Engel E Liu X CEO and board chair roles: to split or not to split? J Corp Finance 2011 17 5 1595 1618 10.1016/j.jcorpfin.2011.09.001 Diallo B Corporate governance, bank concentration and economic growth Emerg Mark Rev 2017 32 28 37 10.1016/j.ememar.2017.05.003 Díaz V Huang Y The role of governance on bank liquidity creation J Bank Finance 2017 77 137 156 10.1016/j.jbankfin.2017.01.003 Dinç IS Politicians and banks: Political influences on government-owned banks in emerging markets J Financ Econ 2005 77 2 453 479 10.1016/j.jfineco.2004.06.011 Djebali N Zaghdoudi K Testing the governance-performance relationship for the Tunisian banks: a GMM in system analysis Financ Innov 2020 6 1 24 10.1186/s40854-020-00182-5 Dong Y Girardone C Kuo JM Governance, efficiency and risk taking in Chinese banking Br Account Rev 2017 49 2 211 229 10.1016/j.bar.2016.08.001 Elkington J Cannibals with forks 2002 Oxford Capstone Publishing Limited Oxford Elyasiani E Zhang L Bank holding company performance, risk, and “busy” board of directors J Bank Finance 2015 60 239 251 10.1016/j.jbankfin.2015.08.022 Erhardt NL Werbel JD Shrader CB Board of director diversity and firm financial performance Corp Gov Int Rev 2003 11 2 102 111 10.1111/1467-8683.00011 Erkens DH Hung M Matos P Corporate governanceinthe 2007–2008 financial crisis: evidence from financial institutions worldwide J Corp Finance 2012 18 2 389 411 10.1016/j.jcorpfin.2012.01.005 Benavent C, Evrard Y (2002) Extension du domaine de l'expérience. Décisions Market 7–11 Fakhfekh M Hachicha N Jawadi F Selmi N Cheffou AI Measuring volatility persistence for conventional and Islamic banks: an FI-EGARCH approach Emerg Markets Rev 2016 27 84 99 10.1016/j.ememar.2016.03.004 Fama EF Jensen MC Separation of ownership and control J Law Econ 1983 26 2 301 325 10.1086/467037 Fernandes C Farinha J Martins FV Mateus C The impact of board characteristics and CEO power on banks’ risk-taking: stable versus crisis periods J Bank Regul 2021 22 319 341 10.1057/s41261-021-00146-4 Forbes DP Milliken FJ Cognition and corporate governance: Understanding boards of directors as strategic decision-making groups Acad Manag Rev 1999 24 3 489 505 10.2307/259138 Fuente JA García-Sanchez IM Lozano MB The role of the board of directors in the adoption of GRI guidelines for the disclosure of CSR information J Clean Prod 2017 141 737 750 10.1016/j.jclepro.2016.09.155 Ghosh S Governance reforms and performance of MENA banks: are disclosures effective? Global Finance J 2018 36 78 95 10.1016/j.gfj.2018.01.002 Giammarino RM Lewis TR Sappington DE An incentive approach to banking regulation J Finance 1993 48 4 1523 1542 10.1111/j.1540-6261.1993.tb04766.x Gilani U Keasey K Vallascas F Board financial expertise and the capital decisions of US banks J Corp Finance 2021 71 102091 10.1016/j.jcorpfin.2021.102091 Gorton G Rosen R Corporate control, portfolio choice, and the decline of banking J Finance 1995 50 5 1377 1420 10.1111/j.1540-6261.1995.tb05183.x Grossman SJ Hart OD Implicit contracts under a symmetric information Q J Econ 1983 90 123 156 10.2307/1885377 Grove H Patelli L Victoravich LM Xu P Corporate governance and performance in the wake of the financial crisis: evidence from US commercial banks Corp Gov Int Rev 2011 19 418 436 10.1111/j.1467-8683.2011.00882.x Gulamhussen MA Santa SF Female directors in bank boardrooms and their influence on performance and risk-taking Global Financ J 2015 28 10 23 10.1016/j.gfj.2015.11.002 Hermalin B Weisbach M Boards of directors as an endogenously determined institution: a survey of the economic literature Econ Policy Rev 2003 9 7 26 Holmstrom B Moral hazard and observability Bell J Econ 1979 10 1 74 91 10.2307/3003320 Houston JF Lin C Lin P Ma Y Creditor rights, information sharing, and bank risk taking J Financ Econ 2010 96 3 485 512 10.1016/j.jfineco.2010.02.008 Iyer R Puri M Understanding bank runs: the importance of depositor-bank relationships and networks Am Econ Rev 2012 102 4 1414 1445 10.1257/aer.102.4.1414 Jensen MC The modern industrial revolution, exit, and the failure of internal control systems J Finance 1993 48 3 831 880 10.1111/j.1540-6261.1993.tb04022.x Jensen MC Meckling WH Theory of the firm: Managerial behavior, agency costs and ownership structure J Financ Econ 1976 3 4 305 360 10.1016/0304-405X(76)90026-X Kanter R Men and women of the corporation 1977 New York Basic Books Karkowska R Acedański J The effect of corporate board attributes on bank stability Port Econ J 2020 19 99 137 10.1007/s10258-019-00162-3 Kaufmann D, Kraay A, Mastruzzi M (2010) The worldwide governance indicators: methodology and analytical issues. World Bank Policy Research Working Paper No. 5430. The World Bank, Washington, DC Kervin JB Methods for business research 1992 New York Harper Collins Khan MS Scheule H Wu E Funding liquidity and bank risk taking J Bank Finance 2017 82 September 203 216 10.1016/j.jbankfin.2016.09.005 Khwaja AI Mian A Do lenders favor politically connected firms? Rent provision in an emerging financial market Q J Econ 2005 120 4 1371 1411 10.1162/003355305775097524 Kiel GC Nicholson GJ Board composition and corporate performance: How the Australian experience informs contrasting theories of corporate governance Corp Gov Int Rev 2003 11 3 189 205 10.1111/1467-8683.00318 Kirkpatrick G The corporate governance lessons from the financial crisis OECD J Financ Market Trends 2009 1 61 87 10.1787/fmt-v2009-art3-en Kosmidou K, Tanna S, Pasiouras F (2005) Determinants of profitability of domestic UK commercial banks: panel evidence from the period 1995–2002. In: Money Macro and Finance (MMF) Research Group Conference, vol 45, pp 1–27 La Porta RL Lopez-de-Silanes F Shleifer A Vishny RW Law and finance J Polit Econ 1998 106 6 1113 1155 10.1086/250042 La Porta R Lopez-de-Silanes F Shleifer A Vishny RW Investor protection and corporate valuation J Finance 2002 57 3 1147 1170 10.1111/1540-6261.00457 Ladipo D Nestor S Bank boards and the financial crisis. A corporate governance study of the 25 largest European banks 2009 London Nesto Advisors Ltd Laeven L Levine R Bank governance, regulation and risk taking J Financ Econ 2009 93 2 259 275 10.1016/j.jfineco.2008.09.003 Lasfer MA The interrelationship between managerial ownership and board structure J Bus Finance Account 2006 33 7–8 1006 1033 10.1111/j.1468-5957.2006.00600.x Lepetit L Nys E Rous P Tarazi A Bank income structure and risk: an empirical analysis of European banks J Bank Finance 2008 32 8 1452 1467 10.1016/j.jbankfin.2007.12.002 Liang Q Xu P Jiraporn P Board characteristics and Chinese bank performance J Bank Finance 2013 37 8 2953 2968 10.1016/j.jbankfin.2013.04.018 Linck JS Netter JM Yang T The determinants of board structure J Financ Econ 2008 87 2 320 328 10.1016/j.jfineco.2007.03.004 Lipton M Lorsch JW A modest proposal for improved corporate governance Bus Lawyer 1992 48 1 59 77 Lombardi R Corporate governance e governo della complessità: un approccio integrato 2012 Napoli Edizioni Scientifiche Italiane Lombardi R Trequattrini R Cuozzo B Cano-Rubio M Corporate corruption prevention, sustainable governance and legislation: first exploratory evidence from the Italian scenario J Clean Prod 2019 217 666 675 10.1016/j.jclepro.2019.01.214 Louhichi A Boujelbène Y Credit risk, managerial behaviour and macroeconomic equilibrium with in dual banking systems: interest-free versus interest-based banking industries Res Int Bus Finance 2016 38 104 121 10.1016/j.ribaf.2016.03.014 Lozano R Towards better embedding sustainability into companies’ systems: an analysis of voluntary corporate initiatives J Clean Prod 2012 25 14 26 10.1016/j.jclepro.2011.11.060 Marie M Kamel H Elbendary I How does internal governance affect banks’ financial stability? Empirical evidence from Egypt Int J Discl Gov 2021 18 240 255 10.1057/s41310-021-00110-8 McInerney-Lacombe N Bilimoria D Salipante PF Vinnicombe S Championing the discussion of tough issues: how women corporate directors contribute to board deliberations Women on corporate boards of directors: International research and practice 2008 Glos Edward Elgar 123 139 Mckinsey Global Institute (2018) A decade after the global financial crisis: what has (and hasn’t) changed? Accessible at: https://www.mckinsey.com/~/media/McKinsey/Industries/Financial%20Services/Our%20Insights/A%20decade%20after%20the%20global%20financial%20crisis%20What%20has%20and%20hasnt%20changed/MGI-Briefing-A-decade-after-the-global-financial-crisis-What-has-and-hasnt-changed.ashx Mehran H, Morrison A, Shapiro J (2011) Corporate governance and banks: what have we learned from the financial crisis? Federal Reserve Bank of New York, Staff Report No. 502 Minton BA, Taillard J, Williamson J (2011) Do independence and financial expertise of the board matter for risk taking and performance? Charles A. Dice Center Working Paper No. 2010–14, Fisher College of Business Working Paper No. 2010-03-014, Available at SSRN: 10.2139/ssrn.1661855 Molyneux P Thornton J Determinants of European bank profitability: a note J Bank Finance 1992 16 6 1173 1178 10.1016/0378-4266(92)90065-8 Murphy KJ (1999) Executive compensation. In: Handbook of labor economics, vol 3, pp 2485–2563 Owen AL Temesvary J The performance effects of gender diversity on bank boards J Bank Finance 2018 90 50 63 10.1016/j.jbankfin.2018.02.015 Palvia A Vähämaa E Vähämaa S Female leadership and bank risk-taking: Evidence from the effects of real estate shocks on bank lending performance and default risk J Bus Res 2020 117 897 909 10.1016/j.jbusres.2020.04.057 Park YW Shin HH Board composition and earnings management in Canada J Corp Finance 2004 10 3 431 457 10.1016/S0929-1199(03)00025-7 Pathan S Faff R Does board structure in banks really affect their performance? J Bank Finance 2013 37 5 1573 1589 10.1016/j.jbankfin.2012.12.016 Pathan S Skully M Wickramanayake J Board size, independence and performance: an analysis of Thai banks Asia-Pac Financ Markets 2007 14 3 211 227 10.1007/s10690-007-9060-y Persson A Rothstein B Teorell J Why anticorruption reforms fail—systemic corruption as a collective action problem Governance 2013 26 449 471 10.1111/j.1468-0491.2012.01604.x Rachdi H El Gaied M L'impact de l'indépendance et de la dualité du Conseil d'Administration sur la Performance des Entreprises : Application au Contexte Américain Revue Libanaise De Gestion Et D'économie 2009 2 3 127 150 10.1016/S1999-7620(09)70018-7 Raouf H, Ahmed H (2020) Risk governance and financial stability: a comparative study of conventional and Islamic banks in the GCC. Global Finance J, 100599 Rosman R Wahab NA Zainol Z Efficiency of Islamic banks during the financial crisis: an analysis of Middle Eastern and Asian countries Pac Basin Financ J 2014 28 76 90 10.1016/j.pacfin.2013.11.001 Sapienza P The effects of government ownership on bank lending J Financ Econ 2004 72 2 357 384 10.1016/j.jfineco.2002.10.002 Schrippe P Duarte Ribeiro JL Preponderant criteria for the definition of corporate sustainability based on Brazilian sustainable companies J Clean Prod 2019 209 10 19 10.1016/j.jclepro.2018.10.001 Shleifer A Vishny RW Large shareholders and corporate control J Polit Econ 1986 94 3 461 488 10.1086/261385 Skully MT (2002) Banking and corporate governance: some prudential issues. Presented in Second International Conference on Banking and Finance in Greece, August 9–11 Spyromitros E Panagiotidis M The impact of corruption on economic growth in developing countries and a comparative analysis of corruption measurement indicators Cogent Econ Finance 2022 10 1 2129368 10.1080/23322039.2022.2129368 Trabelsi MA The impact of the Financial crisis on the global economy:can the Islamic financial system help? J Risk Finance 2011 12 1 15 25 10.1108/15265941111100049 Trinh VQ Elnahass M Cao ND The value relevance of bank cash Holdings: the moderating effect of board busyness J Int Financ Mark Inst Money 2021 73 101359 10.1016/j.intfin.2021.101359 Uddin A Chowdhury MAF Sajib SD Masih M Revisiting the impact of institutional quality on post-GFC bank risk-taking: Evidence from emerging countries Emerg Mark Rev 2020 42 100659 10.1016/j.ememar.2019.100659 Vafeas N Board meeting frequency and firm performance J Financ Econ 1999 53 1 113 142 10.1016/S0304-405X(99)00018-5 Watson WE Kumar K Michaelsen LK Cultural diversity's impact on interaction process and performance: comparing homogeneous and diverse task groups Acad Manag 1993 36 590 602 10.2307/256593 Van Greuning H Brajovic-Bratanovic S Analyse et gestion du risque bancaire : un cadre de référence pour l'évaluation de la gouvernance d'entreprise et du risque financier 2004 Washington The World Bank 1 402 Vilanova L Financial distress, lender passivity and project finance: the case of Eurotunnel Lender Passiv Project Finance Case Eurotunnel 2006 10.2139/ssrn.675304 Wiersema MF Bantel KA Top management team demography and corporate strategic change Acad Manag J 1992 35 1 91 121 10.2307/256474 World Bank (2020) Enhancing government effectiveness and transparency: the fight against corruption. Accessible at: https://documents1.worldbank.org/curated/en/235541600116631094/pdf/Enhancing-Government-Effectiveness-and-Transparency-The-Fight-Against-Corruption.pdf Worrell MD, Maechler AM, Mitra MS (2007) Decomposing financial risks and vulnerabilities in Eastern Europe (No. 7–248). International Monetary Fund Xu G Yano G How does anti-corruption affect corporate innovation? Evidence from recent anti-corruption efforts in China J Comp Econ 2017 45 3 498 519 10.1016/j.jce.2016.10.001	0	PMC9734402	NO-CC CODE	2022-12-14 23:28:28	no	Rev Manag Sci. 2022 Dec 3;:1-38	utf-8	null	null	null	oa_other
==== Front Neural Comput Appl Neural Comput Appl Neural Computing & Applications 0941-0643 1433-3058 Springer London London 8029 10.1007/s00521-022-08029-z S.I. : Neural Computing for IOT based Intelligent Healthcare Systems Threat Object-based anomaly detection in X-ray images using GAN-based ensembles http://orcid.org/0000-0002-8613-0361 Kolte Shreyas [email protected] 1 Bhowmik Neelanjan 2 Dhiraj 3 1 grid.418391.6 0000 0001 1015 3164 Birla Institute of Technology and Science, Pilani, India 2 grid.8250.f 0000 0000 8700 0572 Department of Computer Science, Durham University, Durham, UK 3 CSIR-CEERI, Pilani, India 9 12 2022 116 10 10 2021 29 10 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 problem of detecting dangerous or prohibited objects in luggage is a very important step during the implementation of Security setup at Airports, Banks, Government buildings, etc. At present, the most common techniques for detecting such dangerous objects are by using intelligent data analysis algorithms such as deep learning techniques on X-ray imaging or employing a human workforce for inferring the presence of these threat objects in the obtained X-ray images. One of the major challenges while using deep-learning methods to detect such objects is the lack of high-quality threat image data containing the “dangerous” objects (objects of interest) versus the non-threat image data in practical scenarios. So, to tackle this data scarcity problem, anomaly detection techniques using normal data samples have shown great promise. Also, among the available Deep Learning Strategies for anomaly detection for computer vision applications, generative adversarial networks have achieved state-of-the-art results. Considering these insights, we adopted a newly proposed architecture known as Skip-GANomaly and devised a modified version of it by using a UNet++ style generator which performed better than Skip-GANomaly, getting an AUC of 94.94% on Compass-XP, a public X-ray dataset. Finally, for targeting better latent space exploration, we combine these two architectures into an Ensemble, which gives another boost to the performance, getting an AUC of 96.8% on the same Compass-XP, a public X-ray dataset. To further validate the effectiveness of ensemble-based architecture, its performance was tested on patch-based training data on a subset of randomly chosen images of another huge public X-ray dataset named as SIXray, and obtained an AUC of 75.3% on this reduced dataset. To demonstrate the prowess of the discriminator and to bring some explainability to the working of our ensemble, we have used Uniform Manifold Approximation and Projection to plot the latent-space vectors for the dangerous and non-dangerous objects of the test-set; this analysis indicates that the Ensemble learns better features for separating the anomalous class from non-anomalous with respect to the individual architectures. Thus, our proposed architecture provides state-of-the-art results for threat object detection. Most importantly, our models are able to detect threat objects without ever being trained on images containing threat objects. Keywords Generative adversarial networks (GANs) Anomaly detection GANomaly Skip-GANomaly Ensemble of GANs X-ray images Threat-object detection Compass-XP SixRay ==== Body pmcIntroduction Anomaly Detection poses the following problem statement: Given a set of data points/observations in a vector space, the task is to detect the outlier points or those points which, with some metric, deviate from “normal” behavior. Several traditional techniques have been employed for anomaly detection, starting from simple statistical techniques like the median of the data points or trimmed-mean methodologies and going to Density-based techniques, one-class Support Vector Machines, etc. However, most of these techniques are either purely unsupervised techniques or purely machine learning techniques, not involving deep learning or other sophisticated methodologies with inherent feature extraction options. The two major areas where Anomaly Detection in computer vision is required are Security applications and Industrial applications. In security-related applications, the task is to detect threat objects in luggage carried in airports or other offices, firms, and institutions, whereas, in the Industrial domain, the task is to detect damaged pieces or incorrectly manufactured goods in factories. In the field of security applications, numerous methods have been developed involving Convolutional Neural Networks, which mainly focused on the problem as an image classification problem with the labeled training data set consisting of images of both threat and non-threat objects. A major drawback of these models is the scarcity of threat object instances in the available datasets, i.e., Bags/luggage containing such threat objects is extremely rare as compared to those which do not contain such objects. Also, the supervised class of deep learning methods needs a large amount of class-balanced labeled data in order to train the model effectively. To handle this problem of nonavailability of threat data, deep neural networks such as Generative Adversarial Network (GAN) [1]-based methodologies are developed where it is possible to define a training set using healthy samples containing only the benign object instances, i.e., non-threat objects such that the model will learn the feature space of normal data samples and all threat objects if get encountered in test data will be considered anomaly instance by the model. We discuss GANs and Anomaly detection methods based on them in the following section. The major contributions of this paper are the following:- ∗ Development of a modified architecture based on existing GAN-based anomaly detection architecture for detecting X-Ray images containing threat objects in airport security. ∗ Development of an Ensemble architecture for detecting X-Ray images containing threat objects in airport security. The developed ensemble-based architecture achieves a state-of-the-art result on the Compass-XP dataset, nearly equaling the performance of human annotators on the dataset and also proves to be effective in detecting threat instances on another much complex SIXray dataset. The paper is organized into the following sections:- ∗ Section 2 describes the related work literature on threat detection for detecting anomalies in X-Ray Images. ∗ Section 3 describes the 2 datasets used for testing some of the existing methods and our proposed methods. ∗ Section 4 describes existing methodologies and how we developed new architectures based on these methodologies. ∗ Section 5 describes the results of the developed models on two security datasets and their discussions. ∗ Section 6 consists of the conclusion of the paper along with ideas for future work. Related work Generative Adversarial Networks (GANs) [1], introduced in 2014 has the following basic paradigm: There are 2 networks called as, a Generator and a Discriminator. The Generator network is trained to generate data that mimics real data (e.g., images, text, etc.). The Discriminator is trained to distinguish between data generated by the generator and real data. The two networks are made to compete against each other in the training phase (hence the term “Adversarial Networks”). The Generator tries to maximize a certain metric of similarity between the data it generates and the real data, while the Discriminator tries to minimize a certain loss function based on how well it could distinguish between real data and data generated by the discriminator. Also worth noting is that the Generator receives a input as a vector Z from a Vector Space (known as the Latent Space) and generates data using this input vector, and the Discriminator obviously receives the particular instances of real and Generator-generated data. Hence, when a Generator’s training is completed, the latent space contains a distribution of data that would be close to the distribution of the real data that was fed into the discriminator. All of these aforementioned features of GANs can be used in several ways to suit a particular application. Equation (1) shows the goal of the GAN during training.1 minGmaxD(Ex∼pdata(x)[log(D(x))]+Ez∼pz(z)[log(1-D(G(z)))]) In 2017, the model AnoGAN [2] was introduced as one of the first attempts to use GANs for Anomaly Detection in Image Data. The basic methodology was the following: To train the GAN on “healthy,” i.e., non-anomalous images. When an image is fed to the network, the first step is to find the best possible encoding for the image in the Latent Space using an Iterative Algorithm. Once the encoding is found, the Generator constructs an image based on the encoding. Finally, the generated image and the original image that was fed to the encoder are passed to the Discriminator, which tries to distinguish between the two; Also note that the Discriminator is trained to maximize a certain “difference score” between the original and the generated image. The training aims to Train the Generator to produce images that are indistinguishable from actual healthy images (i.e., the Discriminator should fail to distinguish between the produced and actual images). Thus, the Generator learns the distribution of “healthy” images and this is encoded into the Latent Space. Now, when an unseen image is fed to the generator, it will be encoded into the latent space and will be constructed as if it belonged to the distribution of the “healthy” (non-anomalous) images. The Discriminator then receives both these images and finds out the “difference score” between the images. If the difference score is greater than a certain threshold, the image is classified as anomalous based on the fact that it significantly deviates from the distribution of “healthy” images as learned by the Generator. The Generator is a DC-GAN decoder and the Discriminator is a Convolutional Classifier. The technical goal for the model is to learn the distribution, {X} of the “healthy” images, i.e., given a “healthy” image with latent space encoding {x}, to find the best latent space vector {z} in the latent space {Z} that, when fed to the generator, would allow the generator to generate an image most similar to the “healthy” image and closest to {x}. The algorithm is to optimize (using gradient descent) over K steps by iterating over k=1,2,…,K to find the best-generated image (by finding the best latent space vector as described before). The best image generated will minimize the Residual Loss as mentioned in Eq. (2):-2 LR(zk)=∑k=1K\|x-G(zk)\|. An important point worth noting here is the method used for finding the encoding, that is the Iterative Method, was slow. This greatly increased the Inference time for finding whether an image is anomalous. This led to the introduction of f-AnoGAN, a more efficient architecture that helped reduce the inference time by upgrading the DC-GAN in ano-GAN to a Wasserstein-GAN and adding a Convolutional Auto-encoder to map the image to the latent space (i.e., the iterative process is replaced by a learned mapping). The next methodology introduced was EGBAD (Efficient GAN-Based Anomaly Detection) [3], a Bi-GAN [4]-based architecture that eliminated the Iterative Process and replaced it with an Encoder-Decoder-based architecture which encoded images to map to a vector in a Latent Space and a Decoder which reconstructs the Image. The Discriminator receives as an input two vectors Z1 and Z2 from the latent space. Z1 corresponds to the encoding of the real image and Z2 corresponds to that of the Generator-generated Image. The Discriminator does not have this information and is thus expected to find out which vector corresponds to which image. The most important advantage of EGBAD over AnoGAN was the huge improvement in Inference Time. Following this, GANomaly [5] was introduced. This method consists of a GAN along with an Adversarial Autoencoder and has a pipeline similar to that of EGBAD. An Image would pass through the Encoder-Decoder architecture and thus it’s Latent Space Vector Z will be produced along with the reconstructed Image. Then, an adversarial auto-encoder would produce the Latent Space Vector of the Reconstructed Image. These inputs will then be used in training the Generator The Encoder-Decoder Network as well as the Discriminator (which is the same as that of DCGAN [6]). The model when tested on datasets like MNIST and CIFAR-10 produced better results than both AnoGAN and EGBAD while also having a reduced Inference time than EGBAD and much faster than AnoGAN. Later on, Skip-GANomaly [7] was introduced. This model removed the extra adversarial encoder present in GANomaly, and instead used a U-Net [8] style Generator with Skip-Connections between the Encoder and Decoder of the Generator, and again had the same DCGAN-based Discriminator. The Anomaly Score introduced by Skip-GANomaly is shown in Eq. (3) as follows:-3 A=λ∗Rec+(1-λ)∗Lat Here, Rec stands for the Reconstruction Score, which is simply a pixel-by-pixel difference of the Original (Input) image of the generator and the reconstructed image, while Lat stands for a value derived from the difference in the Latent Space Vectors of the two images. The parameter λ has been derived from experimentation and the best value found out was λ = 0.9. The final prediction of whether or not an object is anomalous is determined by finding the threshold that gives the largest AUC under the ROC. Other methods which tackled related problems include ComboGan_Xray [9] and Ensemble GAN models [10] for Anomaly Detection. ComboGan_Xray [9] worked on different combinations of generators and discriminators from different GAN architectures, e.g., An Auto-encoder acting as the Generator plus(+) the Discriminator of DCGAN (dubbed the “AE+DCGAN” network) and a network with the Generator of BiGAN and the Discriminator of α-GAN (dubbed the “BiGAN + α-GAN” network. These networks were tried on X-ray Images of human hands for anomaly detection and yielded better results than vanilla GANs (i.e., GANs of pure architecture rather than combinations from different architectures). Hence, ensemble_GAN [10] introduced the methodology of training and using multiple GANs of the same architecture for anomaly detection. MRI-GAN [11] and 3D-MRIGAN [12] focused on anomaly detection in MRI and 3D MRI Images, respectively, and were based on AnoGAN and f-AnoGAN, respectively. More recently, Jensen et. al. [13] developed transfer learning strategies using pre-trained convolution neural networks for feature extraction from a custom-made dataset of 16-bit and 8-bit fuel-cell X-ray images for the detection of 11 classes of anomalies. They used balanced accuracy as the metric of model evaluation. Among the most recent methods involved in Anomaly detection using GANs or other forms of adversarial networks are RANDGAN [14], WeaklyAD [15], and AnoSeg [16]. RANDGAN was implemented for the detection of COVID-19 from chest X-ray images; the unknown (anomalous) class was the Covid-19 class of images and used transfer-learning-based image segmentation as a pre-processing step. WeaklyAD is a spectral-constrained GAN for hyperspectral anomaly detection, i.e., the model is trained to detect anomalies in an image by generating images with homogenization of background (non-anomalous class) and anomaly saliency. AnoSeg focused on the task of anomaly segmentation for detecting defects in large-scale industrial manufacturing processes using 3 novel techniques combined: hard augmentation, self-supervised learning, and pixel-wise adversarial losses. Datasets The primary aim of adopting the threat detection algorithms is to raise the alarm on every instance of threat item in the scanned image data and the performance of such algorithms majorly depends on the quality of data used for their training. However, the availability of threat item instances is very limited due to the lesser chances of their occurrence in the normal scene as compared to nondangerous (benign) item instances. So, training a threat detection algorithm using a skewed dataset having fewer threat item instances and more normal instances may lead to poor detection performances or suffers from biasing effects of abundant class instances. So, in order to deal with this data skewness, the threat item detection algorithms can be trained using normal data instances and any threat item instance will therefore be treated as an anomaly by the trained model thereafter. Two such public datasets have been explored where there are sufficient cases of normal object instances and thus the model can be trained using only the normal object samples and their performance can be evaluated after the model gets trained by the instances of threat item class. Compass-XP We have used the COMPASS-XP Dataset [17], which consists of X-Ray images of Luggage Bags with one Item in each bag, for our models. There are 6 types of Images in the dataset: Original Photograph, Gray-Scale, False Duo-Color, and three other types based on different densities. In totality, there are 1901 unique images for each of the 6 types (Thus, making it a total of 11,406 images). Each of these images belongs to a particular class and is labeled Dangerous or Non-dangerous based on whether the class of objects is prohibited or non-prohibited to be carried in Airplanes, etc. There are 334 Non-dangerous classes (examples of such classes include cardigans and other clothes, torches, etc.) and 35 dangerous classes (examples include lighters, knives, etc.). The total “Dangerous” Images are 258 while the total “Non-Dangerous” images are 1643. The research paper [17] which introduced this dataset also tried several methods for detecting and classifying dangerous objects in the dataset including several Image Processing and Machine Learning methods. Their best-achieved result was a median AUC of 81-83%, achieved using Deep-Learning Methods (Convolutional Neural Networks for Binary Classification). It is also worth noting that [17] had divided the Compass-XP dataset into a 4:1 training and testing ratio with both training and testing set consisting of Dangerous and Non-Dangerous objects when training their Deep-Learning models for achieving their best results. Also worth noting is the fact that human experts were also tested on the dataset, and their predictions achieved an average AUC of 97%. Some sample images belonging to the normal (i.e., non-dangerous) classes from this dataset are shown in Fig. 1.Fig. 1 Nonthreat category Sample Images from the CompassXP dataset SIXray The SIXray [18] dataset contains 10,59,231 X-ray images from six common categories of prohibited items, namely, gun, knife, wrench, pliers, scissors, and hammer. Unlike the Compass-XP dataset, the SIXRay dataset has large images with multiple objects present in a single bag. Due to computational constraints in terms of available memory of GPUs, we selected a subset of randomly selected nonthreat and images containing threat objects from the dataset. We then generated a dataset by extracting random patches of size 256 × 256 from the images to generate a total of 90k non-threat patch-based images and 10k patch-based images containing threat objects. The 90k images were divided into subsets of size 80k and 10k, respectively, for the training and the test set. Thus, our training set consisted of 80k nonthreat images and the testing set consisted of 10k images containing threat objects and 10k nonthreat images. We used this dataset for training all the methodologies, i.e., Skip-GANomaly, modified Skip-GANomaly and their Ensemble-based architectures. Some sample images belonging to the normal (i.e., non-dangerous) classes from this dataset are shown in Fig. 2.Fig. 2 Nonthreat Patch Image Samples from the SIXray dataset Methodologies Among the state-of-the-art implementations of GANs for Anomaly detection, we narrowed down our initial studies to two architectures: GANomaly [5] and SkipGANomaly [7]. After some initial comparative experimentation on these methods (described in Sect. 3.2), we decided to move forward with SkipGANomaly as the base model for developing the modified version of it inorder to obtain improved performance. Sections 4.3, 4.4, and 4.5 describes the SkipGANomaly method, Modified SkipGANomaly (proposed), and an Ensemble Network of these two methodologies, respectively. Before this, however, we proceed to describe our training conditions on the two datasets (Compass-XP and SIXray) in Sect. 4.1. Approaches and Dataset Details Due to the adopted training methodologies, the models needed no “dangerous” (positive) class images for the purpose of training. This adopted training methodology has dual benefits first it removes the requirement of obtaining the training sample data for all the classes which is very difficult in this case and second the intraclass variance in the samples of the majority class maximum benefits the model training process. Compass-XP dataset Our Training conditions for the Compass-XP dataset were as follows: The Test set consisted of all of the 258 “Dangerous” images along with 258 “Non-Dangerous” images, where the “Non-Dangerous” images were one each from 258 randomly chosen “Non-Dangerous” classes. The rest of all the Non-Dangerous images were put into the training set. Since the training set size was not significantly larger than the Test-Set size, we augmented the training set with transformation techniques such as horizontal and vertical flips and rotation, such that not only the training set gets augmented, but also the bias created by the higher representation of some classes with respect to others would be removed. Thus, we increased the training set size from 1643 unique images to about 10,500 images. SIXRay dataset Since the SIXRay dataset was already large enough as compared to Compass-XP, no augmentation was needed for the training set. It consists of 100k images, out of which 10k were of the dangerous categories (i.e., they contained an object from one of the six dangerous categories mentioned in Sect. 3.2) and the remaining 90k were nonthreat images. Hence, a random and approximately equal subset of 10k images was chosen from this set of 90k images to be part of the test set and the remaining 80k images were chosen to serve as the training set. Initial experiments We experimented with the two most recently introduced models, GANomaly and Skip-GANomaly for the detection of Dangerous objects in X-ray images. For comparing which of the two is better, we performed some initial experiments with GANomaly and Skip-GANomaly being trained and tested on different batch sizes (Batch Normalization was not used, the batch sizes were changed simply based on computing environment constraints) and different Image sizes (All input images would be resized to a particular image size) are summarized in Table 1. To find the best suited batch size, we trained the models for fixed number of epochs for all the experiments. The values in Table 1 represents the AUC of the ROC.Table 1 Initial experiments for performance comparison of GANomaly and Skip-GANomaly Image size, batch size GANomaly AUC Skip-GANomaly AUC 32 × 32, 256 0.478 0.487 64 × 64, 64 0.493 0.512 128 × 128, 16 0.538 0.589 256 × 256, 4 0.594 0.710 512 × 512, 1 0.634 0.789 From this, we concluded that the Skip-GANomaly performs significantly better than GANomaly on this dataset. We thus dedicated further efforts to training Skip-GANomaly for getting the best possible results. Skip-GANomaly (U-Net-based generator) As stated earlier in the Introduction section, Skip-GANomaly has a U-Net-based generator. We trained the model on the dataset for a larger number of epochs (35–40 epochs). We again tried varied batch sizes and image sizes. The results are summarized in Table 2. As is evident from Table 2, we used 2 Image Sizes. These were chosen based on the fact that most of the images (greater than 95%) were of the size range of 200–600 pixels in one dimension and 400–900 in the other dimension. Also, the fact that the images contained only a single object meant that slight reductions in size would lead to little or no loss in the features. We used different batch sizes (Batch normalization was used here), and obtained different results. The best results were obtained on an image size of 256 × 256 and a batch size of 16. Whenever an increase in batch size gave improved results, we tried increasing the batch size. We chose an initial batch size of 8 because of the fact that when constructing the training set, a large fraction of the images were augmented 8 times.Table 2 Initial experiments for determination of correct batch-size and image-size combination Image size, batch size AUC 256 × 256, 8 0.742 256 × 256, 16 0.948 256 × 256, 32 0.715 512 × 512, 8 0.804 52 × 512, 16 0.795 Bold values indicate the highest AUC achieved by a model Fig. 3 Basic SkipGANomaly architecture [7] Skip-GANomaly with a modified generator Skip-GANomaly [7] as shown in Fig. 3 uses a UNet-based generator. The UNet++ [19] architecture is a modified version of UNet, primarily consisting of multiple Nested UNet models of several sizes, created by modifying the skip-pathways. By adopting the UNet++ based generator, a modified SkipGANomaly based architecture as shown in Fig. 4 was designed. This allowed the following two advantages: The foremost aim was reducing the semantic gap between the encoder and decoder sub-networks of UNet, and a secondary aim of being allowed to use multiple levels of features for predicting the final, thus allowing fast inference from obtained features for predicting the segmentation. Of these aims, our major inspiration behind selecting UNet++ to build a modified generator was the same as the foremost aim, i.e., to reduce the Semantic Gap between features learned by the encoder and decoder sub-networks. The method achieved a slight improvement over the original Skip-GANomaly. Fig. 4 illustrates the modified architecture. In this figure, “x” represents the input image for training/inference, and the italicized “x” is the reconstructed image from the generator.Fig. 4 Modified SkipGANomaly architecture Ensemble network The modified generator performed slightly better than the original Skip-GANomaly architecture. However, to produce even better results, we constructed an ensemble of the 2 architectures for finding out the anomalous objects. The ensemble consists of the two generators (The generator of the original SkipGANomaly and the generator of the modified architecture, i.e., the UNet++ architecture) and the two discriminators. Each generator can be connected to each discriminator, as can be seen in Fig. 5. The reason we chose to form an ensemble for improving the AUC is because of the successful results of [10], where ensembles of a uniform individual architecture were trained and improved results were obtained. However, our ensemble consists of 2 different types of generators, and also we first load each component of the ensemble with the pre-trained weights of the individual architectures (hence justifying the need for 2 discriminators instead of 1). Another reason is that when the discriminators receive inputs from both these generators, their ability to map images to the latent space is boosted. This will be demonstrated by our Uniform Manifold Approximation and Projection (U-MAP) analysis in Sect. 5. As we were initializing the ensemble with the pre-trained weights of the trained individual architectures, we have used a smaller training set (removing all augmented images at 45° rotation) of the size of about 5000 images for Compass-XP. During training, the training set is first divided into batches of images. Then, for each batch, a generator and a discriminator network are randomly chosen from the 2 choices available for each. Then, these chosen networks are trained. The algorithm for testing is also similar. The Algorithm 1 shown describes the training of the ensemble. For the Testing method, the test set remained the same as before. We also reduced the learning rate to 10-6, so that the training would be stable but slow. The model achieved stable performance after 5 epochs of training. The final AUC achieved was 96.8%, significantly better than the modified architecture. Algorithm 2 shown describes the testing method. The block level representation of the proposed ensemble-based architecture is shown in Fig. 5.Fig. 5 High-level view of the ensemble architecture Results and analysis As adopted by [5, 7, 10], we have also chosen Area under Curve (AUC) as the main parameter of evaluation for all our methodologies. Apart from this, to demonstrate the superiority of the features learned by the ensemble, we selected the U-MAP methodology to visually analyze the latent space vectors computed by the discriminator. To explain briefly, UMAP is an algorithm for dimension reduction based on manifold learning techniques and ideas from topological data analysis. It provides a very general framework for approaching manifold learning and dimension reduction, and can also provide specific concrete realizations. The points plotted in the scatterplot are the arithmetic means of the features of batches of images. The label “0” (blue) indicates anomalous images taken from the test set and the label “1” indicates their corresponding reconstructed images. We chose U-MAP over t-SNE and Principle Component Analysis (PCA) because PCA only captures Linear features/components, whereas t-SNE is a better visualization algorithm rather than one which finds actually useful components. Hence, U-MAP captures better components than t-SNE does. Summary of results of models on Compass-XP and SIXRay Shown in Table 3 is the results achieved by the 3 architectures on the 2 datasets on which experiments were performed. As mentioned earlier, for SIXray dataset, the whole dataset was not used for training purpose due to the limited availability of computation resources. So, a subset of randomly chosen 80k nonthreat images was chosen from the full nonthreat dataset of SIXray and from this patches of images were obtained which forms the final training data. All the three models were then trained using this set of generated training dataset.Table 3 Final results on the 3 architectures and 2 datasets Model Compass-XP AUC (%) SIXRay AUC (%) SkipGANomaly 94.8 66.7 Modified SkipGANomaly 94.94 69.2 Ensemble 96.8 75.3 Bold values indicate the highest AUC achieved by a model From the above results, we wish to point out the following:- 1. The modified SkipGANomaly architecture performs better than the original one for both the datasets: We attribute this to more richly learned features by the UNet++ based generator. This can be observed, especially for the more challenging dataset SIXRay, where the improvement over the original SkipGANomaly is more significant. 2. The ensemble performs significantly better than the individual networks: For both datasets, the ensemble brings a significant performance boost, especially with Compass-XP where it nearly equals the AUC achieved by human annotators (96.8% by ensemble vs. 97% by humans). Training plots, UMAP and generated images on Compass-XP In this section, we display some of the images generated by the 3 architectures that we tried, the training plots of the 3 architectures, and a U-MAP [20] analysis of the feature vectors (i.e., the latent space vectors) of the images taken in batches. The original SkipGANomaly architecture Figure 6 shown represents some of the images generated by the architecture. The images look promising in terms of object similarity to the real ones.Fig. 6 Some sample images generated by the basic Skip-GANomaly architecture The following (Fig. 7) is the training curve for SkipGANomaly. The “Best-AUC” curve indicates the Best AUC recorded yet.Fig. 7 Training plot for Skip-GANomaly with U-Net-based generator Figure 8 depicts the UMAP plot for this architecture. Clearly, there is no pattern whatsoever between the anomalous objects’ features and their reconstructions’ features.Fig. 8 UMAP plot on the Compass-XP dataset for the original Skip-GANomaly architecture. The red colored dots (class 0) indicate the nonanomalous (nonthreat) class and the blue colored dots (class 1) indicate the anomalous (threat object) class (colour figure online) SkipGANomaly with modified generator The following (Fig. 9) is the training plot for the Modified SkipGANomaly architecture. Note that the learning curve is smoother than the one for the basic architecture.Fig. 9 Training plot for Skip-GANomaly with modified (Nested-UNet-based) generator The following (Fig. 10) is some images generated by the Modified Architecture. The generated images are very close to the real-life objects.Fig. 10 Sample images generated by the modified architecture The following (Fig. 11) is the UMAP plot for the Modified Architecture. Note again that there is no pattern for the points.Fig. 11 U-MAP scatter-plot on the Compass-XP dataset for the modified SkipGANomaly architecture. The red colored dots (class 0) indicate the nonanomalous (nonthreat) class and the blue colored dots (class 1) indicate the anomalous (threat object) class Ensemble network The following (Fig. 12) is the training plot for the Ensemble Network. Evidently, the model reached its peak performance AUC of 96.8% in 5 epochs, after which the performance dropped (because the learning algorithm could not find a better solution), hence we retained the network weights at this point. (Also note that the scale for this graph is such that AUC ranges from 0.86 to 1 on the Y-axis).Fig. 12 Training plot for the ensemble of the 2 architectures The following (Fig. 13) is some images generated by the Ensemble Architecture. Here, it is clearly visible that model is able to generate more clearer objects and the objects are much similar to their real-life X-ray scans.Fig. 13 Images generated by the ensemble network The following (Fig. 14) is the U-MAP plot for the Features extracted for each batch by the Ensemble Architecture. Note that as compared to the plots of the individual architectures, the plot here shows a very good separation between the reconstructed and the original images. Apart from two batches each of original and reconstructed image features that are misplaced, a clear linear separation exists between the original and reconstructed features.Fig. 14 U-MAP Scatter-Plot on the Compass-XP dataset for the ensemble of the 2 architectures. The red colored dots (class 0) indicate the nonanomalous (nonthreat) class and the blue colored dots (class 1) indicate the anomalous (threat object) class (colour figure online) Training plots and sample generated images for SIXRay dataset The SIXRay dataset is far more challenging than the CompassXP dataset, because of 3 major reasons:- 1. The size of the data: Whereas Compass-XP has close to 2000 unique images, the SIXRay subset we considered 80k patch-based images for training. Also, these images were large in size and hence we had to stick with training on 256 × 256 patches of the images for training the models. 2. Multiple objects per bag as against Compass-XP. 3. Occlusion of objects: It is difficult to extract features of threat objects when these are occluded by translucent/opaque non-threat objects. Despite these challenges, the models could achieve the following results. The original SkipGANomaly architecture The following (Fig. 15) is the training plot for Skip-GANomaly on the SIXray datasetFig. 15 Training plot for the ensemble architecture on the SIXRay dataset The following (Fig. 16) represents some sample images generated by the SkipGANomaly architecture on the SIXray dataset.Fig. 16 Sample images generated by SkipGANomaly on SIXray SkipGANomaly with modified generator The following (Fig. 17) is the training plot for Skip-GANomaly with modified SkipGAnomaly architecture on the SIXray dataset.Fig. 17 Training plot for Skip-GANomaly with modified (Nested-UNet-based) generator on the SIXRay dataset The following (Fig. 18) represents some sample images generated by the Modified SkipGANomaly architecture on the SIXray dataset.Fig. 18 Sample images generated by modified SkipGANomaly on SIXray Ensemble network The following (Fig. 19) is the plot of the ensemble’s training on the SIXRay dataset. The network was trained for 250 epochs, and the plot is the condensed plot taking into account the best AUC of every group of 20 consecutive epochs. The maxima, 75.3% was reached at the 37th epoch, after which there was a value close to the maxima near epoch 227, where (73.1%), but the performance then deteriorated and hence training was stopped.Fig. 19 Training plot for the ensemble architecture on the SIXRay dataset The following (Fig. 20) represents some sample images generated by the Ensemble architecture on the SIXray dataset.Fig. 20 Sample images generated by ensemble architecture on SIXray The following (Fig. 21) is the UMAP plot for the features extracted by the Ensemble on the SIXRay dataset. It is clearly evident that the feature vectors form 2 separate clusters: The blue cluster points (anomalous class) being scattered from top-left to bottom right and the red cluster points (nonanomalous class) being scattered from top-right to bottom-left. They intersect near the center of the graph thus not forming linearly separable clusters for that class of images but are very well separated in the rest of the plot, thus indicating that our Ensemble Architecture learns features very well.Fig. 21 U-MAP scatter-plot on SIXRay dataset for the ensemble of the 2 architectures. The red colored dots (class 0) indicate the nonanomalous (nonthreat) class and the blue colored dots (class 1) indicate the anomalous (threat object) class (colour figure online) Conclusion We have presented two new approaches to tackle the problem of detecting prohibited items in luggage Xray images. On experimenting with popular architectures for anomaly detection on X-ray images, We found that SkipGANomaly performs far better than the State-of-the-art results on the CompassXP dataset, even when our chosen test set is far more challenging. Following this, our first modification to the architecture (The Modified Generator being modeled after a Nested-UNet-like architecture), helped us to achieve better results than the original SkipGANomaly. Then, we combined the 2 architectures into an Ensemble which surpassed the previous performance metric obtained by the modified architecture. This method achieved an AUC of 96.8% on the dataset which, as has been stated in the Dataset Information section, is nearly as good a result as that of the human experts’ predictions (which was 97% on average). Also, the U-MAP scatterplot for the ensemble is testimony to the fact that the ensemble learns more useful features that separate a non-anomalous object from an anomalous one. The second experiment was performed on far more challenging SIXray dataset where the ensemble-based architecture generates an AUC of 75.3%. For both of these case studies, the AUC obtained was greater than the state-of-the art anomaly detection model, i.e., SkipGANomaly which also validates the improved performance of our ensemble model. Since we could demonstrate the performance of the Ensemble network on only a subset of the SIXRay dataset due to computational constraints, a future objective remains to demonstrate the applicability of the ensemble network on the entire SIXRay dataset. Data availability We, the authors, declare that the datasets generated during and/or analyzed during the current study will be made available from the corresponding author on reasonable request. Declarations Conflict of interest We, the authors, certify that we have NO affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers’ bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge, or beliefs) in the subject matter or materials discussed in this manuscript. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Goodfellow I et al (2014) Generative adversarial nets. In: NIPS’14: Proceedings of the 27th international conference on neural information processing systems, vol 2, pp 2672–2680. http://papers.nips.cc/paper/5423-generative-adversarial-nets.pdf 2. Schlegl T et al (2017) Unsupervised anomaly detection with generative adversarial networks to guide marker discovery. In: Proceedings of Information Processing in Medical Imaging (IPMI) (2017), lecture notes in computer science, vol 10265, pp 146–157 3. Zenati H et al (2018) Efficient GAN-based anomaly detection. In: ICDM. arXiv:1812.02288 4. Donahue J et al (2017) Adversarial feature learning. In: ICLR. arXiv:1605.09782 5. Ackay S et al (2018) Semi-supervised anomaly detection via adversarial training. In: Asian Conference on Computer Vision (ACCV). Lecture notes in computer science, vol 11363, pp 622–637. arxiv:1805.06725v3 6. Radford A et al (2016) Unsupervised representation learning with deep convolutional generative adversarial networks. In: ICLR. arXiv:1511.06434 7. Ackay S et al (2019) Skip-GANomaly: skip conected and adversarially trained encoder-decoder anomaly detection. In: International Joint Conference on Neural Networks (IJCNN). IEEE 8. Ronneberger O et al (2015) Convolutional neural networks for biomedical image segmentation. In: International conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), pp 234–241 9. Davletshina D et al (2020) Unsupervised anomaly detection for X-ray images. arXiv:2001.10883 10. Han X et al (2021) GAN ensemble for anomaly detection. In: AAAI. arXiv:2012.07988v1 [cs.LG]. 14 Dec 2020 11. Han C Rundo L Murao K MADGAN: unsupervised medical anomaly detection GAN using multiple adjacent brain MRI slice reconstruction BMC Bioinform 2021 22 31 10.1186/s12859-020-03936-1 12. Bengs M, Behrendt F, Laves M-H, Krüger J, Opfer R, Schlaefer A (2022) Unsupervised anomaly detection in 3D brain MRI using deep learning with multi-task brain age prediction. In: Proceedings of SPIE 12033, Medical Imaging 2022: Computer-Aided Diagnosis, 1203314 13. Jensen S et al (2022) Deep learning-based anomaly detection on X-ray images of fuel cell electrodes. In: Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (2022) 14. Motamed S Rogalla P Randgan KF Randomized generative adversarial network for detection of COVID-19 in chest X-ray Sci Rep 2021 11 8602 10.1038/s41598-021-87994-2 33883609 15. Jiang T Xie W Li Y Lei J Du Q Weakly supervised discriminative learning with spectral constrained generative adversarial network for hyperspectral anomaly detection IEEE Trans Neural Netw Learn Syst 2021 10.1109/TNNLS.2021.3082158 16. Song J, Kong K, Park Y-I, Kim, S-G, Kang S-J (2021) AnoSeg: anomaly segmentation network using self-supervised learning 17. Matthew C, Griffin LD (2019) Limits on transfer learning from photographic image data to X-ray threat detection. J X-ray Sci Technol 27(6):1007–1020 18. Miao C et al (2019) SIXray: a large-scale security inspection X-ray benchmark for prohibited item discovery in overlapping images. In: CVPR (2019) 19. Zhou Z et al (2019) UNet++: a nested U-Net architecture for medical image segmentation. In: D. Stoyanov et al. (eds) DLMIA 2018/ML-CDS 2018, LNCS 11045, pp 3–11. 10.1007/978-3-030-00889-5_1 20. Leland M et al (2018) UMAP: uniform manifold approximation and projection for dimension reduction. arXiv:1802.03426	0	PMC9734403	NO-CC CODE	2022-12-14 23:28:28	no	Neural Comput Appl. 2022 Dec 9;:1-16	utf-8	Neural Comput Appl	2,022	10.1007/s00521-022-08029-z	oa_other
==== Front J Child Adolesc Trauma J Child Adolesc Trauma Journal of Child & Adolescent Trauma 1936-1521 1936-153X Springer International Publishing Cham 505 10.1007/s40653-022-00505-x Original Article An Exploratory Mixed-method Descriptive Analysis of Youth Coping during the First Wave of the COVID-19 Pandemic in Quebec http://orcid.org/0000-0002-4531-5124 Hébert Martine [email protected] 1 Jean-Thorn Arianne 2 Malchelosse Katherine 2 1 grid.38678.32 0000 0001 2181 0211 Département de sexologie, Université du Québec à Montréal, C.P. 8888, Succursale Centre-Ville, Montréal, Québec H3C 3P8 Canada 2 grid.38678.32 0000 0001 2181 0211 Département de psychologie, UQAM, Montréal, Québec Canada 3 12 2022 114 22 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. This study presents an exploratory mixed-method descriptive analysis of psychological distress, challenges encountered and coping strategies of youth during the first wave of COVID-19. A total of 4 914 Quebec youth, aged 14 to 25 recruited through social media completed an online survey on the impact of the pandemic on their daily life, psychological distress and post-traumatic growth. They were also invited to answer two open-ended questions regarding the difficulties they experienced and their coping strategies. Overall, 26.6% of youth showed serious psychological distress and 20.3% displayed probable PTSD symptoms related to the COVID-19. Comparative analyses revealed that youth identifying as gender and sexual minorities were more vulnerable to distress during the first wave of the pandemic. While few sociodemographic variables distinguish youth reporting post-traumatic growth from those who do not, the former group was more likely to seek and receive social support. Qualitative data highlighted difficulties that were common to many respondents including lack of social contacts, the lockdown itself, and problems related to disruptions of educational and academic activities. Youth also mentioned three main strategies they used to cope: maintaining social contacts, engaging in leisure activities and physical exercise. While many youths have encountered compelling challenges during the lockdown of the first wave of COVID-19, some report having relied on efficient coping strategies to deal with the situation. Support services should be implemented to help the youth cope with the pandemic while considering their diverse needs. Keywords COVID-19 Youth Mixed-methods Mental health Coping strategies Canada Research Chairs, Government of Canada950-230791 Hébert Martine ==== Body pmcThe coronavirus outbreak began in Wuhan, China in December 2019. On January 30, 2020, the World Health Organization declared a health emergency and on March 11, 2020, a pandemic. Since then, over 621 million cases and 6 million deaths have been confirmed all around the world (Worldwide Health Organization, 2022). Since March 2020, the COVID-19 pandemic and related sanitary measures have been affecting the mental health of individuals all over the planet. An earlier study on the mental health of adults exposed to COVID-19 in China found that living in the hot spots of the epidemic, more exposure to the media, and less cognitive and prosocial coping strategies were associated with more mental health problems (Guo et al., 2020). Another study found that 7% of a sample of 285 citizens from Wuhan and surrounding cities showed post-traumatic stress symptoms (Liu et al., 2020). Studies conducted in Italy found high levels of post-traumatic stress symptoms, ranging from 7.9% to 29.5% of different samples (Casagrande et al., 2020; Castelli et al., 2020; Forte et al., 2020). The psychosocial impacts of the COVID-19 pandemic are present worldwide. As shown in a review, exposure to the virus, forced quarantine and lockdowns, as well as media exposure are associated to anxiety, acute panic, obsessive behavior, depression, stigmatization, racism and xenophobia (Dubey et al., 2020; Racine et al., 2020). These impacts may also be salient in adolescents and young adults who are in a fragile period of social and personal exploration, that the pandemic and associated consequences might hinder. Indeed, a study conducted in China following the COVID-19 outbreak found that 40.4% of youth aged 14 to 35 were likely to develop a psychological health problem (Liang et al., 2020). In Canada, a few studies have been conducted with youth, browsing the consequences and the challenges related to the pandemic (Courtney et al., 2020; Hawke et al., 2020). One of these studies, conducted among youth with a history of addictions, showed a significant deterioration between prepandemic mental health and intrapandemic mental health. Furthermore, participants reported unmet needs for help and disruptions of health services (Hawke et al., 2020). The detrimental impact of the COVID-19 pandemic could also be particularly salient among gender-diverse (i.e., transgender, queer) and sexually diverse (i.e., gay, bisexual or asexual orientation) youth who are already struggling with their access to health and social services and experiences of violence and stigma (Dysart-Gale, 2010; Hatzenbuehler & Pachankis, 2016). A study conducted by Hawke et al. (2021) showed that gender and sexual diversity (GSD) youth reported more disruptions in mental health services and less support from their families than cisgender youth. While the bulk of studies conducted to date has underscored the negative impact associated with COVID-19, some studies have also explored the possible positive adaptation of the population in the face of the pandemic (Barzilay et al., 2020; Shanahan et al., 2020). Indeed, in the face of adversity, some individuals may display a profile of resilience (Chen & Bonanno, 2020), use effective coping strategies or be able to navigate towards supportive resources to deal with difficult times and maintain a positive adaptation. Some may even develop post-traumatic growth, which is the ability to find meaning in stressful experiences and gain a positive insight into one's perception of themselves, others and the world (Tedeschi & Calhoun, 2004). While resilience develops during exposure to adversity, post-traumatic growth may be an outcome of that experience. In a study conducted by Barzilay et al. (2020), a high level of resilience was found to be associated with fewer COVID-19-related concerns among 3 042 respondents. In a cohort study of 768 participants (Shanahan et al., 2020), the coping strategies associated to lower levels of distress during the COVID-19 pandemic were maintaining a daily routine, physical activity and making use of strategies such as positive reappraisal or reframing. To date, Canada counts more than 4 million confirmed cases within its borders (Worldwide Health Organization, 2022). In Quebec, the first case was confirmed on February 27, 2020, which marks the beginning of the first wave of COVID-19 outbreaks in Quebec. As of March 11, 2020, and over the course of several months, the Prime Minister held a daily press briefing to keep the population informed. That day, the Quebec government announced the first health measures: mandatory 14-day isolation for travelers, a ban on gatherings of more than 250 people, and encouragement of teleworking. On March 16th, daycares, schools and Canadian borders were closed. On March 22nd, the first lockdown in Quebec was announced. On March 24th, all non-essential services (restaurants, bars, entertainment centers) were closed and on April 5th, this measure was extended. It was not until May 6th that a gradual resumption of services was initiated following the drop in outbreaks. In Quebec, the second wave of cases began in the fall of 2020 and other sanitary measures were undertaken to reduce the contamination curve. Against this backdrop, this exploratory study, relying on a mixed-methods design, aimed to assess the impact of the first wave of the COVID-19 pandemic on Quebec youth, the difficulties they encountered and the strategies they used to deal with the situation. The primary objective of this study was to assess the frequency of psychological distress and PTSD symptoms related to the COVID-19 pandemic among cisgender youth and GSD youth as well as to document the perceived impact of the pandemic on their daily lives and routines and to observe possible post-traumatic growth. The second objective was to explore, through qualitative responses, the difficulties encountered by youth and the strategies they used to get through challenging times during the lockdown of the first wave of COVID-19 in Quebec. The purpose of this exploratory study was to provide an initial portrait of the psychosocial adaptation of young people at the beginning of the pandemic. Methodology Participants and Procedure Participants were recruited through an online questionnaire, which was posted on various social networks (Facebook, Instagram, Snapchat, Reddit). Recruitment took place between April 21 and May 25, 2020, during mandatory confinement in Quebec. The Qualtrics platform was used to create the online questionnaire. This platform enables the use of parameters allowing anonymity for participants by giving them a random alphanumeric code. On social networks, the project was presented to participants as a study of adaptation during the COVID-19 pandemic. After completing the questionnaire, as a financial compensation, participants could enter a contest to win one of the $50 gift cards. The study design was approved by the ethics committee of the Université du Québec à Montréal. Informed consent was obtained electronically from participants after the nature of the procedures had been thoroughly described at the beginning of the questionnaire. In the province where the study was conducted, minors aged 14 and over can consent to research and no parental consent is required. The average time to complete the survey was 20 min (SD = 34). There were 7 123 answers registered on the Qualtrics platform. Following preliminary analyses, 489 participants showed invalid responses due to duplicates, similar responses throughout the questionnaire, or too short a completion time. Also, 2 177 participants had incomplete responses on the measures considered for the present analyses. The final convenience sample thus involved 4 914 Francophone youth aged 14 and 25 from the province of Quebec. The average age of participants was 19.94 (SD = 3.14). The majority of participants identified themselves as cisgender girls (64.2%), 33.5% as cisgender boys, and 2.3% identified themselves as gender diversity (trans, gender queer). A total of 78.1% of the sample reported heterosexual orientation, 6.2% homosexual orientation, 11.4% reported being bisexual, and 4.3% were questioning their sexual orientation. The vast majority of participants (91.6%) were of Quebec or Canadian origin. Participants reported being in school (31.1%), in school but also working part-time (41.0%), working full time (22.0%) or neither presently in school or working (5.9%). Of those in school, 22.6% attended high school, 30.1% attended CEGEP or a professional training center and 19.3% attended university. Measures Sociodemographic Variables Participants reported sociodemographic information (age, sexual orientation, occupation, etc.) at the beginning of the survey. COVID-19-related Items These 21 items were specifically designed for this study and inspired by prior studies addressing the COVID-19 pandemic (Satici et al., 2020; Yildirim & Solmaz, 2020). They were used to assess how the pandemic and the confinement measures affected the behaviors and habits of youth. The items addressed several areas of daily life such as school, sleep, family and romantic relationships, entertainment, and fear and anxiety related to the virus. Participants could answer each item with a 4-point Likert scale ranging from Not true (0) to Mostly true (3). All COVID-related items are presented in Table 2. Primary Care PTSD Screen for DSM-5 (PC-PTSD-5) PTSD symptoms were measured by the Primary Care PTSD Screen for DSM-5 (PC-PTSD-5; Prins et al., 2016). The instrument usually starts with an item conceived to assess whether the participant has ever experienced a specific trauma. In the present study, the 5-item scale was answered in reference to the exposure to the COVID-19 pandemic. Participants were asked: “In relation to COVID-19, in the last month, have you …had nightmares about the event(s) or thought about the event(s) when you did not want to?”, “Tried hard not to think about the event(s) or went out of your way to avoid situations that reminded you of the event(s)?”, “Been constantly on guard, watchful, or easily startled?”, “Felt numb or detached from people, activities, or your surroundings?” and “Felt guilty or unable to stop blaming yourself or others for the events(s) or any problems the event(s) may have caused?”. The participant could answer by yes or no to each item and the total score ranged from 0 to 5. A cut-off of 3 was identified as optimally sensitive to probable PTSD (Prins et al., 2016). In the present study, the internal consistency of the instrument measured by Cronbach’s alpha was adequate (α = .70). Kessler Screening Scale (K6) Psychological distress was measured using Kessler screening scale (Kessler et al., 2002). The scale contains 6 items and participants had to answer how often they felt “nervous”, “hopeless”, “restless or fidgety”, “so depressed that nothing could cheer you up”, “that everything was an effort” and “worthless”. Response options range from Rarely (1) to Always (4) with the total score ranging from 4 to 24. We used the same cut-off as Kessler et al. (2003) and McGinty et al. (2020) to qualify serious psychological distress (scores from 13 to 24). The scale showed good internal consistency (α = .83). Post-traumatic Growth (PGI) Four items related to post-traumatic growth were drawn from the Life Paths version (Hamby et al., 2015) of the Posttraumatic Growth Inventory (Tedeschi & Calhoun, 1996). Items were: “I changed my priorities about what is important in life”, “I have a greater sense of closeness with others”, “Now I know that I can handle hard times” and “I have discovered that I am stronger than I thought I was”. Participants were invited to answer each item with a 4-point Likert scale ranging from Not true (1) to Mostly true (4). The continuous score ranged from 1 to 4. A dichotomous score was created where a score of ≥ .5 standard deviation of the mean was used to assess the presence of post-traumatic growth. The scale showed good internal consistency (α = .73). Social Support Social support was measured using the social support items from the Resilience Portfolio Questionnaire (Hamby et al., 2018). The Social support seeking scale (5 items) aims to measure efforts to obtain help and attitudes towards asking for help while the Social support received scale (6 items) is designed to assess help or encouragement provided by others in times of distress (Hamby et al., 2020). The items of the Social support seeking scale were “Talking it out with someone helps me when I’m upset”, “It helps me to discuss ideas with someone when I have a problem”, “I feel better when I talk to people about what’s going on”, “I talk to someone to help me solve problems”, “I ask people to help me make tough decisions” and “Talking to someone who has been through the same thing helps me”. The five items of the second scale were “Someone was there for me when I was having a hard time”, “Someone gave me a place where I could get away for a while”, “Someone helped me get my mind off things”, “Someone went with me to get some help” and “Someone comforted me”. These items are answered on a 4-point Likert scale ranging from Not true (1) to Mostly true (4). Two continuous scores were created ranging from 1 to 4, one for Social support seeking scale and one for Social support received scale. From each score, a dichotomous score was created. To be considered receiving or seeking social support, a score of ≥ 0.5 standard deviation from the mean was required. The two subscales showed high internal consistency (Social support seeking scale: α = .90; Social support received scale: α = .90). Open-ended Questions In order to gather strategies and challenges encountered during the pandemic, respondents were asked two open-ended questions: “What do you find most difficult in the current situation?” and “What strategies are helpful to you in the current situation?”. Two research assistants coded the responses. Data Analysis Quantitative Data Descriptive analyses were conducted to assess the frequency of probable PTSD symptoms and psychological distress and the impact of the pandemic on the habits of participants in our sample. To compare cisgender heterosexual youth to GSD youth, we grouped the gender and sexual orientation variables and divided the sample into three categories: 1) cisgender heterosexual boys (CH boys), 2) cisgender heterosexual girls (CH girls), and 3) GSD youth. The third category included youth who identified as transgender, queer, gay, lesbian, bisexual, or asexual. Chi- square analyses were conducted to compare the three groups on their level of clinical distress and possible PTSD as well as on the COVID-19-related items. A logistic regression was then conducted on posttraumatic growth to assess if sociodemographic variables and social support were significant predictors. Qualitative Data A thematic analysis was conducted to explore the qualitative data. Of the total sample, 4 405 participants responded to the question on the difficulties they encountered and 4 284 answered the question on the useful strategies. To classify these responses, two research assistants individually categorized the first 200 answers to each question by grouping them under themes. An answer could appear in more than one category. Then, the 2nd author of the manuscript (AJT) identified similar categories between the two assistants. Responses that had been rated differently were discussed to identify the appropriate category. Research assistants then assigned a category to all the remaining responses to each question. Along the way, if a recurring theme seemed to emerge, a team discussion was held to decide whether it was appropriate to add a category. Each response was rated according to whether the theme was present in the response (score of 1) or not (score of 0). A total of 20 categories were created for each question. In these analyses, the three most popular themes in each question, i.e., the most recurrent themes in the responses, will be illustrated. These themes were identified on the basis of descriptive data (i.e., frequencies). Results Psychological Distress and PTSD Symptoms Overall, 26.6% of the sample showed serious psychological distress. Applying the cut-off score proposed by Prins et al. (2016), findings revealed that 20.3% displayed probable PTSD symptoms. Chi-square analyses were performed on the percentage of cases reaching clinical levels of psychological distress and probable PTSD by groups. Results of the chi-square analyses and adjusted standardized residuals are presented in Table 1. Values greater than 1.96 flag observed values that are significantly different than expected. Inspection of these values revealed that CH boys were less likely to report serious psychological distress levels while GSD youth were more likely to achieve scores reflecting serious psychological distress. Results regarding probable PTSD symptoms showed that CH boys obtained lower rates than expected values, and both CH girls and GSD youth obtained higher values.Table 1 Percentage (adjusted residuals) of Participants Reaching Clinical Levels CH boys n = 1 364 CH girls n = 2 449 GSD youth n = 1 101 Total sample n = 4 914 χ2 Serious psychological distress 16.3% (-10.2) 26.1% (-0.7) 40.4% (11.8) 26.6% 182.42* Possible PTSD 11.6% (-9.4) 22.0% (2.9) 27.3% (6.6) 20.3% 102.10* * p < .001 Perceived Impact COVID-19 and Confinement Measures The questionnaire related to perceived impact of the pandemic first assessed whether participants respected the quarantine measures and findings showed that the vast majority (97.2%) of respondents respected the imposed measures. Overall, a high proportion of youth was afraid that a member of their family could be infected (71.8%) or that they could be infected themselves (59.1%). Exploration of possible differences between groups revealed that CH boys were less likely to be impacted while CH girls were more likely to be. Items assessing the perceived impact of COVID-19 revealed that the majority of participants in the sample were affected by the social distancing measures (61.1%), the pandemic itself (76.1%) and reported that thinking about the pandemic impacted their daily mood (73.6%), their concentration (67.2%) and their sleep (57%). For all items, comparative analyses revealed that both CH girls and GSD youth obtained results reflecting greater impact relative to CH boys. Globally, more than half of youth (56.5%) reported being worried about the impact of the pandemic on their school and academic trajectory. The confinement measures also had a negative impact as 40% of participants mentioned experiencing more conflicts in their family and 29.6% more conflicts in their romantic relationships. Inspection of adjusted standardized residuals suggest that GSD youth reported greater conflicts in their family since the pandemic while CH girls were more likely to report greater conflicts in their romantic relationships relative to youth in the other groups. The pandemic and related confinement measures also affected daily routines, as 58% of the total sample mentioned eating more junk food, 29.6% endorsed items related to an increase in their alcohol consumption and 13% in drug consumption. As can be expected, the vast majority of youth (93.2%) spent more time on the internet and social networks, and 60.3% more time playing online games. Comparative analyses on these sets of items did not identify a consistent pattern. CH girls were more likely and CH boys less likely to spend more time on social media. However, CH boys were found to be more likely (82.5%) to play online games than CH girls (48%). Compared to CH boys and girls, GSD youth were more likely to report drinking more alcohol since the start of the confinement. Both GSD youth and CH boys were more likely to report using more drugs compared to CH girls. The items that assess possible positive aspects of the confinement highlight that, overall, close to two thirds of participants reported having more time with members of their family (65.2%), to develop new hobbies (62.4%) while a bit more than half (54.9%) mentioned exercising more since the confinement. Also 37.9% of participants considered they had more quality time with their romantic partner and 30% mentioned they had started volunteering and offering help. Analysis of possible differences between groups indicates that CH girls were more likely to report these positive aspects while no differences between groups were noted for the items relating to the development of new hobbies (Table 2).Table 2 Percentage of Participants Endorsing Items as Being True Item CH boys n = 1 364 CH girls n = 2 449 GSD youth n = 1 101 Total sample n = 4 914 χ2 COVID-related I have respected the quarantine measures 94.9% (-6.0) 98.3% (4.8) 97.5% (0.7) 97.2% 37.51 * I am afraid that someone in my family is infected 61.4% (-10.0) 77.0% (8.0) 73.2% (1.2) 71.8% 106.26* I am afraid of being infected with COVID-19 43.9% (-13.4) 67.4% (11.8) 59.5% (.3) 59.1% 199.83 * Mental health I am troubled by the social distancing measures 45.2% (-14.1) 67.9% (9.8) 65.5% (3.4) 61.1% 201.17 * Thinking about the pandemic affects my daily mood 57.9% (-15.4) 79.9% (10.1) 78.7% (4.4) 73.6% 237.69 * I have difficulty concentrating 58.7% (-7.8) 68.6% (2.1) 74.6% (5.9) 67.2% 73.46 * I am troubled by the COVID-19 pandemic 62.6% (-13.7) 82.4% (10.4) 78.7% (2.3) 76.1% 194.58 * I have trouble sleeping 43.0% (-12.3) 61.3% (6.1) 64.6% (5.8) 57.0% 153.39 * Relationships/School I am very worried about my school career 53.2% (-3.0) 57.4% (1.2) 58.8% (1.7) 56.5% 9.34 In my family there is more conflict because of the quarantine 36.4% (-3.2) 40.4% (0.6) 43.4% (2.6) 40.0% 12.80 Within my relationship, I experience more conflict 23.4% (-5.8) 32.7% (4.8) 30.2% (0.5) 29.6% 36.31* Coping I spend more time on the internet and on social networks 90.9% (-4.0) 94.5% (3.6) 93.2% (0.0) 93.2% 17.79 * I play more online games 82.5% (19.7) 48% (-17.5) 60.1% (-0.1) 60.3% 433.98* I eat more junk food 55.8% (-2.0) 58.4% (0.6) 59.9% (1.5) 58.0% 4.60 I've been drinking more alcohol since the start of the quarantine 28.4% (-1.1) 28.2% (-2.1) 34.1% (3.7) 29.6% 13.76 I've been using more drugs since the beginning of the quarantine 15.2% (2.9) 9.2% (-7.8) 18.5% (6.2) 13.0% 66.55 * Positive aspects I have more quality time with my family members 63.9% (-1.2) 68.0% (4.2) 60.5% (-3.7) 65.2% 20.46* I have developed new hobbies 61.1% (-1.2) 63.7% (1.8) 61.3% (-0.9) 62.4% 3.21 I exercise more 55.7% (0.7) 58.0% (4.4) 46.9% (-6.1) 54.9% 38.54 * I have more quality time with my boyfriend/girlfriend 28.3% (-8.5) 45.5% (10.9) 32.7% (-4.0) 37.9% 124.80 * I volunteer or have offered to help someone 26.4% (-3.5) 32.7% (4.1) 28.5% (-1.2) 30.0% 18.22 * * p < .05, p < .01, * p < .001 Posttraumatic Growth Items assessing post-traumatic growth revealed that for some youth, the pandemic brought them to the realization that they are now able to handle difficult times (71.7%). COVID-19 also contributed to them changing their priorities about what is important in life (67.4%) and to discover that they were stronger than they thought they were (54.7%). Close to half of participants also agreed that they now had a greater sense of closeness with others (47.1%). A hierarchical logistical regression was conducted to identify possible predictors of posttraumatic growth. Results are presented in Table 3. In the analysis, sociodemographic variables were entered in the first block as control variables and the two indicators of social support were included in the second block. The final model was significant (χ2(2) = 169.66, p < .001). Results revealed that ethnicity, seeking social support and social support received were significant predictors of posttraumatic growth. Participants who reported belonging to the Quebec or Canadian cultural group were more likely to report posttraumatic growth. In addition, youth who reported high scores on the social support received and seeking social support scales were more likely to show posttraumatic growth.Table 3 Results of the Hierarchical Logistic Regression Predicting Post-traumatic Growth Variables B S.E Wald’s p OR 95% CI Age: 14–18 .11 .07 2.16 .141 1.11 .97–1.27 (Reference group: 18 – 25) Occupation: not studying or working .22 .14 2.56 .110 1.25 .95–1.65 (Reference group: Studying and/or working) Group: CH girls .04 .08 0.29 .591 1.05 .89–1.22 Group: GSD youth -.16 .10 2.66 .103 0.85 .70–1.03 (Reference group: CH boys) Ethnicity: Other .37 .12 10.07 .002 1.45 1.15–1.82 (Reference group: Quebec or Canadian cultural group) No social support requested .44 .07 36.89 < .001 1.55 1.35–1.79 (Reference group: social support requested) No social support received .63 .07 77.32 < .001 1.88 1.63–2.16 (Reference group: social support received) Final model: χ2(7) = 188.95, p < .001; OR odds ratio Qualitative Data With respect to the difficulties encountered during the beginning of the pandemic, the four most recurring themes were concerns related to social contacts (50.5%), the confinement measures per se (13%) and education (10.6%). Among the participants who had concerns with social contacts, several felt a general lack of socialization and human physical contact while many others were missing their loved ones (family, friends and romantic partner). The participants who reported difficulty with the lockdown itself wrote about challenges in two main spheres: limited outings and being locked in alone. Among the statements from participants who had education-related challenges, three preoccupations were common: issues related to distance learning, concentration or motivation at home and the school being closed. The three categories are presented in Table 4 with citations for each subcategory.Table 4 Sample Quotes – What is Most Difficult in the Current Situation? Main Themes Description Quotes Social contacts (50.5%) Human contact in general Lack of socialization in general is one of the difficulties encountered during confinement. Not seeing people, not having a social circle outside of their family was highlighted by participants as being challenging “To no longer see people other than my family” “Lack of human contact on a daily basis” “No longer have social contact with the general population” “What I find the most difficult is not seeing my friends because with friends the relationship is different than that with your family members. I talk about things that I am passionate about, things we have in common” “What I find most difficult is not having my last moments with my friends with whom I spent all my high school years. Not having a prom makes me cry because I’ve been waiting for this moment since I started high school” Not being able to see those they love The inability to see loved ones, friends, family or romantic partners was considered a concern during lockdown “Not being able to go see my dad who is sick” “Not being able to hug my loved ones, not being able to go and spend time with my friends at home or in a bar as well” “Not being able to see my boyfriend because we don't live together and not being able to see my sisters because they no longer live with my mother. Also, I find it difficult not being able to have physical contact with people at work and not being able to get close to them” “Not being able to visit my mother whom I love more than anything. She can no longer spend time with my 10-month old baby. Not being able to see my friends and their children growing up so fast. These are moments that will not come back. Not being able to see my grandfather who recently had surgery Lockdown (13%) Limited outings «Not being able to go out» is considered to be one of the hardest things to experience during the pandemic “Being locked in my house” “Being cut off from the outside world” “[…] I am trapped in my own home with a lazy father who lets my mother do all the work, and a brother who has anger management problems. The clutter is suffocating me, so I spend most of my time in my bedroom, which is the only room that is not a mess. The mess is so big that you have to sneak into the house. There are smells. There are mice living in the walls. The insects invade the house and my parents don't seem to be bothered” “What I find difficult is not being able to leave my house. Being at the end of the semester, I do homework all day, but I can't reward myself by doing an activity such as going to a movie or a restaurant” “No possibility to isolate oneself to calm down after a conflict or a fight. I can't go outside, I can't go to a friend's house. The fact that that we don't have time to be alone make the problems worse” Being alone Being in isolation alone is also a significant negative aspect of lockdown “Staying 24/7 alone in my apartment without anybody” “I find the most difficult thing is to be with myself 24 h a day” “Not being able to go outside and see people… I feel like I'm trapped inside and not being able to see people scares me because when I'm alone at home I'm scared of what's going through my head” “To be alone in my 1 1/2 apartment when the rest of my family and my boyfriend are in another province” “Lack of human contact. Staying alone in my apartment. Having got separated, it's hard to adjust to being alone, considering that I can't see my friends who might be able to support me in this transition” Education (10.6%) Distance learning Completing a school year at a distance either through videoconferencing or self-directed learning is a challenge for some participants. The lack of institutional supervision is also highlighted “I hate distance learning” “The CEGEP now requires me to teach the subject myself and has tried to offer the same performance as when I had teachers to support me and answer my questions […]” “The distance school, the overload of university work and the indifference of teachers to the situation” “Staying at home and not seeing my friends anymore. I find it sad that I couldn't really live the end of my CEGEP at the CEGEP. I also find it sad that I won't be starting university in person. It's like I'm missing out on part of education and it's also difficult because I don't know when everything will go back to the way it was before” Difficulty concentrating and being motivated Many participants also have difficulty concentrating and finding the motivation to do assignments and studies at home “Find the motivation to continue and finish my session at the university and not spend whole days in front of my computer” “The overload of school work when it is very difficult to concentrate at home” “I find it really difficult to study and concentrate at home; I usually manage to do all my work and study at the library or in a restaurant” School on pause The youth also found it difficult to have school on break “Not being able to continue my schooling” “Having to drop out of school because I am not able to go to school remotely” “Being unable to go to school” In terms of the strategies most frequently used by participants, maintaining social contact was seen as the optimal strategy by many participants (33.1%). Among these, many reported taking the opportunity to spend more time with the people they lived with, while others used social networks to maintain ties despite isolation. Several participants mentioned that enjoying leisure activities kept them busy and that lockdown allowed them to take more time for themselves (27.7%). Engaging in a lot of sports or physical activity was also a strategy reported by many participants (8.8%). These strategies are presented in greater detail in Table 5, illustrated with quotations from the database.Table 5 Sample Quotes – What Strategies are Helpful for You in the Current Situation? Main Themes Description Quotes Social contacts (33.1%) Interactions with close ones who live with them For some participants, contact with those with whom they live (love partner, family, roommates) is a strategy for feeling better in times of pandemic “I live with my girlfriend so we encourage each other, we cook a lot, we take advantage of the good weather (when it is there). She encourages me a lot in my studies” “I try to make the most of the time I have with my family and my lover” “Having my roommates with whom I share more things now also helps the time pass quickly until we can join our respective families” Maintain contact trough social networks, videoconference and phone calls The use of social networks, videoconferencing platforms and telephone calls serve as a strategy for maintaining contact with loved ones “I did some video conferences with my friends on Zoom. I often text my family by messenger” “Frequent calls with family and friends, wondering how things are going and supporting each other” “Do a lot of Facetime with my father. I also started writing him letters that I send by mail and we correspond that way. It's even better than talking on the phone about me. It puts a balm on our days” “Talking with my teachers helped me a lot. Several of them helped me. We need to reflect on how we tackle this adversity and learn about this situation” “Chat with my friends and talk to people I don’t know on common interest groups online” Leisure (27.7%) Time for hobbies A popular strategy to get through the pandemic is to keep busy with hobbies and to take time for oneself “Have small projects like drawing, gardening, building stuff or playing video or online board games with my friends. These are often the best times I spend during confinement” “I try to occupy my mind as much as possible. I read, I play the piano, I write, I draw, I learn Spanish, I learn about something else…” “To take my mind off things by drawing, reading, listening to shows or movies, playing video games or taking walks with my dog” “I continue to live my life and keep myself busy. I also like to think that what I am going through is an experience that will make me a better person. I have started new activities like cooking, playing games, and reading” “I take the time to dance, sing, read books, listen to music, watch series, escape in videogames. And all of these things in mindfulness! Always stay conscious about the chance we have (most of us) to access all of things in a time of crisis” Sports (8.8%) Time for sports Physical exercise is another pandemic coping strategy that many participants used “I train almost every day, which helps my mood and energy level” “Physical training helps me regain the upper hand” “I am much more physically active than I used to be. It helps to clear my mind” “Enjoy the free time we have to do more physical activities” Discussion The COVID-19 pandemic and associated health measures have significant psychosocial consequences. Previous studies have shown that the stress caused by the spread of the virus as well as adaptation to a new and more restrictive lifestyle has caused great distress to the general population (Dubey et al., 2020; Guo et al., 2020). Some populations, including youth, may face particular challenges, as the pandemic could affect their social emancipation and identity (Courtney et al., 2020; Hawke et al., 2020). Through a mixed-methods design, the present exploratory study had two objectives. The first objective was to document the impact of the first wave of the pandemic on the mental health of youth in Quebec and identify how their habits were affected. The second objective was to identify what difficulties youth have encountered, and the strategies they used to cope with the situation. Our results show that overall, 26.6% of youth presented serious psychological distress, which is similar to previous studies conducted in the general population during the pandemic (Casagrande et al., 2020; McGinty et al., 2020). Our results also revealed that 20.3% achieved scores qualifying probable PTSD symptoms, which is similar to the study by Castelli et al. (2020). However, studies show a great variability in rates (7% to 29%). This variability may be related to different factors, including the different scales used, the impact of the pandemic varying from one country to another, or from one cultural context to another (Casagrande et al., 2020; Forte et al., 2020; Liang et al., 2020; Liu et al., 2020). Our findings revealed that cisgender heterosexual boys are less likely to show distress while youth identifying as gender and sexual minorities are more vulnerable to distress. These findings corroborate those of Hawke et al. (2020) and suggest that sexually and gender-diverse youth are at greater risk in a pandemic context. In addition to restricting access to health care or psychological services, health recommendations and lockdowns may force them to stay at home with their families. In some cases, these families may not be respectful of their identities and the lockdown may keep them away from the LGBTQ+ community which offers support. This social isolation, in the context of a search for identity and self-assertion, may be related to increased distress. The COVID-19 pandemic is a major stress associated with a fear of infection, as overall 59% of young people are afraid of being infected and 72% fear that a family member may be infected. In addition, youth report that during confinement, the pandemic has affected their daily mood, sleep and concentration and has had an impact on their habits. Indeed, COVID-19 and subsequent confinement measures appear to have resulted in increased consumption of junk food, alcohol and, to a lesser extent, drugs. Several studies have found that there is a change in the population's eating habits and in their consumption of alcohol and drugs during the pandemic (Biddle et al., 2020; Kim et al., 2020; Ruiz-Roso et al., 2020). For some individuals, this is an opportunity to reduce their intake, while for others, the confinement may be driving them to consume more. One study showed that some families had more time to cook healthy foods, but the results showed no improvement in the quality of the overall diet (Ruiz-Roso et al., 2020). Also, for some more vulnerable individuals with a history of problematic substance use, confinement may increase the risk of relapse (Kim et al., 2020). Half of the youth had concerns about their education, which can be explained by the reorganization of institutions to adapt to the health measures related to the pandemic. During the lockdown of the first wave of COVID-19 in Quebec, schools were forced to close. As a result, the education system underwent major changes to adapt to distance learning, leaving youth without structured support and guidance for several weeks. The results of the present study also highlight the diversity of reactions and perceptions of the impact of the pandemic. For example, some youth report that being locked in has been an adversity that allowed to consolidate bonds with family members in some way, and that they have had the opportunity to spend more quality time with their romantic partner. It should be noted that these are outcomes more often reported by cisgender heterosexual girls and less likely reported by GSD youth. Indeed, GSD youth are more likely to report that the situation may have exacerbated existing family conflicts. Qualitative data on the difficulties encountered by youth revealed three recurring themes. Lack of social contact was the most cited difficulty. Some participants missed their loved ones while others felt a general lack of socialization. These results are not surprising considering that social support is an important protective factor among youth (Chu et al., 2010) and that the lockdown increased isolation for some specific populations (Van Gelder et al., 2020). The qualitative data from our study reflects well on the results of the pandemic-related items of the survey. Responses to items related to the impact of the pandemic on habits also showed that the majority of the sample were troubled by the social distancing measures related to the pandemic. The testimonies also highlight the frustration of young people of missing out on landmark events such as graduation, childbirth or the beginning of higher education. The lockdown itself was also considered particularly challenging for some participants. Indeed, some youth felt trapped because they were not able to go out and others, living alone, found it difficult to be alone with one’s thoughts. Studies show that this lockdown-related distress may be caused by several elements. Confinement may increase the risk of victimization for the LGBTQ+ community (Silliman & Bosk, 2020), people living with domestic violence (Van Gelder et al., 2020) or abuse. For others presenting with pre-existing mental health issues or physical disabilities, confinement is also challenging as the availability and provision of support services may be impacted. The third most cited difficulty was related to education. Answers to open-ended questions revealed that many youth had issues adjusting to distance education and the lack of supervision. Others faced difficulties in concentration and experienced a significant drop in motivation, while some worried about having their schooling on break. This is consistent with the quantitative responses to items related to the impact of the pandemic on habits, as the majority of participants reported being worried about their educational trajectory and having difficulty concentrating. Although many of the participants faced challenges associated with the pandemic, many also found strategies that helped them adapt to this new reality. According to the responses to the open-ended question, maintaining social contact was the most prevalent strategy. Several participants took the opportunity to spend time with their loved ones living in the same household, which is in line with the quantitative data indicating that many participants spent more time with their families and romantic partners. Others maintained contact through social media, phone calls, and videoconferencing. The quotes also highlight the benefits of support from various sources such as teachers, family and friends as well as strangers sharing similar interests on online support groups. The other strategies used were engaging in leisure activities (e.g. reading, painting, playing piano, learning a new language) and physical exercise. These strategies mentioned by the participants also fit the items on the impact of the pandemic on daily life. Indeed, more than half of the sample reported discovering new hobbies and getting more exercise. The qualitative data do not allow us to assess the quality of the strategies used and the link to participants' psychological distress. However, previous studies show that some factors may promote resilience, such as social support (Bonanno et al., 2007; Lai et al., 2015) and finding distractions to laugh and have fun (Keltner & Bonanno, 1997). Also, Shanahan et al. (2020) found that physical exercise as a coping strategy during the COVID-19 pandemic was associated with reduced distress. These results are consistent with the strategies employed by participants in this study. Adverse life events may also serve to foster post-traumatic growth. Tedeschi and Calhoun (1996, 2004) propose that post-traumatic growth may occur when an event comes to question the individual’s beliefs about themselves, others and the world, and reflections on the situation allow for growth in recognizing personal strengths (Robles-Bello et al., 2020). A significant proportion of youth realized they could handle hard times and discovered they were stronger than they thought they were. As one youth mentioned: “I also like to think that what I am going through is an experience that will make me a better person”. In exploring factors related to post-traumatic growth, the results of the hierarchical logistic regression highlight the importance of social support in the youth's environment. Our findings show that, during the first wave of the pandemic, the critical factors related to more posttraumatic growth are seeking and receiving social support. These findings are similar to those obtained in the literature which highlight social support being related to higher psychological functioning (Chu et al., 2010; Hamby et al., 2020). In fact, one study shows that youth who receive social support, but also seek it, show the most psychosocial strength (Hamby et al., 2020). Therefore, in conceptualizing resources for youth in the context of a pandemic, it would be important to offer support to more vulnerable youth (ethno-cultural, sexual, and gender diversity) to compensate for the realities that some families may not offer them an optimal level of social support. However, it would also be important to ensure that follow-up is offered when a youth request support and that organizations are available to meet their needs since we also want to encourage seeking social support among youth. Limitations This study needs to be considered in the light of certain limitations. First, the use of a convenience sample limits the generalizability of the results. The sample is exclusively French-speaking and composed of individuals from Quebec, therefore the findings are not necessarily representative of youth in other countries. Second, the cross-sectional design does not allow to establish a causal relationship over time, which means that these results are strictly representative of the lockdown of the first wave of COVID-19 in Quebec and further studies are needed to assess the evolution of the situation during the subsequent waves of the pandemic. Finally, it is impossible to determine whether PTSD and psychological distress symptoms related to the pandemic are due to fear of the virus itself or to health measures (confinement, social distancing, restriction of access to services), as the items are general and do not specifically measure either one or the other. The PTSD scale used in this study is designed to screen individuals with probable PTSD, and further assessment with a structured interview or a more comprehensive self-report measure is generally warranted following a positive screen. In addition, the rates of probable PTSD might have been overestimated given that the life-threat trauma criterion was not directly assessed. Despite these limitations, this study documents a wide range of behaviors, attitudes, and thoughts related to the pandemic among a large sample of youth. The richness of the qualitative data allowed us to add to the quantitative data and describe well the experiences of the participants during the confinement from both a positive and negative perspective. The present study offers some cues to better understand the challenges faced by youth, which may be useful, namely for professionals working with vulnerable populations such as gender-diverse youth. For example, being aware that not all youth benefit from support from their peers and families, may facilitate a more personalized approach in implementing resources during the pandemic. Further data on vulnerable youth, such as youth with a history of trauma, is warranted. Research must also continue assessing how COVID-19 impacts the lives of youth, as adaptation to the pandemic may reflect an important indicator of resilience in individuals. Future studies should also rely on standardized measures of the impacts of COVID-19 considering the virus may be part of everyday life due to its many mutations. Conclusion In conclusion, this study allowed for an exploration of the difficulties encountered and strategies used by youth during the lockdown of the first wave of COVID-19 in Quebec. Youth faced specific challenges and some experienced psychological distress and probable PTSD symptoms that need to be considered. It is also essential to highlight their adaptability, as some youth have used coping strategies or benefited from support in the face of adversity. While the pandemic has negative psychological consequences among the population, it is also important to understand what helps youth overcome the challenges. Such information could offer cues for the design of services addressing the needs of all youth. Acknowledgements Authors wish to thank Hélène Demers and Laurie Fortin for data codification. Our thanks are also extended to the youth who participated in the survey. Funding This study was supported by the Canada Research Chair in Interpersonal Traumas and Resilience (CRSH #950–230791). 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 Barzilay R Moore TM Greenberg DM DiDomenico GE Brown LA White LK Gur RC Gur RE Resilience, COVID-19-related stress, anxiety and depression during the pandemic in a large population enriched for healthcare providers Translational Psychiatry 2020 10 1 291 10.1038/s41398-020-00982-4 32820171 Biddle, N., Edwards, B., Gray, M., & Sollis, K. (2020). Alcohol consumption during the COVID-19 period: May 2020. Australian National University Centre for Social Research & Methods. Retrieved July 11, 2022, from https://openresearch-repository.anu.edu.au/bitstream/1885/213196/1/Alcohol_consumption_during_the_COVID-19_period.pdf Bonanno GA Galea S Bucciarelli A Vlahov D What predicts psychological resilience after disaster? The role of demographics, resources, and life stress Journal of Consulting and Clinical Psychology 2007 75 5 671 682 10.1037/0022-006X.75.5.671 17907849 Casagrande M Favieri F Tambelli R Forte G The enemy who sealed the world: Effects quarantine due to the COVID-19 on sleep quality, anxiety, and psychological distress in the Italian population Sleep Medicine 2020 75 12 20 10.1016/j.sleep.2020.05.011 32853913 Castelli L Di Tella M Benfante A Romeo A The Spread of COVID-19 in the Italian Population: Anxiety, Depression, and Post-traumatic Stress Symptoms The Canadian Journal of Psychiatry 2020 65 10 731 732 10.1177/0706743720938598 32588644 Chen S Bonanno GA Psychological adjustment during the global outbreak of COVID-19: A resilience perspective Psychological Trauma: Theory, Research, Practice, and Policy 2020 12 S1 S51 S54 10.1037/tra0000685 32538658 Chu PS Saucier DA Hafner E Meta-analysis of the relationships between social support and well-being in children and adolescents Journal of Social and Clinical Psychology 2010 29 6 624 645 10.1521/jscp.2010.29.6.624 Courtney D Watson P Battaglia M Mulsant BH Szatmari P COVID-19 Impacts on child and youth anxiety and depression: Challenges and opportunities The Canadian Journal of Psychiatry 2020 65 10 688 691 10.1177/0706743720935646 32567353 Dysart-Gale D Social justice and social determinants of health: Lesbian, gay, bisexual, transgendered, intersexed, and queer youth in Canada Journal of Child and Adolescent Psychiatric Nursing 2010 23 1 23 28 10.1111/j.1744-6171.2009.00213.x 20122085 Dubey S Biswas P Ghosh R Chatterjee S Dubey MJ Chatterjee S Lahiri D Lavie CJ Psychosocial impact of COVID-19 Diabetes & Metabolic Syndrome: Clinical Research & Reviews 2020 14 5 779 788 10.1016/j.dsx.2020.05.035 Forte G Favieri F Tambelli R Casagrande M COVID-19 Pandemic in the Italian Population: Validation of a post-traumatic stress disorder wuestionnaire and prevalence of PTSD Symptomatology International Journal of Environmental Research Public Health 2020 17 11 4151 10.3390/ijerph17114151 32532077 Guo J Feng XL Wang XH van Ijzendoorn MH Coping with COVID-19: Exposure to COVID-19 and negative impact on livelihood predict elevated mental health problems in Chinese adults International Journal of Environmental Research and Public Health 2020 17 11 3857 10.3390/ijerph17113857 32485859 Hamby, S., Grych, J., & Banyard, V. L. (2015). Life Paths measurement packet: Finalized scales. Sewanee, TN: Life Paths Research Program. Retrieved July 11, 2022, from https://lifepathsresearch.org/wp-content/uploads/2021/10/Life-Paths-Measurement-Packet-finalized-scales-Aug-2015-Color-front-and-back.pdf Hamby S Taylor E Mitchell K Jones L Newlin C Is it better to seek or to receive? A dual-factor model of social support International Journal of Child and Adolescent Resilience/revue Internationale De La Résilience Des Enfants Et Des Adolescents 2020 7 1 5 17 10.7202/1072584ar Hamby S Taylor E Smith A Blount Z Resilience Portfolio Questionnaire manual: Scales for youth 2018 Life Paths Research Center Hatzenbuehler ML Pachankis JE Stigma and minority stress as social determinants of health among lesbian, gay, bisexual, and transgender youth: Research evidence and clinical implications Pediatric Clinics of North America 2016 63 6 985 997 10.1016/j.pcl.2016.07.003 27865340 Hawke LD Barbic SP Voineskos A Szatmari P Cleverley K Hayes E Relihan J Daley M Courtney D Cheung A Darnay K Henderson JL Impacts of COVID-19 on youth mental health, substance use, and well-being: A rapid survey of clinical and community samples The Canadian Journal of Psychiatry 2020 65 10 701 709 10.1177/0706743720940562 32662303 Hawke LD Hayes E Darnay K Henderson J Mental health among transgender and gender diverse youth: An exploration of effects during the COVID-19 pandemic Psychology of Sexual Orientation and Gender Diversity 2021 8 2 180 187 10.1037/sgd0000467 Keltner D Bonanno GA A study of laughter and dissociation: Distinct correlates of laughter and smiling during bereavement Journal of Personality and Social Psychology 1997 73 4 687 702 10.1037/0022-3514.73.4.687 9325589 Kessler RC Andrews G Colpe LJ Hiripi E Mroczek DK Normand SL Walters EE Zaslavsky AM Short screening scales to monitor population prevalences and trends in non-specific psychological distress Psychological Medicine 2002 32 6 959 976 10.1017/s0033291702006074 12214795 Kessler RC Barker PR Colpe LJ Epstein JF Gfroerer JC Hiripi E Zaslavsky AM Screening for serious mental illness in the general population Archives of General Psychiatry 2003 60 2 184 189 10.1001/archpsyc.60.2.184 12578436 Kim JU Majid A Judge R Crook P Nathwani R Selvapatt N Lovendoski N Manousou P Thursz M Dhar A Lewis H Vergis N Lemoine M Effect of COVID-19 lockdown on alcohol consumption in patients with pre-existing alcohol use disorder The Lancet Gastroenterology & Hepatology 2020 5 10 886 887 10.1016/S2468-1253(20)30251-X 32763197 Lai BS Tiwari A Beaulieu BA Self-Brown S Kelley ML Hurricane Katrina: Maternal depression trajectories and child outcomes Current Psychology: A Journal for Diverse Perspectives on Diverse Psychological Issues 2015 34 3 515 523 10.1007/s12144-015-9338-6 Liang L Gao T Ren H Cao R Qin Z Hu Y Li C Mei S Post-traumatic stress disorder and psychological distress in Chinese youths following the COVID-19 emergency Journal of Health Psychology 2020 25 9 1164 1175 10.1177/1359105320937057 32627606 Liu N Zhang F Wei C Jia Y Shang Z Sun L Wu L Sun Z Zhou Y Wang Y Liu W Prevalence and predictors of PTSS during COVID-19 outbreak in China hardest-hit areas: Gender differences matter Psychiatry Research 2020 287 112921 1 7 10.1016/j.psychres.2020.112921 McGinty EE Presskreischer R Han H Barry CL Psychological distress and loneliness reported by US adults in 2018 and April 2020 JAMA 2020 324 1 93 94 10.1001/jama.2020.9740 32492088 Prins A Bovin MJ Smolenski DJ Marx BP Kimerling R Jenkins-Guarnieri MA Kaloupek DG Schnurr PP Kaiser AP Leyva YE Tiet QQ The Primary Care PTSD Screen for DSM-5 (PC-PTSD-5): Development and evaluation within a veteran primary care sample Journal of General Internal Medicine 2016 31 10 1206 1211 10.1007/s11606-016-3703-5 27170304 Racine N Cooke JE Eirich R Korczak DJ McArthur B Madigan S Child and adolescent mental illness during COVID-19: A rapid review Psychiatry Research 2020 292 113307 10.1016/j.psychres.2020.113307 32707216 Robles-Bello MA Sánchez-Teruel D Valencia Naranjo N Variables protecting mental health in the Spanish population affected by the COVID-19 pandemic Current Psychology 2020 41 5640 5651 10.1007/s12144-020-01132-1 33106742 Ruiz-Roso MB de Carvalho Padilha P Mantilla-Escalante DC Ulloa N Brun P Acevedo-Correa D Arantes Ferreira Peres W Martorell M Aires MT de Oliveira Cardoso L Carrasco-Marín F Paternina-Sierra K Rodriguez-Meza JE Montero PM Bernabè G Pauletto A Taci X Visioli F Dávalos A Covid-19 confinement and changes of adolescent’s dietary trends in Italy, Spain, Chile, Colombia and Brazil Nutrients 2020 12 6 1807 10.3390/nu12061807 32560550 Satici B Gocet-Tekin E Deniz ME Satici SA Adaptation of the Fear of COVID-19 Scale: Its association with psychological distress and life satisfaction in Turkey International Journal of Mental Health and Addiction 2020 19 6 1980 1988 10.1007/s11469-020-00294-0 32395095 Silliman RC Bosk EA Vulnerable youth and the COVID-19 pandemic Pediatrics 2020 146 1 1 3 10.1542/peds.2020-1306 Shanahan L Steinhoff A Bechtiger L Murray AL Nivette A Hepp U Ribeaud D Eisner M Emotional distress in young adults during the COVID-19 pandemic: Evidence of risk and resilience from a longitudinal cohort study Psychological Medicine 2020 52 5 824 833 10.1017/S003329172000241X 32571438 Tedeschi RG Calhoun LG The posttraumatic growth inventory: Measuring the positive legacy of trauma Journal of Traumatic Stress 1996 9 3 455 471 10.1002/jts.2490090305 8827649 Tedeschi RG Calhoun LG TARGET ARTICLE: "Posttraumatic Growth: Conceptual Foundations and Empirical Evidence" Psychological Inquiry 2004 15 1 1 18 10.1207/s15327965pli1501_01 Van Gelder N Peterman A Potts A O'Donnell M Thompson K Shah N Oertelt-Prigione S COVID-19: Reducing the risk of infection might increase the risk of intimate partner violence EClinicalMedicine. 2020 21 1 2 10.1016/j.eclinm.2020.100348 Worldwide Health Organization. (2022). WHO Coronavirus (COVID-19) Dashboard. Retrieved November 20, 2022, from https://covid19.who.int Yildirim M Solmaz F COVID-19 burnout, COVID-19 stress and resilience: Initial psychometric properties of COVID-19 Burnout Scale Death Studies 2020 46 3 524 532 10.13140/RG.2.2.23216.58883 32915702	0	PMC9734407	NO-CC CODE	2022-12-14 23:28:28	no	J Child Adolesc Trauma. 2022 Dec 3;:1-14	utf-8	J Child Adolesc Trauma	2,022	10.1007/s40653-022-00505-x	oa_other
==== Front World J Pediatr World J Pediatr World Journal of Pediatrics 1708-8569 1867-0687 Springer Nature Singapore Singapore 36484871 655 10.1007/s12519-022-00655-w Review Article Fecal microbiota transplantation in childhood: past, present, and future Gu Xu 1 Chen Zhao-Hong 2 Zhang Shu-Cheng [email protected] 1 1 grid.412467.2 0000 0004 1806 3501 Department of Pediatrics, Shengjing Hospital of China Medical University, 36 Sanhao Street Heping District, Shenyang, 110004 China 2 grid.412467.2 0000 0004 1806 3501 Department of Radiology, Shengjing Hospital of China Medical University, Shenyang, China 9 12 2022 110 18 6 2022 13 11 2022 © Children's Hospital, Zhejiang University School of 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. Background Fecal microbiota transplantation (FMT) has been well described in the treatment of pediatric diseases; however, the latest updates regarding its use in children are unclear and the concepts involved need to be revisited. Data sources We performed advanced searches in the MEDLINE, EMBASE, and Cochrane databases using the keywords “Fecal microbiota transplantation OR Fecal microbiota transfer” in the [Title/Abstract] to identify relevant articles published in English within the last five years. To identify additional studies, reference lists of review articles and included studies were manually searched. Retrieved manuscripts (case reports, reviews, and abstracts) were assessed by the authors. Results Among the articles, studies were based on the mechanism (n = 28), sample preparation (n = 9), delivery approaches (n = 23), safety (n = 26), and indications (n = 67), including Clostridium difficile infection (CDI) and recurrent C. difficile infection (rCDI; n = 21), non-alcoholic fatty liver disease (NAFLD; n = 10), irritable bowel syndrome (IBS; n = 5), inflammatory bowel disease (IBD; n = 15), diabetes (n = 5), functional constipation (FC; n = 4), and autism spectrum disorder (ASD; n = 7). Conclusions Concepts of FMT in pediatric diseases have been updated with respect to underlying mechanisms, methodology, indications, and safety. Evidence-based clinical trials for the use of FMT in pediatric diseases should be introduced to resolve the challenges of dosage, duration, initiation, and the end point of treatment. Keywords Autism spectrum disorder Children Clostridium difficile infection Fecal microbiota transplantation Functional constipation Safety http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China 30700917 81570465 Zhang Shu-Cheng ==== Body pmcIntroduction Fecal microbiota transplantation (FMT) is a method that transfers stool from a healthy donor to a recipient to restore the intestinal microbiota environment and achieve a therapeutic benefit. FMT was first recorded in the Jin Dynasty in ancient China. A Chinese physician, Hong Ge, elaborated the effect of stool by mouth on patients with food poisoning or severe diarrhea. In the “Compendium of Materia Medica,” written by Shi-Zhen Li in the Ming Dynasty, over 20 FMT methods were documented for treating gastrointestinal diseases such as food poisoning, diarrhea, fever, vomiting, constipation, and abdominal pain [1]. FMT was also applied in veterinary medicine in Europe in the sixteenth century. Additional therapeutic use of human excretions was described in Europe in the eighteenth and nineteenth centuries and in World War II, during which gut bacteria were administered to German soldiers suffering from dysentery in the North African campaign [2]. More scientifically, in 1958, Eismann successfully utilized fecal transplantation via enemas in four patients for the treatment of severe pseudomembranous colitis [3]. Three of the four patients recovered and were discharged from the hospital after several days, while the fourth patient died from non-intestinal-associated diarrhea. Taken together, these results suggested the clinical value of FMT [3]. In early 2011, FMT was proposed to treat gastrointestinal diseases[4]. Physicians called for the use of FMT by colonoscopy, gastroscopy, and gastroduodenal catheterization for Clostridium difficile infection (CDI) instead of surgery in an effort to reduce deaths. Since that time, the number of studies focusing on FMT has increased rapidly. In 2012, Khorouts et al. carried out the first study using standard cryopreserved bacteria [5]. In 2013, Nood et al. reported that FMT was successful in the treatment of a recurrent C. difficile infection (rCDI) in a randomized controlled trial at the University of Amsterdam [6]. At the same time, the Food and Drug Administration (FDA) approved the use of FMT in humans [7]. In the pediatric population, the first use of FMT can be traced to Massachusetts General Hospital in the United States in 2010. Russell et al. reported a two-year-old child with rCDI whose symptoms resolved completely 36 hours after FMT administration and no recurrences or adverse events (AEs) occurred during the six-month follow-up period [8]. Although the enormous potential of FMT is apparent, FMT-related AEs have been identified in the published literature [9–13]. Therefore, the acceptance, use, and safety of FMT are still under investigation. Mechanism underlying fecal microbiota transplantation The goal of FMT is to re-establish the intestinal flora by normalizing the amount and activity of immune and inflammatory responses, neurotransmitters and vasoactive substances, and energy metabolism [14]. The diversity of the microbiota prevents the colonization and overgrowth of pathogens when homeostasis is achieved in the gastrointestinal tract. FMT can make the composition of the gut microbiota similar to that of the donor so that the proportion and diversity of beneficial bacteria is balanced [15]. FMT can reduce intestinal permeability and maintain the integrity of the epithelial barrier by increasing the production of short-chain fatty acids, thereby reducing the severity of the disease [14, 16, 17]. Moreover, gut microbiota can activate the humoral immune response and induce the synthesis of immunoglobulin A (IgA), immunoglobulin G (IgG), and immunoglobulin M (IgM) through the Toll-like receptor (TLR) pathway, thus protecting the intestinal mucosa [14]. FMT inhibits the secretion of proinflammatory cytokines and promotes T helper 1 (Th1) cell differentiation, T cell activity, leukocyte adhesion, and immune-stimulating factors [7]. FMT also reduces intestinal pH and increases the adhesion of bacteria to H2O2, inhibiting the transport of pathogens [18]. All these findings serve as the presumed mechanism underlying FMT effectiveness [14–18]. Indications Advanced searches in the MEDLINE, Embase, Cochrane Library, and Cochrane IBD Group Specialized Register databases with the terms “Fecal Microbiota Transplantation OR Fecal Microbiota Transfer” in the [Title/Abstract] field were performed. Based on the research strategy, a bibliometric analysis was performed on the use of FMT in children in the last five years, and 67 articles reported the indications of FMT in children (Fig. 1). The acquired literature included Clostridium difficile infection (CDI) and recurrent Clostridium difficile infection (rCDI; n = 21), non-alcoholic fatty liver disease (NAFLD; n = 10), irritable bowel syndrome (IBS; n = 5), inflammatory bowel disease (IBD; n = 15), diabetes (n = 5), functional constipation (FC; n = 4), and autism spectrum disorder (ASD; n = 7). The references, study design, interventions, and results of the indicated diseases for pediatric FMT were retrieved and are summarized in Table 1.Fig. 1 Literature related to fecal microbiota transplantation (FMT) includes indications, mechanism, donor exclusion, sample preparation, and delivery methods Table 1 Current potential indications for pediatric FMT Diseases The first author Year Identifier (PMID) Design Intervention Numbers References CDI Khoruts A 2010 20048681 Case report Colonoscopy 1 9 Hamilton MJ 2012 22290405 Case control study Colonoscopy 43 5 Van Nood E 2013 23323867 Case control study Nasoduodenal tube 32 6 Wang J 2015 25798243 Case report Nasal jejunal 1 21 Kronman MP 2015 25162365 Case report Nasogastric tube 10 10 Walia R 2014 25162365 Case report Colonoscopy 2 22 Kahn SA 2012 23211865 Case report Nasogastric tube 1 12 Russell G 2010 20547640 Case report Nasogastric tube 1 8 Hourigan SK 2019 31660343 Case report Colonoscopy 9 23 IBD Karolewska-Bochenek K 2018 29151253 Case report Nasoduodenal tube 10 24 Kunde S 2013 23542823 Case report Enema 10 25 Goyal A 2018 29361092 A prospective trial Endoscopy 20 28 Cho S 2019 30320666 A retrospective trial Colonoscopy 8 29 Moutinho BD 2019 30632438 Case report Colonoscopy 1 30 Shimizu H 2016 27324973 Case report Colonoscopy 1 31 Hourigan SK 2015 26198180 Case report Colonoscopy 8 27 IBS Johnsen PH 2018 29100842 RCT Colonoscopy 90 33 FC Tian H 2016 26751143 A pilot study Nasojejunal tube 24 35 NAFLD Philips CA 2017 27816755 A pilot study Nasoduodenal tube 8 43 ASD Kang DW 2017 28122648 An open-label study Oral and colonoscopy 18 45 Li N 2021 34737978 An open-label study Capsule and colonoscopy 56 46 Chen Y 2022 35105621 RCT Capsule 318 47 Diabetes Leiva-Gea I 2018 30224347 Case-control study Oral 43 52 Solito A 2021 34229463 RCT Oral 101 56 FMT fecal microbiota transplantation, CDI Clostridium difficile infection, IBD inflammatory bowel disease, IBS irritable bowel syndrome, FC functional constipation, NAFLD non-alcoholic fatty liver disease, ASD autism spectrum disorder Established indications Clostridium difficile infection Clostridium difficile infection (CDI) and recurrent Clostridium difficile infection (rCDI) are considered the most suitable indications for pediatric FMT [19]. The incidence of rCDIs has been reported to be as high as 90%, some of which are appropriate for traditional therapy [5, 6, 9, 20]. The youngest reported child receiving FMT for CDI was a 13-month-old infant [21]; several other reports involved children over three years of age [10, 12, 22]. These cases were successful, and FMT for CDI has been associated with improved growth in young children [22]. Russell reported the first case of a two-year-old child with rCDI for whom the donor was transported through a nasogastric tube to the child’s small intestine in 2010 [8]. The CDI symptoms resolved completely 36 hours after FMT administration, and there were no recurrences or adverse reactions during six months of follow-up [8]. Suchitra selected nine pairs of donor receptors with an average age of 10 years in 2019 to examine the efficacy of FMT in pediatric CDIs [23]. Three days after FMT treatment, CDI-related symptoms of all recipients were alleviated, and no recurrences occurred. During the follow-up period, one patient had long-term C. difficile-negative diarrhea and intermittent incontinence, which was mild and different in nature compared with CDI symptoms before FMT [23]. Furthermore, a multicenter retrospective cohort study was also conducted on the largest sample size of CDI trials in children [35]. Of the 335 children, 271 (80.9%) were cured by the traditional therapeutic regimen. In the remaining 64 children with rCDIs, 19 (53.1%) were cured after FMT treatment, reaching an overall success rate of 88.6%. Inflammatory bowel disease The efficacy and safety of FMT in pediatric IBD has been confirmed in several studies [24–27]. Katarzyna assessed the effectiveness of a two-week FMT course in 10 children (10–17 years of age) with moderate-to-severe IBD by freshly prepared FMT via a nasoduodenal tube and found that a short, intensive course of FMT has a beneficial effect on ulcerative colitis (UC) and Crohn’s disease (CD) [24]. Goyal et al. reported 21 patients with IBD refractory to medical therapy who underwent a single FMT by upper and lower endoscopy with a median age of 12 years; 57% and 28% demonstrated clinical responses one and six months after FMT administration. Two patients with CDI were in full remission at six months [28]. Similarly, in 2019, Cho reported a 75% rate response in eight patients with IBD three months after FMT [29]; however, contrary results have also been reported. In a case report of a 17-year-old male with refractory UC, clinical improvement lasted for only one month before symptoms recurred. A second implementation of FMT also led to no improvements [30]. In another case involving an 11-year-old girl, the first FMT led to exacerbation of UC symptoms, while repeated procedures allowed her to remain in remission with a minimal dose of steroids [31]. There are two points that may contribute to the variation in FMT in childhood IBD. First, the pathogenesis of IBD in children may differ from that in adults. Second, most parents do not allow their children to be research subjects for safety reasons, thus resulting in poor compliance and a high dropout rate in the pediatric population. Potential indications Irritable bowel syndrome In 2019, a systematic review and meta-analysis reported the efficacy of FMT in the IBS through a total of 33 randomized clinical trials (RCTs) involving 4321 patients [32]. The authors pointed out that the clinical symptoms of IBS were alleviated after FMT treatment. Other meta-analyses and cohort studies have shown significant improvement in IBS patients after FMT treatment [33, 34]. FMT has enormous potential in adult IBS. Nevertheless, there is still uncertainty about FMT treatment in pediatric IBS because clinical trials for treating childhood IBS have not been conducted. Functional constipation Functional constipation has also been reported as a potential indication for FMT. In 2016, Tian conducted an open-label study of FMT in the treatment of slow-transit constipation (STC) [35]. In this trial, 24 STC patients (20–74 years of age) were enrolled. FMT was performed on three consecutive days through nasal-jejunal tubes, and the patients underwent follow-up for 12 weeks. Clinical improvement was shown in 50% (12/24) of those recruited, and full remission was found in 37.5% (9/24) with no AEs reported [35]. Similar results regarding the use of FMT in childhood constipation have also been reported. In 2016, de Meij TG reported a significant increase in bacterial species in a study using conventional culture techniques in 28 constipated children compared with 14 healthy children [36]. By comparing the fecal flora between eight constipated children and 14 healthy children, the authors observed an increase in the abundance of bifidobacteria in constipated subjects [36]. Together, these results indicated that FMT has enormous potential for treating childhood constipation [37]. Fatty liver disease Non-alcoholic fatty liver disease (NAFLD) severity is closely related to the dysregulation of intestinal bacteria and changes in metabolic function. Studies have shown differences in the composition of bacteria in the feces of NAFLD and healthy patients [38]. Another study involving non-alcoholic steatohepatitis (NASH) in children mentioned that compared with the control group, the content of bacterial components in NASH children was different. A meta-analysis confirmed that at normal transaminase levels, the laboratory indices of the probiotic experimental treatment group improved significantly compared to the placebo group [39]. Reducing the number of harmful microbiota can also increase the concentration of butyrate in the cecum and the expression of the intestinal tight junction protein (ZO-1). The increased butyrate and tight junction protein, ZO-1, is beneficial because butyrate is the energy source of colonic motility, and ZO-1 can repair the mucosal barrier of the colonic epithelium [40]. A meta-analysis confirmed that probiotics reduced the number of harmful microbiota and increased the level of butyrate and ZO-1 [41]. It has been reported that with the recovery of intestinal flora, the symptoms of portal hypertension and other hepatic symptoms significantly improved [42]. Moreover, the implementation of FMT has not increased the incidence of AEs in any NASH/NAFLD patients [43]. On the basis of existing clinical and experimental data, FMT has therapeutic potential in NASH/NAFLD [43]. Autism spectrum disorder Gut flora and its metabolites play an important role in the pathophysiology of ASD [44–47]. In 2017, Kang adopted a modified FMT regimen involving 18 participants with ASD (7–16 years of age) [45]. An improvement was observed in 89% of the participants with respect to the symptoms of diarrhea, constipation, dyspepsia, and abdominal pain, but the autism symptoms were not significantly reduced [45]. Subsequently, the author performed a two-year follow-up evaluation, and in 2019, the ASD symptoms of the participants were re-evaluated [45]. Interestingly, a significant improvement in behavior symptoms was observed in all participants compared to the baseline measurements, suggesting the effectiveness of FMT in ASD [48]. Nevertheless, the remission of ASD symptoms of the participants could not be completely attributed to FMT because the improvement of behavioral symptoms occurred two years later. The brain-intestine axis may be a means of communication between the brain and intestinal flora. Several studies have shown how the gut flora may alter brain function [49–51]. Diabetes To determine the differences in intestinal flora among children with metabolic diseases, Isabel Leiva-Gea published a study in 2018 comparing 15 diabetic and 13 healthy children [52]. Leiva-Gea reported that the intestinal flora in children with type 1 diabetes differed in classification and function from healthy subjects, and there were fundamental differences in non-autoimmune diabetes models. In another study, Vrieze administered FMT to patients with metabolic syndrome, and insulin sensitivity improved significantly after FMT administration, suggesting the feasibility of FMT in metabolic diseases [53–55]. Solito assessed the effects of probiotic supplementation on weight and metabolism in 101 obese and insulin-resistant young children in a cross-over, double-blind, randomized controlled trial [56]. The study demonstrated that eight weeks of intervention was safe, well tolerated, and efficacious in improving insulin sensitivity and supporting weight loss. Methodology Donor screening Stool samples from a healthy donor are the first requirement for FMT. Consent from a parent is required for FMT involving a child. Donor samples must be safe and reliable and cannot introduce iatrogenic diseases. Donor exclusion criteria are shown in Table 2 [19, 33, 53, 57–61]. Based on domestic and international donor screening standards, the main direction and focus of screening FMT donors are the factors that affect the quality and efficacy of the fecal bacteria solution of the donor, such as gastrointestinal tract, infectious diseases, use of drugs, and immunologic conditions. We also do not currently understand how to select an ideal FMT recipient or how other underlying conditions might impact the response [62–64]. Current screening protocols for FMT may be insufficient. Little is known about what qualifies as an effective or safe donor and does not account for the possibility of gut microbiota perturbations. Similarly, in the pediatric population, the donor criteria must fulfill the above-described standards, but specific tests should be added to the criteria for a pediatric donor. Factors that may affect child development need to be taken into account. When considering a donor, it is necessary to exclude child-specific diseases, such as attention-deficit hyperactivity disorder, autism, and other inherited metabolic disorders, because all these diseases have the potential to increase the risk of additional diseases in the FMT recipient.Table 2 Disease screening and laboratory test for donor candidates Categories History of diseases Infectious diseases: AIDS, hepatitis, tuberculosis Gastrointestinal diseases: CDI, IBD, IBS, constipation, gastrointestinal surgery in the past 6 mon Immune system diseases or immunomodulatory treatment Tumors Metabolic syndrome and/or obesity Inherited metabolic diseases: phenylketonuria, lysosomal storage disorders, glycogen storage disease, favism Use of antibiotics within 3 mon Serological screening Hepatitis series virus antibody: hepatitis A, B, C, I and II virus surface antigen Epstein–Barr virus (IgG and IgM) and cytomegalovirus (IgG and IgM.) AIDS Bacterial test: syphilis reagin test; TPPA/TPHA; tuberculosis Routine blood examination Liver and renal function C-reactive protein, erythrocyte sedimentation rate and anti-streptolysin O test Insulin and blood glucose Stool test Viral test: viruses associated with diarrhea (RT-PCR): rotavirus, norovirus, astrovirus Parasitic test: Ascaris, Ancylostoma duodenale, Strongyloides stercoralis, Giardia lamblia, Entamoeba histolytica, Trichuris trichiura, Clonorchis sinensis, Blastocystis hominis Bacterial tests: Helicobacter pylori; Salmonella spp., Shigella spp., Vibrio spp., Campylobacter spp., Yersinia enterocolitica and Aeromonas spp. Additional fecal test: fecal white blood cell, Occult blood Additional test Abdominal ultrasound Abdominal and chest (posteroanterior) radiography COVID-19 tests (only for pandemic period): nasopharyngeal swab, serology for SARS-CoV-2, stool testing for SARS-CoV-2 AIDS acquired immune deficiency syndrome, CDI Clostridium difficile infection, IBD inflammatory bowel disease, IBS irritable bowel syndrome, IgG immunoglobulin G, IgM immunoglobulin M, TPPA/TPHA Treponema pallidum particle agglutination assay/Treponema pallidum hemagglutination assay, RT-PCR reverse transcription-polymerase chain reaction, COVID-19 corona virus disease 2019, SARS-CoV-2 severe acute respiratory syndrome coronavirus 2 Preparation Stool that is used for FMT can be fresh or frozen [19, 65–70]. Fresh stool should be disposed of within six hours of donation and stored at room temperature for further treatment. The feces are thoroughly stirred with standard sterile sodium chloride, and the filtrate is drawn into a syringe and injected into the gastrointestinal channel of the recipient. Another type of frozen feces is made by collecting feces from a group of pre-screened donors and storing frozen feces in the feces bank in equal aliquots. Final disposal is in storage at −80 °C. On the day of FMT, the fecal suspension is defrosted in a warm water bath (37°C), then dissolved in normal saline to obtain the expected volume of the suspension. The infusion is performed within six hours after defrosting [7]. Notably, repeated defrosting and freezing should be avoided. Regardless of the methods used to obtain the fecal liquid, the principle of asepsis must be considered in the process of making fecal bacteria liquid for FMT, and the influence of air oxidation on fecal bacteria should be prevented. It is often not possible to determine the amount of fecal bacteria liquid that the patient needs. Compared with the relatively large amount of the infusion, the risk of failure when the infusion amount is < 50 g is more than fourfold higher [20]. The determination of the amount of fecal bacteria solution warrants further experimental determination [60], and there is also no clear consensus on the best method of preservation [67, 68, 71, 72]. Delivery approaches We consider differences in treatment and the preparation of fecal specimens, and patient acceptance of the different delivery methods. The current delivery approaches for FMT include the following: (1) nasogastric, nasoduodenal, or nasal-jejunal tube; (2) capsule; (3) colonoscopy (stool deposited into the right colon or terminal ileum); (4) oral; and (5) enema [2, 73]. Table 3 shows there are notable differences among delivery methods. Colonoscopy is a good option for the delivery of FMT both in children and adults [72, 74–77]. Compared with other methods, the treatment effects of colonoscopy are better in pediatric patients; however, colonoscopy is invasive, requires sedation, has the standard risks of colonoscopy, and the effectiveness may be limited within the colon (i.e., not the entire intestine) [78]. Capsule technology, which has emerged in recent years, is an effective delivery approach for pediatric FMT and can overcome the psychological problems of “oral feces” [71], but it is very expensive and has high technical requirements that exceed the capabilities of most hospitals. Specifically, an oral capsule is unsuitable in children because an oral capsule has the risks of becoming lodged in the esophagus and aspirated. At the same time, the effectiveness of an oral capsule is tentative because the number of fecal bacteria contained might not meet the requirements for FMT. A nasogastric/nasoduodenal/nasal-jejunal tube is easy to use and has the lowest technical requirements. Retrograde colonic enemas via an anal tube are also a widely used method of delivery. Enemas are easy to perform at home, even in pediatric patients. However, enemas are only effective for colon diseases because the transplanted microbiota may not be distributed throughout the entire intestine. Each FMT method has advantages and disadvantages. Clinicians should, therefore, select the appropriate approach based on the purpose and technical capabilities.Table 3 Advantages and disadvantages of the different delivery approaches for FMT Approaches Advantage Disadvantage Nasogastric/nasoduodenal/nasojejunal Avoids sedation Low cost Discomfort of tube placement Risk of vomiting and aspiration Inability to evaluate mucosa or take biopsies Capsules No sedation risk Less invasive Can be administered in office setting Expensive Less effective than colonoscopy Capsule burden Risks of vomiting and aspiration Colonoscopy Ability to evaluate mucosa and take biopsies Most effective route for treatment of rCDI Invasive and requires sedation Standard risks of colonoscopy (discomfort, perforation, bleeding) Expensive Enema Low cost Less invasive and avoids sedation Easy to carry out in office or at home Donor stool does not reach the entire colon and limited to distal colon Less effective than other routes for rCDI FMT fecal microbiota transplantation, rCDI recurrent Clostridium difficile infection Safety of fecal microbiota transplantation Although FMT has shown excellent therapeutic effects in pediatric diseases, it is worth noting that many AEs have been reported [34, 79]. The most common AEs included abdominal pain, gastrointestinal flatulence, diarrhea, constipation, fever, nausea, and vomiting [80, 81]. Serious complications, such as sedation-induced aspiration, perforation, bleeding, toxic megacolon with sepsis and peritonitis, fatal aspiration pneumonia, and death under anesthesia, have also been reported [80]. Potential risks for pediatric FMT include infectious diseases, obesity, diabetes, atherosclerosis, cancer, NAFLD, and asthma [80]. In the pediatric population, especially in newborns, specific AEs include belching, abdominal distention, abdominal pain, vomiting, diarrhea, fever, or transient CRP elevation [19, 28, 48, 82]. Kumagai reported that the AEs in clinical course of UC in a child who received FMT was transient fever and abdominal pain [83]. From 2013 to 2018, Zhang focused on AEs in the short and long terms in pediatric FMT patients. Only a few children developed AEs in the short term, while few AEs occurred during the long-term follow-up. Indeed, no fatal AEs associated with FMT have been reported in children [19, 82, 84–87]. Perspectives and future FMT has become widely practiced over recent years, and interest in FMT has surged among pediatricians and patients. Although the therapeutic effect of FMT in adults is satisfactory, the clinical practice of FMT in pediatrics needs to be improved and supplemented. The gap of FMT in the treatment of pediatric diseases reminds pediatricians that they need to consider many challenges, such as the required dosage, duration, onset, and end point of treatment. Future pediatric guidelines/studies should specify established indications versus potential/scientific indications. Additional larger, controlled, and prospective studies are needed to clarify both the safety and efficacy of FMT in pediatrics. Author contributions ZSC:writing–review and editing. GX: formal analysis, writing–original draft. CZH: data curation. CZH and GX contributed equally as the first author. Funding This project was funded by the National Natural Science Foundation of China (30700917, 81570465) and Minsheng Foundation of Joint Research Project of Liaoning Province（2021JH2/10300129）. Data availability Not required. Declarations Conflict of interest No financial benefits have been received from any party related directly or indirectly to the subject of this article. Ethical approval 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. Zhang F Luo W Shi Y Fan Z Ji G Should we standardize the 1700-year-old fecal microbiota transplantation? Am J Gastroenterol 2012 107 11 1755 10.1038/ajg.2012.251 23160295 2. Vindigni SM Surawicz CM Fecal microbiota transplantation Gastroenterol Clin North Am 2017 46 1 171 185 10.1016/j.gtc.2016.09.012 28164849 3. Eiseman B Silen W Bascom GS Kauvar AJ Fecal enema as an adjunct in the treatment of pseudomembranous enterocolitis Surgery 1958 44 5 854 859 13592638 4. Palmer R Fecal matters Nat Med 2011 17 2 150 152 10.1038/nm0211-150 21297602 5. Hamilton MJ Weingarden AR Sadowsky MJ Khoruts A Standardized frozen preparation for transplantation of fecal microbiota for recurrent Clostridium difficile infection Am J Gastroenterol 2012 107 761 767 10.1038/ajg.2011.482 22290405 6. van Nood E Vrieze A Nieuwdorp M Fuentes S Zoetendal EG de Vos WM Duodenal infusion of donor feces for recurrent Clostridium difficile N Engl J Med 2013 368 407 415 10.1056/NEJMoa1205037 23323867 7. Wang JW Kuo CH Kuo FC Wang YK Hsu WH Yu FJ Fecal microbiota transplantation: review and update J Formos Med Assoc 2019 118 S23 S31 10.1016/j.jfma.2018.08.011 30181015 8. Russell G Kaplan J Ferraro M Michelow IC Fecal bacteriotherapy for relapsing Clostridium difficile infection in a child: a proposed treatment protocol Pediatrics 2010 126 e239 e242 10.1542/peds.2009-3363 20547640 9. Khoruts A Dicksved J Jansson JK Sadowsky MJ Changes in the composition of the human fecal microbiome after bacteriotherapy for recurrent Clostridium difficile-associated diarrhea J Clin Gastroenterol 2010 44 354 360 10.1097/MCG.0b013e3181c87e02 20048681 10. Kronman MP Nielson HJ Adler AL Giefer MJ Wahbeh G Singh N Fecal microbiota transplantation via nasogastric tube for recurrent Clostridium difficile infection in pediatric patients J Pediatr Gastroenterol Nutr 2015 60 23 26 10.1097/MPG.0000000000000545 25162365 11. Chang JY Antonopoulos DA Kalra A Tonelli A Khalife WT Schmidt TM Decreased diversity of the fecal microbiome in recurrent Clostridium difficile-associated diarrhea J Infect Dis 2008 197 435 438 10.1086/525047 18199029 12. Kahn SA Young S Rubin DT Colonoscopic fecal microbiota transplant for recurrent Clostridium difficile infection in a child Am J Gastroenterol 2012 107 1930 1931 10.1038/ajg.2012.351 23211865 13. Pierog A Mencin A Reilly NR Fecal microbiota transplantation in children with recurrent Clostridium difficile infection Pediatr Infect Dis J 2014 33 1198 1200 10.1097/INF.0000000000000419 24853539 14. Shen ZH Zhu CX Quan YS Yang ZY Wu S Luo WW Relationship between intestinal microbiota and ulcerative colitis: mechanisms and clinical application of probiotics and fecal microbiota transplantation World J Gastroenterol 2018 24 5 14 10.3748/wjg.v24.i1.5 29358877 15. Dutta SK Girotra M Garg S Dutta A von Rosenvinge EC Maddox C Efficacy of combined jejunal and colonic fecal microbiota transplantation for recurrent Clostridium difficile Infection Clin Gastroenterol Hepatol 2014 12 1572 1576 10.1016/j.cgh.2013.12.032 24440222 16. Smits LP Bouter KE de Vos WM Borody TJ Nieuwdorp M Therapeutic potential of fecal microbiota transplantation Gastroenterology 2013 145 946 953 10.1053/j.gastro.2013.08.058 24018052 17. Morrison DJ Preston T Formation of short chain fatty acids by the gut microbiota and their impact on human metabolism Gut Microbes 2016 7 189 200 10.1080/19490976.2015.1134082 26963409 18. Yang YX Chen X Gan HT Toll-like receptor 2 regulates intestinal inflammation by controlling integrity of the enteric nervous system: why were TLR3's roles not tested? Gastroenterology 2014 146 1428 10.1053/j.gastro.2014.01.069 24681177 19. Gurram B Sue PK Fecal microbiota transplantation in children: current concepts Curr Opin Pediatr 2019 31 623 629 10.1097/MOP.0000000000000787 31169545 20. Gough E Shaikh H Manges AR Systematic review of intestinal microbiota transplantation (fecal bacteriotherapy) for recurrent Clostridium difficile infection Clin Infect Dis 2011 53 994 1002 10.1093/cid/cir632 22002980 21. Wang J Xiao Y Lin K Song F Ge T Zhang T Pediatric severe pseudomembranous enteritis treated with fecal microbiota transplantation in a 13-month-old infant Biomed Rep 2015 3 173 175 10.3892/br.2014.403 25798243 22. Walia R Garg S Song Y Girotra M Cuffari C Fricke WF Efficacy of fecal microbiota transplantation in 2 children with recurrent Clostridium difficile infection and its impact on their growth and gut microbiome J Pediatr Gastroenterol Nutr 2014 59 565 570 10.1097/MPG.0000000000000495 25023578 23. Hourigan SK Ahn M Gibson KM Perez-Losada M Felix G Weidner M Fecal transplant in children with Clostridioides difficile gives sustained reduction in antimicrobial resistance and potential pathogen burden Open Forum Infect Dis 2019 6 379 10.1093/ofid/ofz379 24. Karolewska-Bochenek K Grzesiowski P Banaszkiewicz A Gawronska A Kotowska M Dziekiewicz M A two-week fecal microbiota transplantation course in pediatric patients with inflammatory bowel disease Adv Exp Med Biol 2018 1047 81 87 10.1007/5584_2017_123 29151253 25. Kunde S Pham A Bonczyk S Crumb T Duba M Conrad H Jr Safety, tolerability, and clinical response after fecal transplantation in children and young adults with ulcerative colitis J Pediatr Gastroenterol Nutr 2013 56 597 601 10.1097/MPG.0b013e318292fa0d 23542823 26. Caldeira LF Borba HH Tonin FS Wiens A Fernandez-Llimos F Pontarolo R Fecal microbiota transplantation in inflammatory bowel disease patients: a systematic review and meta-analysis PLoS ONE 2020 15 e0238910 10.1371/journal.pone.0238910 32946509 27. Hourigan SK Chen LA Grigoryan Z Laroche G Weidner M Sears CL Microbiome changes associated with sustained eradication of Clostridium difficile after single faecal microbiota transplantation in children with and without inflammatory bowel disease Aliment Pharmacol Ther 2015 42 741 752 10.1111/apt.13326 26198180 28. Goyal A Yeh A Bush BR Firek BA Siebold LM Rogers MB Safety, clinical response, and microbiome findings following fecal microbiota transplant in children with inflammatory bowel disease Inflamm Bowel Dis 2018 24 410 421 10.1093/ibd/izx035 29361092 29. Cho S Spencer E Hirten R Grinspan A Dubinsky MC Fecal microbiota transplant for recurrent Clostridium difficile infection in pediatric inflammatory bowel disease J Pediatr Gastroenterol Nutr 2019 68 343 347 10.1097/MPG.0000000000002172 30320666 30. Moutinho BD Baima JP Rigo FF Saad-Hossne R Rodrigues J Romeiro FG Fecal microbiota transplantation in refractory ulcerative colitis - a case report J Int Med Res 2019 47 1072 1079 10.1177/0300060518821790 30632438 31. Shimizu H Arai K Abe J Nakabayashi K Yoshioka T Hosoi K Repeated fecal microbiota transplantation in a child with ulcerative colitis Pediatr Int 2016 58 781 785 10.1111/ped.12967 27324973 32. Asha MZ Khalil SFH Efficacy and safety of probiotics, prebiotics and synbiotics in the treatment of irritable bowel syndrome: a systematic review and meta-analysis Sultan Qaboos Univ Med J 2020 20 e13 e24 10.18295/squmj.2020.20.01.003 32190365 33. Johnsen PH Hilpusch F Cavanagh JP Leikanger IS Kolstad C Valle PC Faecal microbiota transplantation versus placebo for moderate-to-severe irritable bowel syndrome: a double-blind, randomised, placebo-controlled, parallel-group, single-centre trial Lancet Gastroenterol Hepatol 2018 3 17 24 10.1016/S2468-1253(17)30338-2 29100842 34. Green JE Davis JA Berk M Hair C Loughman A Castle D Efficacy and safety of fecal microbiota transplantation for the treatment of diseases other than Clostridium difficile infection: a systematic review and meta-analysis Gut Microbes 2020 12 1 25 10.1080/19490976.2020.1854640 35. Tian H Ding C Gong J Ge X McFarland LV Gu L Treatment of slow transit constipation with fecal microbiota transplantation: a pilot study J Clin Gastroenterol 2016 50 865 870 10.1097/MCG.0000000000000472 26751143 36. de Meij TG de Groot EF Eck A Budding AE Kneepkens CM Benninga MA Characterization of microbiota in children with chronic functional constipation PLoS ONE 2016 11 e0164731 10.1371/journal.pone.0164731 27760208 37. Ohkusa T Koido S Nishikawa Y Sato N Gut microbiota and chronic constipation: a review and update Front Med (Lausanne) 2019 6 19 10.3389/fmed.2019.00019 30809523 38. Yatsunenko T Rey FE Manary MJ Trehan I Dominguez-Bello MG Contreras M Human gut microbiome viewed across age and geography Nature 2012 486 222 227 10.1038/nature11053 22699611 39. Clemente MG Mandato C Poeta M Vajro P Pediatric non-alcoholic fatty liver disease: recent solutions, unresolved issues, and future research directions World J Gastroenterol 2016 22 8078 8093 10.3748/wjg.v22.i36.8078 27688650 40. Wang HB Wang PY Wang X Wan YL Liu YC Butyrate enhances intestinal epithelial barrier function via up-regulation of tight junction protein claudin-1 transcription Dig Dis Sci 2012 57 3126 3135 10.1007/s10620-012-2259-4 22684624 41. Suk KT Kim DJ Gut microbiota: novel therapeutic target for nonalcoholic fatty liver disease Expert Rev Gastroenterol Hepatol 2019 13 193 204 10.1080/17474124.2019.1569513 30791767 42. Garcia-Lezana T Raurell I Bravo M Torres-Arauz M Salcedo MT Santiago A Restoration of a healthy intestinal microbiota normalizes portal hypertension in a rat model of nonalcoholic steatohepatitis Hepatology 2018 67 1485 1498 10.1002/hep.29646 29113028 43. Philips CA Pande A Shasthry SM Jamwal KD Khillan V Chandel SS Healthy donor fecal microbiota transplantation in steroid-ineligible severe alcoholic hepatitis: a pilot study Clin Gastroenterol Hepatol 2017 15 600 602 10.1016/j.cgh.2016.10.029 27816755 44. Li Q Zhou JM The microbiota-gut-brain axis and its potential therapeutic role in autism spectrum disorder Neuroscience 2016 324 131 139 10.1016/j.neuroscience.2016.03.013 26964681 45. Kang DW Adams JB Gregory AC Borody T Chittick L Fasano A Microbiota transfer therapy alters gut ecosystem and improves gastrointestinal and autism symptoms: an open-label study Microbiome 2017 5 10 10.1186/s40168-016-0225-7 28122648 46. Li N Chen H Cheng Y Xu F Ruan G Ying S Fecal microbiota transplantation relieves gastrointestinal and autism symptoms by improving the gut microbiota in an open-label study Front Cell Infect Microbiol 2021 11 759435 10.3389/fcimb.2021.759435 34737978 47. Chen Y Xueying Z Jiaqu C Qiyi C Huanlong Q Ning L FTACMT study protocol: a multicentre, double-blind, randomised, placebo-controlled trial of faecal microbiota transplantation for autism spectrum disorder BMJ Open 2022 12 e051613 10.1136/bmjopen-2021-051613 48. Kang DW Adams JB Coleman DM Pollard EL Maldonado J McDonough-Means S Long-term benefit of microbiota transfer therapy on autism symptoms and gut microbiota Sci Rep 2019 9 5821 10.1038/s41598-019-42183-0 30967657 49. De Palma G Lynch MD Lu J Dang VT Deng Y Jury J Transplantation of fecal microbiota from patients with irritable bowel syndrome alters gut function and behavior in recipient mice Sci Transl Med 2017 9 eaaf6397 10.1126/scitranslmed.aaf6397 28251905 50. Bellono NW Bayrer JR Leitch DB Castro J Zhang C O'Donnell TA Enterochromaffin cells are gut chemosensors that couple to sensory neural pathways Cell 2017 170 185 98e16 10.1016/j.cell.2017.05.034 28648659 51. Bohorquez DV Shahid RA Erdmann A Kreger AM Wang Y Calakos N Neuroepithelial circuit formed by innervation of sensory enteroendocrine cells J Clin Invest 2015 125 782 786 10.1172/JCI78361 25555217 52. Leiva-Gea I Sanchez-Alcoholado L Martin-Tejedor B Castellano-Castillo D Moreno-Indias I Urda-Cardona A Gut microbiota differs in composition and functionality between children with type 1 diabetes and MODY2 and healthy control subjects: a case-control study Diabetes Care 2018 41 2385 2395 10.2337/dc18-0253 30224347 53. Vrieze A Van Nood E Holleman F Salojarvi J Kootte RS Bartelsman JF Transfer of intestinal microbiota from lean donors increases insulin sensitivity in individuals with metabolic syndrome Gastroenterology 2012 143 913 916e7 10.1053/j.gastro.2012.06.031 22728514 54. Backhed F Manchester JK Semenkovich CF Gordon JI Mechanisms underlying the resistance to diet-induced obesity in germ-free mice Proc Natl Acad Sci U S A 2007 104 979 984 10.1073/pnas.0605374104 17210919 55. Turnbaugh PJ Ley RE Mahowald MA Magrini V Mardis ER Gordon JI An obesity-associated gut microbiome with increased capacity for energy harvest Nature 2006 444 1027 1031 10.1038/nature05414 17183312 56. Solito A Bozzi Cionci N Calgaro M Caputo M Vannini L Hasballa I Supplementation with Bifidobacterium breve BR03 and B632 strains improved insulin sensitivity in children and adolescents with obesity in a cross-over, randomized double-blind placebo-controlled trial Clin Nutr 2021 40 4585 4594 10.1016/j.clnu.2021.06.002 34229263 57. Kelly CR Kahn S Kashyap P Laine L Rubin D Atreja A Update on fecal microbiota transplantation 2015: indications, methodologies, mechanisms, and outlook Gastroenterology 2015 149 223 237 10.1053/j.gastro.2015.05.008 25982290 58. Woodworth MH Neish EM Miller NS Dhere T Burd EM Carpentieri C Laboratory testing of donors and stool samples for fecal microbiota transplantation for recurrent clostridium difficile infection J Clin Microbiol 2017 55 1002 1010 10.1128/JCM.02327-16 28077694 59. Collins SM Kassam Z Bercik P The adoptive transfer of behavioral phenotype via the intestinal microbiota: experimental evidence and clinical implications Curr Opin Microbiol 2013 16 240 245 10.1016/j.mib.2013.06.004 23845749 60. Cammarota G Ianiro G Kelly CR Mullish BH Allegretti JR Kassam Z International consensus conference on stool banking for faecal microbiota transplantation in clinical practice Gut 2019 68 2111 2121 10.1136/gutjnl-2019-319548 31563878 61. Cammarota G Ianiro G Tilg H Rajilic-Stojanovic M Kump P Satokari R European consensus conference on faecal microbiota transplantation in clinical practice Gut 2017 66 569 580 10.1136/gutjnl-2016-313017 28087657 62. Khoruts A Rank KM Newman KM Viskocil K Vaughn BP Hamilton MJ Inflammatory bowel disease affects the outcome of fecal microbiota transplantation for recurrent Clostridium difficile infection Clin Gastroenterol Hepatol 2016 14 1433 1438 10.1016/j.cgh.2016.02.018 26905904 63. Khanna S Pardi DS Kelly CR Kraft CS Dhere T Henn MR A novel microbiome therapeutic increases gut microbial diversity and prevents recurrent Clostridium difficile infection J Infect Dis 2016 214 173 181 10.1093/infdis/jiv766 26908752 64. Newman KM Rank KM Vaughn BP Khoruts A Treatment of recurrent Clostridium difficile infection using fecal microbiota transplantation in patients with inflammatory bowel disease Gut Microbes 2017 8 303 309 10.1080/19490976.2017.1279377 28102756 65. Kassam Z Dubois N Ramakrishna B Ling K Qazi T Smith M Donor screening for fecal microbiota transplantation N Engl J Med 2019 381 2070 2072 10.1056/NEJMc1913670 31665572 66. Keller JJ Ooijevaar RE Hvas CL Terveer EM Lieberknecht SC Hogenauer C A standardised model for stool banking for faecal microbiota transplantation: a consensus report from a multidisciplinary UEG working group United European Gastroenterol J 2021 9 229 247 10.1177/2050640620967898 67. Costello SP Hughes PA Waters O Bryant RV Vincent AD Blatchford P Effect of fecal microbiota transplantation on 8-week remission in patients with ulcerative colitis: a randomized clinical trial JAMA 2019 321 156 164 10.1001/jama.2018.20046 30644982 68. Sood A Mahajan R Singh A Midha V Mehta V Narang V Role of faecal microbiota transplantation for maintenance of remission in patients with ulcerative colitis: a pilot study J Crohns Colitis 2019 13 1311 1317 10.1093/ecco-jcc/jjz060 30873549 69. Paramsothy S Kamm MA Kaakoush NO Walsh AJ van den Bogaerde J Samuel D Multidonor intensive faecal microbiota transplantation for active ulcerative colitis: a randomised placebo-controlled trial Lancet 2017 389 1218 1228 10.1016/S0140-6736(17)30182-4 28214091 70. Moayyedi P Surette MG Kim PT Libertucci J Wolfe M Onischi C Fecal microbiota transplantation induces remission in patients with active ulcerative colitis in a randomized controlled trial Gastroenterology 2015 149 102 109e6 10.1053/j.gastro.2015.04.001 25857665 71. Kao D Roach B Silva M Beck P Rioux K Kaplan GG Effect of oral capsule- vs colonoscopy-delivered fecal microbiota transplantation on recurrent Clostridium difficile infection: a randomized clinical trial JAMA 2017 318 1985 1993 10.1001/jama.2017.17077 29183074 72. Nicholson MR Mitchell PD Alexander E Ballal S Bartlett M Becker P Efficacy of fecal microbiota transplantation for Clostridium difficile infection in children Clin Gastroenterol Hepatol 2020 18 612 6191 10.1016/j.cgh.2019.04.037 31009795 73. Paramsothy S Borody TJ Lin E Finlayson S Walsh AJ Samuel D Donor recruitment for fecal microbiota transplantation Inflamm Bowel Dis 2015 21 1600 1606 10.1097/MIB.0000000000000405 26070003 74. Seward E Recent advances in colonoscopy F1000Res 2019 8 1028 10.12688/f1000research.18503.1 75. Nambu R Hagiwara SI Kakuta F Hara T Shimizu H Abukawa D Current role of colonoscopy in infants and young children: a multicenter study BMC Gastroenterol 2019 19 149 10.1186/s12876-019-1060-7 31429721 76. Kellermayer R Fecal microbiota transplantation: great potential with many challenges Transl Gastroenterol Hepatol 2019 4 40 10.21037/tgh.2019.05.10 31231707 77. Zhong M Buch H Wen Q Long C Cui B Zhang F Colonic transendoscopic enteral tubing: route for a novel, safe, and convenient delivery of washed microbiota transplantation in children Gastroenterol Res Pract 2021 2021 6676962 10.1155/2021/6676962 33510783 78. Kim SY Kim HS Park HJ Adverse events related to colonoscopy: global trends and future challenges World J Gastroenterol 2019 25 190 204 10.3748/wjg.v25.i2.190 30670909 79. Park SY Seo GS Fecal microbiota transplantation: is it safe? Clin Endosc 2021 54 157 160 10.5946/ce.2021.072 33827154 80. Wang S Xu M Wang W Cao X Piao M Khan S Systematic review: adverse events of fecal microbiota transplantation PLoS ONE 2016 11 e0161174 10.1371/journal.pone.0161174 27529553 81. Saha S Mara K Pardi DS Khanna S Long-term safety of fecal microbiota transplantation for recurrent Clostridioides difficile infection Gastroenterology 2021 160 1961 1969.e3 10.1053/j.gastro.2021.01.010 33444573 82. Zhang XY Wang YZ Li XL Hu H Liu HF Li D Safety of fecal microbiota transplantation in Chinese children: a single-center retrospective study World J Clin Cases 2018 6 1121 1127 10.12998/wjcc.v6.i161.1121 30613670 83. Kumagai H Yokoyama K Imagawa T Inoue S Tulyeu J Tanaka M Failure of fecal microbiota transplantation in a three-year-old child with severe refractory ulcerative colitis Pediatr Gastroenterol Hepatol Nutr 2016 19 214 220 10.5223/pghn.2016.19.3.214 27738605 84. Kelly CR Ihunnah C Fischer M Khoruts A Surawicz C Afzali A Fecal microbiota transplant for treatment of Clostridium difficile infection in immunocompromised patients Am J Gastroenterol 2014 109 1065 1071 10.1038/ajg.2014.133 24890442 85. Baxter M Ahmad T Colville A Sheridan R Fatal aspiration pneumonia as a complication of fecal microbiota transplant Clin Infect Dis 2015 61 136 137 10.1093/cid/civ247 25805303 86. Hogenauer C Kump PK Krause R Tempered enthusiasm for fecal transplantation? Clin Infect Dis 2014 59 1348 1349 10.1093/cid/ciu567 25034424 87. Goldenberg SD Batra R Beales I Digby-Bell JL Irving PM Kellingray L Comparison of different strategies for providing fecal microbiota transplantation to treat patients with recurrent Clostridium difficile infection in two English hospitals: a review Infect Dis Ther 2018 7 71 86 10.1007/s40121-018-0189-y 29450831	36484871	PMC9734408	NO-CC CODE	2022-12-14 23:28:28	no	World J Pediatr. 2022 Dec 9;:1-10	utf-8	World J Pediatr	2,022	10.1007/s12519-022-00655-w	oa_other
==== Front Acad Psychiatry Acad Psychiatry Academic Psychiatry 1042-9670 1545-7230 Springer International Publishing Cham 36471234 1734 10.1007/s40596-022-01734-w Faculty Viewpoint Surviving and Mastering the Real World: Reflections from Year 1 Post Training in Academic Psychiatry http://orcid.org/0000-0003-3448-806X Ouyang Jessica Xiaoxi [email protected] 1 Railey Michael T. 2 1 grid.411667.3 0000 0001 2186 0438 Georgetown University Medical Center, Washington, DC USA 2 grid.262962.b 0000 0004 1936 9342 Saint Louis University School of Medicine, St. Louis, MO USA 5 12 2022 12 24 7 2022 17 11 2022 © The Author(s), under exclusive licence to American Association of Chairs of Departments of Psychiatry, American Association of Directors of Psychiatric Residency Training, Association for Academic Psychiatry and Association of Directors of Medical Student Education in Psychiatry 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 pmcIn the middle of the COVID-19 pandemic, after moving to a new city in a new medical system, I (JXO) finally became a new attending psychiatrist. Despite the well-traveled road from medical student to faculty, the day-to-day adjustments required as a new faculty member in academic medicine were challenging to me. I was surprised, but relieved, when my mentor (MTR) expressed similar feelings of unpreparedness when he made his first transition. The following is a starting compilation of shared and remarkably similar reflections to help others beginning their own new roles to adapt more comfortably and thrive. Avoid taking workplace disagreements and “unreasonable” requests personally; instead, pause, and use “wise mind” [1]. Do not be surprised if you are asked to perform a series of tasks outside of your skillset and role. It may even be instinctive to feel confused and irritated. However, stepping back to ask, “What is the goal of the request?” is necessary. People with whom you may disagree still have a need to be seen, heard, and understood by you. In fact, they are your partners in problem-solving. By being curious about the end goal, we can get into collaborative problem-solving mode. Here are examples of gracious responses: “It sounds like there is a need for additional support in this area of the project. Taking this on would make it hard for me to accomplish [list duties] our team needs. How about I help to brainstorm a better solution?” or “That sounds interesting. Let me think about it some more and come back to this at the next meeting.” Reframing the goal of the original request can bring out a more collaborative resolution and is a critical skill to develop at any institution. Unfortunately, it will not develop overnight, but practice will lead to more satisfying results. Boundaries involving personal limitations and skillset limitations are important to reconcile to help define your scope of work and appropriate your time properly. In other words, know what should and should not be asked of you. This can be tricky, especially as a new attending who is still building a reputation, but having defined boundaries is vital for the longevity and work satisfaction of the new attending physician. There is a culture in academics of saying yes to everything, for that is the “can-do” and “team player” attitude that everyone is seeking. If you say yes to everything and do not respect or enforce your boundaries, your burnout might actually be your problem. Agreeableness and collegiality are not the same as saying yes and then becoming overwhelmed; being resentful does not help anyone. Having mentors and other supportive relationships will be invaluable here. Camaraderie and relationship development require effort and time to be cultivated. Intentionality is necessary to create your community when it is not built-in. Consider these practical steps: First, take advantage of what exists. Join the mentoring initiative or early career group from a professional organization or your institution. Second, invest in your peers (e.g., five psychiatrists were hired around the same time during the pandemic, and a peer group for supervision and community was born). Third, know your interests outside of your job and use them to build comradery with colleagues. Finally, maintain your connections with loved ones, because not only will this help you stay resilient but also these communities are your roots. Remind the people from whose relationships you derive joy and life that they are important, and show them that it is true. Discover yourself, your purpose, and your needs as you transition. Caring for your whole person may mean sometimes challenging conventional medicine with “You are not only a physical being, but emotional, intellectual, social, and spiritual; all these areas need nurturing. What is going well in these areas, and what do you need to work on?” But we as physicians often neglect to care for one or more domains of our personhood. For us, a daily spiritual practice is a need to remind us of our purposes. Needs are different from wants: “I want to binge Netflix, but I need to sleep. I want to distract my conflicts away, but I need to confront and resolve them.” Being honest and reflective is an invaluable routine that has helped us win battles with not only ourselves but the struggles against broken systems and difficult encounters. For only when we are filled can we pour out to others. Email management is a major task that all faculty face. Try to schedule email catchup times. Emails may contain hidden gems (particularly those from the faculty development office) like invitations to events or unique opportunities. Find and review your contract and benefits a few months in, because your understanding of what you signed and what you want is much better than when you first signed it. You will be glad you did; I learned that part of my pre-tax dollars can fund public transportation and that employees receive free coaching. Growth in your new position does not require perfection, but it does require self-awareness and intentionality. David Steindl-Rast, a Jesuit priest, said, “It is not joy that makes us grateful; it is gratitude that makes us joyful” [2]. During overwhelming and confusing times, it was almost certain I had forgotten to practice gratitude, rushing and thinking “I’m getting behind!” Yet it is when I slow down to re-ground myself, in and with my purpose, my relationships, and gratitude practice, I am the least frantic and the most joyful in my work. In closing, these reflections and lessons we learned are our attempt to begin to give language to the unwritten curriculum from some real-world classes. The materials and learning methods will vary in this next series of experiences post training, but we hope that the ideas described herein can spark a greater conversation in remembering our humanity as we go onto the next phase of the adventure. Declarations Disclosures 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. Linehan MM. Activating “wise mind”. In: Cognitive-behavioral treatment of borderline personality disorder. New York: The Guilford Press; 1993. pp. 214–6. 2. Steindl-Rast D Gratefulness, the heart of prayer 1984 New York Paulist Press	36471234	PMC9734410	NO-CC CODE	2022-12-14 23:28:28	no	Acad Psychiatry. 2022 Dec 5;:1-2	utf-8	Acad Psychiatry	2,022	10.1007/s40596-022-01734-w	oa_other
==== Front Mycopathologia Mycopathologia Mycopathologia 0301-486X 1573-0832 Springer Netherlands Dordrecht 686 10.1007/s11046-022-00686-x Mycopathologia IMAGE Generalized Tinea Incognito Developing from “Mask Tinea” Russo Roberto 12 Trave Ilaria 12 http://orcid.org/0000-0003-3108-4123 Cozzani Emanuele [email protected] 12 Parodi Aurora 12 1 grid.5606.5 0000 0001 2151 3065 DISSAL - Section of Dermatology, University of Genoa, Via Pastore 1, 16132 Genoa, Italy 2 Dermatology Unit, San Martino Polyclinic Hospital, Genoa, Italy 9 12 2022 14 19 7 2022 23 10 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. ==== Body pmcTinea incognito is a cutaneous eruption due to the application of topical steroids on tinea lesions, resulting in changes in their appearance. “Mask tinea” has been proposed as a term to describe cases of tinea faciei related to the use of protective face masks in the setting of SARS-CoV-2 pandemic [1, 2]. A 53-year-old, otherwise healthy woman presented with a generalized, itchy skin eruption. Lesions had started on their right cheek one month before: the patient showed a photograph of an arciform erythematous lesion following the profile of her face mask. She applied topical mometasone as her general practitioner suspected contact dermatitis. Then, the eruption spread to other districts, and the patient kept on applying mometasone daily. At the moment of our examination, a papular, scaly eruption was present on the right cheek, reaching the lower eyelid, and on the upper lip (Fig. 1). Contralaterally, ear showed scaling erythema and edema; also, there was an area of alopecia with desquamation in the parietal region, and a palpable, painful occipital lymphadenopathy (Fig. 2). The right hand was also involved, with papules coalescing in a scaly plaque at the wrist, and scaling erythema involving dorsum and palm (Fig. 3). Also, few papules were starting to the develop on the left knee. There was no sign of onychomycosis or tinea pedis, possible sources of contamination. Scales from right cheek, right palm and from the parietal alopecia were collected; coltures from all the samples demonstrated the presence of Trychophyton rubrum (Fig. 4). Histological examination with PAS staining of a skin biopsy from right wrist revealed hyphae and spores. Terbinafine 250 mg daily was administered for 4 weeks with complete resolution.Fig. 1 Papular, scaly eruption on face Fig. 2 Contralateral parietal alopecia and erythema and edema of ear Fig. 3 Scaly eruption involving hands and wrists Fig. 4 Trychophyton rubrum found on scales from the skin region involved (direct microscopic examination by potassium hydroxide 20% and methylene blue; cottony, white, and radial colony which produce reddish-brown pigment) As well as causing a frequent worsening of facial dermatoses such as acne, rosacea and seborrheic dermatitis, the use of face masks in the setting of the SARS-CoV-2 pandemic seems to have led to an increase in the incidence of tinea faciei. The potential of masks to favour the development of tinea faciei may be explained by their occlusive effect, with subsequent creation of a humid environment attributable to increased sweating and the breath itself. Also, a persistent mechanic microtrauma on skin might be postulated. In particular, traumatic triggers (i.e., scratching) may allow microbes to penetrate the cutaneous barrier and cause lesions with a mechanism called pseudo-Koebnerization. Our patient admitted the use of a face mask > 8 h/day without changing it. Maybe because of scratching, the eruption spread to other cutaneous districts, becoming generalized. The patient continued the steroid applications, so the appearance of lesions was altered, not showing the classical “ringworm” form. It was not possible to determine with certainty the source of infection, but the patient acknowledged she was taking care of stray cats. In conclusion, we want to highlight the possibility of tinea faciei, an infrequent (at least until 2020) disease, in the presence of an erythematous eruption on face, especially when unilateral, in patients wearing face masks. Patients using face masks for many hours/day should be advised to change their mask frequently, and carefully clean their skin. Funding No funds, grant or other support was received. Declarations Conflict of interest The authors have no competing interests to declare that are relevant to the content of this article. Ethical approval No ethical approval was required by our institution. Informed consent The patient gave written informed consent for publication. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Roberto Russo and Ilaria Trave contributed equally to this study. ==== Refs References 1. Bortoluzzi P Boneschi V Veraldi S "Mask" tinea: an increasing infection during COVID-19 pandemic Mycopathologia 2022 187 1 141 142 10.1007/s11046-021-00612-7. 34964931 2. Agarwal A Hassanandani T Das A Panda M Chakravorty S 'Mask tinea': tinea faciei possibly potentiated by prolonged mask usage during the COVID-19 pandemic Clin Exp Dermatol 2021 46 1 190 193 10.1111/ced.14491. 33098693	36494590	PMC9734412	NO-CC CODE	2022-12-14 23:28:28	no	Mycopathologia. 2022 Dec 9;:1-4	utf-8	Mycopathologia	2,022	10.1007/s11046-022-00686-x	oa_other
==== Front Natl Acad Sci Lett Natl Acad Sci Lett National Academy Science Letters. National Academy of Sciences, India 0250-541X 2250-1754 Springer India New Delhi 1194 10.1007/s40009-022-01194-8 Short Communication A Paradigm Shift is Expected in Ethnobiology: Challenges and Opportunities Post-COVID-19 Sharma Alpy Uniyal Sanjay Kr. [email protected] grid.417640.0 0000 0004 0500 553X Environmental Technology Division, CSIR-Institute of the Himalayan Bioresource Technology, Palampur, 176061 Himachal Pradesh India 7 12 2022 14 23 4 2022 18 10 2022 21 10 2022 © The Author(s), under exclusive licence to The National Academy of Sciences, India 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. Documentation of the folk knowledge of indigenous communities forms an integral part of the subject “ethnobiology”. Pursuing leads obtained through ethnobiological documentation has played a key role in maintaining human health and wellbeing. The current pandemic that we are passing through is expected to strengthen the subject with many challenges and opportunities. In this paper, we highlight the avenues and the role of the subject in the times to come. We strongly believe a paradigm shift in ethnobiology is lurking around the corner. Keywords Ethnobiologists Indigenous communities Plants Traditional knowledge ==== Body pmcThe current pandemic is expected to bring about changes in the human use of natural resources and societal relationships. Herein, with many challenging opportunities, ethnobiology is anticipated to become central to human ecology. Ethnobiology refers to studying human use of resources for livelihood, their linkages, beliefs, and knowledge; especially in landscapes inhabited by indigenous rural people [1]. Over the years, the discipline has seen a major change from being qualitative in the past to being more quantitative and validation oriented now [2, 3]. From being local, which is key, to be cross-cultural and global, it now forms the basis of prospection and pursuing leads that have played a crucial role in human livelihood. Reserpine from Rauvolfia serpentina and Quinine from Cinchona pubescens are potent examples of this [4]. There is absolutely no doubt that the current pandemic is going to bring about further changes in the use of natural resources, societal relationships, and scientific innovations, wherein ethnobiology and ethnobiologists are expected to play a central role. Therefore, ethnobiologists across the globe have started brainstorming about the challenges as well as the post-COVID-19 future of ethnobiology [5, 6]. The pandemic has its roots in human over-exploitation of nature and the act of being the “master” [7]. However, even today, indigenous societies treat themselves as a part of nature [8, 9], and the recent calls such as “our solutions are in nature” emphasize the role of nature in human wellbeing. We, therefore, are looking towards the rich traditional legacy for finding a cure for COVID-19. However, enhancing immunity is of immediate importance for passing through these tough times. This is coming up in a big way and is guided by the noted health benefits imparted by many traditionally used spices such as Cinnamomum tamala, Curcuma longa, and Zingiber officinale. This is being aggressively promoted by many Governments for enhancing immunity, and the potential of traditional medicines to fight against the virus is being explored [10, 11]. This also calls for validation of claims, rightful interventions, and disease management as was evident in case of Coronil [12]. All of these prioritize traditional knowledge vis-à-vis nature conservation which is so central to ethnobiology, and scientifically recording these is the prime task of ethnobiologist. Thus, ethnobiology is becoming pivotal. Nonetheless, as ethnobiology demands personal interactions with communities for eliciting knowledge, this is expected to become challenging. Communities may now become averse to visitors [13] and may hold them as culprits behind the disease outbreak, which in turn may limit knowledge sharing. It is to be noted that the implications of the current pandemic are far more grave for the indigenous communities who are relatively poor and are mostly at the receiving end [14]. How this will affect knowledge documentation remains to be seen. While we may have technology at our disposal, establishing emotional connections may now be more tiresome. If there are challenges, there are opportunities. A paradigm shift appears to be in waiting for ethnobiology. Indigenous knowledge systems and their documentation are now being recognized at multiple podiums [15], thereby having much more visibility and support. Therefore, scientific collaborations and resources for ethnobiological studies are expected to be majorly promoted. Many indigenous communities (e.g. bushmeat hunting tribes, fishermen, snake charmers, etc.) are directly involved in handling wild fauna. They are much more exposed to zoonotic viruses; understanding their characteristics and strategies is now being prioritized [5]. Recent reviews on COVID-19 highlight focussing on ethnomedicines, particularly the anti-viral plants [10, 11]. Thus, the pandemic has channelized us to work with indigenous communities for leads and possible cures [16]. During our interactions with the communities (Fig. 1), some of the plant species that were prominently reported by the local people to relieve COVID-19-related symptoms (respiratory problems, fever, cough, and cold) are listed in Table 1. The list does not include common species such as Curcuma longa, Foeniculum vulgare, Zinbiger officinale, etc. Herein, use of Bunium persicum and Carum carvi is noteworthy as these are traditional spices of the Himalayan region. While Tinospora cordifolia is a known immunity booster; Thymus linearis, Viola canescens, and Viola pilosa are known to help in shortness of breath (Fig. 2). These species, therefore, should be further explored with the involvement of the local communities who should be partners in the journey and benefits should also flow to them [17]. Also, the scientific validation, standardization, purity, and demonstration of their efficacy ought to be done for authenticity [18]. This has been a major lacuna in traditional medicines that impedes its growth and acceptance [19].Fig. 1 Ethnobiological interactions for documenting plant uses Table 1 Some of the species used by the communities of western Himalaya to manage symptoms of COVID-19 Species Local name Family Part used Uses Achillea millefolium Gurma Asteraceae Flower The decoction is used during cough and cold Aconitum heterophyllum Patish Ranunculaceae Root The dried root powder is used against cold and fever Angelica glauca Chora Apiaceae Root The decoction is used for curing cold and cough Allium humile Jangli pyaj Amaryllidaceae Whole plant Decoction of the dried plant is used against cough and chest pain Bunium persicum Kala jeera Apiaceae Seed The decoction is used during fever and cold. It is used for boosting immunity Carum carvi Shahi jeera Apiaceae Seed The decoction is used against cold and cough. Believed to boost immunity Cassiope fastigiata Sallu Ericaceae Whole plant Decoction is used during fever, cold, and cough Ephedra gerardiana Buchyur Ephedraceae Aerial part The decoction is used in fever and respiratory problems Inula racemosa Manu Asteraceae Root Root decoction is used against fever, cough and breathlessness Origanum vulgare Bantulsi Lamiaceae Leaf The decoction is taken to relieve cough and fever Picrorhiza kurrooa Karu Plantaginaceae Root Root decoction is used during fever Pleurospermum brunonis Losar Apiaceae Leaf The dried leaves are consumed raw during body pain and cold Rhododendron anthopogon Ballu Ericaceae Leaf The decoction is used during cold, fever, and asthma Rosa macrophylla Jangli gulaab Rosaceae Flower The flower decoction is used during cold and fever Taraxacum officinale Dudhi Asteraceae Aerial part The decoction is used in fever Tinospora cordifolia Giloy Menispermaceae Stem Decoction during fever. Also used for building immunity Thymus linearis Vanajwain Lamiaceae Whole plant Decoction is given for the treatment of cough, cold, and fever Viburnum cotinifolium Khimota Adoxaceae Fruit Consumed to overcome weakness and enhance immunity Viola canescens Napalu Violaceae Flower The decoction of the flower is used for curing cough and cold Viola pilosa Vanaksha Violaceae Whole plant The decoction is used for curing cold, cough, and respiratory problems Fig. 2 Some of the species and parts used by the communities—a Achillea millefolium, b Aconitum heterophyllum, c Ephedra gerardiana, d Taraxacum officinale, e Thymus linearis, f Viola canescens, g root/rhizome of Picrorhiza kurrooa, h seeds of Bunium persicum, i above-ground parts of Allium humile Until late, we believed what concerns others may not impact us. Today, none has been able to escape the wrath of the virus. Consequently, studying human relationships and social connectivity is receiving fresh impetus which overall caters to the Sustainable Development Goals. The outmigration of the rural population is a serious issue that not only pressurizes the already saturated towns but also leads to the erosion of folk knowledge [20]. However, now, reverse migration is being noted [20]. The pandemic has forced people to return to their roots, this is expected to help revive the erstwhile practices, and ultimately, it augurs well for folk knowledge. The dependence of indigenous people on natural resources is very high. They, thus, have devised traditional conservation practices (e.g. sacred groves, taboos, etc.) that impose restrictions on accessing resources. Recognizing that most of the biodiversity exists outside the protected areas, these practices play a crucial role in managing ecosystems. Studies on traditional practices and their basis are expected to see a resurgence. While we have been awed by the swaying exotic trees, it is time when urban landscapes come out of the realm of beauty alone. At times, exotics have become a menace that is hard to control [6]. Herein factoring native multiple-use plants (Ocimum, Azadirachta, Ficus, etc.) in designing urban landscapes will not only cater to pollution alleviation but will also introduce them to the new generation. This would certainly be guided by the location and the ecological characteristics of the species. It is an upcoming area of landscape designing and restoration, especially the urban green areas (parks, avenue trees, etc.). Thus, the pandemic is expected to bring a paradigm shift in the human relationship with nature, wherein the importance of ethnobiology and ethnobiologists would be much more pertinent. Acknowledgements The authors are thankful to the Director CSIR-IHBT for facilities and support; and many people who unknowingly sowed the seeds of this manuscript. Dr. Vikas Kumar is thanked for providing some of the plant photographs (Fig. 2d, e, f). Funding No funding was received for this work. Data Availability Not applicable to this article as no datasets were generated or analysed during the current study. Declarations Conflict of Interest The authors have no conflict of interest. Informed Consent Not Applicable. Significant Statement: The current pandemic is expected to bring about changes in the human use of natural resources and societal relationships. Herein, with many challenging opportunities, ethnobiology is anticipated to become central to human ecology. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Albuquerque UP Alves RRN Introduction to ethnobiology 2016 Berlin Springer 2. Phillips O Gentry AH The useful plants of Tambopata, Peru: II. Additional hypothesis testing in quantitative ethnobotany Econ Bot 1993 47 33 43 10.1007/BF02862204 3. Albuquerque UP Lucena RF Monteiro JM Florentino AT Cecília de Fátima CBR Evaluating two quantitative ethnobotanical techniques Ethnobot Res Appl 2006 4 051 060 10.17348/era.4.0.51-60 4. Fabricant DS Farnsworth NR The value of plants used in traditional medicine for drug discovery Environ Health Perspect 2001 109 69 75 11250806 5. Franco FM Bussmann RW Rising to the occasion: outlining Ethnobiologists’ response to the coronavirus (COVID-19) pandemic Ethnobot Res Appl 2020 20 1 4 6. Vandebroek I Pieroni A Stepp JR Hanazaki N Ladio A Alves RR Picking D Delgoda R Maroyi A van Andel T Quave CL Reshaping the future of ethnobiology research after the COVID-19 pandemic Nat Plants 2020 6 723 730 10.1038/s41477-020-0691-6 32572213 7. Volpato G Fontefrancesco MF Gruppuso P Zocchi DM Pieroni A Baby Pangolins on my plate: possible lessons to learn from the COVID19 pandemic J Ethnobiol Ethnomed 2020 16 1 12 10.1186/s13002-020-00366-4 31924218 8. Bhagwat SA Rutte C Sacred groves: potential for biodiversity management Front Ecol Environ 2006 4 519 524 10.1890/1540-9295(2006)4[519:SGPFBM]2.0.CO;2 9. Huang L Tian L Zhou L Jin C Qian S Jim CY Lin D Zhao L Minor J Coggins C Yang Y Local cultural beliefs and practices promote conservation of large old trees in an ethnic minority region in southwestern China Urban For Urban Green 2020 49 126584 10.1016/j.ufug.2020.126584 10. Lakshmi SA Mohamed R Shafreen B Priya A Shunmugiah KP Ethnomedicines of Indian origin for combating COVID-19 infection by hampering the viral replication: using structure-based drug discovery approach J Biomol Struct Dyn 2021 39 4594 4609 10.1080/07391102.2020.1778537 32573351 11. Benarba B Pandiella A Medicinal plants as sources of active molecules against COVID-19 Front Pharmacol 2020 11 1189 10.3389/fphar.2020.01189 32848790 12. Sujatha V The politics of medicine in a pandemic Econ Political Wkly 2021 56 30 19 24 13. Sharma S (2020) Malana bars outsiders till August, to fine offenders Rs 51,000. Times of India. http://timesofindia.indiatimes.com/articleshow/75447138.cms?utm_source=contentofinterest&utm_medium=text&utm_campaign=cppst 14. Ferrante L Fearnside PM Protect indigenous peoples from COVID-19 Science 2020 368 251 10.1126/science.abc0073 15. Díaz S Demissew S Carabias J Joly C Lonsdale M Ash N Larigauderie A Adhikari JR Arico S Báldi A Bartuska A The IPBES conceptual framework—connecting nature and people Curr Opin Environ Sustain 2015 14 1 16 10.1016/j.cosust.2014.11.002 16. Iwuoha VC Aniche ET COVID-19 lockdown and physical distancing policies are elitist: towards an indigenous (Afro-centred) approach to containing the pandemic in sub-urban slums in Nigeria Local Environ 2020 25 631 640 10.1080/13549839.2020.1801618 17. Varma RV Access and benefit sharing in India: challenges ahead Biodiversity for sustainable development 2017 Cham Springer 87 96 18. Bhosale VV Banerjee D Scientific validation of herbal medicine Herbal medicine in India 2020 Singapore Springer 573 579 19. Taylor JLS Rabe T McGaw LJ Jäger AK Van Staden J Towards the scientific validation of traditional medicinal plants Plant Growth Regul 2001 34 1 23 37 10.1023/A:1013310809275 20. Singh SK Patel V Chaudhary A Mishra N Reverse migration of labourers amidst COVID-19 Econ Political Wkly 2020 55 32 33	0	PMC9734414	NO-CC CODE	2022-12-14 23:28:28	no	Natl Acad Sci Lett. 2022 Dec 7;:1-4	utf-8	Natl Acad Sci Lett	2,022	10.1007/s40009-022-01194-8	oa_other
==== Front Biophys Rev Biophys Rev Biophysical Reviews 1867-2450 1867-2469 Springer Berlin Heidelberg Berlin/Heidelberg 1028 10.1007/s12551-022-01028-3 Review Protein binding sites for drug design http://orcid.org/0000-0003-0160-3375 Konc Janez 1 http://orcid.org/0000-0003-4067-0116 Janežič Dušanka [email protected] 2 1 grid.454324.0 0000 0001 0661 0844 Theory Department, National Institute of Chemistry, Hajdrihova 19, SI-1001 Ljubljana, Slovenia 2 grid.412740.4 0000 0001 0688 0879 Faculty of Mathematics, Natural Sciences and Information Technologies, University of Primorska, Glagoljaška 8, SI-6000 Koper, Slovenia 9 12 2022 19 12 10 2022 1 12 2022 © International Union for Pure and Applied Biophysics (IUPAB) and 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. Drug development is a lengthy and challenging process that can be accelerated at early stages by new mathematical approaches and modern computers. To address this important issue, we are developing new mathematical solutions for the detection and characterization of protein binding sites that are important for new drug development. In this review, we present algorithms based on graph theory combined with molecular dynamics simulations that we have developed for studying biological target proteins to provide important data for optimizing the early stages of new drug development. A particular focus is the development of new protein binding site prediction algorithms (ProBiS) and new web tools for modeling pharmaceutically interesting molecules—ProBiS Tools (algorithm, database, web server), which have evolved into a full-fledged graphical tool for studying proteins in the proteome. ProBiS differs from other structural algorithms in that it can align proteins with different folds without prior knowledge of the binding sites. It allows detection of similar binding sites and can predict molecular ligands of various types of pharmaceutical interest that could be advanced to drugs to treat a disease, based on the entire Protein Data Bank (PDB) and AlphaFold database, including proteins not yet in the PDB. All ProBiS Tools are freely available to the academic community at http://insilab.org and https://probis.nih.gov. Keywords Structural proteome Protein binding sites Prediction ProBiS http://dx.doi.org/10.13039/501100004329 Javna Agencija za Raziskovalno Dejavnost RS N1-0142 L7-8269 N1-0209 J1-1715 J1-9186 BI-JP/16-18-002 Konc Janez Janežič Dušanka ==== Body pmcIntroduction In developing new drugs and vaccines, the pharmaceutical industry is increasingly turning to molecular modeling, a field in science with potential to shorten the drug discovery process, which studies the properties of molecules by recreating them as models on computers (Martinez-Mayorga et al. 2020). In this paper, we will describe our newly developed molecular modeling tools that enable studying of protein binding sites, which are the targets of most drugs, and enable the prediction of their biochemical functions, and ligands that could be potentially used as drugs. The research questions addressed by these tools are important for the entire pharmaceutical industry, since the rational development of new drugs is only possible if the properties of protein binding sites and their ligands are known and well characterized. In recent decades, researchers have made discoveries that have revolutionized our understanding of cell structure and function, most notably the fact that proteins are capable of mutual communication and adapt their functions to current conditions in the cell (Phizicky and Fields 1995; Jones and Thornton 1996). Proteins have been found to form protein complexes by binding temporarily or permanently to each other or to other ligands. The binding occurs at a site on the protein surface called a binding site. Interactions between proteins are crucial for the functioning of biological systems, as they influence the function of proteins and ensure their self-regulation. Protein binding sites can form complexes with small molecules, such as receptor ligands and enzyme substrates, or with large molecules, such as nucleic acids, peptides, and other proteins, the nature of the binding being determined by the specific physicochemical properties of their surfaces (Fig. 1). The prerequisite for two proteins to interact is a complementary pattern of interactions on their surfaces or binding sites, so that, the proteins are attracted to each other (Weiner et al. 1982). In modeling protein interactions, the surface plays an important role, while the interior of the proteins plays a lesser role, since only the surface amino acids contribute to the attractive bonds with which the protein binds to its partners (Schmitt et al. 2002).Fig. 1 Binding site and different possible ligands on a protein The binding sites for small molecules, nucleic acid proteins, ions, and certain water molecules change slowly during evolution (Abrusán and Marsh 2018) and are conserved in the structures of related proteins found in the Protein Data Bank (PDB) (Berman et al. 2003; Kinjo et al. 2017). If we know the structure of a protein but not its binding sites, we can find binding sites on that protein by comparing it to the approximately 190,000 known structures in the PDB; similarities found between the structure studied and those from the database generally coincide with the binding sites on those proteins, and if the function of similar proteins in the database is known, the comparison also allows us to predict the function of the proteins studied (Konc et al. 2013). Many computational tools have been reported for binding site analysis and prediction (Kinoshita et al. 2002; Kinoshita and Nakamura 2005; Salentin et al. 2015; Jakubec et al. 2022), some of which are based on clique-finding algorithms (Ren et al. 2010; Chartier and Najmanovich 2015), and conservation of hot-spot structure may be insufficient for detection (Cukuroglu et al. 2012; Chen et al. 2012). Phylogenetic protein sequences have been used to detect the conservation of surface residue sequences (Glaser et al. 2003). Other traditional approaches take advantage of the fact that the 3D structure is evolutionarily more conserved than the residue sequence. The function of a protein can be determined by finding at least one structurally similar protein whose function is already known in the PDB. These methods compare the overall 3D shape of proteins or protein folds. We have developed protein binding site tools (ProBiS) consisting of web servers, databases, and protein and ligand binding site prediction algorithms, all based on a graph theoretic algorithm, a fast and improved maximum clique algorithm that we developed in 2007 (Fig. 2) (Konc and Janezic 2007). The ProBiS Tools have been independently validated and are widely used in pharmaceutical research (Ehrt et al. 2016, 2018; Fu et al. 2016; Vankayala et al. 2017; Ramatenki et al. 2017; Bancroft et al. 2019). One such application is the development of novel biological drugs, where a combination of ProBiS structure comparison and molecular dynamics simulations was used to reduce unwanted side effects of an antibody-based drug, ipilimumab, without compromising its stability or increasing its immunogenicity (Lešnik et al. 2020). The ProBiS algorithm compares local physicochemical and geometric properties of protein surface structures to identify common amino acid motifs independent of protein folding (Konc and Janežič 2010). This algorithm is unique because it does not require the binding site on the protein to be known in advance, but instead compares the entire surface of the protein in question to the surfaces of other proteins and identifies similar binding sites based on the detectable local surface similarities. An extension of this algorithm, the ProBiS-ligands web server (Konc and Janežič 2014), predicts the interactions and positions of ligands with a given protein based on the detection of similarities within protein binding sites in the PDB. Ligands that bind to similar binding sites identified by this algorithm are transferred to the query protein binding site by rotating and translating their coordinates based on the superposition of binding sites, with each group of ligands of the same type representing a predicted binding site.Fig. 2 A maximum clique problem. Maximum clique (red) is the largest fully connected subgraph within a graph In the following sections, we provide an overview of the development of ProBiS Tools and present examples of their capabilities. What is ProBiS? The ProBiS algorithm is computer software that allows the prediction of binding sites and the corresponding ligands for a given protein structure (Konc and Janežič 2010). It was originally developed in 2010 and has since been expanded into several ProBiS web servers and databases, all under the name ProBiS Tools that:enable rapid determination of binding sites for the entire PDB are, to the best of our knowledge, currently the only tools that can accurately determine the type of ligand for a predicted binding site are currently the only ones that allow the determination of binding sites and ligands for AlphaFold proteins not yet included in the PDB (Varadi et al. 2022). What is maximum clique algorithm? A maximum clique problem is an NP-hard problem for which there is most likely no polynomial solution. A maximum clique algorithm finds the largest fully connected subgraph (a clique) in an undirected graph, i.e., the one with the most vertices. We have developed an algorithm for finding a maximum clique in an undirected graph that is up to 100 times faster than the best comparable algorithm (Konc and Janezic 2007; Depolli et al. 2013; Reba et al. 2022). Protein graphs Proteins can be represented as protein graphs (Konc and Janežič 2007, 2010; Depolli et al. 2013). In protein graphs, the vertices have spatial coordinates, and they are located at the geometric centers of the functional groups of the amino acids of the protein surface. The vertices are labeled with five different colors corresponding to the five physicochemical properties, i.e., acceptor, donor, π-π-stacking, aliphatic, and acceptor–donor, of the protein surface amino acids at the resolution of the functional groups. Two vertices ui and uj of a protein graph G are adjacent, i.e., an edge (ui, uj) ∈ E(G) exists between them if the distance (ui, uj) is less than 15 Å. How can a maximum clique detect the local similarity of two proteins? A pair of protein graphs can be compared by finding a maximum clique, i.e., the clique with the most vertices, in their product graph, where the maximum clique represents the superposition that aligns the most vertices of the compared protein graphs (Konc and Janežič 2007, 2010; Depolli et al. 2013). The protein product graph of two protein graphs G1 and G2 is defined by the set of vertices V(G1, G2) = V(G1) × V(G2). Each vertex of the protein product graph (ui, vi) consists of two subvertices: a subvertex from the first protein graph (ui ∈ G1) and a subvertex from the second protein graph (vi ∈ G2). In general, a protein product graph has x × y vertices if the respective protein graphs have x and y vertices; however, we reduce its size by considering as product graph vertices only those where the two subvertices have identical colors, i.e., identical physicochemical properties, and similar neighborhoods. We connect two protein product graph vertices (ui, vi) and (uj, vj), where (ui, uj) ∈ E(G1) and (vi, vj) ∈ E(G2), by inserting an edge between them if \|distance(ui, uj) − distance(vi, vj)\| is less than 0.5 Å, which means that the distances between the respective first and second subvertices in both protein graphs must be nearly equal. A maximum clique in the protein product graph constructed in this way represents the largest similarity between the two compared protein graphs in terms of physicochemical and geometric properties and allows us to identify pairs of similar binding sites and other similar surface regions in proteins independent of their protein folds. ProBiS Tools development We have developed new methodological solutions for the prediction and study of protein binding sites and their ligands based on graph theoretical approaches combined with molecular simulations (Fig. 3). These are:Fig. 3 Timeline of ProBiS Tools development MaxCliqueDyn algorithm We have developed a new algorithm for finding a maximum clique in an undirected graph (http://insilab.org/maxclique), in which we have improved approximate coloring algorithm that is used by the maximum clique algorithm to provide bounds to the size of the maximum clique (Konc and Janezic 2007). We then extended this algorithm to include dynamically varying bounds to adapt the maximum clique search to the type of input graph (Depolli et al. 2013; Reba et al. 2022). We show that by applying tighter, more computationally expensive upper bounds on a fraction of the search space, it is possible to reduce the time to find the maximum clique. This resulting algorithm is significantly faster (between 10 × and 100 ×) than comparable algorithms. ProBiS algorithm The ProBiS algorithm (http://insilab.org/probis-algorithm) enables local structural matching of entire protein surface structures against a large database of protein structures in a reasonable amount of time (Konc and Janežič 2007, 2010). The comparison includes geometry and physicochemical properties and is performed at the amino acid functional group level. The algorithm compares the query protein to each of the database proteins, using the maximum clique algorithm (Konc and Janezic 2007; Depolli et al. 2013), which allows it efficiently to detect the largest similar subgraphs of compared protein graphs. For each pairwise comparison of the query protein with a database protein, the algorithm generates multiple local alignments of the surface regions found in both proteins; no attempt is made to align the proteins globally, and similar folding is not a requirement for a relationship between the two proteins. Because no assumptions are made about the localization of binding sites prior to comparison, ProBiS can discover new binding sites and suggest ligands that might host these binding sites. Due to the high computational cost, the comparison of such a number of proteins on a local level and in such a short time has not been possible before, even with high-performance computers. ProBiS-ligands web server The ProBiS-ligands web server (http://probis.cmm.ki.si) predicts the binding of ligands to a protein structure (Konc and Janežič 2014). Given a protein structure or binding site, ProBiS-ligands first identify template proteins in the PDB that have similar binding sites. Based on the superimpositions of the query protein and the similar binding sites found, the server then transfers the ligand structures from these sites to the query protein. Such ligand prediction supports many activities, such as drug repurposing (Štular et al. 2016). In addition to identifying protein ligands, the ProBiS-ligands web server can also be used to accurately identify structurally similar binding sites in protein structures or structural evolutionary conservation values. ProBiS-CHARMMing web server Unlike the ProBiS web servers, ProBiS-CHARMMing (https://probis.nih.gov) is hosted at the National Institutes of Health, Bethesda, MD, USA (Konc et al. 2015). ProBiS-CHARMMing provides all the features of the ProBiS web servers, plus molecular modeling capabilities, and allows minimization of predicted ligands and their binding sites and calculation of their interaction energies. This is achieved by integrating ProBiS with the CHARMMing web server at https://charmming.org (Miller et al. 2008; Brooks et al. 2009). The main strength of the ProBiS CHARMMing web server is that it is able to remove steric clashes between predicted ligands and proteins, which can be the cause of unrealistic models, and also remove the lack of energy-based scores to assess the strength of ligand binding. The web interface also facilitates the creation of CHARMM-friendly protein–ligand systems, including CHARMM input scripts for further modeling. The server can be used to predict energy-minimized holo protein structures, that is, protein–protein, protein-small molecule, and protein-ion complexes with unliganded (apo) protein structures as queries, and provides an interactive environment where users can explore the predicted protein–ligand complexes and calculate and compare their energy properties. GenProBiS The GenProBiS web server (http://genprobis.insilab.org) links sequence variants to protein structures and also to protein–protein, protein-nucleic acid, protein-compound, and protein-metal ion binding sites (Konc et al. 2017). This server enables intuitive visual exploration of extensive mapped variants, such as human cancer-associated somatic missense mutations and nonsynonymous single nucleotide polymorphisms from 21 species, within predicted binding site regions for approximately 80,000 PDB protein structures. It also enables the discovery of potentially deleterious sequence variants and the development of new hypotheses for drug discovery, e.g., to explain the sensitivity of a particular drug to a specific mutation in a protein binding site. ProBiS H2O We have developed the ProBiS H2O plugin for PyMOL (http://insilab.org/probis-h2o) that allows rapid identification of conserved water ligands in a protein structure or protein binding site using experimental protein structures from the PDB or a set of custom protein structures available to the user (Jukic et al. 2017). Identifying conserved water sites in protein structures is a challenging task that has applications in molecular docking and protein stability prediction. Using a protein structure, binding site, or single water molecule as a query, ProBiS H2O collects similar proteins from the PDB and performs local or binding site-specific superimpositions of the query structure with similar proteins using the ProBiS algorithm. It collects the experimental water molecules from similar proteins and transfers them to the query protein. The transferred water molecules are clustered according to their mutual proximity, identifying discrete sites in the query protein with high water conservation. ProBiS H2O MD The ProBiS H2O MD plugin (http://insilab.org/probis-h2o-md) is an extension of the ProBiS H2O approach and allows the identification of conserved waters as ligands from molecular dynamics trajectories of proteins in water (Jukič et al. 2020). It uses snapshots of a protein in water from a MD trajectory to identify conserved water sites and allows visualization of the identified conserved water sites on a protein. ProBiS-Dock database This is a web server and interactive web repository of small ligand–protein binding sites (http://probis-dock-database.insilab.org) for drug design of more than 1.4 million small ligand–protein binding sites in the PDB, which allows these binding sites to be ranked according to their druggability (Konc et al. 2021). A new druggability score is used to measure the suitability of a binding site for drug development. It is defined as the extent to which the binding site is currently used in drug development, as reflected by the proportion of PDB structures of this and similar binding sites bound with ligands with druglike properties. The druglike nature of a ligand is measured by the molecular complexity of the ligand, which takes into account the elemental composition and the number of rings in the compound’s structure. This helps screen out binding sites that bind to small molecules with simple structures that could bind nonspecifically to many proteins and favors binding sites that bind to molecules that are similar to most existing drugs. Another unique feature of the database is the division of ligand binding site into compound (substrate-competitive) and cofactor (cofactor-competitive), which may be particularly suitable for drug design, where typically inhibitors against a substrate or against a cofactor are developed. ProBiS-Dock algorithm ProBiS-Dock (http://insilab.org/probisdock) is a hybrid multitemplate homology algorithm for flexible docking enabled by protein binding site comparison (Konc et al. 2022). It is a small molecule docking (and inverse docking) approach based on predicted binding sites that enables flexible docking of small ligands to flexible protein binding sites. The ProBiS-Dock algorithm can be used in drug development for new drug candidate discovery, drug repositioning, and off-target effects. It complements the ProBiS-Dock Database in the sense that its input, the prepared binding sites, can be obtained from that database. The algorithm treats small molecules and proteins as fully flexible entities and allows conformational changes in both after ligand binding. A new scoring function is described that consists of a binding site-specific scoring function (ProBiS-Score) and a general statistical scoring function. This allows the scoring function to adapt to each protein binding site in the PDB. ProBiS-Dock enables rapid docking of small molecules to proteins and has been successfully validated in silico against standard benchmarks. It enables the search for new active ligands by leveraging existing knowledge in the PDB. The potential of the software for drug discovery has been confirmed in vitro by the discovery of new inhibitors of human indoleamine 2,3-dioxygenase 1, an enzyme that is an attractive target for cancer therapy (Dolšak et al. 2021). ProBiS-Fold web server This web server and database (http://probis-fold.insilab.org) enables annotation of human structures from the AlphaFold database (Varadi et al. 2022) with no corresponding structure in the PDB to discover new druggable binding sites (Konc and Janežič 2022). It contains predictions of binding sites and their corresponding ligands from the whole human structural proteome (Fig. 4). The predicted binding sites are divided into protein, peptide, nucleic acid, small molecule, further subdivided into compound (for substrate/agonist competitive ligands) and cofactor (for cofactors and cofactor-competitive ligands) binding sites, conserved water, metal ion, and glycan-binding sites according to the type of ligand they bind. In contrast to our previous approach, peptide ligand binding sites are detected separately from protein binding sites because peptides are an important new class of drugs that are distinct from proteins. For ion and water ligands, only biologically relevant metal ions and conserved water molecules are considered. Conserved water molecules are those found in more than 10 PDB structures bound to a similar motif and have a high conservation score greater than 0.6. Biologically relevant metal ions are those found in more than 10 PDB structures at the same location. A total of 149,960 binding sites were predicted for the entire human structural proteome. Importantly, binding sites were identified on protein structures for small molecules that do not have a corresponding structure in the PDB; 573 of these binding sites are highly druggable and 921 other sites are druggable as judged by our druggability score. These represent a novel pool of binding sites for previously unknown protein structures that could enter pharmaceutical pipelines. ProBiS-Fold is an extension of the ProBiS-ligands (Konc and Janežič 2014) and the ProBiS-CHARMMing web interface (Konc et al. 2015) for prediction and optimization of ligands in protein binding sites, as well on a recent addition, the GenProBiS web server (Konc et al. 2017) and in particular, it is an extension of the ProBiS-Dock Database (Konc et al. 2021) with protein models from the recently developed AlphaFold database (Jumper et al. 2021; Varadi et al. 2022) which provides open access to protein structure predictions for the human proteome and 20 other key organisms (DeepMind, Google, https://alphafold.ebi.ac.uk) is thus opening up completely new possibilities for drug research on virtually the entire human proteome as well as on proteomes of other species. The ProBiS-Fold web server enables the characterization of binding sites for novel protein targets and greatly increases the number of potential protein targets that could be used in drug discovery.Fig. 4 ProBiS-Fold web server as an extension of ProBiS-Dock Database with AlphaFold DB Using ProBiS-Fold, we can predict binding sites for protein structures predicted by AlphaFold, which may or may not already have a structure in the PDB. The predicted binding sites are classified by the type of ligand they bind, and the server also allows construction of complexes of the protein with predicted ligands. An example of the output is shown in Fig. 5, in which angiotensin-converting enzyme 2 was used as a query and for which ProBiS-Fold predicted the ligand of the spike protein of SARS-CoV-2 and a corresponding binding site.Fig. 5 ProBiS-Fold web server results page shows a 3D view of a predicted binding site (blue surface) for the SARS-CoV-2 spike protein (gray cartoon) on a human angiotensin-converting enzyme 2 (ACE2) protein model from the AlphaFold database (model confidence-colored cartoon). Binding site residues on ACE2 are CPK-colored sticks. The list of predicted ligands for this binding site is below the viewer. Links to all the different predicted binding sites (protein, compound, cofactor, glycan, metal ion, and peptide) for the ACE2 protein are on the left Conclusions We have developed ProBiS Tools for protein binding site detection and ligand prediction and characterization. The newly developed ProBiS-Fold web server is the latest addition to the suite of tools. It annotates the AlphaFold human protein structure database of more than 24,000 predicted protein structures with ligand binding sites and sites for post-translational modifications, and 3D structures of ligands that bind to these sites, using a structure-based, comparative approach, and for the first time makes it possible to examine structures in the AlphaFold Database for which there is no corresponding structure in the PDB and to predict in detail where the binding sites are located, to which ligands they bind, and whether the binding sites are suitable for drug development. It can show the reliability of the AlphaFold structure, especially at the binding sites. As a world first, the binding sites are categorized into protein, peptide, nucleic acid, small molecule (substrate and cofactor competitive), metal ion, conserved water, and glycan types, depending on which ligands they bind. All of our past, present, and future web servers and tools are freely available to academic users at http://insilab.org and at https://probis.nih.gov. Acknowledgements We gratefully acknowledge the support of NVIDIA Corporation with the donation of Tesla GPUs. Author contribution J.K. and D.J. conceptualized the project and performed all aspects of the research process and wrote the entire manuscript. Both authors read and approved the final version of the manuscript. Funding This work was supported by the Slovenian Research Agency project grants N1-0142, L7-8269, N1-0209, J1-1715 and J1-9186 and a bilateral grant BI-JP/16–18-002. Data availability Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study. Code availability Not applicable. Declarations Ethics approval Not applicable. Consent to participate Not applicable. Consent for publication Not applicable. Conflict of interest The authors declare no competing interests. This review article is in honor of Prof. Haruki Nakamura’s 70th birthday. Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Abrusán G Marsh JA Ligand binding site structure influences the evolution of protein complex function and topology Cell Rep 2018 22 3265 3276 10.1016/j.celrep.2018.02.085 29562182 Bancroft AJ Levy CW Jowitt TA The major secreted protein of the whipworm parasite tethers to matrix and inhibits interleukin-13 function Nat Commun 2019 10 2344 10.1038/s41467-019-09996-z 31138806 Berman H Henrick K Nakamura H Announcing the worldwide Protein Data Bank Nat Struct Mol Biol 2003 10 980 980 10.1038/nsb1203-980 Brooks BR Brooks CL Mackerell AD CHARMM: The biomolecular simulation program J Comput Chem 2009 30 1545 1614 10.1002/jcc.21287 19444816 Chartier M Najmanovich R Detection of binding site molecular interaction field similarities J Chem Inf Model 2015 55 1600 1615 10.1021/acs.jcim.5b00333 26158641 Chen YC Wright JD Lim C DR_bind: a web server for predicting DNA-binding residues from the protein structure based on electrostatics, evolution and geometry Nucleic Acids Res 2012 40 W249 W256 10.1093/nar/gks481 22661576 Cukuroglu E Gursoy A Keskin O HotRegion: a database of predicted hot spot clusters Nucleic Acids Res 2012 40 D829 D833 10.1093/nar/gkr929 22080558 Depolli M Konc J Rozman K Exact parallel maximum clique algorithm for general and protein graphs J Chem Inf Model 2013 53 2217 2228 10.1021/ci4002525 23965016 Dolšak A Bratkovič T Mlinarič L Novel selective IDO1 inhibitors with isoxazolo[5,4-d]pyrimidin-4(5H)-one scaffold Pharmaceuticals 2021 14 265 10.3390/ph14030265 33804161 Ehrt C Brinkjost T Koch O Impact of binding site comparisons on medicinal chemistry and rational molecular design J Med Chem 2016 59 4121 4151 10.1021/acs.jmedchem.6b00078 27046190 Ehrt C Brinkjost T Koch O A benchmark driven guide to binding site comparison: an exhaustive evaluation using tailor-made data sets (ProSPECCTs) PLoS Comput Biol 2018 14 e1006483 10.1371/journal.pcbi.1006483 30408032 Fu X Zhang G Liu R Mechanistic study of human glucose transport mediated by GLUT1 J Chem Inf Model 2016 56 517 526 10.1021/acs.jcim.5b00597 26821218 Glaser F Pupko T Paz I ConSurf: identification of functional regions in proteins by surface-mapping of phylogenetic information Bioinformatics 2003 19 163 164 10.1093/bioinformatics/19.1.163 12499312 Jakubec D Skoda P Krivak R PrankWeb 3: accelerated ligand-binding site predictions for experimental and modelled protein structures Nucleic Acids Res 2022 50 W593 W597 10.1093/nar/gkac389 Jones S Thornton JM Principles of protein-protein interactions Proc Natl Acad Sci USA 1996 93 13 20 10.1073/pnas.93.1.13 8552589 Jukic M Konc J Gobec S Janezic D Identification of conserved water sites in protein structures for drug design J Chem Inf Model 2017 57 3094 3103 10.1021/acs.jcim.7b00443 29155577 Jukič M Konc J Janežič D Bren U ProBiS H2O MD approach for identification of conserved water sites in protein structures for drug design ACS Med Chem Lett 2020 11 877 882 10.1021/acsmedchemlett.9b00651 32435399 Jumper J Evans R Pritzel A Highly accurate protein structure prediction with AlphaFold Nature 2021 596 583 589 10.1038/s41586-021-03819-2 34265844 Kinjo AR Bekker G-J Suzuki H Protein Data Bank Japan (PDBj): updated user interfaces, resource description framework, analysis tools for large structures Nucleic Acids Res 2017 45 D282 D288 10.1093/nar/gkw962 27789697 Kinoshita K Nakamura H Identification of the ligand binding sites on the molecular surface of proteins Protein Sci 2005 14 711 718 10.1110/ps.041080105 15689509 Kinoshita K Furui J Nakamura H Identification of protein functions from a molecular surface database, eF-site J Struct Func Genom 2002 2 9 22 10.1023/A:1011318527094 Konc J Janezic D An improved branch and bound algorithm for the maximum clique problem MATCH Commun Math Comput Chem 2007 58 569 590 Konc J Janežič D Protein−protein binding-sites prediction by protein surface structure conservation J Chem Inf Model 2007 47 940 944 10.1021/ci6005257 17388583 Konc J Janežič D ProBiS algorithm for detection of structurally similar protein binding sites by local structural alignment Bioinformatics 2010 26 1160 1168 10.1093/bioinformatics/btq100 20305268 Konc J Janežič D ProBiS-ligands: a web server for prediction of ligands by examination of protein binding sites Nucleic Acids Res 2014 42 W215 W220 10.1093/nar/gku460 24861616 Konc J Janežič D ProBiS-Fold approach for annotation of human structures from the AlphaFold Database with no corresponding structure in the PDB to discover new druggable binding sites J Chem Inf Model 2022 10.1021/acs.jcim.2c00947 Konc J Hodošček M Ogrizek M Structure-based function prediction of uncharacterized protein using binding sites comparison PLOS Comput Biol 2013 9 e1003341 10.1371/journal.pcbi.1003341 24244144 Konc J Miller BT Štular T ProBiS-CHARMMing: web interface for prediction and optimization of ligands in protein binding sites J Chem Inf Model 2015 55 2308 2314 10.1021/acs.jcim.5b00534 26509288 Konc J Skrlj B Erzen N GenProBiS: web server for mapping of sequence variants to protein binding sites Nucleic Acids Res 2017 45 W253 W259 10.1093/nar/gkx420 28498966 Konc J Lešnik S Škrlj B Janežič D ProBiS-Dock Database: a web server and interactive web repository of small ligand–protein binding sites for drug design J Chem Inf Model 2021 61 4097 4107 10.1021/acs.jcim.1c00454 34319727 Konc J Lešnik S Škrlj B ProBiS-Dock: a hybrid multitemplate homology flexible docking algorithm enabled by protein binding site comparison J Chem Inf Model 2022 62 1573 1584 10.1021/acs.jcim.1c01176 35289616 Lešnik S Hodošček M Podobnik B Konc J Loop grafting between similar local environments for fc-silent antibodies J Chem Inf Model 2020 60 5475 5486 10.1021/acs.jcim.9b01198 32379970 Martinez-Mayorga K Madariaga-Mazon A Medina-Franco JL Maggiora G The impact of chemoinformatics on drug discovery in the pharmaceutical industry Expert Opin Drug Discov 2020 15 293 306 10.1080/17460441.2020.1696307 31965870 Miller BT Singh RP Klauda JB CHARMMing: a new, flexible web portal for CHARMM J Chem Inf Model 2008 48 1920 1929 10.1021/ci800133b 18698840 Phizicky EM Fields S Protein-protein interactions: methods for detection and analysis Microbiol Rev 1995 59 94 123 10.1128/mr.59.1.94-123.1995 7708014 Ramatenki V Dumpati R Vadija R Identification of new lead molecules against UBE2NL enzyme for cancer therapy Appl Biochem Biotechnol 2017 182 1497 1517 10.1007/s12010-017-2414-7 28185054 Reba K Guid M Rozman K Exact maximum clique algorithm for different graph types using machine learning Mathematics 2022 10 97 10.3390/math10010097 Ren J Xie L Li WW Bourne PE SMAP-WS: a parallel web service for structural proteome-wide ligand-binding site comparison Nucleic Acids Res 2010 38 W441 W444 10.1093/nar/gkq400 20484373 Salentin S Schreiber S Haupt VJ PLIP: fully automated protein–ligand interaction profiler Nucleic Acids Res 2015 43 W443 W447 10.1093/nar/gkv315 25873628 Schmitt S Kuhn D Klebe G A new method to detect related function among proteins independent of sequence and fold homology J Mol Biol 2002 323 387 406 10.1016/S0022-2836(02)00811-2 12381328 Štular T Lešnik S Rožman K Discovery of Mycobacterium tuberculosis InhA inhibitors by binding sites comparison and ligands prediction J Med Chem 2016 59 11069 11078 10.1021/acs.jmedchem.6b01277 27936766 Vankayala SL Kearns FL Baker BJ Elucidating a chemical defense mechanism of Antarctic sponges: a computational study J Mol Graph Model 2017 71 104 115 10.1016/j.jmgm.2016.11.004 27894019 Varadi M Anyango S Deshpande M AlphaFold Protein Structure Database: massively expanding the structural coverage of protein-sequence space with high-accuracy models Nucleic Acids Res 2022 50 D439 D444 10.1093/nar/gkab1061 34791371 Weiner PK Langridge R Blaney JM Electrostatic potential molecular surfaces Proc Natl Acad Sci USA 1982 79 3754 3758 10.1073/pnas.79.12.3754 6285364	0	PMC9734416	NO-CC CODE	2022-12-14 23:28:28	no	Biophys Rev. 2022 Dec 9;:1-9	utf-8	Biophys Rev	2,022	10.1007/s12551-022-01028-3	oa_other
==== Front Public Transp Public Transport 1866-749X 1613-7159 Springer Berlin Heidelberg Berlin/Heidelberg 309 10.1007/s12469-022-00309-0 Original Research A supervised machine learning model for imputing missing boarding stops in smart card data Shalit Nadav [email protected] 1 http://orcid.org/0000-0002-6075-2568 Fire Michael [email protected] 1 http://orcid.org/0000-0002-4169-7129 Ben-Elia Eran [email protected] 2 1 grid.7489.2 0000 0004 1937 0511 Data4Good Lab, Department of Software and Information Systems Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel 2 grid.7489.2 0000 0004 1937 0511 GAMESLab, Department of Geography and Environmental Development, Ben-Gurion University of the Negev, Beer-Sheva, Israel 7 12 2022 133 4 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. Public transport has become an essential part of urban existence with increased population densities and environmental awareness. Large quantities of data are currently generated, allowing for more robust methods to understand travel behavior by harvesting smart card usage. However, public transport datasets suffer from data integrity problems; boarding stop information may be missing due to imperfect acquirement processes or inadequate reporting. This study introduces a supervised machine learning method to impute missing boarding stops based on ordinal classification using GTFS timetable, smart card, and geospatial datasets. A new metric, Pareto Accuracy, is suggested to evaluate algorithms where classes have an ordinal nature. The results are based on a case study in the city of Beer Sheva, Israel, consisting of one month of smart card data. We show that our proposed method is robust to irregular travelers and significantly outperforms well-known imputation methods without the need to mine any additional datasets. The data validation from another Israeli city using transfer learning shows the presented model is general and context-free. The implications for transportation planning and travel behavior research are further discussed. Keywords Machine learning Smart card Boarding stop imputation Public transport Missing data Pareto accuracy http://dx.doi.org/10.13039/100007225 Ministry of Science and Technology The Ministry of Science and Technology of the People’s Republic of China3-15741 Ben-Elia Eran ==== Body pmcIntroduction Public transport (PT) is an integral part of everyday life in many cities. The gradual shift of the global population over the past century to urban areas is markedly increasing people’s dependence on PT for their daily mobility needs (Petrović et al. 2016). PT is a complex system that is based on physical elements of stops, vehicles, routes, and other temporal and spatial elements (Ceder 2016). The PT system consists of regularly scheduled vehicle trips open to all paying passengers, with the capacity to carry multiple passengers whose trips may have different origins, destinations, and purposes (Walker 2012). PT is ideal when passengers regard its service as punctual and regular (Walker 2012). With the growth in the number of cars on urban roads, PT improvements have become an essential part of traffic congestion mitigation strategies and are vital in promoting sustainable transportation (Al Mamun and Lownes 2011). Although understanding the patterns of PT use is crucial to its planning, this task remains a significant challenge in practice and research.Fig. 1 Predicting missing boarding stops algorithm overview Numerous studies in recent years have examined the behavior of PT travelers (Li et al. 2018) in efforts to address this challenge. Habitual travel behavior is of great interest to transportation planners, and its analysis can help improve demand predictions and justify necessary upgrades to PT supply (Briand et al. 2017). This analysis can also contribute to improvements in PT (service/planning/upgrades) with respect to the management of COVID transmission, in terms of providing better information on crowded areas, such as bus stops, which is critically important to the global issue of public health. To this end, transportation planners typically use travel behavior surveys (Stopher and Greaves 2007). While these surveys statistically reflect travel behavior correctly, they are also expensive, time-consuming, and often unable to generate sufficient amounts of data relative to the size of the population, and would need significant changes in scope to cover recent COVID concerns. Conversely, data harvested from smart cards can generate millions of records compared to a typical sample ranging from 2,500 to 10,000 households using surveys (Maeda et al. 2019). Smart cards, also known as automatic fare collection (AFC), provide an efficient and cost-saving alternative to the manual fare collection method (Jang 2010; Chen and Fan 2018). In addition to fulfilling fare collection needs, as a bi-product, smart card transactions also generate geocoded timestamps that record every passenger’s boardings, line transfers, and sometimes alightings for a wide range of PT vehicles (bus, tram, train, or metro) (Pelletier et al. 2011; Faroqi et al. 2018). These records are generated for almost the entire passenger population (Pelletier et al. 2011; Faroqi et al. 2018). Such information is a treasure trove for travel behavior analyses, especially for extracting passengers’ spatio-temporal travel patterns (e.g., Origin–Destination matrices or path choice (Wang et al. 2011)). Nevertheless, common statistical inference methods applied in surveys are of little practical use for understanding the travel patterns of an entire population. Therefore, different methods are required. Kandt and Batty (2021) proclaimed a new area of urban research defined by advances in big data analytics, with smoother decision making and a deeper understanding of urban systems. A massive increase in the volumes, velocities and varieties of big data have also been paralleled by recent developments in the data science field. New data mining tools and robust cloud computing capabilities (Li et al. 2015, 2018) create new opportunities to analyze travel behavior patterns at the individual level, over extended periods, and in large urban areas (Ma et al. 2017). The availability of big data has a vast potential to improve the quality of transportation planning and research, and by applying big data analytics and data mining methods, this task has become much more feasible (Ma et al. 2013). However, similar to the case in other domains, the veracity of such datasets remains questionable (Ben-Elia et al. 2018). Smart card datasets, in particular, may suffer from integrity problems, such as incorrect or missing values, e.g., when operators only record partial data. For example, in Yan et al. (2019), boarding stop information was completely missing, and only time stamps remained intact in the dataset. A common solution for such problems is to replace the missing or erroneous data by utilizing alternative publicly accessible data. One possible solution is to use official PT timetables to impute the missing data in missing boarding stop information. One popular source for such data comes from the General Transit Feed Specification (GTFS), first created in 2006 by Google (Google, 2016), defined as a standard file format for storing PT schedules and associated geographic information (Ma et al. 2012). GTFS contains the complete schedules and routes of every PT line planned for each day of the month in tabular formats together with corresponding geographic shapefiles and is widely used in over 750 urban regions across the world (Hadas 2013; Antrim et al. 2013). Nonetheless, PT running times and arrival times at stops are never perfectly aligned with their official timetables, where PT is not always punctual, even in developed countries. For example, Cats and Loutos (2016) found that only 10% of all arrivals were within an interval of 15 s. This issue becomes more acute, especially when PT vehicles–mainly buses–also share the same road space with private and commercial vehicles (i.e., mixed traffic). While this issue is less severe in major urban areas in developed countries where rail-based and PT bus preemption infrastructure is widespread and right-of-way strongly enforced, this is not the reality everywhere. For example, in Israel (an official OECD member), buses accounted nationally for 85% of PT trips in 2019, with more than 2M passengers served daily. The country suffers from a shortage of adequate PT infrastructure (namely, too few priority lanes—14 m per capita, compared to 300 m in the EU), thus resulting in poor PT service punctuality (Ceder 2004). As shown later, this fact makes schedule-based imputation a poor substitute for boarding stop prediction. A second solution is to discard such data by simply removing missing records or those that do not align with a prescribed hypothesis (Tao et al. 2014). Nonetheless, discarding data can be regarded as a reasonable solution only when that share of the missing data is small. However, when the missing portion is substantial, the whole dataset could be compromised and discarded. This scenario can render certain urban areas effectively blind vis-a-vis smart card data. A third option is to complement the missing data by combining different datasets. In this respect, either automatic vehicle location (or AVL), which uses installed GPS transponders to locate PT vehicles and estimate real-time arrival times at designated stops; or automatic passenger counters (or APC), which use infrared or laser technologies to estimate boarding and alighting passenger numbers, have been used in combination with smart card data (Shalaby and Farhan 2004; Mazloumi et al. 2010; Khiari et al. 2016). Yet, such data is neither always available (Chen and Fan 2018; Yan et al. 2019), nor efficient, as considerably more errors may well be introduced in the process (Luo et al. 2018). These two facts likely reduce their suitability for data imputation. Moreover, even when such data sources exist, matching between them is somewhat challenging. For example, Lou et al. (2018) had no vehicle trip identification (ID), making it impossible to match with AVL records. A further difficulty is that missing data can vary by city or between different operators (Laña et al. 2018). In some cities, data integrity is regarded as very strong, and consequently, boarding and alighting imputation tasks are very good (Munizaga et al. 2014). In contrast, in other cities where data sources are lacking, data integrity can also be flawed. The lack of common standards and methods for data handling and processing across the PT modes and sectors has been identified as a main problem hindering efficient utilization of smart cards and other PT-related data. Such a data interoperability requires developing a standardized application approach that will allow data mining tools and models to be tested and implemented as asserted by Covic and Voß (2019). In this respect heuristic methods, such as ML, can be regarded as a viable solution to perform data imputation tasks (Yan et al. 2019). To this end, our aim is a general and context-free boarding stop imputation method. Specifically, we address use cases where data quality is considered too insufficient to impute by cross-inference and without the need to harvest any other data than what is necessary. While still providing valuable insights for transportation planners, we consider this of particular relevance for developing countries where the traveler population is mostly PT-dependent. We established a general boarding stop imputation method to improve the quality and integrity of PT datasets by predicting missing or corrupted travelers’ records in smart card data. Namely, to the best of our knowledge, we developed the first machine learning (ML) algorithm for predicting passengers’ boarding stops (see Fig. 1). Our algorithm is based on features extracted by harvesting three big data sources, the planned GTFS schedule data, smart card (AFC) data, and geospatial (GIS) data. We applied a machine learning model to these features to predict boarding stops based on the notion of embedding (see Sect. 3). To train and evaluate our algorithm’s performance, we utilized a real-world smart card dataset from the city of Beer Sheva in Israel that consists of over a million trips taken by more than 85,000 passengers. Since the boarding stops are embedded, they also become ordered, and therefore, the problem we addressed is ordinal classification. Accordingly, we also propose a new method of evaluation that shows the percentage in each error dimension that we define as Pareto Accuracy, which is more interpretable and allows for better comparison between imputation models. We show that our model performed significantly better than a naïve prediction model based on harvesting GTFS data alone (aka schedule-based) and other imputation methods. In this study, we succeeded to generate a model which is both wholly generic and has considerably higher accuracy and recall values than other tested imputation methods (see Sect. 4). Additionally, we demonstrated that we obtain similar prediction results in an entirely different city using our method. Moreover, we show how other imputation methods are not always applicable, while our methodology can be applied with a broader scope. Our study’s overall focus is to improve the integrity of public transport data. Specifically, our study provides the following two main contributions: We present a novel prediction model for imputing missing boarding stops using supervised learning. Moreover, our proposed model is generic and transferable, i.e., it can be trained on one city’s data and then impute missing data in another municipality. We propose a new metric—Pareto Accuracy—for evaluating public transport metrics that are more interpretable and allow broader comparisons between imputation models. The rest of the paper is organized as follows: In Sect. 2, we review related work on smart card usage, missing data imputation, and ML applications in transportation research. Section 3 describes the use case, experimental framework, and methods used to develop the ML model and the extraction of its features. In Sect. 4, we present the results of the ML model and compare its performance to other known solutions. In Sect. 5, we discuss the implications of the findings and the study’s limitations and present our conclusions and future research directions. Related work We provide an overview of relevant studies by first presenting smart card research in general, followed by studies that have utilized smart card data with machine learning to perform predictive analytics. We then give an overview of the field of missing data imputation. Lastly, we present studies in the field of ordinal classification. Smart card analytics The smart card system was introduced as a smart and efficient AFC system in the early 2000s (Chien et al. 2002) and has since become an increasingly popular payment method (Bagchi and White 2005; Trépanier et al. 2007). In particular, smart cards have also become an increasingly popular source of big data for research and policy making (Agard et al. 2006; Jang 2010). For example, smart card data is used for exploring travel behavior, determining travel patterns, measuring the performance of PT services, locating critical transfer points, and analyzing crowdedness effects on route choice (Bryan and Blythe 2007; Jang 2010; Alguero 2013; Zhao et al. 2017; Li et al. 2018; Yap et al. 2020). Recently, smart card datasets were used to study travel behavior changes to travel behavior as a result of the COVID-19 pandemic (Almlöf et al. 2020; Orro et al. 2020; Zhang et al. 2021). Comprehensive literature reviews of smart card usage were provided by Pelletier et al. (2011), Schmöcker et al. (2017), and Faroqi et al. (2018). Initially, smart card research applied rather classic statistical methods and descriptive analytics. Devillaine et al. (2012) inferred the location, time, duration, and designation of PT users’ activities using rules derived from smart card data and work and study schedules. The main research challenge evident in the literature was to estimate origin–destination (OD) matrices which describe the spatial distribution of travel demand between locations during different periods of the day (Chu and Chapleau 2008; Wang et al. 2011; Munizaga and Palma 2012; Gordon et al. 2013). OD matrices are also crucial inputs to perform the three stages in PT network design, namely: route design, frequency (headway) setting, and timetabling (Guihaire and Hao 2008). Before the advent of smart cards, these matrices were only derived and validated based on some representative sample of travelers (Chen et al. 2016). However, as noted, surveys often lack sufficient spatial and temporal coverage. Various studies have demonstrated the advances in OD estimation with smart card data (Chu and Chapleau 2008; Wang et al. 2011; Munizaga and Palma 2012; Gordon et al. 2013). Nevertheless, with the introduction of smart cards, new problems in OD estimation appeared. Namely, many PT agencies adopted a TAP (Transit Access Protocol) IN system where only boarding stop information is recorded. In contrast, the availability of alighting stop information “TAP IN+TAP OUT” systems allows for the OD matrix to be derived using more straightforward approaches. Alighting stop information is necessary for many tasks such as route loading profiles, market research, and improvements in service planning. However, under TAP IN, the destination must be somehow predicted (Trépanier et al. 2007; Faroqi et al. 2018). In addition, combining smart card data with a smaller scale travel behavior survey for validation purposes is a useful approach to better understand passengers’ daily travel patterns (Wang et al. 2011). Nonetheless, OD analyses inherently assume that PT passengers travel routinely back and forth from/to the same locations. Recent findings suggest this assumption does not necessarily hold, and some share of PT passengers are quite flexible (Huang et al. 2018) or use PT infrequently (Benenson et al. 2019). Therefore, a simple OD estimation will possibly result in PT planning that is mismatched with actual demand patterns. Traditional analysis methods do not take advantage of the full potential of the added value of big data. At the same time, rapid growth in power and cost reduction in computational technologies provide new opportunities, both in terms of the availability of the massive amount of data collected and the development of more novel algorithms (Welch and Widita 2019). Agard et al. (2006) obtained travel behavior indicators that identify daily travel patterns and clustering of major user groups. Kieu et al. (2015) applied a density-based spatial clustering application with noise (DBSCAN) algorithm to cluster passengers and identify classes of passengers for strategic planning improvements. Ma et al. (2013) used smart card data to cluster the travel patterns of PT riders to characterize commuter profiles. In this respect, the literature shows a shift toward harvesting the prognostic nature of ML to yield better predictive analytics highlighting the growing emphasis on using smart card data for analytical purposes. This shift underscores the change from the more straightforward analyses conducted in the past to the more comprehensive analysis done today. Hagenauer and Helbich (2017) compared several ML classifiers and showed both their predictive power and ability to uncover travelers’ mode choices via feature importance analysis. For example, they showed that the trip distance was the most important predicting factor, while the temperature was only a key feature for predicting bicycle use. In 2018, Palacio (2018) showed that ML predictions are much more accurate than traditional linear models that were sub-optimal both in terms of R-square and MSE. In the following year, Traut and Steinfeld (2019) combined smart card data with crime records to assist agencies in identifying insecure and dangerous PT stops. Chen et al. (2016), who inferred mode and route choices, stress the need for cross-disciplinary collaborations between data scientists and transportation planners to exploit the information withheld in the data. Further evidence of the prominence of big data analytics in PT research can be found in several review papers such as Fonzone et al. (2016), Namiot and Sneps-Sneppe (2017), Anda et al. (2017), Li et al. (2018), and Milne and Watling (2019). Deep learning algorithms have also been utilized to address PT issues using smart card data. Deep learning is a sub-field of ML that automatically creates feature engineering, and its methods are state-of-the-art in many domains. Examples of such implementations include forecasting passenger destinations (Toqué et al. 2016; Jung and Sohn 2017), predicting multimodal passenger flows (Toqué et al. 2017), improving passenger segmentation (Dacheng et al. 2018), inference of passenger employment status (Zhang and Cheng 2020), and using standard deep network and long- and short-term memory networks, inference of demographics using convolutional neural networks (Zhang et al. 2019). Missing data imputation Incomplete data is a universal problem, and the application of different imputation methods will often yield different results. Therefore, to preserve reproducibility, they must be adequately addressed (Saunders et al. 2006). This problem is, notably, relevant for transportation planning, e.g., in the case of road traffic analysis (Qu et al. 2009). Incomplete data is a well-known problem in the data mining literature where a significant amount of data can be missing or incorrect. Lakshminarayan et al. (1996) elucidated both the severity of this issue as well as recommended applying ML techniques toward its solution rather than classical statistical methods. Batista and Monard (2003) assert that missing data imputation must be carefully handled to prevent bias from being introduced. Moreover, they show that the most common methods, such as mean or mode imputation, are not always optimal. One example we found in the PT literature is from Kusakabe and Asakura (2014). They used a Naïve Bayesian model for data imputation and analysis of PT to understand continuous long-term changes in trip attributes. They showed both the power of smart card data and the usefulness of missing data imputation in this field. Their method of imputation, however, is not reported in sufficient detail to be understood or replicated. Several techniques to optimize missing data imputation showed the importance attributed to this area of research (Bertsimas et al. 2017). Moreover, even state-of-the-art deep learning methods have been applied to this problem (Garg et al. 2018; Costa et al. 2018; Camino et al. 2019). These implementations were performed on a variety of datasets and problems, such as classification of continuous attributes (breast cancer and default credit card classification); images (Camino et al. 2019; Garg et al. 2018); and regression (Camino et al. 2019). Insofar as this field of study has not been operationalized for PT data, further examination is warranted, particularly when considering the issue of completing missing data to provide better information on crowded PT areas, as it pertains to the spread of COVID. In many imputation tasks, including PT, ML methods outperform standard methods significantly when the missing portion increases (Saunders et al. 2006; Laña et al. 2018; Echaniz et al. 2020; Yan et al. 2019). Additionally, standard imputation methods are too sensitive to the ratio of missing data and infrequent or ‘irregular’ users of the PT network (Van Lint et al. 2005). Conversely, ML-based imputation showed stable results regardless of the missing ratio (Laña et al. 2018). As noted previously, one solution is to impute the missing boarding stops using complementary datasets such as AVL or APC. However, AVL data are not always available (Chen and Fan 2018), whereas combining several datasets (i.e., AVL, AFC, APC, GTFS, etc.) can introduce more errors, and make it much harder to match them perfectly (Luo et al. 2018). Ordinal classification Classification is a form of supervised ML that aims to generalize a hypothesis from a given set of records. It learns to create h(xi)→yi where y has a finite number of classes (Kotsiantis et al. 2007). The basic metrics for classification are sensitivity, specificity, and accuracy (Jiao and Du 2016). Accuracy is the percentage of observations classified correctly, specificity is the percentage of true negatives classified correctly, and sensitivity is the percentage of true positives classified correctly. A classification task becomes ordered when the classes have some inherent order between them. There are a variety of metrics to evaluate supervised learning algorithms (Liu et al. 2014). Each metric has its advantages and limitations. This study introduces the Pareto Accuracy (see Sect. 3.3), suitable for assessing the constructed classifiers’ performances in imputing missing boarding stops based on ordinal classification. Ordinal classification is a form of multi-class classification where the classes exhibit some natural ordering (such as cold, warm, and hot), but not necessarily numerical traits for each class. Rather than being chosen based on the traditional metrics discussed above, a classifier may be chosen based on the severity of its errors (Gaudette and Japkowicz 2009). Additionally, classic modeling techniques will sometimes perform suboptimally since ML models assume there is no order between classes. In such tasks, e.g., the well-known Boston housing and breast cancer datasets, different models that take advantage of ordinal information are preferred (Frank and Hall 2001). In this case, additional metrics are proposed to calculate such tasks differently, such as regression metrics like Mean Absolute Error (MAE) and the Mean Square Error (MSE) and even their own metric, the Ordinal Classification Index (Cardoso and Sousa 2011). Notwithstanding, as noted below, these approaches neither fit our data nor our needs. Therefore, we developed a different and novel performance metric (see Sect. 3.3). Methods The main goal of our study is to use ML algorithms to improve the integrity of PT data. Specifically, we develop a supervised learning-based model to impute missing boarding stops in any given smart card dataset. Moreover, our goal is to construct a generic model that will be fully transferable to other datasets to impute missing data in different contexts without further adjustments. To maintain these generic objectives, we had to contend with two significant challenges: First, we could only incorporate generic properties in our model. For instance, our model cannot include the actual line number of a bus route specific to a particular city. Moreover, since supervised ML algorithms can only predict classes they were initially trained upon, classification classes must remain the same across datasets, e.g., bus stop #14 in a specific city is an irrelevant feature for other cities. Therefore, once more, a different numerical representation is applied by embedding (see Sect. 3.2). Second, we develop a genuinely generic model that can also be applied to other geographical contexts in which it was not initially trained. The model must also undergo a process of transfer learning (Torrey and Shavlik 2010) that entails the transfer of relevant knowledge by fine-tuning a model on a “novel” dataset. In our case, our model underwent the process of transfer learning using a dataset on which it had not been trained before.Fig. 2 Modeling methodology overview The missing boarding stop values were imputed using the following methodology (see Figs. 1 and 2): First, we preprocessed and cleaned the smart card dataset that we utilized in this study (see Sect. 3.1). Next, we extracted various features from two other datasets: (a) the GTFS timetable data; and (b) open municipal geospatial data. In addition, we converted boarding stops from their original identifiers to embedded numerical representations based on GTFS data (see Sect. 3.2). Afterward, we applied ML algorithms to estimate a model that can predict the missing boarding stops. We used SHAP (SHapley Additive exPlanations) values for determining feature importance (Lundberg and Lee 2017),1 i.e., which features make the most substantial contribution to the predictive power of the model (see Fig. 8). We also evaluated the performance of our model using a novel performance metric called Pareto Accuracy. Then, based also on common metrics, we evaluated our model relative to a schedule-based model estimated only on GTFS timetable data. Finally, we compared our model to several other comparative models (e.g., passenger history, temporal proximity, or semi-random guessing) that were previously used in the literature. Below, we describe each step of our approach in more detail. Datasets and data preprocessing As noted above, we used three datasets: The Smart card dataset—“Rav Kav” is the Israeli AFC system applying the TAP protocol, allowing PT passengers to pay for their trip using their smartcards anywhere in the country. Rav-Kav operates a nationwide TAP IN for buses and rail that codes information on unique passenger identifiers, traveler types (such as student or senior travelers), boarding stops, boarding timestamps, fares, discount attributes, and unique trip identifiers of the line at that time. For rail trips only, TAP OUT also records alighting stops and times. During the period 2018/9, circa 2M boardings were recorded per day in the entire country. GTFS—a GTFS feed, as described above, consists of rail/bus schedules and timetables, stops, and routes of every PT trip planned for every day of the month. In Israel since 2012, the GTFS feed has been published daily online by the Ministry of Transport, providing schedules of 36 bus and rail operators, encompassing 7,800 route-direction-pattern alternatives served by 28,000 bus and rail stations. The GTFS feed aligns with the smart card dataset as described below. This study utilized the GTFS dataset to enrich the feature space and convert boarding stop records into an embedded numerical value. Geospatial information—we derived a variety of geospatial attributes from municipal GIS databases. To obtain a dataset suitable for constructing the prediction model, we were required to remove any record that lacked a boarding stop or a trip ID (a unique identifier of a trip provided by a specific and unique PT operator) from the smart card dataset. Next, we joined the smart card dataset with the GTFS dataset by matching the trip ID attributes. Lastly, we joined the geospatial dataset with the smart card dataset using the GTFS dataset, which contains all the geographic coordinates of each PT route. Feature extraction and machine learning model construction ML performance is highly correlated to the quality of the feature space, and therefore, including more features results in better model performance (Gudivada et al. 2017). While the smart card data contains the PT line and boarding time of each passenger, it lacked several essential data, such as the duration that had elapsed since that line left the origin depot, the time remained until arrival to the final destination, the total number of stops, and other relevant trip attributes. Moreover, the smart card data is missing physical geospatial characteristics, such as the number of traffic lights on the PT route that more likely increases traffic congestion and consequent delays, and could well strengthen model performance. Overall, three features were extracted using the smart card dataset, five features using the GTFS dataset, three features using the geospatial dataset and four from combined GTFS and smart card datasets. From the 41 features we initially tested in total, we selected 15 to include in our model based on stepwise selection and feature importance analysis (see Table 1) by exploiting the SHAP values Lundberg and Lee (2017).Table 1 Extracted features Dataset Feature Explanation Municipal geospatial records Addresses_average The number of addresses listed on the route Street_light_average The number of streetlights on the route Traffic_Lights_average The number of traffic lights on the route GTFS Number_of_points The number of points in a shapefile in GTFS per route Average_distance_per_stop The total length of the route divided by the number of points Average_time_per_stop The total expected travel time of the route divided by the number of points Average_points_to_stops The number of points in a shapefile in GTFS per route divided by the number of points Time_diff_of_trip Total travel time GTFS and smart card Time_from_boarding_to_last_stop Time from boarding time to expected last stop of the route Time_from_departure_to_boarding Time from route departure time to boarding time Predicted_sequence GTFS prediction sequence of the most likely stop Hourly_expected_lateness The average lateness per hour (based on training data) Smart card Boardingtime_Seconds_from_midnight Timestamp of boarding to a numerical value in seconds from midnight Boardingtime_weekday The day of the week in which the boarding occurred Is_weekend Is it a weekend? To construct the prediction model, we used the GTFS dataset to create a schedule-based prediction. This naive prediction reflects the transit vehicle’s position along a line according to the GTFS schedule. Namely, let Si be the sequence number of the boarding stop based on the GTFS schedule and let Ai be the actual boarding stop sequence number. Then, we define Di as Di=Ai-Si. Our prediction model goal was to predict Di by utilizing the variety of features presented in the previous section. For instance, consider a passenger who boarded a line at the third stop, i.e., Ai=3, but the transit vehicle was scheduled to arrive at the second stop at the designated time. The schedule-based prediction would be 2, i.e., Si=2, the stop where it was supposed to be at that time. Then, the difference is Di=Ai-Si=3-2=1, and this is the class the algorithm will predict.Fig. 3 Evaluation process overview Subsequently, we performed the following steps to construct the prediction model: First, we selected several well-known classification algorithms. Namely, we used Random Forest (Singh et al. 2016), Logistic Regression (Singh et al. 2016), and XGBoost (Chen and Guestrin 2016). Second, we split our dataset into training and testing datasets (see Fig. 3). Due to the temporal nature of the data, we used a logical splitting, the training dataset consisted of the first three weeks of data (75%), and a testing dataset consisted of the last week of data (25%).2 Figure 4 shows the distributions of the embedded boarding stops by the computed difference between the actual and schedule-based sequences (Di) for the training and testing subsets. No apparent differences between the two distributions are evident. Third, for both the training and testing datasets, we extracted all the 15 features mentioned above. Fourth, we constructed the prediction models using each one of the selected algorithms. Lastly, we compared the generated models and selected the one with the best performance based on the Pareto Accuracy metric (see Sect. 3.3).Fig. 4 Train and test boarding stop histogram Model evaluation We evaluated each model and compared it to the schedule-based method on the test dataset using common metrics: accuracy, recall, precision, F1 (see Appendix A for definitions), and the new metric we developed: Pareto Accuracy. We used the following variables for our novel Pareto Accuracy metric: Let pi be the predicted sequence of stopi, ai be the actual sequence, and di be the absolute difference between them. Let l be the limit of acceptable difference for imputation, i.e., if an error of one stop is tolerated, such as for neighborhood segmentation, then l=0. Let Xi be an indicator defined as:Xi=1ifdi<=l0otherwise. We define Pareto Accuracy as follows:PAl=∑i=1nXin. The PA metric is a generalization of the accuracy metric. Namely, PA0 is the well-known accuracy metric. Unlike other ordinal classification methods, the primary advantage of using the PA metric is to evaluate the accurate dimension of error while being extremely robust to outliers (by setting parameter l). Moreover, this metric is highly informative since its outcome value can be interpreted easily; for example, 0.6 means that 60% of the predictions had at most l difference from true labels. For example, let us consider a set of eight observations of embedded boarding stops {-2,0,3,20,-3,4,3,2}, where each observation is a simulated boarding by a passenger where each number (Di) in the set represents the difference between expected (Si) and actual boarding stops (Ai). With a value of 20, the fourth observation is an outlier, which might occur due to some fault in the decoder device of the public transport operator. We do not want to predict it, as it is naturally unpredictable. We seek a metric that will be both resilient to outliers, as they are unpredictable, and still account for the true dimension of the errors (see Sect. 2.3). Let us compare two classifiers, A and B. Classifier A predicted the following boarding stops {-2,0,4,3,-2,3,2,2}, while Classifier B predicted {3,0,3,7,1,1,3,2}. Classifier A is a more useful classifier since, in general, its predicted values are closer to the actual values, i.e., its variance is very small, which makes it more reliable. However, when using the classical accuracy and RMSE metrics, Classifier B has a higher accuracy and RMSE values than Classifier A, with accuracy values of 50% vs. 37.5%, and RMSE values of 5.2 vs. 6. By using the Pareto Accuracy (PA1), we obtain a more accurate picture in which Classifier A clearly outperforms Classifier B (87.5% vs. 50%). Here, we see a case where metrics used for both classical classification (accuracy) and ordinal classification (RMSE) do not reflect the actual performance of each classifier. In addition to the metrics, to evaluate the performance of our model and to compare it to the schedule-based model, we also performed a spatial analysis by plotting heatmaps and a temporal analysis using hours and day of the week (see Sect. 4.3). The analysis entailed comparing boarding stops that were predicted well, i.e., at accuracies of 50% or above. Lastly, to enrich our understanding of the nature and patterns of PT, we produced and analyzed feature importance by exploiting the SHAP values method, considered as constituting a unified framework for interpreting predictions based on game theory (Lundberg and Lee 2017). The values are the average of the marginal contributions across all permutations. Procedure We evaluated the above methodology by applying it to the smart card data of the city of Beer Sheva, Israel. With about 200,000 inhabitants, Beer Sheva is the largest city in the southern part of Israel. It presents an interesting use case given its relatively remote location, making it more isolated from a traffic perspective. Additionally, it has a sparse PT network that is easier to model. Furthermore, it has complete passenger boarding stop information, and road traffic in the city is not prone to heavy congestion. We utilized a smart card dataset consisting of over 1M records (after preprocessing, about 92% of the smart card records remained) from over 85,000 distinct travelers for one month during November and December 2018. Based on pre-analysis of the smart card data, the boarding profile per day of the week, the number of boardings, and the number of users recorded show regular patterns of use throughout this period, both for weekdays (Sun–Thu) and weekends (Fri–Sat). As evident from Fig. 5a, b the average usage profile for Beer Sheva users is quite stable across weeks and working days (Sun–Thu). The top figure shows the boardings and users per day for the one month of data. The bottom figure shows the boarding profile by day of the week. Next, we used a GTFS feed containing over 27,000 stops and over 200,000 PT trips in Israel for the equivalent period as the smart card data included all the operators (or agencies in GTFS tables) in the country. The dataset also included a detailed timetable for every PT trip. Lines and stops for the city of Beer Sheva were sorted by operator and geographic coordinates. All selected routes were bus lines. In total, there were about 650 stops selected in the study area.Fig. 5 Average usage profile for Beer Sheva users Based on the literature, there are different reports on the sizes of data sets used from several months of data for one city (e.g., Agard et al. 2006; Hasan et al. 2013) to more common studies between one week to four weeks of data (e.g., Chu and Chapleau 2010; Munizaga et al. 2014). Usually, longer studies tend to focus on more limited scales—lines or stations or small cities. Faroqi et al. (2018) noted this problem, especially if only one week of data is used. In this case, an inherent assumption is that travels are regular between days. However, as we have shown there is also the problem of irregular travelers that are commonly discarded (e.g., in OD analysis, Munizaga et al. 2014). Given the above, we consider that one month of data is most likely sufficient for our purposes. We also used a geospatial dataset from the municipal open GIS portal that contained a variety of geographical attributes of the city of Beer Sheva, such as traffic light locations, built-area densities, and more. We then extracted the 15 features from the above datasets. We converted the boarding stops from their Beer Sheva identifiers to numerical values (i.e., embedding). Lastly, we estimated an ML algorithm to classify the boarding stops and evaluated the classifier’s performance as described earlier. Model validation, comparative imputation methods and robustness As mentioned, one of our primary goals was to develop a generic model that can be applied in any city. To that end, we validated our model based on the data of the peripheral city of Kiryat Gat situated 43 km north of Beer Sheva and outside of the metropolitan region. We applied the method of transfer learning (Torrey and Shavlik 2010), entailing the transfer of relevant knowledge by fine-tuning a model on a "novel" dataset, i.e., a set of data on which it did not train. Other than allowing our model to train more to prove our hypothesis, we split the data initially into intervals of 10 days for the transfer learning task and then into intervals of 20 days for the evaluation. The main advantage of our modeling approach is that no ground truth is necessary to apply the model. This advantage is related to the fact that training is enabled without using domain-specific labels, i.e., when data integrity is poor, and no complementary data is available. We test this assertion by comparing the ML model to other possible imputation models. Such methods, specifically passenger history and temporal closeness, can, in some cases, provide very accurate predictions, mainly when data integrity is high. However, it is important to note that they have some essential limitations. The passenger history method requires passengers have multiple observations in the dataset, which is not always available when dealing with irregular travelers or to split the research data and utilize fewer data records. Additionally, the temporal closeness method is susceptible to data integrity and sparse rides. Passenger history and temporal closeness were applied using the following two algorithms: In addition, we also evaluated a semi-random classifier as a lower end imputation method using the following algorithm: Where: S—Smart Card dataset Hi,r,t—is the history of passenger i in route r and time period t Hi,b,r,t—is the most frequent boarding stop b of passenger i in route r and time period t Pi—is the ML prediction for observation i Tj—is the timestamp of observation j Bj—is the boarding stop of observation j Model robustness was validated by examining model performance on irregular passengers in comparison to the comparative imputation methods, given that simple imputation methods are ineffective when considering irregular travelers (Van Lint et al. 2005). Therefore, we examined model performance for predicting the boarding stop of one-time travelers in Beer Sheva, i.e., passengers who boarded once and did not return with PT on the same day. These observations are usually discarded because they do not contribute to OD estimation (Munizaga et al. 2014). Results The results are presented in the following order: First, we describe some properties of the data we used, showing its suitability for the developed methodology. Second, we describe the estimated ML model and its performance in comparison to the schedule-based model. Third, we analyze the performance between the two models both temporally and spatially. Fourth, we show the validation of the ML model on the use case of the city of Kiryat Gat, using transfer learning. Fifth, we compare our model to other alternative imputation model specifications mentioned earlier. Lastly, we examine prediction robustness. Data properties We began the analyses by exploring the processed data. First, we examined the degree of lateness in the smart card data compared to the timetable data in the GTFS feed for the city of Beer Sheva. For every PT trip, the time difference between planned and actual arrival times was computed for every stop on each line (see Fig. 6). As can be observed in Fig. 6, the density function shows both incidents of early arrival and lateness between about 500 s (8 min) early to 1000 s (16 min) late. This result suggests that the data is very suitable for applying our method. Moreover, it can be estimated that the schedule-based model using only GTFS timetable data will be less accurate.Fig. 6 The density of lateness in seconds in Beer Sheva Second, we investigated the distribution of the missing boarding stop information in the smart card data. Figure 7 presents the mean proportion of missing boarding stops per trip of the top three PT operators in Israel. This distribution is not random. If boarding stops were missing at random, the mean would be expected to be around 0 with a long tail. However, as the density function is far from that shape, we can deduce that boarding stops are indeed not missing at random.Fig. 7 The ratio of missing boarding stops per operator. Operator 3 is the largest, and 5, 18 are the second and third largest PT operators Model training and performance We trained several classifiers and evaluated their performances. Among the trained classifiers, the XGBoost classifier presented the best performance (see Table 2). We compared the classifiers using the common metrics as described before. Additionally, we evaluated our Pareto Accuracy metric based on error sizes of 1, 2, i.e., PA1, PA2. Any larger gap would typically be deemed unacceptable in terms of level-of-service and because these error sizes are highly correlated with PAi for i>2. One significant advantage of embedding is the calculation speed, which was an average of 15.9±0.023 s on about 300 K observations.Table 2 Classifier performances (test) Algorithm Accuracy Recall Precision F1 AUC PA1 PA2 Schedule based 0.209 0.209 0.212 0.209 0.590 0.470 0.643 Logistic regression 0.205 0.205 0.097 0.102 0.573 0.474 0.654 Random forest 0.368 0.368 0.348 0.353 0.666 0.672 0.818 XGBoost 0.410 0.410 0.393 0.394 0.765 0.712 0.843 The highest obtained result for each metric is marked in bold The SHAP values to evaluate the effect of each feature are presented in Fig. 8 (see also definitions in Table 1) . Here we can note: (a) by far the most important feature for the prediction is created by the predicted sequence, which shows it is highly correlated to actual patterns and is very useful for classification (i.e., schedule-based); (b) other than the first two SHAP features, the following four are temporal, which is commonsensical given that the different periods have varied impacts on traffic (such as the morning peak) and as a bus progresses along its route, stochastic events accumulate and the variance increases; (c) although geospatial features are not of the highest importance, they are not trivial, and thus, we conclude that certain physical attributes can influence the nature of our problem, e.g., denser areas can engender more congestion; and (d) the two least significant features pertain to the day of the week, from which we can assert that daily PT routines remained quite stable in our case study.Fig. 8 Feature importance using SHAP values In Fig. 9, we present Pareto Accuracy between the ML model and the schedule-based one. It shows that the results are stable even for higher values than 1. Therefore, we can conclude that the proposed model outperforms the schedule-based model.Fig. 9 Pareto accuracy comparison between ML and schedule-based models (test) Spatial and temporal analyses In addition to the aggregated results, we analyzed the model performance both temporally (see Fig. 10) and spatially (see Fig. 11). The temporal analysis shows that, in terms of accuracy, our proposed model outperformed the schedule-based method on both a daily and an hourly basis.3 Moreover, the spatial analysis showed similar results, and the stops where the predictions were ranked ’good’, i.e., over 50% accuracy, were plotted. Two major insights can be derived from these analyses: First, the ML model predicts considerably more stops than the schedule-based model. Second, the schedule-based model renders good predictions mainly for the central stops (train stations, main roads, or industrial zones). However, when the model is applied to non-central locations, it is suboptimal, in stark contrast to the ML model, making good predictions across all locations.Fig. 10 Temporal performance of models (test)—a daily, b hourly Fig. 11 Heatmaps of boarding stops with prediction accuracy of over 50% (test) Model validation As noted, we performed the model validation for the nearby city of Kiryat Gat. Evidently, the ML model performed remarkably better than the schedule-based model (see Table 3). Figure 12 showing the Pareto Accuracy for different values of error size showing the ML model is consistently better. Figure 13 presents a performance comparison in Kiryat Gat showing the temporal analysis—accuracy by day of the week and on an hourly basis for weekdays which shows similar properties to the trained model, the ML model demonstrated higher accuracy compared to the schedule-based model. Figure 14 presents the spatial analysis revealing once more that the ML model predicts more stops with higher accuracy.Table 3 Classifier performances for model validation Algorithm Accuracy Recall Precision F1 AUC PA1 PA2 Schedule based 0.253 0.253 0.224 0.234 0.599 0.404 0.550 Logistic regression 0.202 0.202 0.789 0.317 0.392 0.388 0.521 Random forest 0.221 0.221 0.594 0.246 0.578 0.423 0.588 XGBoost 0.438 0.438 0.441 0.419 0.685 0.668 0.802 The highest obtained result for each metric is marked in bold Fig. 12 Pareto accuracy comparison of models between ML and schedule-based models (validation) Fig. 13 Temporal performance of models (validation) Fig. 14 Heatmaps of boarding stops with predicted accuracy of over 50% (validation) Comparative imputation methods Table 4 shows the results of the comparisons to alternative imputation methods. While the predicted accuracy of the two alternative methods is similar, the disadvantages of the aforementioned methods are more evident in the lower share of the population than can be predicted compared to the ML model. The semi-random classifier naturally demonstrates that it is far from trustworthy in the case of hierarchical PT networks.Table 4 Results of comparative imputation methods Method Percent predicted (%) Accuracy for predicted observations (%) Proposed XGBoost 100 41 Passenger history 82 59 Temporally close passengers 52 59 Semi-random guessing 100 11 It is important to note that while the accuracy of our proposed method is lower, it is far more robust, both in terms of percentage of population predicted and on irregular travelers, which other suggested methods are incapable of predicting (see Sect. 4.6). For example, in predicting using historical records, we cannot predict a new passenger or a new route. For using temporal closeness, the prediction will be extremely sensitive to sparse routes. Robustness to irregular travelers While the personal history method can indeed be relevant as evident in Table 4, as noted above (see Sect. 3.5) model robustness was evaluated by examining performance for predicting the boarding stop of one-time travelers. As shown in Table 5, the results clearly show (see the first row in Table 5) that the ML model is robust and capable of predicting missing stops even for irregular or new passengers that have no historical pattern. Additionally, as noted earlier, the suggested methods are very limited. The evaluation of passengers they do not predict is clearly shown below in Table 5 (see the second and the third row).Table 5 Results of the ML model (XGBoost) on one-time travelers and passengers not predicted by comparative methods Passenger type Accuracy Recall Precision F1 AUC PA1 PA2 One-time 0.408 0.408 0.394 0.390 0.767 0.703 0.838 Not predicted by method 1 0.419 0.419 0.407 0.402 0.772 0.706 0.835 Not predicted by method 2 0.348 0.348 0.336 0.330 0.744 0.671 0.822 Discussion and conclusions In this study, we showed that by mining smart card data and extracting timetable data, we could construct a passenger boarding stop prediction model, which surpasses the traditional schedule-based method. Our research revealed that applying machine learning techniques improves the integrity of PT data, which can significantly benefit the field of transportation planning and operations. From the results, we can deduce the following conclusions: First, our methodology for feature extraction and machine learning model construction demonstrates several noteworthy advantages: (a) the ML algorithm generates a generic model that can be used with other smart card datasets since the labels (i.e., numeric representations) are always aligned in all datasets; (b) by embedding the boarding stops, our method ensures that the number of distinct labels is relatively small and a significant computation time reduction can be accomplished; (c) boarding stop use is inherently imbalanced, as some stops are frequently used while others are used rarely. Our proposed methodology is able to accurately classify many classes despite the inherent imbalances, thus contributing to unpredictability reduction; (d) the method is data lean and requires only mining a smart card dataset and a GTFS feed (or any compatible timetable dataset) without the need to process any other datasets; (e) the ML model is entirely complementary to other imputation methods including the schedule-based method as well as passenger history or temporal closeness; and (f) the method provides a robust model capable of dealing even with irregular or unpredictable passengers. Second, our model (applying the XGBoost algorithm) produced the highest performance, with 41% accuracy and 71% PA1, whereas the schedule-based method achieved only 21% accuracy and 47% PA1. Even for larger error sizes, the ML model outperformed the schedule-based one. Moreover, the schedule-based method was able to render good predictions only for a few main stops compared to the ML model, which predicted well across all stops. This dependency on centrality was clearly visible in the spatial analysis of the stops that were well-predicted. This result confirms our conjecture that the schedule-based imputation approaches can be significantly improved by using ML methods. Furthermore, we also found that complex methods, such as ensemble, resulted in much better model performance than simple algorithms, such as logistic regression. In future research, we intend to test the performance of additional prediction algorithms, such as Deep Neural Networks (Jung and Sohn 2017; Liu and Chen 2017). Third, from the SHAP values (Fig. 8), the following can be noted: The temporal features (created by the timetable from the GTFS feeds) are indeed crucial for the operation of the ML model. Geospatial features, however, were less important. Accordingly, we estimated a model trained without the geospatial features (see Table B.1 in Appendix B). In comparison to the richer model, the performance is somewhat worse. Therefore, we assert that such information is considered useful: Firstly, to understand patterns in a given city, for instance, which spatial attribute is more closely correlated with lateness or earliness. Secondly, it can help the transfer learning process in a new city, i.e., if the model was trained on city A, and will be used to predict city B, using the spatial features will produce a more robust model to the difference between those cities. Fourth, we showed that the ML model is transferable (see Sect. 3.2) and able to provide strong and consistent results when validated on another city while outperforming the schedule-based imputation method. Nonetheless, our method, given its generic nature, is not entirely comparable with methods of dissimilar nature, such as those presented in Table 4 which cannot be straightforwardly transferred to another context. Since, to the best of our knowledge, no other imputation method shows such transferability, robustness, and generic nature other than the schedule-based imputation, the latter should be regarded as the comparative benchmark until another imputation method is developed. Fifth, we recommend using our model when the lack of data does not allow for other more accurate methods to be used, such as passenger history or temporal closeness. Nonetheless, our model can complement these methods, especially for those records that are overlooked, as shown in Table 4, and thus can utilize more of the scarce data at hand. As noted, our method does not require mining or accessing any additional datasets (like AVL or APC), which are not always available and can increase the extent of errors in the prediction. This observation makes our method extremely suitable for planning purposes in non-auto-dependent and less technologically-orientated societies in developing countries and the Global South (Sohail et al. 2006). Lastly, we introduced a new generalized accuracy metric which we named Pareto Accuracy that allows to better compare between classifiers for ordinal classification problems. This metric is more robust to outliers, easier to interpret, and accounts for the true dimension of errors. In addition, the metric is easy to implement. In the future, we hope to understand how Pareto Accuracy can improve additional ordinal classification use cases. There are a few limitations to the study worth noting. One is that our method requires several constraints to succeed, such as timestamps, trip IDs, and existing trip timetables. These constraints potentially reduce the number of relevant datasets and the number of observations that could be imputed. However, these constraints also preclude the use of the schedule-based method; hence, in practice, our method has little effect on the ability to impute missing data. In addition, the generality of our method can increase bias, as it ignores features that cannot be transferred between datasets. These features, such as having each PT line as a categorical feature, can reduce bias when imputing a specific dataset. Possible extensions include: predicting alighting stops (when the operator does not record TAP out), imputing other attributes of interest such as trip ID or time of day, etc. In the future, we would like to test our model in other cities to verify its generalizations. In addition, we also suggest testing the influence of transfer learning on new datasets. Following a suggestion by one of our Reviewers, we consider it important that researchers also carry out transnational studies where models trained on data from one country are validated on similarly structured data from a least one other country to ensure geographical and cultural robustness. In addition, we suggest that researchers test the method with data from urbanities of different spatial scales to verify robustness to the public network dimensions. To summarize, missing data imputation is a difficult and complex task. On the one hand, one wants as much data as possible for analyses, while on the other hand, data integrity is of critical importance and demands the availability of imputation methods that work well. We assert that the commonly used schedule-based method suffers from a subpar performance in terms of accuracy and other key metrics. It is highly dependent on the centrality of boarding stops. In contrast, we showed that our model outperformed the schedule-based method in all metrics over different temporal periods. It was more robust to the centrality of the imputed stops and irregularity of recorded trips. This makes it a much more suitable method for imputation as it improves data integrity. In addition, our method is based on generic classification and thus can be used in a wide variety of use cases. Appendix A Metrics presented in this paper:Accuracy—Percent of observations that were correctly classified Recall—The number of observations for each class that were correctly classified divided by the total number of distinct observations from this class. Final Recall is the weighted average of the above on all classes. Precision—The number of observations for each class that were correctly classified divided by the total number of observations that were predicted within this class. Final Precision is the weighted average of the above on all classes. F1—2 × (Precision × Recall)/(Precision + Recall) AUC—Area under curve (AUC) is the area under the receiver operating characteristic (ROC) curve. This curve, for each class, is the true positive rate as a function of the false positive rate. A weighted average of the areas under the curves of all classes is calculated as the AUC metric. RMSE—Root mean square error (RMSE) is a method for ordinal classification and regression. It sums the square difference from prediction to actual label, then returns the root of the above average. Appendix B See Table 6.Table 6 Model performance with and without geospatial features (XGBoost) Algorithm Accuracy Recall Precision F1 AUC PA1 PA2 Without 0.409 0.409 0.392 0.393 0.770 0.712 0.842 With 0.410 0.410 0.393 0.394 0.765 0.712 0.843 Acknowledgements This research was supported by the Ministry of Science & Technology, Israel, and The Ministry of Science & Technology of the People’s Republic of China (Grant No. 3-15741). We want to thank Prof. Itzhak Benenson (Tel Aviv University) for the preliminary discussions of the research question. We also want to thank Valfredo Macedo Veiga Junior (Valf) for designing the infographic illustration and Sandra Falkenstein for editing and proofreading this paper. Special thanks to Data Scientist Raz Vais, advisor to the Israeli National Public Transport Authority, for help in obtaining and processing the data. Lastly, we thank the Editor in Chief and two anonymous Reviewers for their constructive comments to our paper. 1 “SHAP (SHapley Additive exPlanations) is a game-theoretic approach to explain the output of any machine learning model. It connects optimal credit allocation with local explanations using the classic Shapley values from game theory and related extensions” https://github.com/slundberg/shap. 2 It is frequent practice to split large-scale datasets into test and train datasets (Guyon 1997), where a split of 80%/20% is regarded a common practice. In our case, due to the temporal nature of the dataset, we find it practical and logical to train the classifier on three weeks (75%) of data and evaluate the classifier’s performance on a week (25%) of data. 3 The hourly analysis was done on weekdays when traffic congestion makes the prediction of PT service punctuality likely more complex. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Agard B Morency C Trépanier M Mining public transport user behaviour from smart card data IFAC Proc Vol 2006 39 3 399 404 10.3182/20060517-3-FR-2903.00211 Al Mamun MS Lownes NE A composite index of public transit accessibility J Public Transp 2011 14 2 69 87 10.5038/2375-0901.14.2.4 Alguero P (2013) Using smart card technologies to measure public transport performance: data capture and analysis. Technical Report Industrial Engineering, Universitat Politecnica de Catalunya, Barcelona Almlöf E, Rubensson I, Cebecauer M, Jenelius E (2020) Who is still travelling by public transport during Covid-19? Socioeconomic factors explaining travel behaviour in Stockholm based on smart card data. Working Paper, Integrated Transport Research Lab (ITRL), KTH - Royal Institute of Technology, Stockholm (September 8, 2020) Anda C Erath A Fourie PJ Transport modelling in the age of big data Int J Urban Sci 2017 21 sup1 19 42 10.1080/12265934.2017.1281150 Antrim A, Barbeau SJ (2013) The many uses of GTFS data—opening the door to transit and multimodal applications. Location-Aware Information Systems Laboratory at the University of South Florida Bagchi M White PR The potential of public transport smart card data Transp Policy 2005 12 5 464 474 10.1016/j.tranpol.2005.06.008 Batista GE Monard MC An analysis of four missing data treatment methods for supervised learning Appl Artif Intell 2003 17 5–6 519 533 10.1080/713827181 Ben-Elia E Lyons G Mokhtarian PL Epilogue: the new frontiers of behavioral research on the interrelationships between ICT, activities, time use and mobility Transportation 2018 45 2 479 497 10.1007/s11116-018-9871-x Benenson I, Marinov M, Ben Elia E (2019) Is servicing commuters the goal of the public transport system? In: Geocomputation 2019. The University of Auckland Bertsimas D Pawlowski C Zhuo YD From predictive methods to missing data imputation: an optimization approach J Mach Learn Res 2017 18 1 7133 7171 Briand A-S Côme E Trépanier M Oukhellou L Analyzing year-to-year changes in public transport passenger behaviour using smart card data Transp Res Part C Emerg Technol 2017 79 274 289 10.1016/j.trc.2017.03.021 Bryan H, Blythe P (2007) Understanding behaviour through smartcard data analysis. In: Proceedings of the Institution of Civil Engineers-Transport, vol 160. Thomas Telford Ltd, pp 173–177 Camino RD, Hammerschmidt CA, State R (2019) Improving missing data imputation with deep generative models. arXiv preprint arXiv:1902.10666 Cardoso JS Sousa R Measuring the performance of ordinal classification Int J Pattern Recognit Artif Intell 2011 25 8 1173 1195 10.1142/S0218001411009093 Cats O Loutos G Real-time bus arrival information system: an empirical evaluation J Intell Transp Syst 2016 20 2 138 151 10.1080/15472450.2015.1011638 Ceder A New urban public transportation systems: initiatives, effectiveness, and challenges J Urban Plan Dev 2004 130 1 56 65 10.1061/(ASCE)0733-9488(2004)130:1(56) Ceder A Public transit planning and operation: modeling, practice and behavior 2016 Boca Raton CRC Press Chen Z Fan W Extracting bus transit boarding stop information using smart card transaction data J Mod Transp 2018 26 3 209 219 10.1007/s40534-018-0165-y Chen T, Guestrin C (2016) XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining, pp 785–794 Chen C Ma J Susilo Y Liu Y Wang M The promises of big data and small data for travel behavior (aka human mobility) analysis Transp Res Part C Emerg Technol 2016 68 285 299 10.1016/j.trc.2016.04.005 27182125 Chien H-Y Jan J-K Tseng Y-M An efficient and practical solution to remote authentication: smart card Comput Secur 2002 21 4 372 375 10.1016/S0167-4048(02)00415-7 Chu KKA Chapleau R Enriching archived smart card transaction data for transit demand modeling Transp Res Rec 2008 2063 1 63 72 10.3141/2063-08 Chu KKA Chapleau R Augmenting transit trip characterization and travel behavior comprehension: multiday location-stamped smart card transactions Transp Res Rec 2010 2183 1 29 40 10.3141/2183-04 Costa AF, Santos MS, Soares JP, Abreu PH (2018) Missing data imputation via denoising autoencoders: the untold story. In: International symposium on intelligent data analysis. Springer, pp 87–98 Covic F Voß S Interoperable smart card data management in public mass transit Public Transp 2019 11 3 523 548 10.1007/s12469-019-00216-x Dacheng C, Ruizhi Y, Lei S, Kiat T, Hui D, Whye JKH, Kiong N (2018) Traveler segmentation using smart card data with deep learning on noisy labels. In: Proceedings of ACM KDD conference, vol 10, New York Devillaine F Munizaga M Trépanier M Detection of activities of public transport users by analyzing smart card data Transp Res Rec 2012 2276 1 48 55 10.3141/2276-06 Echaniz E Ho C Rodriguez A dell’Olio L Modelling user satisfaction in public transport systems considering missing information Transportation 2020 47 6 2903 2921 10.1007/s11116-019-09996-4 Faroqi H Mesbah M Kim J Applications of transit smart cards beyond a fare collection tool: a literature review Adv Transp Stud 2018 45 107 122 Fonzone A Schmöcker J-D Viti F New services, new travelers, old models? Directions to pioneer public transport models in the era of big data J Intell Transp Syst 2016 20 311 315 10.1080/15472450.2016.1190553 Frank E, Hall M (2001) A simple approach to ordinal classification. In: European conference on machine learning. Springer, New York, pp 145–156 Garg A, Naryani D, Aggarwal G, Aggarwal S (2018) DL-GSA: a deep learning metaheuristic approach to missing data imputation. In: International conference on sensing and imaging. Springer, New York, pp 513–521 Gaudette L, Japkowicz N (2009) Evaluation methods for ordinal classification. In: Canadian conference on artificial intelligence. Springer, New York, pp 207–210 Gordon JB Koutsopoulos HN Wilson NH Attanucci JP Automated inference of linked transit journeys in London using fare-transaction and vehicle location data Transp Res Rec 2013 2343 1 17 24 10.3141/2343-03 Gudivada V Apon A Ding J Data quality considerations for big data and machine learning: going beyond data cleaning and transformations Int J Adv Softw 2017 10 1 1 20 Guihaire V Hao J-K Transit network design and scheduling: a global review Transp Res Part A Policy Pract 2008 42 10 1251 1273 10.1016/j.tra.2008.03.011 Guyon I (1997) A scaling law for the validation-set training-set size ratio. AT &T Bell Laboratories 1(11) Hadas Y Assessing public transport systems connectivity based on Google transit data J Transp Geogr 2013 33 105 116 10.1016/j.jtrangeo.2013.09.015 Hagenauer J Helbich M A comparative study of machine learning classifiers for modeling travel mode choice Expert Syst Appl 2017 78 273 282 10.1016/j.eswa.2017.01.057 Hasan S Schneider CM Ukkusuri SV González MC Spatiotemporal patterns of urban human mobility J Stat Phys 2013 151 1 304 318 10.1007/s10955-012-0645-0 Huang J Levinson D Wang J Zhou J Wang Z-J Tracking job and housing dynamics with smartcard data Proc Natl Acad Sci 2018 115 50 12710 12715 10.1073/pnas.1815928115 30455293 Jang W Travel time and transfer analysis using transit smart card data Transp Res Rec 2010 2144 1 142 149 10.3141/2144-16 Jiao Y Du P Performance measures in evaluating machine learning based bioinformatics predictors for classifications Quant Biol 2016 4 4 320 330 10.1007/s40484-016-0081-2 Jung J Sohn K Deep-learning architecture to forecast destinations of bus passengers from entry-only smart-card data IET Intell Transp Syst 2017 11 6 334 339 10.1049/iet-its.2016.0276 Kandt J Batty M Smart cities, big data and urban policy: towards urban analytics for the long run Cities 2021 109 102992 10.1016/j.cities.2020.102992 Khiari J, Moreira-Matias L, Cerqueira V, Cats O (2016) Automated setting of bus schedule coverage using unsupervised machine learning. In: Pacific-Asia conference on knowledge discovery and data mining. Springer, pp 552–564 Kieu LM Bhaskar A Chung E Passenger segmentation using smart card data IEEE Trans Intell Transp Syst 2015 16 3 1537 1548 10.1109/TITS.2014.2368998 Kotsiantis SB Zaharakis I Pintelas P Supervised machine learning: a review of classification techniques Emerg Artif Intell Appl Comput Eng 2007 160 3 24 Kusakabe T Asakura Y Behavioural data mining of transit smart card data: a data fusion approach Transp Res Part C Emerg Technol 2014 46 179 191 10.1016/j.trc.2014.05.012 Lakshminarayan K, Harp SA, Goldman RP, Samad T (1996) Imputation of missing data using machine learning techniques. In: Proceedings of ACM KDD conference, pp 140–145 Laña I Olabarrieta II Vélez M Del Ser J On the imputation of missing data for road traffic forecasting: new insights and novel techniques Transp Res Part C Emerg Technol 2018 90 18 33 10.1016/j.trc.2018.02.021 Li H Li F Song C Yan Y Towards smart card based mutual authentication schemes in cloud computing KSII Trans Internet Inf Syst (TIIS) 2015 9 7 2719 2735 Li T Sun D Jing P Yang K Smart card data mining of public transport destination: a literature review Information 2018 9 1 18 10.3390/info9010018 Liu L Chen R-C A novel passenger flow prediction model using deep learning methods Transp Res Part C Emerg Technol 2017 84 74 91 10.1016/j.trc.2017.08.001 Liu Y Zhou Y Wen S Tang C A strategy on selecting performance metrics for classifier evaluation Int J Mobile Comput Multimed Commun (IJMCMC) 2014 6 4 20 35 10.4018/IJMCMC.2014100102 Lundberg SM, Lee S-I (2017) A unified approach to interpreting model predictions. In: Advances in neural information processing systems, pp 4765–4774 Luo D Bonnetain L Cats O van Lint H Constructing spatiotemporal load profiles of transit vehicles with multiple data sources Transp Res Rec 2018 2672 8 175 186 10.1177/0361198118781166 Ma X-L Wang Y-H Chen F Liu J-F Transit smart card data mining for passenger origin information extraction J Zhejiang Univ Sci C 2012 13 10 750 760 10.1631/jzus.C12a0049 Ma X Wu Y-J Wang Y Chen F Liu J Mining smart card data for transit riders’ travel patterns Transp Res Part C Emerg Technol 2013 36 1 12 10.1016/j.trc.2013.07.010 Ma X Liu C Wen H Wang Y Wu Y-J Understanding commuting patterns using transit smart card data J Transp Geogr 2017 58 135 145 10.1016/j.jtrangeo.2016.12.001 Maeda TN Shiode N Zhong C Mori J Sakimoto T Detecting and understanding urban changes through decomposing the numbers of visitors’ arrivals using human mobility data J Big Data 2019 6 1 4 10.1186/s40537-019-0168-5 Mazloumi E Currie G Rose G Using GPS data to gain insight into public transport travel time variability J Transp Eng 2010 136 7 623 631 10.1061/(ASCE)TE.1943-5436.0000126 Milne D Watling D Big data and understanding change in the context of planning transport systems J Transp Geogr 2019 76 235 244 10.1016/j.jtrangeo.2017.11.004 Munizaga MA Palma C Estimation of a disaggregate multimodal public transport origin-destination matrix from passive smartcard data from Santiago, Chile Transp Res Part C Emerg Technol 2012 24 9 18 10.1016/j.trc.2012.01.007 Munizaga M Devillaine F Navarrete C Silva D Validating travel behavior estimated from smartcard data Transp Res Part C Emerg Technol 2014 44 70 79 10.1016/j.trc.2014.03.008 Namiot D, Sneps-Sneppe M (2017) A survey of smart cards data mining. In: AIST (Supplement), pp 314–325 Orro A Novales M Monteagudo Á Pérez-López J-B Bugarín MR Impact on city bus transit services of the Covid-19 lockdown and return to the new normal: the case of A Coruña (Spain) Sustainability 2020 12 17 7206 10.3390/su12177206 Palacio SM (2018) Machine learning forecasts of public transport demand: a comparative analysis of supervised algorithms using smart card data. XREAP WP, available at SSRN Pelletier M-P Trépanier M Morency C Smart card data use in public transit: a literature review Transp Res Part C Emerg Technol 2011 19 4 557 568 10.1016/j.trc.2010.12.003 Petrović N, Bojović N, Petrović J (2016) Appraisal of urbanization and traffic on environmental quality. J CO2 Util 16:428–430 Qu L Li L Zhang Y Hu J PPCA-based missing data imputation for traffic flow volume: a systematical approach IEEE Trans Intell Transp Syst 2009 10 3 512 522 10.1109/TITS.2009.2026312 Saunders JA Morrow-Howell N Spitznagel E Doré P Proctor EK Pescarino R Imputing missing data: a comparison of methods for social work researchers Soc Work Res 2006 30 1 19 31 10.1093/swr/30.1.19 Schmöcker J, Kurauchi F, Shimamoto H (2017) An overview on opportunities and challenges of smart card data analysis. Public transport planning with smart card data. CRC, Boca Raton, pp 2–12 Shalaby A Farhan A Prediction model of bus arrival and departure times using AVL and APC data J Public Transp 2004 7 1 41 61 10.5038/2375-0901.7.1.3 Singh A, Thakur N, Sharma A (2016) A review of supervised machine learning algorithms. In: 2016 3rd International conference on computing for sustainable global development (INDIACom). IEEE, pp 1310–1315 Sohail M Maunder D Cavill S Effective regulation for sustainable public transport in developing countries Transp Policy 2006 13 3 177 190 10.1016/j.tranpol.2005.11.004 Stopher PR Greaves SP Household travel surveys: where are we going? Transp Res Part A Policy Pract 2007 41 5 367 381 10.1016/j.tra.2006.09.005 Tao S Rohde D Corcoran J Examining the spatial-temporal dynamics of bus passenger travel behaviour using smart card data and the flow-comap J Transp Geogr 2014 41 21 36 10.1016/j.jtrangeo.2014.08.006 Toqué F, Côme E, El Mahrsi MK, Oukhellou L (2016) Forecasting dynamic public transport origin-destination matrices with long-short term memory recurrent neural networks. In: 2016 IEEE 19th international conference on intelligent transportation systems (ITSC). IEEE, pp 1071–1076 Toqué F, Khouadjia M, Come E, Trepanier M, Oukhellou L (2017) Short & long term forecasting of multimodal transport passenger flows with machine learning methods. In: 2017 IEEE 20th international conference on intelligent transportation systems (ITSC). IEEE, pp 560–566 Torrey L, Shavlik J (2010) Transfer learning. In: Handbook of research on machine learning applications and trends: algorithms, methods, and techniques. IGI Global, pp 242–264 Traut EJ Steinfeld A Identifying commonly used and potentially unsafe transit transfers with crowdsourcing Transp Res Part A Policy Pract 2019 122 99 111 10.1016/j.tra.2019.02.005 Trépanier M Tranchant N Chapleau R Individual trip destination estimation in a transit smart card automated fare collection system J Intell Transp Syst 2007 11 1 1 14 10.1080/15472450601122256 Van Lint J Hoogendoorn S van Zuylen HJ Accurate freeway travel time prediction with state-space neural networks under missing data Transp Res Part C Emerg Technol 2005 13 5–6 347 369 10.1016/j.trc.2005.03.001 Walker J Human transit: how clearer thinking about public transit can enrich our communities and our lives 2012 Washington Island Press Wang W Attanucci JP Wilson NH Bus passenger origin-destination estimation and related analyses using automated data collection systems J Public Transp 2011 14 4 131 150 10.5038/2375-0901.14.4.7 Welch TF Widita A Big data in public transportation: a review of sources and methods Transp Rev 2019 39 6 795 818 10.1080/01441647.2019.1616849 Yan F Yang C Ukkusuri SV Alighting stop determination using two-step algorithms in bus transit systems Transportmetrica A Transp Sci 2019 15 2 1522 1542 10.1080/23249935.2019.1615578 Yap M Cats O van Arem B Crowding valuation in urban tram and bus transportation based on smart card data Transportmetrica A Transp Sci 2020 16 1 23 42 10.1080/23249935.2018.1537319 Zhang Y Cheng T A deep learning approach to infer employment status of passengers by using smart card data IEEE Trans Intell Transp Syst 2020 21 2 617 629 10.1109/TITS.2019.2896460 Zhang Y, Cheng T, Sari Aslam N (2019) Deep learning for demographic prediction based on smart card data and household survey. In: Proceedings of the 27th conference on GIS research UK (GISRUK), vol 2019. Geographic Information Science Research UK (GISRUK) Zhang N Jia W Wang P Dung C-H Zhao P Leung K Su B Cheng R Li Y Changes in local travel behaviour before and during the COVID-19 pandemic in Hong Kong Cities 2021 112 103139 10.1016/j.cities.2021.103139 33589850 Zhao J Qu Q Zhang F Xu C Liu S Spatio-temporal analysis of passenger travel patterns in massive smart card data IEEE Trans Intell Transp Syst 2017 18 11 3135 3146 10.1109/TITS.2017.2679179	0	PMC9734418	NO-CC CODE	2022-12-14 23:28:28	no	Public Transp. 2022 Dec 7;:1-33	utf-8	null	null	null	oa_other
==== Front Br J Cancer Br J Cancer British Journal of Cancer 0007-0920 1532-1827 Nature Publishing Group UK London 36476659 2070 10.1038/s41416-022-02070-4 Review Article A comprehensive systematic review of colorectal cancer screening clinical practices guidelines and consensus statements http://orcid.org/0000-0002-4852-5100 Maes-Carballo Marta [email protected] 123 García-García Manuel 1 Martín-Díaz Manuel 4 Estrada-López Carlos Roberto 1 Iglesias-Álvarez Andrés 5 Filigrana-Valle Carmen Milagros 1 Khan Khalid Saeed 36 Bueno-Cavanillas Aurora 367 1 grid.418883.e 0000 0000 9242 242X Department of General Surgery, Breast cancer Unit, Complexo Hospitalario de Ourense, Ourense, Spain 2 Hospital Público de Verín, Ourense, Spain 3 grid.4489.1 0000000121678994 Department of Preventive Medicine and Public Health, University of Granada, Granada, Spain 4 Hospital Santa Ana de Motril, Granada, Spain 5 grid.11794.3a 0000000109410645 University of Santiago de Compostela, Santiago de Compostela, Spain 6 grid.507088.2 Instituto de Investigación Biosanitaria IBS, Granada, Spain 7 grid.466571.7 0000 0004 1756 6246 CIBER of Epidemiology and Public Health (CIBERESP), Madrid, Spain 7 12 2022 112 12 4 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. High-quality clinical practice guidelines (CPGs) and consensus statements (CSs) are essential for evidence-based medicine. The purpose of this systematic review was to appraise the quality and reporting of colorectal cancer (CRC) screening CPGs and CSs. After prospective registration (Prospero no: CRD42021286156), a systematic review searched CRC guidances in duplicate without language restrictions in ten databases, 20 society websites, and grey literature from 2018 to 2021. We appraised quality with AGREE II (% of maximum score) and reporting with RIGHT (% of total 35 items) tools. Twenty-four CPGs and 5 CSs were analysed. The median overall quality and reporting were 54.0% (IQR 45.7–75.0) and 42.0% (IQR 31.4–68.6). The applicability had low quality (AGREE II score <50%) in 83% of guidances (24/29). Recommendations and conflict of interest were low-reported (RIGHT score <50%) in 62% guidances (18/29) and 69% (20/29). CPGs that deployed systematic reviews had better quality and reporting than CSs (AGREE: 68.5% vs. 35.5%; p = 0.001; RIGHT: 74.6% vs. 41.4%; p = 0.001). In summary, CRC screening CPGs and CSs achieved low quality and reporting. It is necessary a revision and an improvement of the current guidances. Their development should apply a robust methodology using proper guideline development tools to obtain high-quality evidence-based documents. Subject terms Cancer screening Colon cancer ==== Body pmcBackground Colorectal cancer (CRC) is the third most commonly worldwide cancer in both men and women, with 1.9 million new cases and a mortality of 10%, 935,000 patients, per year [1]. Early detection of CRC due to screening programmes, removal of precancerous polyps with colonoscopy and advances in treatment management have decreased CRC incidence and mortality rates [2, 3]. It has been demonstrated that early diagnosis could decrease CRC morbimortality. The 5-year mortality rate of 10% for early-stage increases to 28% for locally advanced disease and 86% for metastatic cancer, according to USA data [4]. Although cancer prevention programmes are undoubtedly important, there is a certain variation in CRC screening guidance documents depending on the source [5]. Screening programmes should be accommodated to risk groups to offer strategies adapted to their risk of developing CRC [6]. Patients and clinicians should assess the patient’s overall health, previous screening history, and preferences to define if screening is appropriate [7]. The years range for CRC screening in the general population should be determined to capture the most significant number of CRC cases while considering the effectiveness and cost-effectiveness of screening tests, regional epidemiology, and expected benefits and harms to the screened population. This implies that CRC screening guidelines show heterogeneity in recommendations and purpose since they are often aimed at particular subgroups. This heterogeneity could be a barrier to standardising care quality and make it hard to follow recommendations [8, 9]. Clinical practice guidelines (CPG) and consensus statements (CS) are evidence-based documents to support high-quality care in specific situations [10–13]. The analysis of the quality (the validity of the recommendations made) and reporting (the rigour of the presentation of the document) are elements that allow practitioners to identify trustworthy guidance documents [14]. Therefore, there is a need to assess recently published CRC screening CPGs and CSs [15]. A decade previously, Simone et al. [16] inspected the quality of CRC guidance documents but with an older tool (AGREE, previous version). Therefore, this review is currently outdated. It focused on hereditary CRC guidance in general (screening, surveillance, and management). That old systematic review [16] only included 17 guidances. Tian et al. [17] published a recent systematic review written in Chinese with only 19 guidances selected and selecting only English and Chinese guidances. There is a need for a broad systematic review focused on CRC screening CPGs and CSs without language or data source limitations. So, given this background, we systematically assessed quality and reporting of all the CRC screening guidances published using current, validated instruments and highlighted each guidance’s strengths and limitations. Materials and methods We conducted a thorough systematic review following prospective registration (Prospero ID: CRD42021286156) and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [18–20] (Appendix S0). Literature search strategy, data sources, study selection and data extraction We completed an exhaustive literature examination of PubMed, EMBASE, Web of Science, Scopus, CDSR and Tripdatabase from January 2018 to November 2021 without language limitations. Our selection criteria for the time period targeted documents published in the last 3 years (from 2018 onwards), following the advice of an extensive systematic review of the methodological handbooks for updating clinical practice guidelines. This systematic review stated that most handbooks that collect recommendations on updating guidances recommended that they should be updated 3-yearly [21]. We used MeSH terms “practice guidelines”, “guidelines”, “consensus”, “colorectal neoplasms”, “colorectal cancer”, “screening”, “quality”, “reporting” and including term variants. The contribution to global colorectal cancer’s scientific production of the professional societies´ country of origin greater than 0.5% was the main criterion for including these professional societies in our systematic review. Scopus was searched on March 10th, 2022, to estimate the scientific production of each country (85932 “Colorectal Cancer and Health” documents). This decision was in line with the previous peer-reviewed published systematic reviews [22–25]. We visited 20 pertinent professional organisations´ websites and four guideline databases: National Comprehensive Cancer Network- NCCN, TRIP database, CMA Infobase, Health Services Technology Assessment Texts-HSTAT, and Scottish Intercollegiate Guidelines Network-SIGN) to conclude the examination. More additional records were searched in the identified publications´ bibliographies to include other essential studies in our review. Appendix S1 shows the search strategy. The inclusion criteria were CPGs, CSs, recommendations, and position statements about CRC screening produced by professional organisations, societies, or government agencies. Exclusion criteria were CPGs and CSs not related to CRC screening and protocols in general. We decided to exclude protocols, programmes that sets out a precise sequence of activities in managing a specific clinical condition, as they did not define how a procedure was executed but why, where, when and by whom the care was given [26]. We also rejected obsolete versions of guidelines updated in more recent years from the same organisation, guidelines for education purposes or only for patients. Three independent reviewers (AIA, CMFV and CREL) confirmed eligibility by checking the titles and abstracts and performed a full-text assessment of the selected studies. Duplicate documents were removed. Disagreements or inconsistencies were resolved by consensus with the input of a fourth reviewer (MMC). Data extraction was carried out independently by three authors (AIA, CMFV and CREL) and collected on an Excel datasheet to compare results. Quality and reporting appraisal AGREE II statement and RIGHT instrument were used in a manner similar to our previously published work [22, 23] to evaluate quality and reporting, respectively (Appendix S2) [27, 28]. Before data extraction, the reviewers had sessions to understand AGREE and RIGHT criteria (items and domains). After independent data extraction, two reviewers (AIA and CREL) discussed their disagreements, and in case of inability to resolve disagreements mutually, an arbitrator (MMC) helped reach a final judgement. AGREE II examined the elements of the guideline development and the recommendation grades. It defined quality as the “trustworthiness that conceivable development biases have been properly managed and recommendations are internally and externally valid” [29]. Twenty-three items were categorised into six domains: scope and purpose (items 1 to 3), stakeholder involvement (items 4 to 6), the rigour of development (items 7 to 14), clarity and presentation (items 15 to 17), applicability (items 18 to 21) and editorial independence (items 22 and 23). Each item scored between 1 (strongly disagree, i.e., when there was no information of the item) and 7 (strongly agree, i.e. when there was a well-constructed description). An arbitrator (MMC) solved disparities between the two analysts (AIA and CREL). The global reviewers´ scores were used to calculate the 0–100% domain quality scores following the AGREE II formula supplied in the tool manual [29]. The overall assessment items were also incorporated: a rating of the overall quality of the guidance and an assessment of whether it will be recommended for use in practice. The overall guideline assessment was gauged as the mean scores of the 6 standardised domains, and a recommendation was made: a CPG or CS was “recommended” when scored >80% [30], “recommended with modifications” if scored 50–80%, and “not recommended” if <49% [31]. RIGHT [28] investigated the reporting of the CPGs and CSs, and categorised it into twenty-two items (thirty-five subitems)that were scored as 1 (reported), 0.5 (partially reported), or 0 (unreported) and were categorised into 7 domains: basic information (items 1 to 4), background (items 5 to 9), evidence (items 10 to 12), recommendations (items 13 to 15), review and quality assurance (items 16 and 17), funding and declaration and management of interests (items 18 and 19), and other information (items 20 to 22). An overall reporting appraisal was counted based on the rate of the total (score >80%: “well-reported”, score = 50–80%: “moderate-reported” and score <50%: “low-reported”). Statistical analysis We conducted a descriptive analysis concerning particular items, domains, and overall scores, expressing the AGREE II and RIGHT scores as a percentage of the maximum possible score. The consistency between “reviewers” was estimated using the intraclass correlation coefficient (ICC), and it was considered excellent when ICC > 0.90 [32]. AGREE II and RIGHT correlation (“r”) was estimated to analyse if quality and reporting of the guidances were associated. The Kruskal–Wallis test was used to compare guidances outcomes (AGREE II and RIGHT scores). We used Stata 16 for analysis. Statistical significance was p < 0.05. Results Study selection A total of 8199 guidances were found from PubMed, EMBASE, Web of Science, Scopus, CDSR and Tripdatabase, and 30 documents from the grey literature (guideline specific databases, professional societies, and the Word Wide Web). After removing 439 duplicated guidances, 7752 were also rejected for not fulfilling the inclusion characteristics required (unsuited population or publication, outdated guidances substituted by an update or inappropriate development group). Thirty-eight of the records were filtered for reviewing titles and abstracts. Finally, 29 documents were included (24 CPGs [7, 33–56] and 5 CSs [57–61]) in quality and reporting full-text assessment. Nine documents were excluded for not accomplishing the criteria (4 conference abstracts, 3 posters, and 2 CPG for education and information purposes only). Figure 1 shows the flow diagram of the study. Table 1 shows the selected studies and their characteristics.Fig. 1 Flow chart of the systematic review. Explanation of the study selection screening. Table 1 Report of the set of guidances analysed in the systematic review (n = 29). Name of the CPG or protocol Abbreviated name Type of document Entity Country Year Publication in a Journal Version Evidence analysis Quality tool referral Type of Cancer Last updated date (months) 1 European guidelines from the EHTG and ESCP for Lynch syndrome: an updated third edition of the Mallorca guidelines based on gene and gender 2021 EHTG/ESCP Lynch syndrome CPG EHTG/ESCP Europe 2021 BJS 3 Systematic review, Delphi consensus Yes CRC 7 2 ACG Clinical Guidelines: Colorectal Cancer Screening 2021 2021 ACG CRC Screening CPG ACG USA 2021 Am J Gastroenteroly 3 Systematic review, GRADE No CRC 9 3 Cancer Screening in the Coronavirus Pandemic Era: Adjusting to a New Situation 2021 COVID Pandemic cancer screening CS ASCO USA 2021 JCO Global Oncology 1 Scoping review No CRC/ Breast/ Cervical 7 4 Screening for Colorectal Cancer. US Preventive Services Task Force Recommendation Statement 2021 USPSTF CRC screening CPG USPSTF USA 2021 JAMA 5 Systematic review No CRC 4 5 Colorectal Cancer Screening Guideline 2021 KFHPW CRC screening CPG KFHPW USA 2021 Not published 2 Not reported No CRC 3 6 NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines). Colorectal Cancer Screening 2021 NCCN CRC screening CPG NCCN USA 2021 Not published 2 Systematic review, consensus No CRC 8 7 Colorectal cancer screening. Clinical practice guideline. Nov 2013 (Revised 2020) 2020 CCAl CRC screening CPG CCAl Canada 2020 Not published 6 Not reported No CRC 23 8 American Society for Gastrointestinal Endoscopy guideline on the role of endoscopy in familial adenomatous polyposis syndromes 2020 ASGE adenomatous polyposis syndromes CPG ASGE USA 2020 Gastrointest Endosc 1 Systematic review, consensus No CRC 18 9 中国早期结直肠癌筛查流程专家共识意见 (2019, 上海) 2019 Chinese CRC screening CS CS SMMU China 2019 CGP 1 Not reported No CRC 25 10 Management of Familial Adenomatous Polyposis in Children and Adolescents: Position Paper From the ESPGHAN Polyposis Working Group 2019 ESPGHAN adenomatous polyposis syndromes CS ESPGHAN Europe 2019 JPGN 1 Systematic literature, GRADE No CRC 33 11 Guidelines for the management of hereditary colorectal cancer from the British Society of Gastroenterology (BSG)/Association of Coloproctology of Great Britain and Ireland (ACPGBI)/ United Kingdom Cancer Genetics Group (UKCGG) 2019 BSG/ACPGBI/UKCGG CRC screening CPG BSG/ ACPGBI/ UKCGG UK 2019 BMJ 1 Systematic review, GRADE, Delphi consensus No CRC 26 12 Colorectal cancer screening with faecal immunochemical testing, sigmoidoscopy or colonoscopy: a clinical practice guideline 2019 MAGIC CRC screening CPG MAGIC UK 2019 BMJ 1 Systematic review, GRADE No CRC 35 13 Colorectal cancer surveillance in inflammatory bowel disease: Practice guidelines and recent developments 2019 CRC in IBD CPG HMS USA 2019 WJG 2 Not reported No CRC 28 14 Colorectal Cancer Screening 2019 USMSTF CRC Screening CPG USMSTF USA 2019 JAMA 1 Systematic review, GRADE No CRC 31 15 Screening for Colorectal Cancer in Asymptomatic Average-Risk Adults: A Guidance Statement From the American College of Physicians 2019 ACP Average-risk CRC screening CPG ACP USA 2019 Ann Intern Med 1 Review Yes CRC 25 16 Clinical practice guideline. Diagnosis and prevention of colorectal cancer. 2018 Update 2019 ACP CRC screening CPG CPG ACP USA 2019 Ann Intern Med 2 Not reported No CRC 29 17 Colorectal cancer screening for patients with a family history of colorectal cancer or adenomas 2019 Ottawa CRC screenin CPG University of Ottawa Canada 2019 CFP 1 Systematic review ad met-analysis, GRADE No CRC 25 18 Recommendations on prevention and screening for colorectal cancer in Hong Kong 2018 Hong Kong CRC screening CPG CEWGCPS China 2018 Hong Kong Med J 3 Not reported No CRC 38 19 Diagnóstico y prevención del Cáncer Colorectal 2018 AEC CRC screening CPG AEG Spain 2018 Not published 2 Review, GRADE No CRC 48 20 Guía de práctica clínica de tamizaje del cáncer colorectal 2018 2018 Uruguay CRC screening CPG Ministerio de Salud Uruguay 2018 Not published 2 Systematic review Yes CRC 48 21 Colorectal Cancer Screening and Prevention 2018 CRC Screening and Prevention CPG AAFP USA 2018 Not published 1 Not reported No CRC 42 22 Cystic Fibrosis Colorectal Cancer Screening Consensus Recommendations 2018 Cystic Fibrosis CRC Screening CS AGA USA 2018 Gastroenterology 1 Systematic review, consensus No CRC 46 23 Colorectal Cancer Screening in Black Men: Recommendations for Best Practices 2018 CRC screening in Black Men CPG NIH USA 2018 Am J Prev Med 1 Review No CRC 48 24 ACR Appropriateness Criteria. Colorectal Cancer Screening 2018 ACR CRC Screening CPG ACR USA 2018 ACR 1 Systematic review, GRADE No CRC 48 25 Early Detection for Colorectal Cancer: ASCO Resource-Stratified Guideline 2018 Early Detection for CRC CPG ASCO USA 2018 JGO 1 Review, consensus Yes CRC 34 26 Colorectal Cancer Screening for Average-Risk Adults: 2018 Guideline Update From the American Cancer Society 2018 ACS Average-risk CRC screening CPG ACS USA 2018 CA Cancer J Clin 2 Systematic review, GRADE No CRC 31 27 Detección temprana, diagnóstico y clasificación por etapas 2018 ACS CRC screening CPG ACS USA 2018 Not published 3 Not reported No CRC 46 28 Clinical Practice Guideline on Screening for Colorectal Cancer in Individuals With a Family History of Non hereditary Colorectal Cancer or Adenoma: The Canadian Association of Gastroenterology Banff Consensus 2018 Banff consensus CS CAG Canada 2018 Gastroenterology 1 Systematic review, GRADE, consensus No CRC 37 29 Revised Australian national guidelines for colorectal cancer screening: family history 2018 Australian CRC screening CPG CCA Australia 2018 MJA 2 Systematic review, consensus No CRC 36 Characteristics of the studies Table 1 revealed the main characteristics of the chosen manuscripts, including the title, year, country, the supported entity for publication, version, evidence analysis, referral of a quality or reporting tool, type of cancer-focused and months passed after the last update was released. The majority of the guidelines were from North America (69%; 20). Five were from Europe (17%), two from Asia (7%) and one from South America and Oceania (3%) (see Appendix S3). The ICC was 0.85 for quality and 0.82 for reporting. Quality assessment The correlation score between AGREE II and RIGHT in the studies was r = 0.97 (Appendix S4). Quality was very heterogeneous, with a median overall rate of 69.0% (IQR 45.7–75.0; range 23.0%-88.0%). Figure 2 and Appendix S5 compiled the results. Almost 50% (13/29; 45%) of the guides were ranked as “recommended with modifications”, 38% (11/29) as “not recommended”, and only 17% (5/29) as “recommended”. Figure 3 illustrates the accomplishment regarding domains. Scope and purpose (domain 1) and clarity of presentation (domain 4) obtained the best quality with 62% (18/29) of the guidances with high quality (scoring >75%), respectively. Average scores (scoring 50–75%) were obtained in stakeholder involvement (domain 2) with 34% (10/29), the rigour of development (domain 3) with also 34% (10/29), and editorial independence (domain 6) with 28% (8/29). Utterly, only domain 5 (applicability) achieved “low” (25–50%) or “very low” (<25%) with 83% (24/29) in these categories. The guidances with more satisfactory quality were five (in order of high to low quality): the ASCO [56], the Spanish [46], the Banff consensus [58], the ACS [54] and the MAGIC [39] CRC guidelines (Appendix S6).Fig. 2 Quality overall score in colorectal screening guidances. Results after using AGREE II statement in each guidance document. Fig. 3 AGREE and RIGHT domains of the selected guidances. Analysis of every guidance document using quality and reporting instruments. Reporting assessment The median overall reporting was 42.0% (IQR 31.4–68.6; range 8.0%–86.0%). Twelve guidances (41.4%) were “recommended with modifications” (scoring >50–80%) while 48.3% (14/29) were “not recommended” (scoring <51%). Only 10.3% (3/29) were “recommended” (scoring >79%). Figure 3 demonstrated the reporting of each domain in the guidances. Basic information (domain 1) was well-reported in 19/29 (66%) of the guidances. Background (domain 2) and review and quality assurance (domain 5) were moderate-reported with 59% (17/29) and 69% (20/29), respectively. The reporting of recommendations (domain 4), funding and declaration and management of interests (domain 6) and other information (domain 7) was scarce with 62% (18/29), 83% (24/29) and 69% (20/29), respectively. The domain median for reporting was 83% (0–100%) in domain 1 (basic information), 63% (0-100%) in domain 2 (background), 40% (0–100%) in domain 3 (evidence), 43% (0–100%) in domain 4 (recommendations), 0% (0–100%) in domain 5 (review and quality assurance), 25% (0–100%) in domain 6 (funding and declaration and management of interests) and finally, 33% (0–100%) in domain 7 (other information). The better reported guidances were the ASCO [56], the Banff consensus [58] and the ACS [54] guidances. Figure 4 and Appendixes S7 and S8 collect this information.Fig. 4 Reporting overall score in colorectal screening guidances. Results after using the RIGHT instrument in each guidance. Focus of the guidances Regarding the focus of the guidance, 18/29 (62.1%) were CPGs and CSs about general CRC screening, 2/29 (6.9%) were focused on average-risk CRC, and 1/29 (3.5%) was about inflammatory bowel disease, and another (3.5%) focused on black men population. Finally, 7/29 (24.1%) were about different sorts of hereditary CRC screening (2/29 (6.9%) for Adenomatous syndrome, 1/29 (3.5%) for Lynch syndrome, and another (3.5%) about CRC related to cyst fibrosis, and finally, 3/29 (10.4%) about general hereditary cancer). Concerning quality, CRC screening guidances focused on hereditary cancer, and the average-risk population had a better score in all the domains than guidances about general colorectal screening. Scope and purpose domain was 94% in hereditary CRC screening guidances while 77% in general guidances, stakeholder involvement was 75% vs. 55%, the rigour of development was 70% vs. 40%, clarity of presentation 87% vs. 80%, and editorial independence 62% vs. 45%. Only the applicability domain reached a 30% overall score in both general and hereditary guidances. Appendix S9 and S10 show the differences in quality domains depending on the type of guidance (general CRC screening, average-risk CRC screening, hereditary CRC screening, inflammatory bowel disease and specific subpopulations CRC screening guidances). Concerning Reporting, CPGs and CSs related to hereditary CRC had a better reporting than general guidances in domains 1 to 5 (basic information 88% vs. 72%, background 77% vs. 54%, evidence 80% vs. 42%, recommendations 57% vs. 44%, and review and assurance 42% vs. 25%). But worse in domain 6 funding and conflict of interest (21% vs. 31%) and domain 7 other information (38% vs. 46%). Appendix S11 and Appendix S12 show the reporting depending on the type of guidance. Factors associated with quality and reporting The guidances underpinned by systematic reviews obtained better quality (68.5% vs. 35.5%; p = 0.001) and reporting than consensus (41.4% vs. 74.6%; p = 0.001). No significant differences were found between CSs and CPGs (AGREE II: p = 0.729; RIGHT: p = 0.954). The origin (AGREE II: p = 0.181; RIGHT: p = 0.162)., the publication in a journal (AGREE II: p = 0.093; RIGHT: p = 0.063)., the year of publication (AGREE II: p = 0.751; RIGHT: p = 0.852)., the version of the guidance (AGREE II: p = 0.427; RIGHT: p = 0.394), the type of cancer (AGREE II: p = 0.114; RIGHT: p = 0.077) or the referral of a quality tool such as AGREE II or RIGHT (AGREE II: p = 0.189; RIGHT: p = 0.189) did not influence quality or reporting. Quality and reporting of the guideline documents stratified by different characteristics were collected in Table 2.Table 2 Elements associated with the quality and reporting of the guidelines. AGREE II RIGHT Variable Median IQR range p-value Median IQR range p-value Type of document CPGs 60.3% 40.0–71.9 p = 0.729 63.2% 47.5–79.6 p = 0.954 CSs 70.6% 31.2–76.1 66.4% 37.1–82.1 Country Europe 71.4% 70.6–74.3 81.4% 76.4–82.9 North America 56.2% 39.1–65.4 p = 0.181 61.4% 45.7–68.6 p = 0.162 Other countries 44.9% 31.2–72.5 55.7% 37.1–77.9 Publication year 2018 62.3% 40.0–80.0 p = 0.751 67.5% 47.1–83.9 p = 0.852 2019 62.3% 41.7–70.7 64.3% 49.3–76.4 2020 48.2% 21.4–75.0 55.0% 22.9–87.1 2021 60.1% 29.0–65.9 62.9% 32.9–68.6 Publication in a journal Yes 63.8% 41.7–74.3 p = 0.093 65.3% 50.7–82.1 p = 0.063 No 35.5% 21.4–66.3 41.4% 22.9–72.9 Version number 1 63.4% 39.5–75.0 p = 0.427 66.4% 50.7–82.1 p = 0.394 2 50.5% 40.4–71.7 58.6% 47.5–77.8 Other 52.4% 21.4–65.9 53.9% 22.9–68.6 Evidence analysis Not reported 35.5% 26.8–39.5 p = 0.001 41.4% 32.1–45.7 p = 0.001 Systematic review 68.5% 60.3–75.5 74.6% 63.2–82.9 Quality tool referral Reported 68.8% 64.9–78.8 p = 0.114 77.8% 68.6–88.2 p = 0.077 Not reported 56.2% 39.1–72.5 61.4% 43.6–77.9 Type of cancer CRC 62.9% 40.0–73.7 p = 0.189 64.3% 47.5–81.8 p = 0.189 Mixed 29.0% 28.9–29.1 32.9% 32.8–33.0 Discussion Main findings This extensive systematic review of CRC screening guidance documents demonstrated a wide variety in quality and reporting. We studied guidances from different countries (5 continents and 8 countries) and languages, which provided an international viewpoint of the present position of screening guidelines for CRC. Analysing quality by AGREE II, almost half of the guides had a moderate quality and needed improvement, and more than a third were classified as not recommended. Concerning reporting examined by RIGHT, most of the guidances had a well-detailed scope and purpose and good clarity of presentation, although applicability was poorly explained. The domains stakeholder involvement, rigour of development and editorial independence were average. More than a third of the guidances were moderate-reported (RIGHT score 50–80%), and almost a third were low-reported (RIGHT score <50%). Basic information was well-reported (RIGHT score >80%); background and review and quality assurance were moderate-reported (RIGHT score 50–80%); the funding reporting, the conflict of interest and other information were low-reported (RIGHT score <50%). The use of systematic reviews was associated with improving quality and reporting of the guidances. No other factors such as the type of guidance (CPGs vs. CSs), the origin, the year of release, the version or the publication in a journal showed a relationship with quality or reporting. Strengths and limitations Our study was a broad systematic review focused on CRC screening with no specific languages and no data source limitations to offer an international perspective of the current situation of screening guidelines for CRC. English and Spanish were the most internationally spoken languages [62], and most organisations offered versions in both languages. Our reviewers were native speakers of both English and Spanish. The diversity of the guidances reviewed is an example of the existing heterogeneity of the publications, and it could be unavoidable as guidances varied in their configuration, background, development, objectives, outputs, regional/local epidemiological situation, etc. [63]. The aim of our systematic review was to analyse quality and reporting of CRC screening guidances in general. The external validity of our systematic review, i.e., the extent to which the study’s findings can be generalised, was not affected by the individual validity of the guidances analysed, and our findings could be reproduced. Our systematic review included CPGs and CSs about CRC screening, although protocols were excluded as they did not accomplish the selection criteria. We must emphasise that some of these countries do have protocols for CRC screening, but they do not provide guidance or recommendations about CRC screening. For a better understanding of the quality and reporting analysis of the guidances, we decided to classify the CPGs and CSs by their main purpose (general CRC screening, average-risk CRC, inflammatory bowel disease, specific populations, and hereditary cancer), giving the reader a better perspective of the current situation in every type of guidances. We studied articles published from 2018 onwards. So, we are aware that CPGs and CSs outside our period of time scope from reputable institutions would have been excluded. Our decision to select a 3 years frame was not arbitrary but evidence-based. An extensive systematic review of literature remarked that most guidance methodological handbooks for updating CPGs recommended a two or 3-year window between updates. We are aware that the update of guidances depends on new improvements available. However, regarding evidence [21], the need for a more extensive analysis would be unnecessary since older guidelines would possibly be now obsolete due to quick advances in CRC and anal cancer. Although the subjective character of the data extraction could introduce bias, CPGs and CSs were assessed by at least two reviewers and an arbitrator in case of disagreements, as AGREE II and RIGHT have recommended in their user manuals [22, 23], increasing the trustworthiness of the data reported. Before using the tools, the reviewers had sessions to learn and unify standards about the process of using AGREE II and RIGHT. The reviewer´s concordance was excellent (ICC > 90%). The reviewers were experienced in systematic reviews, quality health care management [24, 64–66], the analysis of guidances and the use of AGREE and RIGHT [22, 23]. They were also experts in the study of CRC and screening (experienced CRC surgeons or specialists related), so they had the relevant vocabulary to understand the documents included properly. The two validated appraisal instruments used, AGREE II [27] and RIGHT [28], did not guide thresholds or weighting for scoring items and domains. Their instructions suggested avoiding calculating an overall rate for the guidances as it could hide weaknesses in individual domains. We used previously published cut-offs [23, 30, 31] as this approach helps to simplify the analyses. Like other tools, AGREE II and RIGHT have intrinsic boundaries as they do not estimate the strength of recommendations or patient values and choices. We are aware that the interpretation of the results must be handled with caution because, although the guidelines may have similar overall scores, they may differ individually in each domain. This is so because all the domains had the same weight. Implications Guidance documents should supply specific evidence-based advice in high-quality care management. The quality of guidelines is an essential condition in its development [67]. However, the attainment of this requirement does not necessarily convey into implementation, and strict compliance with guidance recommendations (even of the more outstanding quality) does not automatically deliver the most proper care per patient [6]. Nowadays, there is a multiplicity of recommendations in CRC screening guidelines [5]. Screening programmes should be adjusted to risk groups to deliver techniques individualised to their risk of acquiring CRC [68]. Clinicians should inspect the patient’s general health, earlier screening history, and choices and values to offer if screening is appropriate [7]. These diverse subgroups with specific necessities would explain the vast heterogeneity of CRC screening guidances recommendations as they differ in aims and implicated subgroups. High-quality guidance documents are crucial for adequately managing patients. Our systematic review highlighted that quality and reporting of the CRC screening guidance documents had a vast scope for improvements. The debate about weighting and cut-offs of items and domains should be also investigated in the future. Quality was exceptionally poor in the applicability (the description of facilitators and barriers for application, the resources provided for application and the monitoring and auditing criteria) domain, which would merit urgent consideration. The stakeholder involvement, the rigour of development (particularly the external review of the document and an updating procedure) and the editorial independence of the analysed guidances should also enhance their quality (Appendix S13). The formulation of the recommendations was not well-described, and the methodology was not clarified in the majority of the guidances. Primary users of the guideline or the population subgroups were not appropriately reported, and the selection of the guidelines contributors and their roles were not specified. The values and preferences of the target population were not considered in the formulation of each recommendation. CPGs and CSs also did not describe any limitations in their development process nor indicated how any limitations might have affected the validity of the proposals. Guidances did not register any gaps in the evidence or provide future research suggestions. The funding and conflict of interest reporting were very low-reported (see Appendix S14). Guidances that followed systematic review for evidence analysis had obtained better quality and reporting. This finding supported the idea that Systematic reviews are considered the gold standard for evidence-based research [69]. Although CPGs are normally better than CSs [70] in the literature, differences between CPGs and CSs quality and reporting were not significant in our systematic review. This is probably due to the fact that the terms CPGs and CSs are often used interchangeably. Comparing previous systematic reviews about CRC screening guidances, our results highlighted worse quality in all the areas. Only stakeholder involvement has remained similar in recent guidances to 10 years ago. This could be produced by a selection bias. Former studies had probably selected well-known guidances while our study was more recent and no language restricted; hence, we have analysed a third more guidances than these other studies. Appendix S15 shows the characteristics of the studies and a comparison of domains. Comparing CRC and breast cancer screening CPGs and CSs (prior publication by our team) [22], CRC guidances had better quality but worse reporting. The applicability was worst in CRC guidelines, but both types of cancer should improve. The scope and purpose, the stakeholder involvement, the rigour of development, the clarity of presentation, and the editorial independence enclosed better quality in CRC guidances. The reporting was more varied. Although basic information, funding, declaration, and management of interests were better documented in CRC guidances, the evidence, the reporting of recommendations, and the review and quality assurance had more valuable reporting on breast cancer CPGs and CSs. Conclusions CRC screening guidances had a heterogeneous quality and reporting. Half of the analysed CPGs and CSs had an average quality but low reporting that would merit urgent improvement in all their areas. In the future, the development of guidelines should involve a robust process using appropriate guideline development tools at the start of the process to ensure the production of high-quality guidance based on the best available evidence. Supplementary information Legends of the supplementary material Appendix 0 Appendix S1 Appendix S2 Appendix S3 Appendix S4 Appendix S5 Appendix S6 Appendix S7 Appendix S8 Appendix S9 Appendix S10 Appendix S11 Appendix S12 Appendix S13 Appendix S14 Appendix S15 Supplementary information The online version contains supplementary material available at 10.1038/s41416-022-02070-4. Acknowledgements KSK is a Distinguished Investigator funded by the Beatriz Galindo (senior modality) Programme grant given to the University of Granada by the Ministry of Science, Innovation, and Universities of the Spanish Government. Author contributions MMC conceived the work. MMC, AIA, CREL and CMFV compiled and analysed the data for the systematic review. MMC and ABC interpreted the data. MMC wrote the first version of the draft. KSK, ABC, MMD and MGG edited the work critically for important academic content. ABC and KSJ directed the work. All authors consented to the final version of the manuscript. They agreed to be responsible for all elements of the review, providing those questions related to the accuracy or integrity of any part of the work were appropriately investigated and solved. Funding This systematic review was not funded. Data availability The data supporting the results are available from the corresponding author on reasonable request. Materials availability The materials supporting the results are available from the corresponding author on reasonable request. Competing interests The authors declare no competing interests. Ethics approval and 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. Sung H Ferlay J Siegel RL Laversanne M Soerjomataram I Jemal A Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries CA Cancer J Clin 2021 71 209 49 10.3322/caac.21660 33538338 2. Cotton S Sharp L Little J The adenoma-carcinoma sequence and prospects for the prevention of colorectal neoplasia Crit Rev Oncog 1996 7 293 342 10.1615/CritRevOncog.v7.i5-6.10 9467664 3. Zauber AG Winawer SJ O’Brien MJ Lansdorp-Vogelaar I van Ballegooijen M Hankey BF Colonoscopic polypectomy and long-term prevention of colorectal-cancer deaths N Engl J Med 2012 366 687 96 10.1056/NEJMoa1100370 22356322 4. Cancer.Net. Colorectal cancer: statistics. https://www.cancer.net/cancer-types/colorectal-cancer/statistics 2022. 5. Chetroiu D Pop CS Filip PV Beuran M How and why do we screen for colorectal cancer? J Med Life 2021 14 462 7 10.25122/jml-2021-0192 34621368 6. Navarro M Nicolas A Ferrandez A Lanas A Colorectal cancer population screening programs worldwide in 2016: an update World J Gastroenterol 2017 23 3632 42 10.3748/wjg.v23.i20.3632 28611516 7. Force USPSTDavidson KW Barry MJ Mangione CM Cabana M Caughey AB Screening for colorectal cancer: US preventive services task force recommendation statement JAMA 2021 325 1965 77 10.1001/jama.2021.6238 34003218 8. Pavlidis N Towards a convenient way to practice medical oncology Ann Oncol 2007 18 ii3 4 10.1093/annonc/mdm014 17491034 9. Fervers B Philip T Haugh M Cluzeau F Browman G Clinical-practice guidelines in Europe: time for European co-operation for cancer guidelines Lancet Oncol 2003 4 139 40 10.1016/S1470-2045(03)01017-9 12623358 10. Institute of Medicine (US) Committee to Advise the Public Health Service on Clinical Practice Guidelines. In: Field MJ, Lohr KN, editors. Clinical practice guidelines: directions for a new program. Washington (DC): National Academies Press (US); 1990. 11. Browman GP Snider A Ellis P Negotiating for change. The healthcare manager as catalyst for evidence-based practice: changing the healthcare environment and sharing experience Health Pap 2003 3 10 22 10.12927/hcpap..17125 12. Grimshaw JM Russell IT Effect of clinical guidelines on medical practice: a systematic review of rigorous evaluations Lancet 1993 342 1317 22 10.1016/0140-6736(93)92244-N 7901634 13. Woolf SH Grol R Hutchinson A Eccles M Grimshaw J Clinical guidelines: potential benefits, limitations, and harms of clinical guidelines BMJ 1999 318 527 30 10.1136/bmj.318.7182.527 10024268 14. Booth A Searching for qualitative research for inclusion in systematic reviews: a structured methodological review Syst Rev 2016 5 74 10.1186/s13643-016-0249-x 27145932 15. Wouters MW Jansen-Landheer ML van de Velde CJ The quality of cancer care initiative in the Netherlands Eur J Surg Oncol 2010 36 S3 S13 10.1016/j.ejso.2010.06.004 20576399 16. Simone B De Feo E Nicolotti N Ricciardi W Boccia S Quality evaluation of guidelines on genetic screening, surveillance and management of hereditary colorectal cancer Eur J Public Health 2012 22 914 20 10.1093/eurpub/ckr166 22140249 17. Tian JB Wen Y Yang ZY Zheng YD Wu Z Li J [Quality assessment of global colorectal cancer screening guidelines and consensus] Zhonghua Liu Xing Bing Xue Za Zhi 2021 42 248 57 33626611 18. Moher D Liberati A Tetzlaff J Altman DG Group P Preferred reporting items for systematic reviews and meta-analyses: the PRISMA Statement Open Med 2009 3 e123 30 21603045 19. Liberati A Altman DG Tetzlaff J Mulrow C Gotzsche PC Ioannidis JP 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 20. Page MJ McKenzie JE Bossuyt PM Boutron I Hoffmann TC Mulrow CD The PRISMA 2020 statement: an updated guideline for reporting systematic reviews BMJ 2021 372 n71 10.1136/bmj.n71 33782057 21. Vernooij RW, Sanabria AJ, Sola I. Guidance for updating clinical practice guidelines: a systematic review of methodological handbooks. Implement Sci. 2014;9:3. 22. Maes-Carballo M, Mignini L, Martin-Diaz M, Bueno-Cavanillas A, Khan KS. Clinical practice guidelines and consensus for the screening of breast cancer: a systematic appraisal of their quality and reporting. Eur J Cancer Care (Engl). 2021;31:e13540. 23. Maes-Carballo M Mignini L Martin-Diaz M Bueno-Cavanillas A Khan KS Quality and reporting of clinical guidelines for breast cancer treatment: A systematic review Breast 2020 53 201 11 10.1016/j.breast.2020.07.011 32858405 24. Maes-Carballo M Moreno-Asencio T Martin-Diaz M Mignini L Bueno-Cavanillas A Khan KS Shared decision making in breast cancer screening guidelines: a systematic review of their quality and reporting Eur J Public Health 2021 31 873 83 10.1093/eurpub/ckab084 34148093 25. Maes-Carballo M, Munoz-Nunez I, Martin-Diaz M, Mignini L, Bueno-Cavanillas A, Khan KS. Shared decision making in breast cancer treatment guidelines: development of a quality assessment tool and a systematic review. Health Expect. 2020;23:1045–64. 26. Hewitt-Taylor J Melling S Care protocols: rigid rules or useful tools? Paediatr Nurs 2004 16 38 42 10.7748/paed.16.4.38.s27 15160620 27. Brouwers MC Kerkvliet K Spithoff K Consortium ANS The AGREE Reporting Checklist: a tool to improve reporting of clinical practice guidelines BMJ 2016 352 i1152 10.1136/bmj.i1152 26957104 28. Chen Y Yang K Marusic A Qaseem A Meerpohl JJ Flottorp S A reporting tool for practice guidelines in health care: the RIGHT statement Ann Intern Med 2017 166 128 32 10.7326/M16-1565 27893062 29. Brouwers MC Kho ME Browman GP Burgers JS Cluzeau F Feder G AGREE II: advancing guideline development, reporting, and evaluation in health care Prev Med 2010 51 421 4 10.1016/j.ypmed.2010.08.005 20728466 30. Oh MK Jo H Lee YK Improving the reliability of clinical practice guideline appraisals: effects of the Korean AGREE II scoring guide J Korean Med Sci 2014 29 771 5 10.3346/jkms.2014.29.6.771 24932076 31. Hoffmann-Esser W Siering U Neugebauer EAM Lampert U Eikermann M Systematic review of current guideline appraisals performed with the Appraisal of Guidelines for Research & Evaluation II instrument-a third of AGREE II users apply a cut-off for guideline quality J Clin Epidemiol 2018 95 120 7 10.1016/j.jclinepi.2017.12.009 29288133 32. Koo TK Li MY A guideline of selecting and reporting intraclass correlation coefficients for reliability research J Chiropr Med 2016 15 155 63 10.1016/j.jcm.2016.02.012 27330520 33. Alberta CC. Cancer Care Alberta. Colorectal cancer screening. Clinical Practice Guideline \| Nov 2013 (Revised 2020). https://actt.albertadoctors.org/CPGs/Lists/CPGDocumentList/colorectal-cancer-screening-guideline.pdf. 2020. 34. Alberta CC. Colorectal cancer screening. Clinical Practice Guideline \| Nov 2013 (Revised 2020). 2020. 35. Clarke WT Feuerstein JD Colorectal cancer surveillance in inflammatory bowel disease: Practice guidelines and recent developments World J Gastroenterol 2019 25 4148 57 10.3748/wjg.v25.i30.4148 31435169 36. Cubiella J Marzo-Castillejo M Mascort-Roca JJ Amador-Romero FJ Bellas-Beceiro B Clofent-Vilaplana J Clinical practice guideline. Diagnosis and prevention of colorectal cancer. 2018 Update Gastroenterol Hepatol 2018 41 585 96 10.1016/j.gastrohep.2018.07.012 30245076 37. Expert Panel on Gastrointestinal IMoreno C Kim DH Bartel TB Cash BD Chang KJ ACR appropriateness criteria((R)) colorectal cancer screening J Am Coll Radio 2018 15 S56 S68 10.1016/j.jacr.2018.03.014 38. Gupta N Kupfer SS Davis AM Colorectal cancer screening JAMA 2019 321 2022 3 10.1001/jama.2019.4842 31021387 39. Helsingen LM Vandvik PO Jodal HC Agoritsas T Lytvyn L Anderson JC Colorectal cancer screening with faecal immunochemical testing, sigmoidoscopy or colonoscopy: a clinical practice guideline BMJ 2019 367 l5515 10.1136/bmj.l5515 31578196 40. Jenkins MA Ait Ouakrim D Boussioutas A Hopper JL Ee HC Emery JD Revised Australian national guidelines for colorectal cancer screening: family history Med J Aust 2018 209 455 60 10.5694/mja18.00142 30359558 41. Kwaan MR Jones-Webb R Colorectal cancer screening in black men: recommendations for best practices Am J Prev Med 2018 55 S95 S102 10.1016/j.amepre.2018.05.008 30670207 42. Lam TH Wong KH Chan KK Chan MC Chao DV Cheung AN Recommendations on prevention and screening for colorectal cancer in Hong Kong Hong Kong Med J 2018 24 521 6 30318477 43. Monahan KJ Bradshaw N Dolwani S Desouza B Dunlop MG East JE Guidelines for the management of hereditary colorectal cancer from the British Society of Gastroenterology (BSG)/Association of Coloproctology of Great Britain and Ireland (ACPGBI)/United Kingdom Cancer Genetics Group (UKCGG) Gut 2020 69 411 44 10.1136/gutjnl-2019-319915 31780574 44. Network NCC. NCCN guidelines: colorectal cancer screening Version 2.2021. https://www.nccn.org/. 2021. 45. Qaseem A Crandall CJ Mustafa RA Hicks LA Wilt TJ Clinical Guidelines Committee of the American College of Physicians, Screening for colorectal cancer in asymptomatic average-risk adults: a guidance statement from the american college of physicians Ann Intern Med 2019 171 643 54 10.7326/M19-0642 31683290 46. SemFYC AEdG. Diagnóstico y prevención del cáncer colorrectal. ACTUALIZACIÓN 2018. 2018. 47. Seppala TT Latchford A Negoi I Sampaio Soares A Jimenez-Rodriguez R Sanchez-Guillen L European guidelines from the EHTG and ESCP for Lynch syndrome: an updated third edition of the Mallorca guidelines based on gene and gender Br J Surg 2021 108 484 98 10.1002/bjs.11902 34043773 48. Shaukat A Kahi CJ Burke CA Rabeneck L Sauer BG Rex DK ACG Clinical guidelines: colorectal cancer screening 2021 Am J Gastroenterol 2021 116 458 79 10.14309/ajg.0000000000001122 33657038 49. America Cancer Society. Detección temprana, diagnóstico y clasificación por etapas. https://www.cancer.org/es/cancer/cancer-de-pulmon/deteccion-diagnostico-clasificacion-por-etapas.html. 2018. 50. Uruguay MdSd. Ministerio de Salud de Uruguay. Guía de práctica clínica de tamizaje del cáncer colo-rectal 2018. https://www.paho.org/uru/dmdocuments/Guia%20de%20practica%20clinica%20de%20tamizaje%20del%20cancer%20colo-rectal%202018.pdf. 2018. 51. Washington KFHPo. Kaiser Foundation Health Plan of Washington. Colorectal Cancer Screening Guideline. https://wa.kaiserpermanente.org/static/pdf/public/guidelines/colon.pdf. 2021. 52. Wilkins T McMechan D Talukder A Colorectal cancer screening and prevention Am Fam Physician 2018 97 658 65 29763272 53. Wilkinson AN Lieberman D Leontiadis GI Tse F Barkun AN Abou-Setta A Colorectal cancer screening for patients with a family history of colorectal cancer or adenomas Can Fam Physician 2019 65 784 9 31722908 54. Wolf AMD Fontham ETH Church TR Flowers CR Guerra CE LaMonte SJ Colorectal cancer screening for average-risk adults: 2018 guideline update from the American Cancer Society CA Cancer J Clin 2018 68 250 81 10.3322/caac.21457 29846947 55. Yang J Gurudu SR Koptiuch C Agrawal D Buxbaum JL Abbas Fehmi SM American Society for Gastrointestinal Endoscopy guideline on the role of endoscopy in familial adenomatous polyposis syndromes Gastrointest Endosc 2020 91 963 82 10.1016/j.gie.2020.01.028 32169282 56. Lopes G Stern MC Temin S Sharara AI Cervantes A Costas-Chavarri A Early detection for colorectal cancer: ASCO resource-stratified guideline J Glob Oncol 2019 5 1 22 57. Basu P Alhomoud S Taghavi K Carvalho AL Lucas E Baussano I Cancer Screening in the coronavirus pandemic era: adjusting to a new situation JCO Glob Oncol 2021 7 416 24 10.1200/GO.21.00033 33784177 58. Leddin D Lieberman DA Tse F Barkun AN Abou-Setta AM Marshall JK Clinical practice guideline on screening for colorectal cancer in individuals with a family history of nonhereditary colorectal cancer or adenoma: The Canadian Association of Gastroenterology Banff Consensus Gastroenterology 2018 155 1325 47 10.1053/j.gastro.2018.08.017 30121253 59. Hadjiliadis D Khoruts A Zauber AG Hempstead SE Maisonneuve P Lowenfels AB Cystic fibrosis colorectal cancer screening consensus recommendations Gastroenterology 2018 154 736 45 10.1053/j.gastro.2017.12.012 29289528 60. Hyer W Cohen S Attard T Vila-Miravet V Pienar C Auth M Management of familial adenomatous polyposis in children and adolescents: position paper from the ESPGHAN polyposis working group J Pediatr Gastroenterol Nutr 2019 68 428 41 10.1097/MPG.0000000000002247 30585891 61. 中国抗癌协会肿瘤内镜学专业委员会国上国中医中中消中中中金国. 中国早期结直肠癌筛查流程专家共识意见 (2019, 上海) . Natl Med J China. 2019;99. 62. Amano T Gonzalez-Varo JP Sutherland WJ Languages are still a major barrier to global science PLoS Biol 2016 14 e2000933 10.1371/journal.pbio.2000933 28033326 63. Pentheroudakis G Stahel R Hansen H Pavlidis N Heterogeneity in cancer guidelines: should we eradicate or tolerate? Ann Oncol 2008 19 2067 78 10.1093/annonc/mdn418 18662954 64. Maes-Carballo M Gomez-Fandino Y Reinoso-Hermida A Estrada-Lopez CR Martin-Diaz M Khan KS Quality indicators for breast cancer care: a systematic review Breast 2021 59 221 31 10.1016/j.breast.2021.06.013 34298301 65. Maes-Carballo M Munoz-Nunez I Martin-Diaz M Mignini L Bueno-Cavanillas A Khan KS Shared decision making in breast cancer treatment guidelines: development of a quality assessment tool and a systematic review Health Expect 2020 23 1045 64 10.1111/hex.13112 32748514 66. Maes-Carballo M, Gomez-Fandino Y, Estrada-Lopez CR, Reinoso-Hermida A, Khan KS, Martin-Diaz M, et al. Breast cancer care quality indicators in Spain: a systematic review. Int J Environ Res Public Health. 2021;18:6411. 67. Lugtenberg M Burgers JS Westert GP Effects of evidence-based clinical practice guidelines on quality of care: a systematic review Qual Saf Health Care 2009 18 385 92 10.1136/qshc.2008.028043 19812102 68. Hoffmann S Crispin A Lindoerfer D Sroczynski G Siebert U Mansmann U Evaluating the effects of a risk-adapted screening program for familial colorectal cancer in individuals between 25 and 50 years of age: study protocol for the prospective population-based intervention study FARKOR BMC Gastroenterol 2020 20 131 10.1186/s12876-020-01247-6 32370777 69. Pussegoda K Turner L Garritty C Mayhew A Skidmore B Stevens A Systematic review adherence to methodological or reporting quality Syst Rev 2017 6 131 10.1186/s13643-017-0527-2 28720117 70. Jacobs C Graham ID Makarski J Chasse M Fergusson D Hutton B Clinical practice guidelines and consensus statements in oncology-an assessment of their methodological quality PLoS ONE 2014 9 e110469 10.1371/journal.pone.0110469 25329669	36476659	PMC9734419	NO-CC CODE	2022-12-14 23:28:28	no	Br J Cancer. 2022 Dec 7;:1-12	utf-8	Br J Cancer	2,022	10.1038/s41416-022-02070-4	oa_other
==== Front Stoch Environ Res Risk Assess Stoch Environ Res Risk Assess Stochastic Environmental Research and Risk Assessment 1436-3240 1436-3259 Springer Berlin Heidelberg Berlin/Heidelberg 2345 10.1007/s00477-022-02345-5 Original Paper Daily precipitation performances of regression-based statistical downscaling models in a basin with mountain and semi-arid climates http://orcid.org/0000-0001-7006-8340 Şan Murat [email protected] 1 http://orcid.org/0000-0003-2497-5032 Nacar Sinan 2 http://orcid.org/0000-0003-0897-4742 Kankal Murat 3 http://orcid.org/0000-0003-4359-9183 Bayram Adem 4 1 grid.448936.4 0000 0004 0369 6808 Civil Engineering Department, Gümüşhane University, 29100 Gümüşhane, Turkey 2 grid.411550.4 0000 0001 0689 906X Civil Engineering Department, Tokat Gaziosmanpaşa University, 60150 Tokat, Turkey 3 grid.34538.39 0000 0001 2182 4517 Civil Engineering Department, Bursa Uludağ University, 16059 Bursa, Turkey 4 grid.31564.35 0000 0001 2186 0630 Civil Engineering Department, Karadeniz Technical University, 61080 Trabzon, Turkey 4 12 2022 125 9 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. The impacts of climate change on current and future water resources are important to study local scale. This study aims to investigate the prediction performances of daily precipitation using five regression-based statistical downscaling models (RBSDMs), for the first time, and the ERA-5 reanalysis dataset in the Susurluk Basin with mountain and semi-arid climates for 1979–2018. In addition, comparisons were also performed with an artificial neural network (ANN). Before achieving the aim, the effects of atmospheric variables, grid resolution, and long-distance grid on precipitation prediction were holistically investigated for the first time. Kling-Gupta efficiency was modified and used for holistic evaluation of statistical moments parameters at precipitation prediction comparison. The standard triangular diagram, quite new in the literature, was also modified and used for graphical evaluation. The results of the study revealed that near grids were more effective on precipitation than single or far grids, and 1.50° × 1.50° resolution showed similar performance to 0.25° × 0.25° resolution. When the polynomial multivariate adaptive regression splines model, which performed slightly higher than ANN, tended to capture skewness and standard deviation values of precipitations and to hit wet/dry occurrence than the other models, all models were quite well able to predict the mean value of precipitations. Therefore, RBSDMs can be used in different basins instead of black-box models. RBSDMs can also be established for mean precipitation values without dry/wet classification in the basin. A certain success was observed in the models; however, it was justified that bias correction was required to capture extreme values in the basin. Supplementary Information The online version contains supplementary material available at 10.1007/s00477-022-02345-5. Keywords Grid selection MARS PolyMARS Predictor selection Standard triangular diagram Statistical downscaling ==== Body pmcIntroduction The world is warming to dangerous levels due to the increase in the concentration of carbon dioxide and other greenhouse gases (Salman et al. 2018). Despite the decrease in global carbon dioxide emissions during the world lockdown after the onset of the COVID-19 pandemic in March 2019, this decreased amount was compensated in the last half of 2020 (Tollefson 2021). This situation indicates that warming will continue without significant reduction. Liu and Raftery (2021) have shown that if our current trend continues, the target of the Paris agreement to stay below the increase of 2 °C in the warming has a 5% probability according to the pre-industrial level. While if all countries meet the specified conditions in the agreement, this probability increases to the level of 26%. In other words, reducing the gases released into the atmosphere within the framework of the determined commitments will not significantly reduce the impact of climate change. Climate change, which is the change in the mean and variability of the climate over more than decades, is one of the most severe threats to the environment and humanity in this century. It is generally accepted as the reason for the increase in frequency, density, and duration of extreme values such as floods, droughts, forest fires, and heatwaves in various parts of the world (Ahmed et al. 2018; Shiru et al. 2019). And therefore, climate change will impact local or global precipitation and hydrological regime (IPCC 2013; Rashid et al., 2015). The Mediterranean region, including Turkey, is among the areas that will be most affected by climate change (Diffenbaugh and Giorgi 2012; Keupp et al. 2019; IPCC 2021). It is necessary to develop an effective policy for analyzing and understanding the current and possible future climate changes and adaptation to climate change (Noor et al. 2020). The global climate models (GCMs) allow the possibility to study changes using physically-based equations to simulate the effect of greenhouse gas concentrations on the atmosphere and various ocean processes in means and variations (IPCC 2013, 2021). So, the GCMs are essential tools for observing large-scale climate features and climate change under possible future scenarios. However, the GCMs are insufficient due to their coarse resolutions. The climate and hydrology components that directly or indirectly provide water resources management at the local scale require high resolution (Nasseri et al. 2013; Tavakol-Davani et al. 2013; Rudd and Kay 2016). Downscaling methods are used to overcome this obstacle, and act as a bridge between the GCMs and local climate-hydrology components. Downscaling methods are generally divided into statistical and dynamic downscaling. However, as dynamic downscaling methods require high computing power and design and some specialized knowledge, the statistical downscaling (SD) methods are mainly used in hydrological studies thanks to their cheaper and easier applications (Fowler et al. 2007; Chen et al. 2010; Chen et al. 2014; Ekstrom et al. 2015). The SD methods are broadly divided into three parts; transfer function (perfect prognosis), weather generator, and weather pattern approach (Chen et al. 2010; Maraun et al. 2010; Tavakol-Davani et al. 2013; Chen et al. 2014; Hou et al. 2017). The transfer function method is frequently used because it is easy to apply anywhere and/or anytime. It establishes linear or nonlinear relationships between local climatic components and large-scale GCMs (Wilby 1998; Wilby et al. 2004; Hessami et al. 2008; Maraun et al. 2010). Reanalysis datasets establish downscaling methods with transfer functions that downscaled GCMs to a regional scale. Reanalysis datasets consist of a combination of atmospheric data obtained from different sources. Although there are many reanalysis datasets, e.g., JRA, MERRA, ERA-40, CERA, ERA-Interim (0.75° × 0.75°), and NCEP/NCAR (2.5° × 2.5°), which are generally used to predict local atmospheric variables in the SD studies such as; Hessami et al. 2008, Chen et al. 2010, Tavakol-Davani et al. 2013, Liu et al. 2019, Jafarzadeh et al. 2021, Quesada-Chacon et al. 2021, the ERA-5 being the development of the ERA-Interim is a high-resolution reanalysis set emerged in 2019 (Hersbach et al. 2020). However, it has not been used so far for the SD studies for Turkey’s basins. Daily precipitation data are vital for assessing the impact of climate change on small and medium-sized basins in many hydrological models (Frost et al. 2011; Beecham et al. 2014). Daily precipitation is difficult to predict because it contains high spatial and temporal randomness (Beecham et al. 2014; Rashid et al. 2015; Liu et al. 2019). Different SD methods have been used for daily precipitation predictions. For example, Hessami et al. (2008) used ridge regression and a statistical downscaling model for daily precipitation prediction and could not determine superiority. Tavakol-Davani et al. (2013) used hybrid models for precipitation occurrence and daily precipitation amount using multiple linear regression (MLR), multivariate model tree (MT), and multivariate adaptive regression splines (MARS) methods. Contrary to the hybrid models, the other methods did not show remarkable differences in the success of daily precipitation prediction compared to the MLR method. Nasseri et al. (2013) made precipitation predictions by hybridizing the MLR, MT, MARS, k-nearest neighbor, and genetic algorithm-optimized support vector machine models. The study results revealed that the precipitation occurrence and amount could be successfully modeled using the hybrid models of the MLR, MT, and MARS methods. The least absolute shrinkage and selection operator (LASSO), which has punishment parameters but does not use low coefficient predictors in contrast to ridge regression, is considered an alternative to precipitation prediction. Although the LASSO method is not superior to stepwise regression in the study by Gao et al. (2014), it was found to perform better than principal component regression in the study by He et al. (2019). Besides, the choice of predictors is also essential in affecting heavy precipitation (Keupp et al. 2019). The Susurluk Basin in Turkey, having mountain and semi-arid Mediterranean climates, was chosen as the study area because the interplay of dynamic and thermodynamic processes has given rise to the Mediterranean precipitations reported by Keupp et al. (2019). The importance of variable selection for the SD models is even more significant in the Susurluk Basin. While establishing the SD methods, long-distance grids can also affect the local climate in different periods (Wilby and Wigley 2000; Crawford et al. 2007; Borges et al. 2017). Also, resolution differences between GCMs and the predictor set can source additional uncertainty (Amjad et al. 2020). However, no study has examined the most appropriate predictors, resolution, and long-distance grid situations in a holistic way on a daily precipitation scale. As daily precipitation downscaling models, no study comparing MLR, elastic net regression (ENET, which combines ridge regression, and LASSO methods), and MARS has been found in the literature. And also, the polynomial MARS (PolyMARS) method, modified from the MARS method, has not been used as a daily precipitation prediction model. Besides, the SD models have not been sufficiently investigated with the ERA-5 reanalysis for daily precipitation prediction. The study aims to draw a directive holistic analysis framework that selects the most appropriate predictors, grid resolution, and long-distance grid states for the first time. Because regression-based models with ease of use and high interpretability provide advantages over black-box models, it also evaluates daily precipitation SD performances using the MLR, exponential regression (EREG), ENET, MARS, and PolyMARS models with ERA-5 reanalysis data in this study for 1979–2018. Besides, the regression models were compared with the black-box model, namely, artificial neural network (ANN). The study consists of six sections; the study area and datasets in the second section, the methodology in the third section, the results in the fourth section, the discussion in the fifth section, and the conclusion in the last section. The study area and datasets The Susurluk Basin is located between 39° and 40° latitudes and 27°–30° longitudes in north-west Turkey, and covers approximately 24,000 km2 area (Fig. 1a). The altitude of the basin varies between the Marmara Sea and the Uludağ Mountain (~ 2543 m). In Turkey, there are 14 Ramsar sites having international significance, two of which are the Ulubat and Manyas lakes located in the Susurluk Basin (Ramsar 2021). The basin includes important stream watersheds such as Kocaçay, Mustafakemalpaşa, Nilüfer, and Simav. The basin is influenced by the Mediterranean climate, with hot summers and wet winters. However, the Nilüfer Stream Watershed, where the Uludağ Mountain is located, has a mountain climate (Peel et al. 2007; Ozturk et al. 2017). The basin has a mean annual temperature of 12 °C, and receives mean precipitation annual of 640 mm. There are approximately 65 dam lakes of various sizes in the basin, either completed or ongoing, for drinking, irrigation, and energy needs, flood protection (DSI 2020). Water demand is also supplied from many aquifers in the basin (Akbas et al. 2020). Turkey’s most intense mining activities occur in the Balıkesir and Bursa provinces, covering a large part of the basin (MTA 2021). It is one of the basins with the densest population in Turkey (Akbas et al. 2020). The mentioned activities and reasons cause to be under the pressure of high-water consumption in the basin, and reveal the basin’s importance. In the basin, flood frequency is high; fortunately, mortality rates are low (SYGM 2018; Haltas et al. 2021).Fig. 1 a Digital elevation model of the Susurluk Basin, b Monthly total precipitation changes at the meteorology stations from S1 to S9 With daily total precipitation data from 1979 to 2018, nine meteorological stations were selected to represent the basin (Fig. 1a). The precipitation data were obtained from the Turkish State Meteorological Service (TSMS 2021), and monthly total precipitation changes were calculated (Fig. 1b). For the SD model setup, the ERA-5 reanalysis hourly dataset was obtained from European Centre for Medium-Range Weather Forecasts Re-analysis 5 (Hersbach et al. 2020). Methodology The grid resolution, long-distance grid situations, and predictor selection were performed in the ERA-5 dataset before the SD models were established for daily prediction. After applying and comparing regression-based five SD models and ANN, bias correction procedures were performed (Fig. 2).Fig. 2 Flow chart of statistical downscaling procedure applied in this study Predictor, resolution, and grid condition selection methodology Before the SD model setup, predictor selection is an essential step in model accuracy because using physical and logical components provides meaningful connections between predictors and predictand (Wilby et al. 2002; Liu et al. 2013; Borges et al. 2017). Although different methods are used for predictor selection, correlation methods, e.g., Pearson and Spearman, are generally preferred for their simplicity of use. Including different regression methods, their advantages have been recently investigated comparatively (Yang et al. 2018; Liu et al. 2019; Jafarzadeh et al. 2021). The stepwise regression method, which is as if it evaluates all possible models, is generally found successful. However, it does not assess the suitability of all possible parameters. Instead of this method, if the number of predictors is 15 or less, the-all possible regression (APR) method can give more accurate results (Burnham et al. 2011; NCSS 2021). In the study by Okkan and Kirdemir (2016), the APR method was used for the predictor selection, and a single prediction was determined. However, the stationary may not decrease in such an evaluation since no humidity parameter is included in predictors (Crane and Hewitson 1998; Wilby et al. 1998; Hessami et al. 2008). The Mediterranean precipitations result from the interaction of dynamic and thermodynamic processes (Keupp et al. 2019). Therefore, common predictors of reanalysis were determined from the studies by Chen et al. (2010), Nasseri et al. (2013), Beecham et al. (2014), Bettolli and Penalba (2018), Yang et al. (2018), Keupp et al. (2019), and Jafarzadeh et al. (2021) that achieved successful results in predicting daily and heavy precipitations (Table 1).Table 1 ERA − 5 reanalysis predictor variables employed in this study Predictor description Abbreviation Pressure Level (hPa) Geopotential height z_500 500 z_700 700 z_850 850 Relative humidity r_500 500 r_700 700 r_850 850 Meridional wind velocity u_500 500 u_700 700 u_850 850 Zonal wind velocity v_500 500 v_700 700 v_850 850 Mean air temperature at pressure levels t_500 500 t_700 700 t_850 850 Mean air temperature at 2 m tm Surface Mean sea level pressure slp Surface Total precipitation tp Surface Due to more computation time and overfitting, it is not recommended to apply downscaling methods with many predictors (Mujumdar and Nagesh Kumar 2012; Das and Nanduri 2018). Different methods are used to select the predictors from the reanalysis data set, but Spearman correlation is frequently used (Chen et al. 2010; Lin et al. 2017; Bettolli and Penalba 2018; Liu et al. 2019). Because Spearman correlation is less affected by extreme situations than Pearson, it does not contain any assumptions and investigates nonlinear relationships (Krause et al. 2005; Chen et al. 2010; Sen 2020a). Using Spearman correlation at a significance level of 1%, the choice of the predictors was performed with the grid values, in which the related meteorology station is located. In addition to the predictor selection, using more than one grid is vital as remote grids may affect the local climate at different times (Wilby and Wigley 2000; Crawford et al. 2007; Borges et al. 2017). Also, the resolution difference between the predictor set and the GCMs may be an additional cause of uncertainty (Amjad et al. 2020). There are studies evaluating grid resolution, predictor selection, and long-distance grid effect separately, e.g., Borges et al. (2017), Yang et al. (2018), Amjad et al. (2020), Jafarzadeh et al. (2021). However, no study performs holistic evaluation together in daily precipitation prediction. The GCM and reference gridded data are usually overlapped with grid centers (Sarhadi et al. 2016; Salman et al. 2018; Khan et al. 2020). While the original resolution of the ERA-5 set used in this study is 0.25° × 0.25°, the GCMs are ~ 1.00°–3.00°. However, overlapping the GCMs and reanalysis grid centers after converting the GCMs to finer resolution may cause extra uncertainty in the SD models (Amjad et al. 2020). Therefore, the reanalysis data with three different resolutions, i.e., 1.00° × 1.00°, 1.50° × 1.50°, and 2.00° × 2.00°, were obtained by considering the mean and standard deviation of the resolutions of the GCMs in Coupled Model Intercomparison Project sixth phase (CMIP6) in addition to the ERA-5 original resolution. Since distant grids can affect the local climate at different times (Amjad et al. 2020), the mean status of the four grids around the stations by considering the grid centers closest to the station (C2) and the grid mean status of the basin (C3) were also taken into account in addition to the grid situation (C1) in which the station is located (Fig. S1). In the studies by Chen et al. (2014), Borges et al. (2017), Lin et al. (2017), and Saengsawang et al. (2017), the one-day lag values (lag-1) in addition to the current day values (lag-0) of the predictors slightly improved the performance of the models. Thus, lag-1 values of the predictors were also examined with Spearman correlation. Then, the most efficient first 12 predictors were selected by taking the absolute values of Spearman correlation. That is, after the first predictor selection was performed at the ERA-5 original grid, in which the stations are located, Spearman correlation analysis was performed with four different resolutions (0.25° × 0.25°, 1.00° × 1.00°, 1.50° × 1.50°, and 2.00° × 2.00°) (Fig. S2), three different grid conditions (C1, C2, and C3), and the lag-0 and lag-1 values of the predictors. Then, 12 predictors were selected for each station. The APR method was used for grid condition, grid resolution, and predictor selection for the best case. This method tests combinations of all possible regressions for each parameter. The total number of all possible regression models is 2k-1. It can be done with Mallows’ Cp that has been used with success by the previous studies to select the best model (Fistikoglu and Okkan 2011; Okkan and Kirdemir 2016). Mallows’ Cp is used to compare models with different parameters (Mallows 1973; Fistikoglu and Okkan 2011). Mallows’ Cp is calculated as follows:1 Cp=n-kSRi2SRF2-n-2i where n is the number of data, i is the number of model parameters, SRi2 is the sum of residual squares in the model with i parameters, and SRF2 is the sum of residual squares in the whole model. The smaller or equal to i+1 (if possible) the value of Cp, the better a model fit is (Pardoe 2013; STAT462 2021). Statistical downscaling methods Regression-based methods are used in SD studies for ease of use and interpretability (Wilby 1998; Wilby et al. 2004; Hessami et al. 2008; Chen et al. 2010; Tavakol-Davani et al. 2013; Nacar et al. 2022). In this study, after the grid condition, resolution, and predictor selection were performed, regression-based methods with the relatively new ERA-5 reanalysis dataset were compared with a black-box model by establishing SD models. Linear regression, i.e., MLR, EREG, a nonlinear form of MLR, ENET to remove unnecessary variables in MLR, and MARS, a nonlinear form of ENET, and PolyMARS models, which are an improved version of MARS were chosen as regression-based methods. ANN was chosen as a black-box model. Details about those mentioned above are given in the supplementary file. In SD studies by Maraun et al. (2010) and Hertig and Jacobeit (2013), it is assumed that the relationship between predictors and predictand will not change under changing climate conditions. It is also thought that 40 years of data may represent the actual climatic conditions for the area in question, including less frequent climatic events (Khan et al. 2006). Again, it has been stated in some studies (Khan et al. 2006, Huang et al. 2011; Nasseri et al. 2013; Tavakol-Davani et al. 2013) that the long training period increases the performance of the models and the potential of the models to catch rare climatic events. Wilby (1998) also stated that the SD model could perform nonstationary using a long calibration period. Therefore, the 1979–2008 period was reserved for the training set, and the 2009–2018 period for the testing set. Bias correction The GCMs and reanalysis data contain biases due to factors such as the false representation of climatic physical processes, parameter optimization, and black-ocean atmosphere feedbacks, and thus, bias correction is required (Troin et al. 2015; Sippel et al. 2016; Nahar et al. 2017; Amjad et al. 2020; Nguyen et al. 2020). The evaluation may not be helpful for future scenarios if bias correction is not applied (Johnson and Sharma 2015). Considering the study by Amjad et al. (2020), in which the study area covers this study’s area, bias correction was performed since the ERA-5 dataset was determined to contain high bias. It is the simplest to apply the standardization to predictors and predictand before the model is set up for bias correction, and it corrects possible biases in mean and variance (Wilby et al. 2004). Predictors and predictand were standardized with long-term mean and standard deviation as follows:2 x^=x-u¯σ where x is series value, u¯ is the long-term mean of x, σ is the standard deviation of x, and x^ is standardized x. In addition, another bias correction method is quantile mapping (QM), which is frequently used in post-processing in the literature (Ashfaq et al. 2010; Rashid et al. 2015). One of the essential features of this method is that it can correct the number of wet days because the GCMs simulate too many wet days (Gutowski et al. 2003; Maraun, 2016). Cumulative distributions of observed and predicted values are used in the QM given as follows in its simplest form:3 P¨=cdfo-1cdfsimP where P is a raw model output, P¨ is validated model output, cdfsim is cumulative density function for model output, and cdfo-1 is inverse cumulative density function of observation values. This study used the empirical QM, which does not require the parametric distribution assumption. In the empirical QM, empirical cumulative density functions are estimated using empirical percentile tables. The studies by Boe et al. (2007) and Themessl et al. (2012) can be examined for more details. Model evaluation To evaluate predictor, resolution, and grid condition selection on a monthly scale and model selection, the Nash–Sutcliffe efficiency (NSE), the most frequently used model performance criterion in hydrology (Lamontagne et al. 2020), is calculated as follows:4 NSE=1-∑i=1Nxi-yi2∑i=1Nxi-x¯2 where yi is the model result, xi is the observation value, and x¯ is the arithmetic mean of the observation values. Model performances of daily precipitation prediction are generally evaluated separately in terms of basic statistical moments such as the mean, standard deviation (Std), and skewness (Hessami et al. 2008; Chen et al. 2010; Liu et al. 2013; Rashid et al. 2015). Although some studies (Nasseri et al. 2013; Tavakol-Davani et al. 2013) holistically evaluated these parameters, they mostly applied evaluations by increasing the coefficients of the mean. However, extreme conditions and precipitation variations are essential as much as the mean values for hydrological models (Liu et al. 2013; Pour et al. 2014; Rashid et al. 2015). Thus, the Kling-Gupta efficiency (KGE) proposed by Gupta et al. (2009) was modified to holistically evaluate the statistical moments of precipitation. It has been frequently used in hydrological modeling (Lamontagne et al. 2020). Because skewness is taken into account to assess information about rare extreme deviations from the mean (Chen et al. 2010) by substituting the skewness parameter for the correlation parameter, the KGE formulation has evolved as follows:5 KGES=1-SKmSKo-12+u¯mu¯o-12+σmσo-12 where KGES is KGE included skewness parameter, SKo (SKm), u¯o (u¯m), and σo (σm) is the skewness, arithmetic mean, and standard deviation of the observed (modeled) data, respectively. In Eq. (5), the closer to 1 the KGES ((-∞, 1]) value is, the higher the model performance is. It should be noted that the KGES formulation was created by utilizing the Euclidean distance feature of the KGE, and is open to the modeler’s judgment for different purposes. The standard triangular diagram (STD), which has been proposed by Sen (2020b), has been modified and employed to support the holistic numerical assessment. Standard deviation (σ), mean (μ), and serial correlation (ρ) values were used in the original of this method. The calculation steps for the STD graph are as follows: Step I calculates σ, μ, and ρ statistic values for observation and prediction values. Step II rates observation and prediction statistics with each other. The ratio values here should be calculated between 0 and 1. So, it does not have to be observation/prediction or vice versa (0<μr=μ1/μ2<1;0<σr=σ1/σ2<1;0<ρr=ρ1/ρ2<1). Step III sums the ratios: S=μr+σr+ρr Step IV divides each ratio by the sum (S) calculated in step III, and multiplies it by 100 as follows:6 μp=μrS×100 7 σp=σrS×100 8 ρp=ρrS×100 Step V places the proportions found in step IV on the triangular diagram given in Fig. 3.Fig. 3 Standard triangular diagram (Sen, 2020a, b) The diagram is divided into four equal sub-triangles; the fourth region in the middle means that no parameter dominates the others, except for insignificant differences (Fig. 3). If a point falls into the fourth region, the parameters are similar in ratios. If a point is closer to the reference point (R, 33.33% for each parameter), parameter ratios are more similar. If a point falls in the first region, the predicted values are highly correlated with the observed values, but the other two parameters are less harmonious with significant differences than the correlation. Similar comments are also valid for the second and third regions (Sen 2020b). In addition, the contingency table (Table 2) was used to evaluate the performance of each model. This table has four internal partitions calculated according to the following schedule: Misses: the number of observed wet days modeled as dry days, False alarm: the number of observed dry days modeled as wet days, Hits: the true distinctive number of wet events, True negative: true distinctive number of dry events. The critical achievement index (CSI) was applied according to Table 2:9 CSI=HitsMisses+Hits+Falsealarm where CSI is taken one in the best case, and zero the in worst case (Jafarzadeh et al. 2021).Table 2 Contingency table event for statistical downscaling models Modelled precipitation events Observed precipitation events Wet event Dry event Wet event Hits (WW) False alarm (DW) Dry event Misses (WD) Correct negative (DD) Results Predictor, resolution, and grid condition selection results Primarily, the relationship between total precipitation values and the predictors in the ERA-5 set was investigated at a 1% significance level using the Spearman correlation method for the period 1979–2018, and non-significant parameters were excluded from the scope of the study (Table 3).Table 3 Spearman correlation results (absolute values) at a 1% significance level Pressure level/Stations r slp tmean tp u v z t 500 700 850 500 700 850 500 700 850 500 700 850 500 700 850 S1 0.35 0.45 0.57 0.03 0.35 0.57 0.09 0.11 0.09 0.10 0.08 0.03 0.50 0.51 0.38 0.43 0.47 0.45 S2 0.35 0.45 0.56 0.06 0.34 0.56 0.10 0.12 0.10 0.12 0.10 0.05 0.50 0.51 0.40 0.42 0.46 0.44 S3 0.35 0.44 0.53 0.11 0.31 0.54 0.11 0.16 0.15 0.13 0.12 0.09 0.49 0.51 0.42 0.40 0.44 0.41 S4 0.35 0.46 0.55 0.09 0.33 0.54 0.12 0.16 0.18 0.11 0.08 0.06 0.47 0.49 0.39 0.39 0.42 0.40 S5 0.37 0.47 0.58 0.16 0.28 0.56 0.14 0.20 0.16 0.18 0.16 0.14 0.45 0.48 0.43 0.36 0.39 0.35 S6 0.34 0.45 0.55 0.12 0.26 0.54 0.09 0.13 0.16 0.10 0.05 0.02 0.42 0.45 0.38 0.34 0.37 0.34 S7 0.37 0.49 0.56 0.12 0.31 0.55 0.14 0.19 0.24 0.11 0.08 0.06 0.47 0.49 0.40 0.38 0.42 0.38 S8 0.34 0.47 0.58 0.03 0.34 0.58 0.10 0.12 0.14 0.06 0.02 0.02 0.48 0.50 0.38 0.40 0.45 0.43 S9 0.37 0.50 0.61 0.07 0.36 0.56 0.13 0.15 0.21 0.07 0.04 0.01 0.51 0.53 0.39 0.42 0.47 0.45 *Bold italic: Significant Considering the Spearman correlation results, the predictors, u, v, and slp, were excluded from the study since these did not show a meaningful relationship for all stations in general. The predictors with the highest absolute correlation with daily total precipitation values were determined as tp (~ 0.56), r_850 (~ 0.56), and z_700 (~ 0.50). Then by considering the lag-0 and lag-1 values of the predictors for three different grid conditions (Fig. S1) and four different resolutions, i.e., 0.25° × 0.25°, 1.00° × 1.00°, 1.50° × 1.50°, and 2.00° × 2.00° (Fig. S2), 12 predictors were selected by employing Spearman correlation. The situation mentioned above was examined for the Susurluk Basin with different climate characteristics, and the predictor selection was made together with the conditions using the APR method. The application results for the first station as a representation of the APR method are given in Table 4. In the APR method, daily values were used for Cp. Since daily precipitation contains high randomness, the NSE criterion, a combined measure of correlation, bias, and variability (Gupta et al. 2009), was also used as support over monthly total precipitation.Table 4 The all-possible regression results for the first station, the Susurluk Basin, Turkey n Daily Cp Monthly NSE tp r_500 r_700 r_850 z_500 z_700 z_850 t_500 t_700 t_850 tm 1 2447.04 0.73 • 2 149.33 0.79 ∆ • 3 124.42 0.79 ∆ • • 4 91.43 0.79 ∆ • • • 5 65.61 0.78 ∆ • • • ∆ 6 36.68 0.78 ∆ • • • • ∆ 7 16.73 0.78 ∆ • • ∆ • ∆ • 8 10.31 0.79 ∆ • • ∆ ∆ • ∆ • 9 10.35 0.79 ∆ • • ∆ ∆ • ∆ • ∆ 10 10.18 0.79 ∆ • • • ∆ ∆ • ∆ • ∆ 11 11.62 0.79 ∆ • • ∆ • ∆ • ∆ • • ∆ 12 13.00 0.79 ∆ • • • ∆ • ∆ • ∆ • • ∆ •: lag − 1 value, ∆: lag − 0 value In Table 4, Cp increases after a rapid decrease to n = 8. If possible, it is recommended that ‘Cp=~n+1’ should be preferred (STAT462 2021). In the absence of this situation, the smallest value of Cp is evaluated as the most suitable model (Pardoe 2013; STAT462 2021). Therefore, nine variables were chosen for the first station. The predictor frequencies determined for the precipitation prediction in the basin are given in Fig. 4.Fig. 4 Predictor frequency with reference to the all-possible regression method As seen in Fig. 4, the most frequently used predictors are tp, z_500, r_700, and z_700, respectively, for the downscaling model in the basin. After the predictor selection was made with the APR method, grid conditions and resolution selections were comparatively performed using monthly NSE values (Fig. 5).Fig. 5 Grid condition and resolution determination according to monthly NSE values (the red dashed line corresponds to an NSE value of 0.75) In Fig. 5, C1 generally showed the worst performance for all situations at all stations. C2 and C3 had similar performances for all grid resolutions. Although C3 mostly performed well compared to C1, it did not perform as stable and well as C2 at all grid resolutions. The horizontal red dashed line corresponding to an NSE value of 0.75 (> very good performance) was drawn for easier decision-making. Except for the third (NSE = 0.73) and sixth (NSE = 0.61) stations, which have poor performances in all grid resolutions and conditions, it was observed that C2 did not cross below the horizontal red dashed line at 1.50° × 1.50° resolution. For this reason, C2 and 1.50° × 1.50° grid resolutions were chosen because the mean of GCM in CMIP6 resolutions was about 1.50° × 1.50°. So, it can be said that long-distance grids effectively predict precipitation (Wilby and Wigley 2000; Crawford et al. 2007; Borges et al. 2017). Precipitation occurrence results Precipitation occurrence was modeled with precipitation amount without any dry/wet classification. While determining the conditions as wet and dry days, precipitation of ≥ 1 mm was considered a wet and dry class and for otherwise conditions (Frost et al. 2011). The holistic evaluation results of the monthly mean wet days of the stations in the basin are graphically shown in Fig. 6. In addition, the heat map of monthly mean wet days by altitude is given in Fig. 7 for testing (2009–2018) period. The reason for examining the statistics of daily values in a month was to show the behavior of daily precipitation during a month. In other words, the seasonal behavior capture performances of the models were examined.Fig. 6 Downscaling results of the mean number of wet days and daily precipitation statistics per month for the training (1979–2008) and testing (2009–2018) periods Fig. 7 Heat map of downscaling results of the mean number of wet days and daily precipitation statistics by altitude per month for testing (2009–2018) period The EREG model provided the worst prediction of the wet day mean in the basin during training and testing periods, as shown in Fig. 6. The EREG and ENET models did not improve prediction performance compared to the MLR model, and even predicted precipitation occurrence as worse. The PolyMARS, ANN, and MARS, models provided the best models for both periods, with a negligible difference, respectively. The PolyMARS model predicted an error of about 17% for March, with the worst prediction success, considering the testing period. Compared to the study by Chen et al. (2010), this error value was at a better prediction level, in which wet/dry seasons and wet/dry days were classified. As the altitude increases (Fig. 7), the mean of wet days in the months also increases, but the S6 station, which has an altitude of 833, is different from this situation. Although MARS, PolyMARS, and ANN tend to catch the observation values due to the scale width due to the altitude difference, the PolyMARS model becomes more prominent when looking at mid-altitudes for the testing period (Fig. 7). However, although there is a pattern between the observed values and the modeled results, it is seen from Fig. 6 that there are biases, and therefore, bias correction is going to be applied in the next step. To Table 5, the EREG model showed the worst performance among the models with the lowest CSI values. When EREG has the highest average WW value, PolyMARS model has the highest average DD value. That is, PolyMARS correctly captured dry days, while EREG captured wet days. Although the CSI values of PolyMARS and ANN were equal and higher during the training period, PolyMARS showed significantly higher performance during the testing period. So, PolyMARS performed better than the other models in modeling precipitation occurrence.Table 5 Results of contingency table events Stations and parameters Training Testing DD DW WD WW CSI DD DW WD WW CSI S1 MLR 6950 1740 281 1986 0.496 2237 563 114 738 0.522 EREG 6840 1850 265 2002 0.486 2187 613 102 750 0.512 ENET 6730 1960 210 2057 0.487 2152 648 80 772 0.515 MARS 6981 1709 241 2026 0.510 2235 565 93 759 0.536 PolyMARS 7320 1370 340 1927 0.530 2353 447 132 720 0.554 ANN 7268 1422 304 1963 0.532 2296 504 119 733 0.541 S2 MLR 7073 1726 208 1950 0.502 2210 603 89 750 0.520 EREG 6779 2020 172 1986 0.475 2082 731 73 766 0.488 ENET 7077 1722 211 1947 0.502 2210 603 89 750 0.520 MARS 7323 1476 275 1883 0.518 2294 519 110 729 0.537 PolyMARS 7418 1381 272 1886 0.533 2320 493 119 720 0.541 ANN 6667 2132 156 2002 0.467 2134 679 73 766 0.505 S3 MLR 7522 1570 186 1679 0.489 2336 594 43 679 0.516 EREG 7519 1573 185 1680 0.489 2335 595 43 679 0.516 ENET 7531 1561 184 1681 0.491 2338 592 43 679 0.517 MARS 7778 1314 239 1626 0.511 2404 526 60 662 0.530 PolyMARS 7831 1261 248 1617 0.517 2439 491 63 659 0.543 ANN 7938 1154 263 1602 0.531 2548 382 84 638 0.578 S4 MLR 7084 1776 218 1879 0.485 2255 621 73 703 0.503 EREG 7002 1858 198 1899 0.480 2215 661 66 710 0.494 ENET 7087 1773 218 1879 0.486 2256 620 73 703 0.504 MARS 7173 1687 229 1868 0.494 2278 598 73 703 0.512 PolyMARS 7336 1524 267 1830 0.505 2318 558 75 701 0.525 ANN 7118 1742 240 1857 0.484 2248 628 76 700 0.499 S5 MLR 7267 1489 272 1929 0.523 2368 411 120 753 0.586 EREG 7102 1654 232 1969 0.511 2316 463 114 759 0.568 ENET 7266 1490 211 1990 0.539 2315 464 105 768 0.574 MARS 7567 1189 288 1913 0.564 2416 363 140 733 0.593 PolyMARS 7395 1361 259 1942 0.545 2397 382 130 743 0.592 ANN 7386 1370 330 1871 0.524 2350 429 139 734 0.564 S6 MLR 7193 1660 302 1802 0.479 2198 608 117 729 0.501 EREG 7169 1684 297 1807 0.477 2184 622 114 732 0.499 ENET 7033 1820 260 1844 0.470 2145 661 99 747 0.496 MARS 7406 1447 393 1711 0.482 2277 529 132 714 0.519 PolyMARS 7493 1360 390 1714 0.495 2324 482 141 705 0.531 ANN 7137 1716 341 1763 0.462 2195 611 113 733 0.503 S7 MLR 6140 2226 223 2368 0.492 1983 738 74 857 0.513 EREG 5576 2790 195 2396 0.445 1756 965 62 869 0.458 ENET 6149 2217 223 2368 0.493 1985 736 75 856 0.513 MARS 6699 1667 306 2285 0.537 2137 584 107 824 0.544 PolyMARS 6796 1570 344 2247 0.540 2175 546 118 813 0.550 ANN 6697 1669 357 2234 0.524 2121 600 114 817 0.534 S8 MLR 6861 1728 211 2157 0.527 2176 620 51 805 0.545 EREG 6830 1759 206 2162 0.524 2159 637 52 804 0.539 ENET 6862 1727 211 2157 0.527 2178 618 51 805 0.546 MARS 7043 1546 219 2149 0.549 2239 557 65 791 0.560 PolyMARS 7149 1440 274 2094 0.550 2297 499 78 778 0.574 ANN 7226 1363 294 2074 0.556 2320 476 77 779 0.585 S9 MLR 5018 2531 179 3229 0.544 1686 782 113 1071 0.545 EREG 3369 4180 504 2904 0.383 1185 1283 190 994 0.403 ENET 4587 2962 80 3328 0.522 1523 945 75 1109 0.521 MARS 5653 1896 260 3148 0.594 1848 620 159 1025 0.568 PolyMARS 5496 2053 244 3164 0.579 1838 630 143 1041 0.574 ANN 5844 1705 228 3180 0.622 1789 679 143 1041 0.559 The values closer to observation values are given in bold Precipitation statistics results Daily precipitation statistics are essential for the hydrology of current and future periods. The statistics of six different SD model results are given in Table 6.Table 6 The statistical results for the statistical downscaling models Stations and parameters Training Testing Obs MLR EREG ENET MARS PolyMARS ANN Obs MLR EREG ENET MARS PolyMARS ANN S1 Mean 1.82 1.98 1.95 1.92 1.89 1.83 1.87 2.24 2.60 2.54 2.52 2.45 2.35 2.13 Std 6.10 3.89 3.94 3.51 4.09 4.12 4.28 8.22 5.21 5.27 4.69 5.28 5.40 4.67 Skewness 7.50 3.33 4.30 3.13 4.65 4.18 4.37 13.24 3.36 4.28 3.13 4.39 4.02 4.23 KGES − − 0.38 0.08 − 0.58 0.22 0.07 0.17 – − 2.00 − 1.17 2.32 − 1.09 − 1.35 − 1.26 NSE 0.43 0.43 0.42 0.46 0.46 0.46 0.32 0.32 0.32 0.32 0.34 0.34 S2 Mean 1.73 1.87 1.88 1.87 1.76 1.78 1.93 2.11 2.30 2.30 2.29 2.16 2.18 2.16 Std 5.67 3.69 3.71 3.69 4.08 4.03 3.81 6.76 4.41 4.47 4.41 4.84 4.86 4.35 Skewness 5.91 3.38 4.22 3.38 5.26 5.05 4.45 6.86 3.40 4.49 3.41 5.42 5.08 4.99 KGES – 0.07 0.33 0.08 0.59 0.56 0.40 − − 0.15 0.26 − 0.15 0.52 0.47 0.33 NSE 0.45 0.45 0.45 0.52 0.51 0.47 0.41 0.41 0.41 0.44 0.45 0.41 S3 Mean 1.44 1.52 1.52 1.52 1.46 1.45 1.46 1.82 1.88 1.88 1.87 1.86 1.86 1.70 Std 5.14 3.12 3.12 3.12 3.31 3.20 3.30 5.79 3.54 3.54 3.54 3.85 3.66 3.60 Skewness 7.32 3.47 3.49 3.48 4.12 3.34 3.48 5.30 3.39 3.41 3.40 4.04 3.27 3.24 KGES − − 0.28 − 0.28 − 0.28 0.04 − 0.34 − 0.23 – 0.15 0.15 0.15 0.41 0.15 0.12 NSE 0.38 0.38 0.38 0.42 0.39 0.40 0.43 0.43 0.43 0.42 0.43 0.42 S4 Mean 1.44 1.56 1.55 1.56 1.55 1.51 1.67 1.64 1.74 1.73 1.74 1.72 1.66 1.92 Std 4.51 2.91 2.92 2.91 2.97 3.05 3.25 4.70 3.05 3.04 3.05 3.09 3.12 3.43 Skewness 5.26 3.36 3.87 3.36 3.51 3.73 5.17 4.31 3.18 3.60 3.18 3.42 3.80 4.13 KGES − 0.21 0.35 0.21 0.28 0.37 0.59 – 0.35 0.42 0.35 0.42 0.47 0.60 NSE 0.44 0.44 0.44 0.46 0.47 0.49 0.39 0.39 0.39 0.40 0.41 0.41 S5 Mean 2.03 2.26 2.21 2.16 2.06 2.11 1.67 1.88 2.02 1.97 1.95 1.91 1.91 1.88 Std 6.89 5.02 5.10 4.64 5.48 5.31 4.08 5.49 3.99 3.94 3.71 4.21 4.19 4.27 Skewness 6.52 4.11 5.66 3.87 6.27 5.82 5.04 4.84 3.45 4.33 3.25 4.74 5.06 4.60 KGES – 0.30 0.61 0.16 0.74 0.68 0.22 – 0.44 0.59 0.31 0.70 0.69 0.71 NSE 0.56 0.57 0.55 0.64 0.60 0.55 0.45 0.45 0.45 0.47 0.48 0.46 S6 Mean 1.23 1.30 1.30 1.27 1.24 1.25 1.31 1.47 1.53 1.53 1.50 1.45 1.44 1.54 Std 3.78 2.28 2.28 2.05 2.49 2.48 2.27 4.05 2.45 2.44 2.21 2.48 2.46 2.46 Skewness 5.10 3.50 3.73 3.19 4.57 4.26 3.50 4.50 3.26 3.42 2.96 3.48 3.09 3.31 KGES – 0.20 0.25 − 0.04 0.47 0.44 0.19 – 0.24 0.27 0.02 0.30 0.21 0.26 NSE 0.38 0.38 0.37 0.44 0.44 0.40 0.28 0.28 0.28 0.31 0.32 0.30 S7 Mean 1.93 2.09 2.10 2.09 1.98 1.97 1.94 2.08 2.23 2.15 2.23 2.06 2.04 2.21 Std 5.36 3.55 3.60 3.56 3.97 4.01 3.81 5.45 3.62 3.27 3.62 3.72 3.76 3.97 Skewness 5.13 3.07 4.74 3.07 3.85 3.73 3.68 3.99 2.90 3.69 2.90 3.32 3.37 3.17 KGES – 0.16 0.50 0.16 0.52 0.50 0.44 – 0.37 0.33 0.37 0.49 0.52 0.54 NSE 0.47 0.49 0.47 0.56 0.57 0.57 0.41 0.44 0.41 0.48 0.49 0.49 S8 Mean 1.83 1.96 1.96 1.96 1.86 1.86 1.76 2.21 2.37 2.36 2.37 2.25 2.22 2.00 Std 5.48 3.75 3.75 3.75 3.86 3.95 3.85 6.44 4.40 4.40 4.40 4.55 4.52 4.17 Skewness 5.14 3.42 3.62 3.42 3.79 3.70 4.33 5.56 3.32 3.50 3.32 3.99 3.75 4.41 KGES – 0.31 0.37 0.31 0.45 0.45 0.53 – 0.18 0.25 0.18 0.43 0.36 0.39 NSE 0.49 0.49 0.49 0.50 0.52 0.51 0.48 0.48 0.48 0.49 0.50 0.49 S9 Mean 3.94 4.11 4.33 4.09 4.06 4.02 3.86 3.95 4.19 4.38 4.17 4.25 4.08 4.37 Std 9.99 6.94 6.66 6.30 7.06 7.22 7.10 10.21 7.08 6.80 6.45 7.34 7.43 7.74 Skewness 4.81 3.09 3.26 2.83 2.66 2.98 3.05 5.07 2.98 3.01 2.71 2.58 3.23 3.04 KGES – 0.29 0.30 0.09 0.09 0.28 0.30 – 0.17 0.15 − 0.05 − 0.04 0.32 0.25 NSE 0.50 0.44 0.49 0.51 0.53 0.38 0.33 0.38 0.38 0.41 0.39 The values closer to observation values are given in bold All models gave results closer to the observation values, with no significant difference in the mean parameter. The MARS and PolyMARS models gave results closer to the observation values in the mean parameter. All models underestimated Std for the training and testing periods. However, the MARS, PolyMARS, and ANN models gave the prediction values closer to the observation values for both periods in Std. While all models for both periods underestimated skewness associated with the generation of rare extreme values, the MARS and PolyMARS models generally gave results closer to the observation values. Except for the first station, precipitation parameters at all stations were successfully predicted since the KGES values were greater than zero (Table 6). Although the MARS model had a higher performance to the KGES values, the PolyMARS model had higher performance to NSE values, especially during the testing period (Table 6). Besides, in the training and testing sets, the relative error mean of the means of daily precipitation in months is 8% and 11%, respectively (Table S1). While the ANN gave the relative error mean of the means of daily precipitation in months in the training period, PolyMARS showed a high performance with the smallest in the testing period. So, when the precipitation statistics in the basin are evaluated monthly or in whole time series, all models most likely catch the observation values in predicting the mean precipitation. In other words, someone who needs to examine in terms of mean values can use these models under specified conditions. However, it should also be considered that it contains bias in Std and skewness values (Figs. 6 and 7 and Table 6). Although the MARS, PolyMARS, and ANN models were slightly better than the other models in general in both periods for Std and skewness values, these models could not catch the observation values exactly (Figs. 6 and 7). All models could not detect the Std and skewness values of the observation (Fig. 6), but PolyMARS tended to capture high skewness values in summer and autumn at low altitudes during the testing period (Fig. 7). According to Figs. 6 and 7, although there is a certain pattern, there is bias in Std and skewness values. Therefore, bias correction should be performed for daily downscaled precipitations. For the testing period, the observation Std and skewness were high due to the extreme precipitations (~ 250 mm/day) around the Marmara Sea, 7–12 September 2009 (Komuscu et al. 2013; Komuscu and Celik 2013). The PolyMARS model tended to capture these extreme precipitation values in months during both the training and testing period, slightly different from the other models (Fig. 6). However, the MARS model tends to catch these values in terms of the whole time series in both periods (Table 4). For 9 September 2009, the heaviest day of precipitation, the mean absolute error of the EEG and PolyMARS models is smaller than the other models, but PolyMARS tends to capture 248 mm of precipitation at station S1 more than EREG. Therefore, the general distribution of the absolute error of the PolyMARS model in the basin is finally given in Fig. S3 for 9 September 2009. The reason for the high error values in the first and second stations (Fig. S3) is due to the high atmospheric instability of the precipitation values observed during this period and the short-term precipitation intensity caused by the different warming between the land and sea surfaces (Komuscu et al. 2013; Komuscu and Celik 2013). In other words, the model performance is lower in the first and second stations compared to the rest, as the precipitation between these dates contains very high randomness. It can be said that the MARS, PolyMARS, and ANN models, which consider nonlinear relationship, are somewhat better than the other models because it tends to catch extreme rainfalls during the training period. The MARS and PolyMARS models also tend to catch up with these extremes to some extent. These models fail due to the precipitation having a random process (Beecham et al. 2014; Rashid et al. 2015) and the dry days in the daily precipitation series, especially during the summer months. In daily precipitation prediction, the SD methods may model wet days as dry days and vice versa (Beecham et al. 2014). It is seen from Fig. 6 that the ERA-5 contains a high bias for the Std and skewness, and has a drizzle effect, which is the situation of climate models to simulate very few dry days (Gutowski et al. 2003). Bias correction results for the downscaled precipitation As a bias correction method, predictor and predictand values are generally standardized. Although the same method was used in this study, it was observed to be insufficient (Fig. 6). The insufficiency of this method was also revealed in a study by Rashid et al. (2015). Therefore, a second bias correction application was thought appropriate in this study, and the QM method was applied. After the method was applied to the training set on a station basis, it was also applied to the testing set with the same parameters (Figs. S4 and S5). For example, quantile–quantile plots of the PolyMARS precipitation model at the first station with the highest skewness parameter and semi-arid Mediterranean climate, and the last station with mountain climate are given in Fig. 8.Fig. 8 Quantile–quantile plot results of bias correction of the PolyMARS model for the first and last stations, the Susurluk Basin, Turkey The QM method gives successful results up to 50 mm/day for the training and testing periods at the first and last stations with different climatic characteristics (Fig. 8). The QM method also makes more successful corrections in the training period than in the testing period but corrects over/under in heavy precipitations in both periods. So, after the limit of 50 mm/day, there is a decrease in the success of the QM method in heavy precipitations, especially in rare extreme values. For heavy precipitations (≥ 50 mm), the models gave a maximum (mean) relative error of 58% (42%) during the training period and 55% (39%) during the testing period except for the third station before bias correction. However, the maximum (mean) errors occurred with 8% (2%) and 38% (15%), respectively, in the training and testing periods after bias correction (Table 7). So, the QM method gave more successful results in the training period compared to the testing period for heavy precipitations. The STD graph examining the relative dominance of the mean, Std, and skewness parameters over each other over the entire time series of each model in both the training and testing set at each station is given in Fig. 9.Table 7 The relative error (%) values of the quartiles above 50 mm precipitation, the Susurluk Basin, Turkey Stations Bias Training Testing MLR EREG ENET MARS PolyMARS ANN MLR EREG ENET MARS PolyMARS ANN S1 Uncorrected 52.38 43.47 58.02 39.08 45.73 33.30 47.11 39.84 53.52 41.16 41.08 49.90 Corrected 1.45 1.69 1.64 1.72 1.49 0.96 37.97 20.11 36.63 9.74 28.94 17.20 S2 Uncorrected 54.46 46.00 54.44 31.68 38.40 27.60 52.17 43.63 52.16 37.65 42.00 33.40 Corrected 1.44 1.34 1.45 0.85 1.22 0.49 12.87 11.70 12.76 12.01 8.23 6.75 S3 Uncorrected 59.60 59.45 59.57 50.52 63.94 46.30 45.24 44.98 45.20 34.05 50.85 36.30 Corrected 1.36 1.46 1.34 1.35 0.94 1.28 73.80 73.45 73.70 41.51 80.43 45.50 S4 Uncorrected 48.86 39.82 48.85 43.47 45.91 9.05 – – – – – – Corrected 1.67 3.01 1.65 7.43 4.65 1.69 – – – – – – S5 Uncorrected 38.95 32.18 44.83 25.70 25.97 31.10 36.76 32.16 42.43 21.43 25.51 15.00 Corrected 1.30 1.24 1.26 1.38 0.55 1.55 9.40 8.57 9.32 8.84 13.81 22.70 S6 Uncorrected 46.55 40.79 54.42 30.14 21.98 36.10 – – – – – – Corrected 0.00 0.00 0.00 0.00 0.00 0.00 – – – – – – S7 Uncorrected 47.63 22.91 47.62 35.58 38.34 35.80 38.61 25.17 38.62 31.53 31.15 21.80 Corrected 5.71 7.57 5.78 8.18 3.74 5.95 6.29 14.42 6.23 16.64 7.07 8.69 S8 Uncorrected 44.50 41.82 44.49 42.53 43.39 19.40 41.33 38.80 41.32 31.06 35.58 24.50 Corrected 1.17 1.11 1.15 1.68 1.48 0.87 21.11 19.62 21.07 27.74 26.68 10.60 S9 Uncorrected 35.96 36.30 43.11 40.35 35.51 24.90 37.61 39.75 44.58 40.40 35.28 24.50 Corrected 0.71 0.81 0.59 0.74 0.79 0.96 4.12 6.44 4.60 4.77 7.77 10.50 Fig. 9 Standard triangular diagram of the quantile mapping results per model and station from S1 to S9 It is seen from Fig. 9 that all model outputs of all stations do not have any significant advantage over each other in uncorrected bias during the training period. However, the skewness values are less dominant than the mean and Std parameters. The rate of capturing the observation values is low in terms of the skewness parameters of the models. This situation did not change in uncorrected bias during the testing period; even the ENET and ANN models captured the parameter at the first station much lower than the other parameters. After bias correction was applied, all models were gathered on the reference point, i.e., the black point, in terms of Std, mean, and skewness. In the testing period, the bias correction method could not catch as much as the training period, but it can be considered reasonable. Since these findings have shown consistent results with the study by Rashid et al. (2015), it can be said that the bias correction improves the scattering in a limited capacity. Still, it does not catch statistically significant moments. The comparison of wet/dry days after bias correction is shown in Figs. 10 and 11. It can be said that good harmony was achieved for all models in the training period except for March, but it was at an acceptable level during the testing period. Considering the change of bias correction in month’s statistical values (Figs. 10 and 11), it can be said that it shows an improvement for all models on the hits of wet/dry, Std and skewness parameters compared to Figs. 6 and 7. This is because the QM method can capture variations in observations compared to other bias methods.Fig. 10 The quantile mapping results of the mean number of wet days and daily precipitation statistics per month for the training (1979–2008) and testing (2009–2018) periods Fig. 11 Heat map of the quantile mapping results of the mean number of wet days and daily precipitation statistics by altitude per month for the testing (2009–2018) period Compared to the others, the PolyMARS model tends to capture Std and skewness values with minor differences in training and testing periods, in the basin. It has been also observed that precipitation quarters, dry/wet days, and statistical moments can be significantly captured by applying bias correction in addition to the MLR models established with ERA-5 data. However, it should be noted that even though the linear regression-based methods are corrected for bias, the explained variance of predictors is less than other methods, as stated by Chen et al. (2014). Discussion Streamflows are essential for managing disasters such as floods and droughts, building safety and management, and requirements such as water supply, energy generation, and irrigation. In addition, correct streamflow prediction is essential for hydrological models that enable the analysis of the dry–wet periods and transitions of the hydrological and biochemical processes in a basin (Loucks and van Beek 2017; Arora et al. 2020; Maina et al. 2020; Newcomer et al. 2021). Therefore, correct precipitation prediction, the most important parameter affecting the streamflow, is important. Precipitation has been considered the most critical parameter in the studies by Fistikoglu and Okkan (2011), Okkan and Inan (2015), and Okkan and Kirdemir (2016) on the prediction of monthly precipitation and streamflow in the adjacent basin, namely Gediz, with similar climatic conditions. However, the geopotential and humidity parameters, which trigger precipitation and allow for avoiding the stationary assumption (Crane and Hewitson 1998; Wilby et al. 1998; Hessami et al. 2008), were removed from the predictors set in the studies. Besides, the humidity was a parameter controlling extreme precipitation in the Mediterranean Basin (Keupp et al. 2019). The geopotential parameter also significantly affects precipitation in semi-arid basins or dry seasons (Chen et al. 2010; He et al. 2019; Kumar et al. 2021). Similar studies (Turkes 1998; Hertig et al. 2013, 2017; Keupp et al. 2019) have also shown that precipitation in the Mediterranean region controls geopotential height and humidity parameters. High geopotential height stands for stagnant air, while high relative humidity indicates the high moisture required for heavy precipitation (GFA, 2022; GMAFM, 2022). So, there is an inverse relationship between them. Therefore, these parameters are important for precipitation prediction in the SD studies. In this study, precipitation, geopotential, and humidity were effective parameters, and confirmed the abovementioned studies. Besides, although long-distance grids have been emphasized more than the closest grid effect in some previous studies (Wilby and Wigley 2000; Crawford et al. 2007; Borges et al. 2017), the longer distance effect has not been taken into account. However, it was seen that the effect of long-distance grids on the station might decrease depending on the topography in this study (Figs. 1a and 5). At this point, the number of long-distance grids is still an open debate for researchers, but it should be considered. Besides this study, it was stated that the effect of the grids around the station was also important in similar studies (Borges et al. 2017; Herrera et al. 2019; Nacar et al. 2022). Again, considering the coarse resolution and the long-distance grid, there may be a slight improvement in monthly total precipitation. In the study by Amjad et al. (2020), including this study area, the resolution decreases in the case of an average of grids around stations. Still, a slight decrease preserves the bias, and the correlation increases by a small amount. The change of NSE, the combined measure of correlation, bias, and variance (Gupta et al. 2009), has also similar results in this study (Fig. 5). The difference in resolution between the reanalysis and GCMs data can be a source of extra uncertainty during the downscaling phase (Amjad et al. 2020). For this reason, to reduce the possible uncertainties in the relationship between GCMs and ERA-5 data, different grid resolutions were examined in this study. Considering the long-distance effect, the average (1.5° × 1.5°) resolution of the GCMs in CMIP6 was selected as the optimum rather than the best for this basin. This does not mean that fine resolution should not be preferred, but doing so will reduce the uncertainty in future predictions with the GCMs (Amjad et al. 2020). So, the scale (resolution) effect is still an open issue (Chen et al. 2014). In terms of precipitation occurrence and amount, linear regression-based methods, i.e., the MLR and ENET, have the most ineffective performances due to the randomness of precipitation (Rashid et al. 2015). Some studies have attained similar outcomes (Chen et al. 2010; Tavakol-Davani et al. 2013; Chen et al. 2014; Liu et al. 2019). Compared to the EREG model, which is a nonlinear one, the PolyMARS model provided successful results (on the KGES, CSI, and NSE parameters, PolyMARS outperformed the MLR by 66%, 5%, and 9%, respectively) since these models set up regression models within the parts by separating the data into different intervals (Kooperberg et al. 1997; Stone et al. 1997; Yilmaz et al. 2018). In the basin, local convective instability aforementioned by Ozturk (2010) may be the reason for the failure of prediction performances for precipitation values to achieve a very high degree of success. Conclusions In this study, the performances of regression-based statistical downscaling methods, multiple linear regression (MLR), exponential regression (EREG), elastic net regression (ENET), multivariate adaptive regression splines (MARS), and polynomial MARS (PolyMARS), were compared in the Susurluk Basin (Turkey), which have both the mountain climate and semi-arid climate. In addition, an artificial neural network (ANN) was used to examine the performances of regression models according to black-box models. Before comparing the models, the effects of atmospheric variables, grid resolution, and remote grid conditions in the reanalysis dataset were also examined. Then, nine stations were selected, and their daily total precipitation data and atmospheric variables in the ERA-5 reanalysis dataset from 1979 to 2018 were used. While doing the holistic evaluation, the standard triangular diagram, which was recently proposed as a graphical model evaluation metric, was considered. It was also used as a numerical model evaluation metric by modifying the Kling-Gupta efficiency. The conclusions obtained from the study are itemized below:The grids around the station affect precipitation prediction rather than a single grid or basin grids mean. The monthly precipitation prediction performance may decrease from fine resolution to coarse resolution due to the basin characteristics. The most influential parameters are total precipitation, geopotential height, and relative humidity from 18 atmospheric variables selected from the ERA-5 dataset in the basin. All models tend to predict more wet days due to climate models having a drizzle effect. Considering their wet/dry day hit performances, the PolyMARS model, however, shows more success by separating them from the other models. The PolyMARS model is also successful compared to the other models. To the observation values, the model has closer prediction values with a bit of difference in statistical moments, i.e., mean, standard deviation, and skewness. However, the linear models, i.e., ENET and MLR, have worse prediction performances. The ANN model has similar results, although not as much as the PolyMARS model, in monthly and whole time series evaluation of daily precipitation values. In other words, regression-based nonlinear models can be used with reanalysis sets such as ERA-5 in statistical downscaling studies rather than the black-box model. All models perform well in mean values but do not in variance and rare extreme values. Statistical downscaling models can be used by ERA-5 data without any wet-dry classification in the basin. Although the quantile mapping method corrects the number of wet days and statistical parameters, it is less successful in correcting biases in heavy precipitation and variance. This study draws a framework for statistical downscaling. However, some shortcomings and proposed future examinations are given below:The predictors were selected from the ERA-5 reanalysis dataset. The same analyses can be performed using different reanalysis datasets. Thus, the reanalysis dataset that best represents the basin can be determined. The only regression-based models, whose performances were studied separately in the literature, were examined mutually in this study. Still, downscaling performances of the PolyMARS model, which were successful in this study, can be examined or hybridized with rule-based models such as Cubist, M5-tree, and Part in future studies. Downscaling model performances were evaluated on daily precipitation in this study. In future studies, hourly or minute precipitations or other hydrological parameters, e.g., maximum and minimum temperatures, can be downscaled with the models used in this study and/or different models. The bias correction method of quantile mapping was applied in this study. The most appropriate bias correction method can be determined using different procedures. This study has drawn important parameters and flowcharts for statistical downscaling studies. The findings within the scope of this study will form a basis for future climate change scenario modeling and researchers. Besides, by switching to the streamflow with the models obtained, it will help the management of water resources of the basin, and the management of extreme events such as floods and droughts, etc. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 1819 KB) Acknowledgements The authors would like to thank Salih Türk, a Faculty member of Gümüşhane University, for his help in teaching and implementing the software codes. Also, the authors would like to thank the anonymous reviewers for their constructive comments and suggestions, which helped improve the paper. Author contributions MŞ: Conceptualization, Methodology, Software, Writing—Original draft, Investigation, SN: Conceptualization, Methodology, Investigation, Validation, Visualization, Writing—Review & Editing, MK: Conceptualization, Validation, Writing—Review & Editing, Supervision, Project administration, AB: Project administration, Writing—Review & Editing, Visualization. Funding This research received no external funding. 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 Ahmed K Shahid S Nawaz N Impacts of climate variability and change on seasonal drought characteristics of Pakistan Atmos Res 2018 214 364 374 10.1016/j.atmosres.2018.08.020 Akbas A Freer J Ozdemir H What about reservoirs? Questioning anthropogenic and climatic interferences on water availability Hydrol Process 2020 34 5441 5455 10.1002/hyp.13960 Amjad M Yilmaz MT Yucel I Yilmaz KK Performance evaluation of satellite- and model-based precipitation products over varying climate and complex topography J Hydrol 2020 584 124707 10.1016/j.jhydrol.2020.124707 Arora B Burrus M Newcomer M Differential C-Q analysis: a new approach to inferring lateral transport and hydrologic transients within multiple reaches of a mountainous headwater catchment Front Water 2020 10.3389/frwa.2020.00024 Ashfaq M Bowling LC Cherkauer K Influence of climate model biases and daily-scale temperature and precipitation events on hydrological impacts assessment: a case study of the United States J Geophys Res 2010 115 D14116 10.1029/2009JD012965 Beecham S Rashid M Chowdhury RK Statistical downscaling of multi-site daily rainfall in a South Australian catchment using a generalized linear model Int J Climatol 2014 34 3654 3670 10.1002/joc.3933 Bettolli ML Penalba OC Statistical downscaling of daily precipitation and temperatures in southern La Plata Basin Int J Climatol 2018 38 3705 3722 10.1002/joc.5531 Boe J Terray L Habets F Martin E Statistical and dynamical downscaling of the Seine basin climate for hydro-meteorological studies Int J Climatol 2007 27 1643 1655 10.1002/joc.1602 Burnham KP Anderson DR Huyvaert KP AIC model selection and multimodel inference in behavioral ecology: some background, observations, and comparisons Behav Ecol Sociobiol 2011 65 23 35 10.1007/s00265-010-1029-6 Chen S-T Yu P-S Tang Y-H Statistical downscaling of daily precipitation using support vector machines and multivariate analysis J Hydrol 2010 385 13 22 10.1016/j.jhydrol.2010.01.021 Chen J Brissette FP Leconte R Assessing regression-based statistical approaches for downscaling precipitation over North America Hydrol Process 2014 28 3482 3504 10.1002/hyp.9889 Crane RG Hewitson BC Doubled CO2 precipitation changes for the Susquehanna Basin: down-scaling from the Genesis general circulation model Int J Climatol 1998 18 65 76 10.1002/(SICI)1097-0088(199801)18:1<65::AID-JOC222>3.0.CO;2-9 Crawford T Betts N Favis-Mortlock D GCM grid-box choice and predictor selection associated with statistical downscaling of daily precipitation over Northern Ireland Clim Res 2007 34 145 160 10.3354/cr034145 Das J Nanduri UV Assessment and evaluation of potential climate change impact on monsoon flows using machine learning technique over Wainganga River basin, India Hydrol Sci J 2018 63 1020 1046 10.1080/02626667.2018.1469757 de Borges PA Barfus K Weiss H Bernhofer C Extended predictor screening, application and added value of statistical downscaling of a CMIP5 ensemble for single-site projections in Distrito Federal, Brazil Int J Climatol 2017 37 46 65 10.1002/joc.4686 Diffenbaugh NS Giorgi F Climate change hotspots in the CMIP5 global climate model ensemble Clim Change 2012 114 813 822 10.1007/s10584-012-0570-x 24014154 DSI (2020) General directorate of state hydraulic works. https://www.dsi.gov.tr/. Accessed 03 Mar 2022 Ekstrom M Grose MR Whetton PH An appraisal of downscaling methods used in climate change research Wiley Interdiscip Rev Clim Chang 2015 6 301 319 10.1002/wcc.339 Fistikoglu O Okkan U Statistical downscaling of monthly precipitation using NCEP/NCAR reanalysis data for Tahtali River Basin in Turkey J Hydrol Eng 2011 16 157 164 10.1061/(ASCE)HE.1943-5584.0000300 Fowler HJ Blenkinsop S Tebaldi C Linking climate change modelling to impacts studies: recent advances in downscaling techniques for hydrological modelling Int J Climatol 2007 27 1547 1578 10.1002/joc.1556 Friedman JH Multivariate adaptive regression splines Ann Stat 1991 19 1 67 Frost AJ Charles SP Timbal B A comparison of multi-site daily rainfall downscaling techniques under Australian conditions J Hydrol 2011 408 1 18 10.1016/j.jhydrol.2011.06.021 Gao L Schulz K Bernhardt M Statistical downscaling of ERA-Interim forecast precipitation data in complex terrain using LASSO algorithm Adv Meteorol 2014 2014 472741 10.1155/2014/472741 GFA (2022) Guide to the forecasts and analyses. http://www.uni-koeln.de/math-nat-fak/geomet/meteo/winfos/ncep/fcstkey.html Accessed 07 Jun 2022 GMAFM (2022) Guide to the Model Analysis and Forecast Maps. http://wxmaps.org/fcstkey.php Accessed 07 Jun 2022 Gupta HV Kling H Yilmaz KK Martinez GF Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling J Hydrol 2009 377 80 91 10.1016/j.jhydrol.2009.08.003 Gutowski WJ Decker SG Donavon RA Temporal–spatial scales of observed and simulated precipitation in central U.S. climate J Clim 2003 16 3841 3847 10.1175/1520-0442(2003)016<3841:TSOOAS>2.0.CO;2 Haltas I Yildirim E Oztas F Demir I A comprehensive flood event specification and inventory: 1930–2020 Turkey case study Int J Disaster Risk Reduct 2021 56 102086 10.1016/j.ijdrr.2021.102086 He RR Chen Y Huang Q Kang Y LASSO as a tool for downscaling summer rainfall over the Yangtze River Valley Hydrol Sci J 2019 64 92 104 10.1080/02626667.2019.1570210 Herrera S Kotlarski S Soares PMM Uncertainty in gridded precipitation products: influence of station density, interpolation method and grid resolution Int J Climatol 2019 39 3717 3729 10.1002/joc.5878 Hersbach H Bell B Berrisford P The ERA5 global reanalysis Q J R Meteorol Soc 2020 146 1999 2049 10.1002/qj.3803 Hertig E Jacobeit J A novel approach to statistical downscaling considering nonstationarities: application to daily precipitation in the mediterranean area J Geophys Res Atmos 2013 118 520 533 10.1002/jgrd.50112 Hertig E Tramblay Y Regional downscaling of Mediterranean droughts under past and future climatic conditions Glob Planet Change 2017 151 36 48 10.1016/j.gloplacha.2016.10.015 Hessami M Gachon P Ouarda TBMJ Automated regression-based statistical downscaling tool Environ Model Softw 2008 23 813 834 10.1016/j.envsoft.2007.10.004 Hou Y-K Chen H Xu C-Y Coupling a markov chain and support vector machine for at-site downscaling of daily precipitation J Hydrometeorol 2017 18 2385 2406 10.1175/JHM-D-16-0130.1 Huang J Zhang J Zhang Z Estimation of future precipitation change in the Yangtze River basin by using statistical downscaling method Stoch Environ Res Risk Assess 2011 25 781 792 10.1007/s00477-010-0441-9 IPCC (2013) Climate change 2013: The physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change [Stocker, T.F., D. Qin, G.-K. Plattner, et al.]. Cambridge, United Kingdom and New York, NY, USA, IPCC (2021) Climate change 2021: The physical science basis. Contribution of working group I to the sixth assessment report of the intergovernmental panel on climate change [Masson-Delmotte, V., P. Zhai, A. Pirani, et al.]. Cambridge, United Kingdom Jafarzadeh A Pourreza-Bilondi M Khashei Siuki A Ramezani Moghadam J Examination of various feature selection approaches for daily precipitation downscaling in different climates Water Resour Manag 2021 35 407 427 10.1007/s11269-020-02701-6 Johnson F Sharma A What are the impacts of bias correction on future drought projections? J Hydrol 2015 525 472 485 10.1016/j.jhydrol.2015.04.002 Keupp L Hertig E Kaspar-Ott I Weighted multi-model ensemble projection of extreme precipitation in the Mediterranean region using statistical downscaling Theor Appl Climatol 2019 138 1269 1295 10.1007/s00704-019-02851-7 Khan MS Coulibaly P Dibike Y Uncertainty analysis of statistical downscaling methods J Hydrol 2006 319 357 382 10.1016/j.jhydrol.2005.06.035 Khan N Shahid S Ahmed K Selection of GCMs for the projection of spatial distribution of heat waves in Pakistan Atmos Res 2020 233 104688 10.1016/j.atmosres.2019.104688 Komuscu AU Celik S Analysis of the Marmara flood in Turkey, 7–10 September 2009: an assessment from hydrometeorological perspective Nat Hazards 2013 66 781 808 10.1007/s11069-012-0521-x Komuscu AU Celik S Ceylan A Rainfall analysis of the flood event that occurred in Marmara Region on 8–12 September 2009 Turk J Geo Sci 2011 9 209 220 10.1501/Cogbil_0000000125 Kooperberg C Bose S Stone CJ Polychotomous regression J Am Stat Assoc 1997 92 117 127 10.1080/01621459.1997.10473608 Krause P Boyle DP Base F Comparison of different efficiency criteria for hydrological model assessment Adv Geosci 2005 5 89 97 10.5194/adgeo-5-89-2005 Kumar YP Maheswaran R Agarwal A Sivakumar B Intercomparison of downscaling methods for daily precipitation with emphasis on wavelet-based hybrid models J Hydrol 2021 599 126373 10.1016/j.jhydrol.2021.126373 Lamontagne JR Barber CA Vogel RM Improved estimators of model performance efficiency for skewed hydrologic data Water Resour Res 2020 56 1 25 10.1029/2020WR027101 Lin GF Chang MJ Wang CF A novel spatiotemporal statistical downscaling method for hourly rainfall Water Resour Manag 2017 31 3465 3489 10.1007/s11269-017-1679-5 Liu PR Raftery AE Country-based rate of emissions reductions should increase by 80% beyond nationally determined contributions to meet the 2 °C target Commun Earth Environ 2021 2 29 10.1038/s43247-021-00097-8 33899003 Liu W Fu G Liu C Charles SP A comparison of three multi-site statistical downscaling models for daily rainfall in the North China Plain Theor Appl Climatol 2013 111 585 600 10.1007/s00704-012-0692-0 Liu Y Feng J Shao Y Li JL Identify optimal predictors of statistical downscaling of summer daily precipitation in China from three-dimensional large-scale variables Atmos Res 2019 224 99 113 10.1016/j.atmosres.2019.03.022 Loucks DP, van Beek E (2017) Water resource systems planning and management. Springer International Publishing, Cham,10.1007/978-3-319-44234-1 Maina FZ Siirila-Woodburn ER Newcomer M Determining the impact of a severe dry to wet transition on watershed hydrodynamics in California, USA with an integrated hydrologic model J Hydrol 2020 580 124358 10.1016/j.jhydrol.2019.124358 Mallows CL Some comments on Cp Technometrics 1973 15 661 675 10.1080/00401706.1973.10489103 Maraun D Bias correcting climate change simulations—a critical review Curr Clim Chang Reports 2016 2 211 220 10.1007/s40641-016-0050-x Maraun D Wetterhall F Ireson AM Precipitation downscaling under climate change: recent developments to bridge the gap between dynamical models and the end user Rev Geophys 2010 48 RG3003 10.1029/2009RG000314 MTA (2021) Mineral research and exploration general directorate. http://www.mta.gov.tr/eng/. Accessed 03 Mar 2022 Mujumdar PP Nagesh Kumar D Floods in a changing climate: Hydrologic modeling 2012 Cambridge Cambridge University Press Nacar S Kankal M Okkan U Evaluation of the suitability of NCEP/NCAR, ERA-Interim and ERA5 reanalysis data sets for statistical downscaling in the Eastern Black Sea Basin Turkey Meteorol Atmos Phys 2022 134 39 10.1007/s00703-022-00878-6 Nahar J Johnson F Sharma A Assessing the extent of nonstationary biases in GCMs J Hydrol 2017 549 148 162 10.1016/j.jhydrol.2017.03.045 Nasseri M Tavakol-Davani H Zahraie B Performance assessment of different data mining methods in statistical downscaling of daily precipitation J Hydrol 2013 492 1 14 10.1016/j.jhydrol.2013.04.017 NCSS (2021) All Possible Regressions. https://ncss-wpengine.netdna-ssl.com/wp-content/themes/ncss/pdf/Procedures/NCSS/All_Possible_Regressions.pdf . Accessed 03 Mar 2022 Newcomer ME Bouskill NJ Wainwright H Hysteresis patterns of watershed nitrogen retention and loss over the past 50 years in United States hydrological basins Global Biogeochem Cycles 2021 10.1029/2020GB006777 Nguyen H Mehrotra R Sharma A Assessment of climate change impacts on reservoir storage reliability, resilience, and vulnerability using a multivariate frequency bias correction approach Water Resour Res 2020 56 e2019WR026022 10.1029/2019WR026022 Noor M Bin IT Ullah S A non-local model output statistics approach for the downscaling of CMIP5 GCMs for the projection of rainfall in peninsular Malaysia J Water Clim Chang 2020 11 944 955 10.2166/wcc.2019.041 Okkan U Inan G Statistical downscaling of monthly reservoir inflows for Kemer watershed in Turkey: use of machine learning methods, multiple GCMs and emission scenarios Int J Climatol 2015 35 3274 3295 10.1002/joc.4206 Okkan U Kirdemir U Downscaling of monthly precipitation using CMIP5 climate models operated under RCPs Meteorol Appl 2016 23 514 528 10.1002/met.1575 Ozturk MZ Cetinkaya G Aydin S Climate types of Turkey according to Köppen-Geiger climate classification J Geog 2017 35 17 27 Ozturk MZ Comparative climate of Uludağ (Zirve) and Bursa meteorology stations Turkish Geogr Rev 2010 55 13 24 Pardoe I Regression model building II Applied Regression Modeling 2013 Hoboken, NJ, USA John Wiley & Sons Inc 189 242 Peel MC Finlayson BL McMahon TA Updated world map of the Köppen-Geiger climate classification Hydrol Earth Syst Sci 2007 11 1633 1644 10.5194/hess-11-1633-2007 Pour S Harun S Shahid S Genetic programming for the downscaling of extreme rainfall events on the east coast of Peninsular Malaysia Atmosphere (Basel) 2014 5 914 936 10.3390/atmos5040914 Quesada-Chacon D Barfus K Bernhofer C Climate change projections and extremes for Costa Rica using tailored predictors from CORDEX model output through statistical downscaling with artificial neural networks Int J Climatol 2021 41 211 232 10.1002/joc.6616 Ramsar (2021) Ramsar sites information service. https://www.ramsar.org/wetland/turkey. Accessed 12 Dec 2021 Rashid MM Beecham S Chowdhury RK Statistical downscaling of CMIP5 outputs for projecting future changes in rainfall in the Onkaparinga catchment Sci Total Environ 2015 530–531 171 182 10.1016/j.scitotenv.2015.05.024 Rudd AC Kay AL Use of very high resolution climate model data for hydrological modelling: estimation of potential evaporation Hydrol Res 2016 47 660 670 10.2166/nh.2015.028 Saengsawang S Pankhao P Kaprom C Sriwongsitanon N Projections of future rainfall for the upper Ping River Basin using regression-based downscaling Adv Clim Chang Res 2017 8 256 267 10.1016/j.accre.2017.11.001 Salman SA Shahid S Ismail T Selection of climate models for projection of spatiotemporal changes in temperature of Iraq with uncertainties Atmos Res 2018 213 509 522 10.1016/j.atmosres.2018.07.008 Sarhadi A Burn DH Johnson F Water resources climate change projections using supervised nonlinear and multivariate soft computing techniques J Hydrol 2016 536 119 132 10.1016/j.jhydrol.2016.02.040 Sen Z Up-to-date statistical essentials in climate change and hydrology: a review Int J Glob Warm 2020 22 392 10.1504/IJGW.2020.111515 Sen Z Model efficiency performance assessment through a standard triangular diagram (STD) Model Earth Syst Environ 2020 10.1007/s40808-020-00993-6 Shiru MS Shahid S Chung E-S Alias N Changing characteristics of meteorological droughts in Nigeria during 1901–2010 Atmos Res 2019 223 60 73 10.1016/j.atmosres.2019.03.010 Sippel S Otto FEL Forkel M A novel bias correction methodology for climate impact simulations Earth Syst Dyn 2016 7 71 88 10.5194/esd-7-71-2016 STAT 462 (2021) Best subsets regression, adjusted R-Sq, Mallows Cp. https://online.stat.psu.edu/stat462/node/197/. Accessed 03 Mar 2022 Stone CJ Hansen MH Kooperberg C Truong YK Polynomial splines and their tensor products in extended linear modeling Ann Stat 1997 25 1371 1425 10.1214/aos/1031594728 SYGM (2018) Susurluk Basin flood management plan. https://www.tarimorman.gov.tr/SYGM/Belgeler/Taşkın Yönetim Planları/4) SUSURLUK HAVZASI TAŞKIN YONETIM PLANI.pdf. Accessed 03 Mar 2022 Tavakol-Davani H Nasseri M Zahraie B Improved statistical downscaling of daily precipitation using SDSM platform and data-mining methods Int J Climatol 2013 33 2561 2578 10.1002/joc.3611 Themessl MJ Gobiet A Heinrich G Empirical-statistical downscaling and error correction of regional climate models and its impact on the climate change signal Clim Change 2012 112 449 468 10.1007/s10584-011-0224-4 Tollefson J COVID curbed carbon emissions in 2020—but not by much Nature 2021 589 343 343 10.1038/d41586-021-00090-3 33452515 Troin M Velázquez JA Caya D Brissette F Comparing statistical post-processing of regional and global climate scenarios for hydrological impacts assessment: a case study of two Canadian catchments J Hydrol 2015 520 268 288 10.1016/j.jhydrol.2014.11.047 TSMS (2021) Turkish State Meteorological Service. https://mgm.gov.tr/. Accessed 03 Mar 2022 Turkes M Influence of geopotential heights, cyclone frequency and southern oscillation on rainfall variations in Turkey Int J Climatol 1998 18 649 680 10.1002/(SICI)1097-0088(199805)18:6<649::AID-JOC269>3.0.CO;2-3 Wilby R Statistical downscaling of daily precipitation using daily airflow and seasonal teleconnection indices Clim Res 1998 10 163 178 10.3354/cr010163 Wilby RL Wigley TML Precipitation predictors for downscaling: observed and general circulation model relationships Int J Climatol 2000 20 641 661 10.1002/(SICI)1097-0088(200005)20:6<641::AID-JOC501>3.0.CO;2-1 Wilby RL Wigley TML Conway D Statistical downscaling of general circulation model output: a comparison of methods Water Resour Res 1998 34 2995 3008 10.1029/98WR02577 Wilby RL Dawson CW Barrow EM SDSM - A decision support tool for the assessment of regional climate change impacts Environ Model Softw 2002 17 145 157 10.1016/s1364-8152(01)00060-3 Wilby RL, Charles SP, Zorita E, et al (2004) Guidelines for use of climate scenarios developed from statistical downscaling methods. http://www.ctn.etsmtl.ca/cours/mgc921/dgm_no2_v1_09_2004.pdf. Accessed 03 Mar 2022 Yang C Wang N Wang S Zhou L Performance comparison of three predictor selection methods for statistical downscaling of daily precipitation Theor Appl Climatol 2018 131 43 54 10.1007/s00704-016-1956-x Yilmaz B Aras E Nacar S Kankal M Estimating suspended sediment load with multivariate adaptive regression spline, teaching-learning based optimization, and artificial bee colony models Sci Total Environ 2018 639 826 840 10.1016/j.scitotenv.2018.05.153 29803053	0	PMC9734427	NO-CC CODE	2022-12-14 23:28:28	no	Stoch Environ Res Risk Assess. 2022 Dec 4;:1-25	utf-8	Stoch Environ Res Risk Assess	2,022	10.1007/s00477-022-02345-5	oa_other
==== Front Sport Sci Health Sport Sci Health Sport Sciences for Health 1824-7490 1825-1234 Springer Milan Milan 1013 10.1007/s11332-022-01013-z Research Comparison between the effects of tibialis posterior versus fibularis longus Kinesio taping on foot posture, physical performance, and dynamic balance in young women with flexible flatfoot http://orcid.org/0000-0001-6745-6414 Tahmasbi Alireza [email protected] http://orcid.org/0000-0002-1186-9997 Shadmehr Azadeh [email protected] http://orcid.org/0000-0002-0184-8930 Attarbashi Moghadam Behrouz [email protected] http://orcid.org/0000-0002-9827-3108 Fereydounnia Sara [email protected] grid.411705.6 0000 0001 0166 0922 Physical Therapy Department, School of Rehabilitation, Tehran University of Medical Sciences, Tehran, Iran 6 12 2022 18 8 9 2022 30 9 2022 © The Author(s), under exclusive licence to Springer-Verlag Italia S.r.l., 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. Purpose The purpose of the study is to compare the effectiveness of the tibialis posterior Kinesio taping and fibularis longus Kinesio taping on the foot posture, physical performance, and dynamic balance in young women with flexible flatfoot. Methods Twenty-four subjects were recruited for the study. They were randomly divided into groups (A = 12, B = 12). In group A, Kinesio taping was applied on the tibialis posterior, and in group B, Kinesio taping was applied on the fibularis longus and remained for 30 min. Outcome measures were the navicular drop test (NDT), foot posture index (FPI), timed up and go (TUG) test, and Y-balance test. The pre- and post-treatment results were compared for each group; between-group differences were determined as well. Results For group A, NDT, FPI, and TUG test changed significantly (P = 0.01, P = 0.001, P = 0.006, respectively). For group B, the FPI score decreased (P = 0.03), and the Y-balance test in the anterior direction improved significantly (P = 0.01). Any variables have not shown a significant difference between groups (P > 0.05). Conclusion Kinesio taping of the tibialis posterior and fibularis longus can improve foot posture in young women with flexible flatfoot. Also, physical performance and dynamic balance improved by Kinesio taping of the tibialis posterior and the fibularis longus, respectively. In addition to the tibialis posterior, we found that the fibularis longus muscle can be considered a therapeutic target for managing flexible flatfoot in healthy young women. Keywords Flatfoot Kinesio tape Posture Physical performance http://dx.doi.org/10.13039/501100004484 Tehran University of Medical Sciences and Health Services #1400-3-103-55363 Shadmehr Azadeh ==== Body pmcIntroduction Flatfoot, also referred to as “pes planus” or “pronated foot,” is a common orthopedic condition that consists of reduced height of the medial longitudinal arch (MLA), which absorbs shocks and provides stability during dynamic activities [1]. Approximately it is reported that 5–20% of different populations, including healthy subjects and athletes, have this condition [2–4]. Two types of flatfoot are mentioned in scientific literature based on the consistency of the MLA: (1) flexible and (2) rigid. The MLA exists without weight-bearing in the flexible type, but it disappears under loading conditions, like standing. In the rigid type, there is no MLA with and without weight-bearing [5]. MLA is supported by both passive structures, like plantar fascia and plantar ligament, and active structures, like extrinsic and intrinsic foot muscles [6]. Several factors are considered causes of flexible flatfoot, such as ligamentous laxity, muscle weakness, obesity, bone deformity, and family history [7]. However, there is no consensus on the exact cause of this condition. Flexible flatfoot predisposes to other conditions, such as low back pain, patellofemoral pain syndrome, Achilles tendinopathy, plantar fasciitis, medial tibial stress syndrome, and knee osteoarthritis [8, 9]; so, it would be essential to be considered in the assessment and causative management of these musculoskeletal conditions as well. Furthermore, healthy people and athletes should notice foot and ankle deformations like the flexible flatfoot to prevent daily and sports injuries. Kinesio taping, an approach that has gained popularity in recent years among practitioners, is a noninvasive therapy that uses elastic bands with specific textures to treat, modify, and prevent numerous musculoskeletal conditions. Several mechanisms have been proposed to explain its effectiveness. From a neurophysiological perspective, it can increase the activation of mechanical skin receptors, leading to a reflexive contraction of muscle spindles and an increase in the sensitivity of motor units [10]. Moreover, it can increase the blood circulation of muscles by stimulating the autonomic system and inducing peripheral vasodilation [11]. According to previous studies, Kinesio tape can affect foot posture and muscle activity [12]; moreover, it can alter foot pressure, range of motion, and pain perception in people with flexible flatfoot [5, 13]. Previous studies have shown the positive effects of tibialis posterior (TP) facilitation on the foot posture and the dynamic balance in people with the flexible flatfoot [12, 14]. Moreover, in 2020, Sumal et al. stated that the fibularis longus (FL) tendon contributes to the MLA height, and physical therapy interventions targeting this muscle can lead to positive effects on the foot posture [6]; so, we hypothesized that using a facilitatory technique by means of Kinesio tape could result in the improved activity of these muscles. Also, it should be noted that some physiological characteristics, such as soft tissue flexibility, differ among genders, so it should be considered an influential factor in the results of therapeutic interventions on musculoskeletal deformities. Therefore, this study aims to compare the results of Kinesio taping of two important muscles of the foot, the tibialis posterior and fibularis longus, and their effects on foot posture, physical performance, and dynamic balance in young women with flexible flatfoot. Materials and methods Study design This study is a single-blind, parallel-design, randomized clinical trial. It is approved by the ethics committee of Tehran University of Medical Sciences with the approval identification of IR.TUMS.MEDICINE.REC.1400.771 and is registered in the Iranian Registry of Clinical Trials with the registration code IRCT20211018052805N. Participants In this study, 24 women were included. The sample size was calculated using GPower software. Power was considered 85%, the alpha value was 0.05, and the effect size was estimated using foot posture index (FPI) measurements in previous studies [15]. Inclusion criteria were: (1) 18–40-year-old people; (2) navicular drop test (NDT) ≥ 10 mm; (3) foot posture index (FPI) ≥ + 6; (4) positive Jack’s test; and (5) body mass index (BMI) ≤ 25 kg/m2. Exclusion criteria were: (1) ankle injury in the last six months; (2) history of foot surgery; (3) foot injury due to systemic, inflammatory, and infectious diseases; (4) foot deformities including hallux valgus, hammer toe, and claw toe; (5) pregnancy; (6) ankle pain during the study; (7) static standing and walking problems; (8) Kinesio tape sensitivity; and (9) refuse to participate in the study or not adhering to treatment anymore. All subjects were informed about the study procedures and signed the consent form prior to the study (Fig. 1).Fig. 1 CONSORT flowchart of the participants Assessment Procedure All of the subjects were asked to fill out a demographic questionnaire. Jack’s test was used to determine the flexibility of the MLA [5]. Indeed, the type of the flatfoot was examined, while the assessor passively dorsiflexed the 1st metatarsophalangeal joint of the subject in a standing position. MLA was considered flexible if the curve appeared, so the test was positive. If the curve did not appear, it showed a rigid flatfoot, and the test was considered negative. Afterward, NDT and FPI were assessed, and the eligible subjects were included in the study. Next, they were asked to do a timed up and go (TUG) test for their physical performance assessment. Finally, subjects performed a Y-balance test to be assessed for their dynamic balance. Three trials were conducted with 2 min rest intervals between each trial. All of these assessments were reassessed 30 min after intervention. The left leg was selected for the assessments and intervention in all subjects to avoid selection bias [16]. Navicular drop test For this test, the subject sat on a suitable chair with her hips and knees at 90 °, both feet on the ground, and subtalar joint in a neutral position. Navicular tuberosity was found and marked with a removable marker. The assessor marked the height of navicular tuberosity from the ground on an index card. Then, they were asked to stand and put equal weight on both feet. Again, similar measurements were done [17]. The difference in navicular heights in both situations was measured using a ruler. This test was repeated three times, and the average value was calculated. NDT has shown good reliability and validity in previous studies [18]. Foot posture index A 6-item version of the foot posture index was used to assess foot posture while standing [19]. Foot postures are classified based on a scoring system for each portion of the foot from − 2 to + 2. Summation of the scores determines the type of the foot as pronated (FPI ≥ + 6), neutral (0 ≤ FPI ≤ + 5), and supinated (FPI < 0). To determine, the subject was asked to stand with the arms at the side while looking straightforward. The assessor scored each item using a specific checklist. Timed up and go test In this test, the subject stood from a suitable chair and began to walk three meters at a usual speed and then returned and walked through the chair and sat on it [20]. This time was recorded by a stopwatch. Y-balance test Dynamic balance is assessed with the Y-balance test [21]. First, the assessor showed the correct form of performing the tasks. Then, the subjects were asked to complete three maneuvers in each direction (anterior, posteromedial, and posterolateral direction) to avoid the learning effect. Next, three trials were conducted with 2-min rest intervals between them. If any of these errors have occurred, the test was repeated: (1) Subject was not able to stand on one leg; (2) stance leg was moved; (3) reaching foot touched the ground due to loss of balance; and (4) subject could not return to the starting position. The average reaching distance was calculated in each direction, divided by limb length, and multiplied by one hundred (%). To measure the limb length, the subject lay supine. After correcting the pelvic position, a tape measure was placed from the anterior superior iliac spine (ASIS) to the most inferior part of the medial malleolus [14]. Randomization Two concealed envelopes containing blue and red pieces of cardboard were placed in front of the subjects, and they were asked to choose one of them. If they chose the blue or red one, they were assigned to the tibialis posterior Kinesio taping group or the fibularis longus Kinesio taping group, respectively. Subjects were blinded to the group in which they would be allocated. Intervention Tibialis posterior Kinesio taping (group a) For this purpose, the subject lay supine and actively dorsiflexed and everted her ankle. Kinesio tape was applied from half of the tibia bone, passing behind the medial malleolus, and finished at the head of the fifth metatarsal bone [12] (Fig. 2).Fig. 2 Tibialis posterior Kinesio taping Fibularis longus Kinesio taping (group b) The subject was asked to lie supine and actively plantarflex and invert her ankle. Kinesio tape was applied from the head of the fibula, passing behind the lateral malleolus, and finished at the base of the first metatarsal bone [22] (Fig. 3).Fig. 3 Fibularis longus Kinesio taping Using a facilitatory technique, an I-shaped red piece of Kinesio tape (TEMTEX, South Korea) with 35 percent tension was used for both groups [23]. While the subject had returned the muscle to the rest position, the anchor was applied without tension. Subjects rested in a comfortable position for 30 min [23, 24]. Afterward, they were reassessed. All assessments and interventions were performed by a single physiotherapist who was certified in the Kinesio tape method. Statistical analysis SPSS statistics version 24 was used for statistical analysis. The Kolmogorov–Smirnov test was used to assess the normal distribution of data. For intra-group comparison, paired samples T test and Wilcoxon test were used for parametric and nonparametric statistics, respectively. For intergroup comparison, independent T test and Mann–Whitney test were used for parametric and nonparametric statistics. The significance level (P value) was considered less than 0.05. Results Sixty participants were screened, and of them, 36 participants did not meet the inclusion criteria or were excluded. Finally, 24 participants were eligible and enrolled in the study. The subjects did not report any adverse effects during and after the study. The demographic characteristics of the subjects are shown in Table 1. The Kolmogorov–Smirnov (K-S) test showed that the distribution of the weight and BMI was not normal in group A (P = 0.03, and P = 0.04, respectively). Also, K-S test showed that the distribution of the NDT values was not normal in both groups (P = 0.01 and P = 0.03 for groups A and B, respectively), so we used the related nonparametric tests. There was no significant difference between the groups regarding the demographic characteristics (P > 0.05). According to the results, NDT, FPI, and TUG test changed significantly in group A (P = 0.01, P = 0.001, P = 0.006, respectively) (Table 2). For group B, the FPI score decreased (P = 0.03), and the Y-balance test in the anterior direction improved significantly (P = 0.01) (Table 2). Navicular drop decreased after the intervention in group B, but it was not statistically significant (P = 0.09). Any variables have not shown a significant difference between groups (Table 3); however, FPI improved better in group A (P = 0.05, mean difference = − 1.08).Table 1 Demographic characteristics in two groups Variables Group A (n = 12) Group B (n = 12) P value t (z) Power Age (years) 22.25 ± 2.52 24.58 ± 6.02 0.23 − 1.23 0.23 Height (m) 1.63 ± 0.05 1.62 ± 0.04 0.68 0.41 0.08 Weight (kg) a 60 ± 6.98 59 ± 7.36 0.81 − 0.23 0.05 BMI (kg/m2) a 22.42 ± 2.00 22.29 ± 2.46 0.86 − 0.17 0.03 Values are shown as mean ± standard deviation aFor weight and BMI, the Mann–Whitney test was used due to a lack of normal distribution Table 2 Before–after comparison of variables in group A and group B Variable Group A Group B Before mean ± SD After mean ± SD t (z) P value Effect size Power Before mean ± SD After mean ± SD t (z) P value Effect size Power NDT (mm) 10.96 ± 1.34 8.85 ± 2.03 − 2.39 0.01 1.22 – 11.82 ± 2.39 10.35 ± 2.78 − 1.69 0.09 – 0.28 FPI 8.33 ± 1.30 6.50 ± 1.97 − 4.33 0.001* 1.09 – 8.33 ± 1.49 7.58 ± 1.62 − 2.46 0.03* 0.48 – TUG (s) 10.32 ± 1.10 9.56 ± 1.02 − 3.43 0.006* 0.71 – 10.37 ± 1.34 9.85 ± 1.42 1.82 0.09 – 0.15 YBT (A) (%) 72.54 ± 7.16 72.90 ± 6.79 0.295 0.77 – 0.03 76.46 ± 6.62 79.52 ± 6.92 3.03 0.01* 0.45 – YBT (PM) (%) 101.50 ± 15.21 100.90 ± 16.24 − 0.48 0.63 – 0.03 100.80 ± 7.13 103.85 ± 7.71 1.91 0.08 – 0.17 YBT (PL) (%) 88.91 ± 14.90 91.13 ± 16.55 1.65 0.12 – 0.05 87.79 ± 7.43 90.35 ± 8.92 1.52 0.15 – 0.12 SD standard deviation, NDT navicular drop test, FPI foot posture index, TUG timed up and go, YBT (A, PM, PL) Y-balance test (anterior, posteromedial, and posterolateral) aFor NDT, the Wilcoxon test was used due to a lack of normal distribution *P < 0.05 is significant Table 3 Intergroup comparison of variables between groups A and B Variable Mean difference ± SD (mean rank difference) P value t (z) Confidence interval a (sum of ranks) Power Lower Upper NDT b(mm) − 2.08 0.47 − 0.72 137.50 162.50 0.11 FPI − 1.08 ± 0.52 0.05 − 2.07 − 2.16 − 0.002 0.54 TUG (s) − 0.23 ± 0.35 0.52 − 0.65 − 0.97 0.51 0.10 YBT (A) (%) − 2.69 ± 1.59 0.10 − 1.68 − 6.00 0.61 0.39 YBT (PM) (%) − 3.64 ± 2.01 0.08 − 1.81 − 7.82 0.53 0.44 YBT (PL) (%) − 0.33 ± 2.15 0.87 − 0.15 − 4.80 4.12 0.04 SD standard deviation, NDT navicular drop test, FPI foot posture index, TUG timed up and go, YBT (A, PM, PL)Y-balance test (anterior, posteromedial, and posterolateral), a95% confidence interval of the difference bFor NDT, the Mann–Whitney test was used due to a lack of normal distribution Discussion The study investigated the immediate effects of Kinesio taping on two important muscles of the leg, the tibialis posterior and the fibularis longus. The results showed that the foot posture was improved in both groups, though the navicular drop decreased significantly just with the tibialis posterior Kinesio taping. This finding is consistent with the study of Siu et al. in 2019 on runners with flexible flatfoot [12]. They showed that the Kinesio taping of the TP and the transverse arch could immediately reduce the NDT. Since the TP activity provides dynamic stability for the MLA due to its line of action [25], it is possible that the increased activity of this muscle, and the following force generation capacity, is achievable as an immediate effect of Kinesio taping, and hence, we can expect the improved arch height. Moreover, since the TP Kinesio taping procedure passes over the medial side of the foot and covers the navicular bone, it is possible that it can provide mechanical support to the navicular bone and the MLA. However, in 2012, Roman et al. did not find any significant effect of TP Kinesio taping on rear foot pronation 24 h after the application, which questions the prolonged effect of TP Kinesio taping on foot posture [26]. In 2020, Sumal et al. stated that the fibularis longus tendon contributes to the MLA height, and physical therapy interventions targeting this muscle can lead to positive effects on foot posture [6]. Our results showed that the overall foot posture improved with FL Kinesio taping. Although there was no significant difference in FPI values between the two groups, the results were so close to being meaningful in favor of group A. Aguilar et al. in 2015 suggested that Kinesio taping using low-dye technique can lead to better result of FPI compared to the sham Kinesio taping in the pronated foot of amateur runners after 45 min of running [16]. It should be noticed that the mechanical correction of the low-dye technique was more than our method since it used 75% tension of Kinesio tape. Y-balance test in the anterior direction increased significantly in group B. As an ankle evertor, the fibularis longus provides stability in the ankle joint [27], and it is a primary muscle for balance maintenance. Our findings were consistent with the results of Fereydounnia et al. in 2019 in which they investigated the effects of fibularis longus Kinesio taping on dynamic balance in soccer players with and without ankle instability [23]. They showed that Kinesio tape could immediately affect the dynamic balance of subjects. In 2016, Correia et al. found that the Kinesio taping of the FL has no immediate effect on the static balance in young healthy subjects, though the applied tension was only 10%. Our study showed that the FL Kinesio taping with 30% tension can lead to better results in the dynamic balance [22]; however, it is unclear if there is a significant relationship between the static balance and dynamic balance with the same applied tension of the Kinesio tape. The timed up and go test decreased significantly in group A; it might happen due to the improved support of MLA during the load transfer in the gait mechanism as a result of TP facilitation. Moreover, Siu et al. found that Kinesio taping of TP and transverse arch can increase the muscle activity of the tibialis anterior during running, which is an important muscle in the propulsion phase of locomotion [12, 28]. For group B, the result was not significant, which is consistent with the study of Fereydounnia et al. in 2021 on soccer players with and without ankle instability. In that study, Kinesio taping was applied in order to facilitate FL, and gait initiation parameters were measured with the force plate [29]. To our knowledge, it is the first study that has investigated the TUG test as a functional assessment for the physical performance of people with the flexible flatfoot. EMG studies have shown the altered electromyographic activity of TP and FL in the gait mechanism of flatfeet people [30]. As mentioned before, several theories have been explained for the mechanisms of Kinesio tape effectiveness. It can increase blood flow through the autonomic system [11], facilitate muscle activity by increasing the sensory inputs of skin and joint receptors [10], and adjust the length–tension relationship and, subsequently, the force generation capacity of muscles [31]. All of these changes could lead to better function of the target muscles in this study. To the authors’ knowledge, it is the first study that investigated the effects of Kinesio taping on the fibularis longus muscle in flexible flatfoot, so this effect on the mentioned outcome measures can be further investigated in future studies. For example, it is suggested that the prolonged effects of FL Kinesio taping plus the comparison to the control group be addressed in future studies. One limitation of our study is that the subjects were young females, and the results are limited to this population. Moreover, just immediate effects of the intervention were studied, and the long-term effects remained unclear. Also, the lack of the control group disabled us from deducing consistently from the present findings. Another limitation is the small sample size of the study. Due to the COVID-19 pandemic, it was not possible to enlarge the sample size. Conclusion Kinesio taping of the tibialis posterior can improve the foot posture and physical performance of young women with flexible flatfoot. Additionally, Kinesio taping of fibularis longus leads to promising results in the foot posture and dynamic balance in this population. No significant differences were shown between the two groups. In addition to the tibialis posterior, it is suggested that the fibularis longus muscle be considered a therapeutic target for the management of flexible flatfoot in healthy young women. Acknowledgements The authors are grateful to the Department of Physical Therapy, School of Rehabilitation, Tehran University of Medical Sciences, for their support and the laboratory space and equipment. Also, the authors appreciate the participants’ contribution to the study. Author contributions Alireza Tahmasbi, Sara fereydounnia, Azadeh Shadmehr, and Behrouz Attarbashi Moghadam conceptualized the research. Azadeh Shadmehr and Sara Fereydounnia supervised the process. Alireza Tahmasbi gathered the data and wrote the main manuscript. Alireza Tahmasbi and Sara Fereydounnia analyzed the data. All authors reviewed and edited the manuscript. Azadeh Shadmehr managed the provided fund. Funding This article is extracted from the MSc Dissertation of the first author in the Department of Physical Therapy, School of Rehabilitation, Tehran University of Medical Sciences, Iran (Grant: #1400–3-103–55363) Data availability This study is original research, and data were collected from subjects via related procedures by the first author in the biomechanical laboratory of the Physical Therapy Department, School of Rehabilitation, Tehran University of Medical Sciences. Declarations Conflict of interest The authors declare no conflict of interest. Ethical approval This study was approved by the ethics committee of Tehran University of Medical Sciences with the approval identification of IR.TUMS.MEDICINE.REC.1400.771. It is also registered in the Iranian Registry of Clinical Trials with the registration code IRCT20211018052805N. 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 All subjects were informed about the study purpose and procedure and signed a consent form before the study. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Birinci T Demirbas SB Relationship between the mobility of medial longitudinal arch and postural control Acta Orthop Traumatol Turc 2017 51 3 233 237 10.1016/j.aott.2016.11.004 28462802 2. Bhoir T Anap DB Diwate A Prevalence of flat foot among 18–25 years old physiotherapy students: cross sectional study I JBasic Appl Med Res 2014 3 4 7 3. Ganapathy A Sadeesh T Rao S Morphometric analysis of foot in young adult individuals W J Pharm Pharmaceutical Sci 2015 4 8 980 993 4. Bhosale N, Nandala P (2021) Prevalance of flexible flat foot in athletes. Kesari Mahratta Trust 1(1):1–13 5. Karthikeyan J Singh K Govind S Mahalingam K Vamsi S Annamalai P To compare the effectiveness of taping and arch support on the flexible flat foot on a random population I J Forensic Med Toxicol 2020 14 4 7825 6. Sumal AS Jarvis GE Norrish AR Brassett C Whitaker RH The role of the angle of the fibularis longus tendon in foot arch support Clin Anat 2021 34 4 651 658 10.1002/ca.23686 32986255 7. Atik A Ozyurek S Flexible flatfootness North Clin Istanb 2014 1 1 57 10.14744/nci.2014.29292 28058304 8. Neal BS Griffiths IB Dowling GJ Murley GS Munteanu SE Franettovich Smith MM Foot posture as a risk factor for lower limb overuse injury: a systematic review and meta-analysis Journal of foot and ankle research 2014 7 1 1 13 10.1186/s13047-014-0055-4 24423411 9. Gross KD Felson DT Niu J Hunter DJ Guermazi A Roemer FW Association of flat feet with knee pain and cartilage damage in older adults Arthritis Care Res 2011 63 7 937 944 10.1002/acr.20431 10. Oliveira AK Borges DT Lins CA Cavalcanti RL Macedo LB Brasileiro JS Immediate effects of Kinesio Taping® on neuromuscular performance of quadriceps and balance in individuals submitted to anterior cruciate ligament reconstruction: a randomized clinical trial J Sci Med Sport 2016 19 1 2 6 10.1016/j.jsams.2014.12.002 25601016 11. Gómez-Soriano J Abián-Vicén J Aparicio-García C Ruiz-Lázaro P Simón-Martínez C Bravo-Esteban E The effects of Kinesio taping on muscle tone in healthy subjects: a double-blind, placebo-controlled crossover trial Man Ther 2014 19 2 131 136 10.1016/j.math.2013.09.002 24829961 12. Siu W-S Shih Y-F Lin H-C Effects of kinesio tape on supporting medial foot arch in runners with functional flatfoot: a preliminary study Res Sports Med 2020 28 2 168 180 10.1080/15438627.2019.1638258 31262193 13. Wang J-S Um G-M Choi J-H Immediate effects of kinematic taping on lower extremity muscle tone and stiffness in flexible flat feet J Phys Ther Sci 2016 28 4 1339 1342 10.1589/jpts.28.1339 27190479 14. Alam F Raza S Moiz JA Bhati P Anwer S Alghadir A Effects of selective strengthening of tibialis posterior and stretching of iliopsoas on navicular drop, dynamic balance, and lower limb muscle activity in pronated feet: a randomized clinical trial Phys Sportsmed 2019 47 3 301 311 10.1080/00913847.2018.1553466 30517043 15. Unver B Erdem EU Akbas E Effects of short-foot exercises on foot posture, pain, disability, and plantar pressure in pes planus J Sport Rehabil 2019 29 4 436 440 10.1123/jsr.2018-0363 30860412 16. Aguilar MB Abián-Vicén J Halstead J Gijon-Nogueron G Effectiveness of neuromuscular taping on pronated foot posture and walking plantar pressures in amateur runners J Sci Med Sport 2016 19 4 348 353 10.1016/j.jsams.2015.04.004 25956688 17. Elataar FF Abdelmajeed SF Abdellatif NM Mohammed MM Core muscles’ endurance in flexible flatfeet: a cross-sectional study J Musculoskelet Neuronal Interact 2020 20 3 404 32877977 18. Zuil-Escobar JC Martínez-Cepa CB Martín-Urrialde JA Gómez-Conesa A Medial longitudinal arch: accuracy, reliability, and correlation between navicular drop test and footprint parameters J Manipulative Physiol Ther 2018 41 8 672 679 10.1016/j.jmpt.2018.04.001 30573198 19. Redmond AC Crosbie J Ouvrier RA Development and validation of a novel rating system for scoring standing foot posture: the foot posture index Clin Biomech 2006 21 1 89 98 10.1016/j.clinbiomech.2005.08.002 20. Larsson BA Johansson L Johansson H Axelsson KF Harvey N Vandenput L The timed up and go test predicts fracture risk in older women independently of clinical risk factors and bone mineral density Osteoporos Int 2021 32 1 75 84 10.1007/s00198-020-05681-w 33089354 21. Kim J-a Lim O-b Yi C-h Difference in static and dynamic stability between flexible flatfeet and neutral feet Gait Posture 2015 41 2 546 550 10.1016/j.gaitpost.2014.12.012 25560044 22. Correia C Lopes S Gonçalves R Torres R Pinho F Gonçalves P Kinesiology taping does not change fibularis longus latency time and postural sway J Bodyw Mov Ther 2016 20 1 132 138 10.1016/j.jbmt.2015.07.037 26891648 23. Fereydounnia S Shadmehr A Moghadam BA Moghadam ST Mir SM Salemi S Improvements in strength and functional performance after kinesio taping in semi-professional male soccer players with and without functional ankle instability Foot 2019 41 12 18 10.1016/j.foot.2019.06.006 31675595 24. Lemos TV Pereira KC Protássio CC Lucas LB Matheus JPC The effect of Kinesio Taping on handgrip strength J Phys Ther Sci 2015 27 3 567 570 10.1589/jpts.27.567 25931682 25. Willegger M Seyidova N Schuh R Windhager R Hirtler L The tibialis posterior tendon footprint: an anatomical dissection study J Foot Ankle Res 2020 13 1 1 7 10.1186/s13047-020-00392-1 31956341 26. Román MF Méndez AC Cabello MA Effects of treatment with Kinesio Tape for flat feet Fisioterapia 2012 34 1 11 15 27. Sarvestan J Ataabadi PA Svoboda Z Kovačikova Z Needle AR The effect of ankle Kinesio™ taping on ankle joint biomechanics during unilateral balance status among collegiate athletes with chronic ankle sprain Phys Ther Sport 2020 45 161 167 10.1016/j.ptsp.2020.06.007 32781269 28. Lee H-S Lee J-H Kim H-S Activities of ankle muscles during gait analyzed by simulation using the human musculoskeletal model J Exerc Rehabil 2019 15 2 229 10.12965/jer.1938054.027 31111005 29. Fereydounnia S Shadmehr A Moghadam BA Moghadam ST Mir SM Salemi P The effects of lower extremity kinesio taping on temporal and spatial parameters of gait initiation in semi-professional soccer players with and without functional ankle instabilit J Mod Rehabil 2021 15 4 253 264 30. Murley GS Menz HB Landorf KB Foot posture influences the electromyographic activity of selected lower limb muscles during gait J Foot Ankle Res 2009 2 1 1 9 10.1186/1757-1146-2-35 19144200 31. Csapo R Alegre LM Effects of kinesio® taping on skeletal muscle strength—a meta-analysis of current evidence J Sci Med Sport 2015 18 4 450 456 10.1016/j.jsams.2014.06.014 25027771	0	PMC9734429	NO-CC CODE	2022-12-14 23:28:28	no	Sport Sci Health. 2022 Dec 6;:1-8	utf-8	Sport Sci Health	2,022	10.1007/s11332-022-01013-z	oa_other
==== Front Sport Sci Health Sport Sci Health Sport Sciences for Health 1824-7490 1825-1234 Springer Milan Milan 1013 10.1007/s11332-022-01013-z Research Comparison between the effects of tibialis posterior versus fibularis longus Kinesio taping on foot posture, physical performance, and dynamic balance in young women with flexible flatfoot http://orcid.org/0000-0001-6745-6414 Tahmasbi Alireza [email protected] http://orcid.org/0000-0002-1186-9997 Shadmehr Azadeh [email protected] http://orcid.org/0000-0002-0184-8930 Attarbashi Moghadam Behrouz [email protected] http://orcid.org/0000-0002-9827-3108 Fereydounnia Sara [email protected] grid.411705.6 0000 0001 0166 0922 Physical Therapy Department, School of Rehabilitation, Tehran University of Medical Sciences, Tehran, Iran 6 12 2022 18 8 9 2022 30 9 2022 © The Author(s), under exclusive licence to Springer-Verlag Italia S.r.l., 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. Purpose The purpose of the study is to compare the effectiveness of the tibialis posterior Kinesio taping and fibularis longus Kinesio taping on the foot posture, physical performance, and dynamic balance in young women with flexible flatfoot. Methods Twenty-four subjects were recruited for the study. They were randomly divided into groups (A = 12, B = 12). In group A, Kinesio taping was applied on the tibialis posterior, and in group B, Kinesio taping was applied on the fibularis longus and remained for 30 min. Outcome measures were the navicular drop test (NDT), foot posture index (FPI), timed up and go (TUG) test, and Y-balance test. The pre- and post-treatment results were compared for each group; between-group differences were determined as well. Results For group A, NDT, FPI, and TUG test changed significantly (P = 0.01, P = 0.001, P = 0.006, respectively). For group B, the FPI score decreased (P = 0.03), and the Y-balance test in the anterior direction improved significantly (P = 0.01). Any variables have not shown a significant difference between groups (P > 0.05). Conclusion Kinesio taping of the tibialis posterior and fibularis longus can improve foot posture in young women with flexible flatfoot. Also, physical performance and dynamic balance improved by Kinesio taping of the tibialis posterior and the fibularis longus, respectively. In addition to the tibialis posterior, we found that the fibularis longus muscle can be considered a therapeutic target for managing flexible flatfoot in healthy young women. Keywords Flatfoot Kinesio tape Posture Physical performance http://dx.doi.org/10.13039/501100004484 Tehran University of Medical Sciences and Health Services #1400-3-103-55363 Shadmehr Azadeh ==== Body pmcIntroduction Flatfoot, also referred to as “pes planus” or “pronated foot,” is a common orthopedic condition that consists of reduced height of the medial longitudinal arch (MLA), which absorbs shocks and provides stability during dynamic activities [1]. Approximately it is reported that 5–20% of different populations, including healthy subjects and athletes, have this condition [2–4]. Two types of flatfoot are mentioned in scientific literature based on the consistency of the MLA: (1) flexible and (2) rigid. The MLA exists without weight-bearing in the flexible type, but it disappears under loading conditions, like standing. In the rigid type, there is no MLA with and without weight-bearing [5]. MLA is supported by both passive structures, like plantar fascia and plantar ligament, and active structures, like extrinsic and intrinsic foot muscles [6]. Several factors are considered causes of flexible flatfoot, such as ligamentous laxity, muscle weakness, obesity, bone deformity, and family history [7]. However, there is no consensus on the exact cause of this condition. Flexible flatfoot predisposes to other conditions, such as low back pain, patellofemoral pain syndrome, Achilles tendinopathy, plantar fasciitis, medial tibial stress syndrome, and knee osteoarthritis [8, 9]; so, it would be essential to be considered in the assessment and causative management of these musculoskeletal conditions as well. Furthermore, healthy people and athletes should notice foot and ankle deformations like the flexible flatfoot to prevent daily and sports injuries. Kinesio taping, an approach that has gained popularity in recent years among practitioners, is a noninvasive therapy that uses elastic bands with specific textures to treat, modify, and prevent numerous musculoskeletal conditions. Several mechanisms have been proposed to explain its effectiveness. From a neurophysiological perspective, it can increase the activation of mechanical skin receptors, leading to a reflexive contraction of muscle spindles and an increase in the sensitivity of motor units [10]. Moreover, it can increase the blood circulation of muscles by stimulating the autonomic system and inducing peripheral vasodilation [11]. According to previous studies, Kinesio tape can affect foot posture and muscle activity [12]; moreover, it can alter foot pressure, range of motion, and pain perception in people with flexible flatfoot [5, 13]. Previous studies have shown the positive effects of tibialis posterior (TP) facilitation on the foot posture and the dynamic balance in people with the flexible flatfoot [12, 14]. Moreover, in 2020, Sumal et al. stated that the fibularis longus (FL) tendon contributes to the MLA height, and physical therapy interventions targeting this muscle can lead to positive effects on the foot posture [6]; so, we hypothesized that using a facilitatory technique by means of Kinesio tape could result in the improved activity of these muscles. Also, it should be noted that some physiological characteristics, such as soft tissue flexibility, differ among genders, so it should be considered an influential factor in the results of therapeutic interventions on musculoskeletal deformities. Therefore, this study aims to compare the results of Kinesio taping of two important muscles of the foot, the tibialis posterior and fibularis longus, and their effects on foot posture, physical performance, and dynamic balance in young women with flexible flatfoot. Materials and methods Study design This study is a single-blind, parallel-design, randomized clinical trial. It is approved by the ethics committee of Tehran University of Medical Sciences with the approval identification of IR.TUMS.MEDICINE.REC.1400.771 and is registered in the Iranian Registry of Clinical Trials with the registration code IRCT20211018052805N. Participants In this study, 24 women were included. The sample size was calculated using GPower software. Power was considered 85%, the alpha value was 0.05, and the effect size was estimated using foot posture index (FPI) measurements in previous studies [15]. Inclusion criteria were: (1) 18–40-year-old people; (2) navicular drop test (NDT) ≥ 10 mm; (3) foot posture index (FPI) ≥ + 6; (4) positive Jack’s test; and (5) body mass index (BMI) ≤ 25 kg/m2. Exclusion criteria were: (1) ankle injury in the last six months; (2) history of foot surgery; (3) foot injury due to systemic, inflammatory, and infectious diseases; (4) foot deformities including hallux valgus, hammer toe, and claw toe; (5) pregnancy; (6) ankle pain during the study; (7) static standing and walking problems; (8) Kinesio tape sensitivity; and (9) refuse to participate in the study or not adhering to treatment anymore. All subjects were informed about the study procedures and signed the consent form prior to the study (Fig. 1).Fig. 1 CONSORT flowchart of the participants Assessment Procedure All of the subjects were asked to fill out a demographic questionnaire. Jack’s test was used to determine the flexibility of the MLA [5]. Indeed, the type of the flatfoot was examined, while the assessor passively dorsiflexed the 1st metatarsophalangeal joint of the subject in a standing position. MLA was considered flexible if the curve appeared, so the test was positive. If the curve did not appear, it showed a rigid flatfoot, and the test was considered negative. Afterward, NDT and FPI were assessed, and the eligible subjects were included in the study. Next, they were asked to do a timed up and go (TUG) test for their physical performance assessment. Finally, subjects performed a Y-balance test to be assessed for their dynamic balance. Three trials were conducted with 2 min rest intervals between each trial. All of these assessments were reassessed 30 min after intervention. The left leg was selected for the assessments and intervention in all subjects to avoid selection bias [16]. Navicular drop test For this test, the subject sat on a suitable chair with her hips and knees at 90 °, both feet on the ground, and subtalar joint in a neutral position. Navicular tuberosity was found and marked with a removable marker. The assessor marked the height of navicular tuberosity from the ground on an index card. Then, they were asked to stand and put equal weight on both feet. Again, similar measurements were done [17]. The difference in navicular heights in both situations was measured using a ruler. This test was repeated three times, and the average value was calculated. NDT has shown good reliability and validity in previous studies [18]. Foot posture index A 6-item version of the foot posture index was used to assess foot posture while standing [19]. Foot postures are classified based on a scoring system for each portion of the foot from − 2 to + 2. Summation of the scores determines the type of the foot as pronated (FPI ≥ + 6), neutral (0 ≤ FPI ≤ + 5), and supinated (FPI < 0). To determine, the subject was asked to stand with the arms at the side while looking straightforward. The assessor scored each item using a specific checklist. Timed up and go test In this test, the subject stood from a suitable chair and began to walk three meters at a usual speed and then returned and walked through the chair and sat on it [20]. This time was recorded by a stopwatch. Y-balance test Dynamic balance is assessed with the Y-balance test [21]. First, the assessor showed the correct form of performing the tasks. Then, the subjects were asked to complete three maneuvers in each direction (anterior, posteromedial, and posterolateral direction) to avoid the learning effect. Next, three trials were conducted with 2-min rest intervals between them. If any of these errors have occurred, the test was repeated: (1) Subject was not able to stand on one leg; (2) stance leg was moved; (3) reaching foot touched the ground due to loss of balance; and (4) subject could not return to the starting position. The average reaching distance was calculated in each direction, divided by limb length, and multiplied by one hundred (%). To measure the limb length, the subject lay supine. After correcting the pelvic position, a tape measure was placed from the anterior superior iliac spine (ASIS) to the most inferior part of the medial malleolus [14]. Randomization Two concealed envelopes containing blue and red pieces of cardboard were placed in front of the subjects, and they were asked to choose one of them. If they chose the blue or red one, they were assigned to the tibialis posterior Kinesio taping group or the fibularis longus Kinesio taping group, respectively. Subjects were blinded to the group in which they would be allocated. Intervention Tibialis posterior Kinesio taping (group a) For this purpose, the subject lay supine and actively dorsiflexed and everted her ankle. Kinesio tape was applied from half of the tibia bone, passing behind the medial malleolus, and finished at the head of the fifth metatarsal bone [12] (Fig. 2).Fig. 2 Tibialis posterior Kinesio taping Fibularis longus Kinesio taping (group b) The subject was asked to lie supine and actively plantarflex and invert her ankle. Kinesio tape was applied from the head of the fibula, passing behind the lateral malleolus, and finished at the base of the first metatarsal bone [22] (Fig. 3).Fig. 3 Fibularis longus Kinesio taping Using a facilitatory technique, an I-shaped red piece of Kinesio tape (TEMTEX, South Korea) with 35 percent tension was used for both groups [23]. While the subject had returned the muscle to the rest position, the anchor was applied without tension. Subjects rested in a comfortable position for 30 min [23, 24]. Afterward, they were reassessed. All assessments and interventions were performed by a single physiotherapist who was certified in the Kinesio tape method. Statistical analysis SPSS statistics version 24 was used for statistical analysis. The Kolmogorov–Smirnov test was used to assess the normal distribution of data. For intra-group comparison, paired samples T test and Wilcoxon test were used for parametric and nonparametric statistics, respectively. For intergroup comparison, independent T test and Mann–Whitney test were used for parametric and nonparametric statistics. The significance level (P value) was considered less than 0.05. Results Sixty participants were screened, and of them, 36 participants did not meet the inclusion criteria or were excluded. Finally, 24 participants were eligible and enrolled in the study. The subjects did not report any adverse effects during and after the study. The demographic characteristics of the subjects are shown in Table 1. The Kolmogorov–Smirnov (K-S) test showed that the distribution of the weight and BMI was not normal in group A (P = 0.03, and P = 0.04, respectively). Also, K-S test showed that the distribution of the NDT values was not normal in both groups (P = 0.01 and P = 0.03 for groups A and B, respectively), so we used the related nonparametric tests. There was no significant difference between the groups regarding the demographic characteristics (P > 0.05). According to the results, NDT, FPI, and TUG test changed significantly in group A (P = 0.01, P = 0.001, P = 0.006, respectively) (Table 2). For group B, the FPI score decreased (P = 0.03), and the Y-balance test in the anterior direction improved significantly (P = 0.01) (Table 2). Navicular drop decreased after the intervention in group B, but it was not statistically significant (P = 0.09). Any variables have not shown a significant difference between groups (Table 3); however, FPI improved better in group A (P = 0.05, mean difference = − 1.08).Table 1 Demographic characteristics in two groups Variables Group A (n = 12) Group B (n = 12) P value t (z) Power Age (years) 22.25 ± 2.52 24.58 ± 6.02 0.23 − 1.23 0.23 Height (m) 1.63 ± 0.05 1.62 ± 0.04 0.68 0.41 0.08 Weight (kg) a 60 ± 6.98 59 ± 7.36 0.81 − 0.23 0.05 BMI (kg/m2) a 22.42 ± 2.00 22.29 ± 2.46 0.86 − 0.17 0.03 Values are shown as mean ± standard deviation aFor weight and BMI, the Mann–Whitney test was used due to a lack of normal distribution Table 2 Before–after comparison of variables in group A and group B Variable Group A Group B Before mean ± SD After mean ± SD t (z) P value Effect size Power Before mean ± SD After mean ± SD t (z) P value Effect size Power NDT (mm) 10.96 ± 1.34 8.85 ± 2.03 − 2.39 0.01 1.22 – 11.82 ± 2.39 10.35 ± 2.78 − 1.69 0.09 – 0.28 FPI 8.33 ± 1.30 6.50 ± 1.97 − 4.33 0.001* 1.09 – 8.33 ± 1.49 7.58 ± 1.62 − 2.46 0.03* 0.48 – TUG (s) 10.32 ± 1.10 9.56 ± 1.02 − 3.43 0.006* 0.71 – 10.37 ± 1.34 9.85 ± 1.42 1.82 0.09 – 0.15 YBT (A) (%) 72.54 ± 7.16 72.90 ± 6.79 0.295 0.77 – 0.03 76.46 ± 6.62 79.52 ± 6.92 3.03 0.01* 0.45 – YBT (PM) (%) 101.50 ± 15.21 100.90 ± 16.24 − 0.48 0.63 – 0.03 100.80 ± 7.13 103.85 ± 7.71 1.91 0.08 – 0.17 YBT (PL) (%) 88.91 ± 14.90 91.13 ± 16.55 1.65 0.12 – 0.05 87.79 ± 7.43 90.35 ± 8.92 1.52 0.15 – 0.12 SD standard deviation, NDT navicular drop test, FPI foot posture index, TUG timed up and go, YBT (A, PM, PL) Y-balance test (anterior, posteromedial, and posterolateral) aFor NDT, the Wilcoxon test was used due to a lack of normal distribution *P < 0.05 is significant Table 3 Intergroup comparison of variables between groups A and B Variable Mean difference ± SD (mean rank difference) P value t (z) Confidence interval a (sum of ranks) Power Lower Upper NDT b(mm) − 2.08 0.47 − 0.72 137.50 162.50 0.11 FPI − 1.08 ± 0.52 0.05 − 2.07 − 2.16 − 0.002 0.54 TUG (s) − 0.23 ± 0.35 0.52 − 0.65 − 0.97 0.51 0.10 YBT (A) (%) − 2.69 ± 1.59 0.10 − 1.68 − 6.00 0.61 0.39 YBT (PM) (%) − 3.64 ± 2.01 0.08 − 1.81 − 7.82 0.53 0.44 YBT (PL) (%) − 0.33 ± 2.15 0.87 − 0.15 − 4.80 4.12 0.04 SD standard deviation, NDT navicular drop test, FPI foot posture index, TUG timed up and go, YBT (A, PM, PL)Y-balance test (anterior, posteromedial, and posterolateral), a95% confidence interval of the difference bFor NDT, the Mann–Whitney test was used due to a lack of normal distribution Discussion The study investigated the immediate effects of Kinesio taping on two important muscles of the leg, the tibialis posterior and the fibularis longus. The results showed that the foot posture was improved in both groups, though the navicular drop decreased significantly just with the tibialis posterior Kinesio taping. This finding is consistent with the study of Siu et al. in 2019 on runners with flexible flatfoot [12]. They showed that the Kinesio taping of the TP and the transverse arch could immediately reduce the NDT. Since the TP activity provides dynamic stability for the MLA due to its line of action [25], it is possible that the increased activity of this muscle, and the following force generation capacity, is achievable as an immediate effect of Kinesio taping, and hence, we can expect the improved arch height. Moreover, since the TP Kinesio taping procedure passes over the medial side of the foot and covers the navicular bone, it is possible that it can provide mechanical support to the navicular bone and the MLA. However, in 2012, Roman et al. did not find any significant effect of TP Kinesio taping on rear foot pronation 24 h after the application, which questions the prolonged effect of TP Kinesio taping on foot posture [26]. In 2020, Sumal et al. stated that the fibularis longus tendon contributes to the MLA height, and physical therapy interventions targeting this muscle can lead to positive effects on foot posture [6]. Our results showed that the overall foot posture improved with FL Kinesio taping. Although there was no significant difference in FPI values between the two groups, the results were so close to being meaningful in favor of group A. Aguilar et al. in 2015 suggested that Kinesio taping using low-dye technique can lead to better result of FPI compared to the sham Kinesio taping in the pronated foot of amateur runners after 45 min of running [16]. It should be noticed that the mechanical correction of the low-dye technique was more than our method since it used 75% tension of Kinesio tape. Y-balance test in the anterior direction increased significantly in group B. As an ankle evertor, the fibularis longus provides stability in the ankle joint [27], and it is a primary muscle for balance maintenance. Our findings were consistent with the results of Fereydounnia et al. in 2019 in which they investigated the effects of fibularis longus Kinesio taping on dynamic balance in soccer players with and without ankle instability [23]. They showed that Kinesio tape could immediately affect the dynamic balance of subjects. In 2016, Correia et al. found that the Kinesio taping of the FL has no immediate effect on the static balance in young healthy subjects, though the applied tension was only 10%. Our study showed that the FL Kinesio taping with 30% tension can lead to better results in the dynamic balance [22]; however, it is unclear if there is a significant relationship between the static balance and dynamic balance with the same applied tension of the Kinesio tape. The timed up and go test decreased significantly in group A; it might happen due to the improved support of MLA during the load transfer in the gait mechanism as a result of TP facilitation. Moreover, Siu et al. found that Kinesio taping of TP and transverse arch can increase the muscle activity of the tibialis anterior during running, which is an important muscle in the propulsion phase of locomotion [12, 28]. For group B, the result was not significant, which is consistent with the study of Fereydounnia et al. in 2021 on soccer players with and without ankle instability. In that study, Kinesio taping was applied in order to facilitate FL, and gait initiation parameters were measured with the force plate [29]. To our knowledge, it is the first study that has investigated the TUG test as a functional assessment for the physical performance of people with the flexible flatfoot. EMG studies have shown the altered electromyographic activity of TP and FL in the gait mechanism of flatfeet people [30]. As mentioned before, several theories have been explained for the mechanisms of Kinesio tape effectiveness. It can increase blood flow through the autonomic system [11], facilitate muscle activity by increasing the sensory inputs of skin and joint receptors [10], and adjust the length–tension relationship and, subsequently, the force generation capacity of muscles [31]. All of these changes could lead to better function of the target muscles in this study. To the authors’ knowledge, it is the first study that investigated the effects of Kinesio taping on the fibularis longus muscle in flexible flatfoot, so this effect on the mentioned outcome measures can be further investigated in future studies. For example, it is suggested that the prolonged effects of FL Kinesio taping plus the comparison to the control group be addressed in future studies. One limitation of our study is that the subjects were young females, and the results are limited to this population. Moreover, just immediate effects of the intervention were studied, and the long-term effects remained unclear. Also, the lack of the control group disabled us from deducing consistently from the present findings. Another limitation is the small sample size of the study. Due to the COVID-19 pandemic, it was not possible to enlarge the sample size. Conclusion Kinesio taping of the tibialis posterior can improve the foot posture and physical performance of young women with flexible flatfoot. Additionally, Kinesio taping of fibularis longus leads to promising results in the foot posture and dynamic balance in this population. No significant differences were shown between the two groups. In addition to the tibialis posterior, it is suggested that the fibularis longus muscle be considered a therapeutic target for the management of flexible flatfoot in healthy young women. Acknowledgements The authors are grateful to the Department of Physical Therapy, School of Rehabilitation, Tehran University of Medical Sciences, for their support and the laboratory space and equipment. Also, the authors appreciate the participants’ contribution to the study. Author contributions Alireza Tahmasbi, Sara fereydounnia, Azadeh Shadmehr, and Behrouz Attarbashi Moghadam conceptualized the research. Azadeh Shadmehr and Sara Fereydounnia supervised the process. Alireza Tahmasbi gathered the data and wrote the main manuscript. Alireza Tahmasbi and Sara Fereydounnia analyzed the data. All authors reviewed and edited the manuscript. Azadeh Shadmehr managed the provided fund. Funding This article is extracted from the MSc Dissertation of the first author in the Department of Physical Therapy, School of Rehabilitation, Tehran University of Medical Sciences, Iran (Grant: #1400–3-103–55363) Data availability This study is original research, and data were collected from subjects via related procedures by the first author in the biomechanical laboratory of the Physical Therapy Department, School of Rehabilitation, Tehran University of Medical Sciences. Declarations Conflict of interest The authors declare no conflict of interest. Ethical approval This study was approved by the ethics committee of Tehran University of Medical Sciences with the approval identification of IR.TUMS.MEDICINE.REC.1400.771. It is also registered in the Iranian Registry of Clinical Trials with the registration code IRCT20211018052805N. 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 All subjects were informed about the study purpose and procedure and signed a consent form before the study. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Birinci T Demirbas SB Relationship between the mobility of medial longitudinal arch and postural control Acta Orthop Traumatol Turc 2017 51 3 233 237 10.1016/j.aott.2016.11.004 28462802 2. Bhoir T Anap DB Diwate A Prevalence of flat foot among 18–25 years old physiotherapy students: cross sectional study I JBasic Appl Med Res 2014 3 4 7 3. Ganapathy A Sadeesh T Rao S Morphometric analysis of foot in young adult individuals W J Pharm Pharmaceutical Sci 2015 4 8 980 993 4. Bhosale N, Nandala P (2021) Prevalance of flexible flat foot in athletes. Kesari Mahratta Trust 1(1):1–13 5. Karthikeyan J Singh K Govind S Mahalingam K Vamsi S Annamalai P To compare the effectiveness of taping and arch support on the flexible flat foot on a random population I J Forensic Med Toxicol 2020 14 4 7825 6. Sumal AS Jarvis GE Norrish AR Brassett C Whitaker RH The role of the angle of the fibularis longus tendon in foot arch support Clin Anat 2021 34 4 651 658 10.1002/ca.23686 32986255 7. Atik A Ozyurek S Flexible flatfootness North Clin Istanb 2014 1 1 57 10.14744/nci.2014.29292 28058304 8. Neal BS Griffiths IB Dowling GJ Murley GS Munteanu SE Franettovich Smith MM Foot posture as a risk factor for lower limb overuse injury: a systematic review and meta-analysis Journal of foot and ankle research 2014 7 1 1 13 10.1186/s13047-014-0055-4 24423411 9. Gross KD Felson DT Niu J Hunter DJ Guermazi A Roemer FW Association of flat feet with knee pain and cartilage damage in older adults Arthritis Care Res 2011 63 7 937 944 10.1002/acr.20431 10. Oliveira AK Borges DT Lins CA Cavalcanti RL Macedo LB Brasileiro JS Immediate effects of Kinesio Taping® on neuromuscular performance of quadriceps and balance in individuals submitted to anterior cruciate ligament reconstruction: a randomized clinical trial J Sci Med Sport 2016 19 1 2 6 10.1016/j.jsams.2014.12.002 25601016 11. Gómez-Soriano J Abián-Vicén J Aparicio-García C Ruiz-Lázaro P Simón-Martínez C Bravo-Esteban E The effects of Kinesio taping on muscle tone in healthy subjects: a double-blind, placebo-controlled crossover trial Man Ther 2014 19 2 131 136 10.1016/j.math.2013.09.002 24829961 12. Siu W-S Shih Y-F Lin H-C Effects of kinesio tape on supporting medial foot arch in runners with functional flatfoot: a preliminary study Res Sports Med 2020 28 2 168 180 10.1080/15438627.2019.1638258 31262193 13. Wang J-S Um G-M Choi J-H Immediate effects of kinematic taping on lower extremity muscle tone and stiffness in flexible flat feet J Phys Ther Sci 2016 28 4 1339 1342 10.1589/jpts.28.1339 27190479 14. Alam F Raza S Moiz JA Bhati P Anwer S Alghadir A Effects of selective strengthening of tibialis posterior and stretching of iliopsoas on navicular drop, dynamic balance, and lower limb muscle activity in pronated feet: a randomized clinical trial Phys Sportsmed 2019 47 3 301 311 10.1080/00913847.2018.1553466 30517043 15. Unver B Erdem EU Akbas E Effects of short-foot exercises on foot posture, pain, disability, and plantar pressure in pes planus J Sport Rehabil 2019 29 4 436 440 10.1123/jsr.2018-0363 30860412 16. Aguilar MB Abián-Vicén J Halstead J Gijon-Nogueron G Effectiveness of neuromuscular taping on pronated foot posture and walking plantar pressures in amateur runners J Sci Med Sport 2016 19 4 348 353 10.1016/j.jsams.2015.04.004 25956688 17. Elataar FF Abdelmajeed SF Abdellatif NM Mohammed MM Core muscles’ endurance in flexible flatfeet: a cross-sectional study J Musculoskelet Neuronal Interact 2020 20 3 404 32877977 18. Zuil-Escobar JC Martínez-Cepa CB Martín-Urrialde JA Gómez-Conesa A Medial longitudinal arch: accuracy, reliability, and correlation between navicular drop test and footprint parameters J Manipulative Physiol Ther 2018 41 8 672 679 10.1016/j.jmpt.2018.04.001 30573198 19. Redmond AC Crosbie J Ouvrier RA Development and validation of a novel rating system for scoring standing foot posture: the foot posture index Clin Biomech 2006 21 1 89 98 10.1016/j.clinbiomech.2005.08.002 20. Larsson BA Johansson L Johansson H Axelsson KF Harvey N Vandenput L The timed up and go test predicts fracture risk in older women independently of clinical risk factors and bone mineral density Osteoporos Int 2021 32 1 75 84 10.1007/s00198-020-05681-w 33089354 21. Kim J-a Lim O-b Yi C-h Difference in static and dynamic stability between flexible flatfeet and neutral feet Gait Posture 2015 41 2 546 550 10.1016/j.gaitpost.2014.12.012 25560044 22. Correia C Lopes S Gonçalves R Torres R Pinho F Gonçalves P Kinesiology taping does not change fibularis longus latency time and postural sway J Bodyw Mov Ther 2016 20 1 132 138 10.1016/j.jbmt.2015.07.037 26891648 23. Fereydounnia S Shadmehr A Moghadam BA Moghadam ST Mir SM Salemi S Improvements in strength and functional performance after kinesio taping in semi-professional male soccer players with and without functional ankle instability Foot 2019 41 12 18 10.1016/j.foot.2019.06.006 31675595 24. Lemos TV Pereira KC Protássio CC Lucas LB Matheus JPC The effect of Kinesio Taping on handgrip strength J Phys Ther Sci 2015 27 3 567 570 10.1589/jpts.27.567 25931682 25. Willegger M Seyidova N Schuh R Windhager R Hirtler L The tibialis posterior tendon footprint: an anatomical dissection study J Foot Ankle Res 2020 13 1 1 7 10.1186/s13047-020-00392-1 31956341 26. Román MF Méndez AC Cabello MA Effects of treatment with Kinesio Tape for flat feet Fisioterapia 2012 34 1 11 15 27. Sarvestan J Ataabadi PA Svoboda Z Kovačikova Z Needle AR The effect of ankle Kinesio™ taping on ankle joint biomechanics during unilateral balance status among collegiate athletes with chronic ankle sprain Phys Ther Sport 2020 45 161 167 10.1016/j.ptsp.2020.06.007 32781269 28. Lee H-S Lee J-H Kim H-S Activities of ankle muscles during gait analyzed by simulation using the human musculoskeletal model J Exerc Rehabil 2019 15 2 229 10.12965/jer.1938054.027 31111005 29. Fereydounnia S Shadmehr A Moghadam BA Moghadam ST Mir SM Salemi P The effects of lower extremity kinesio taping on temporal and spatial parameters of gait initiation in semi-professional soccer players with and without functional ankle instabilit J Mod Rehabil 2021 15 4 253 264 30. Murley GS Menz HB Landorf KB Foot posture influences the electromyographic activity of selected lower limb muscles during gait J Foot Ankle Res 2009 2 1 1 9 10.1186/1757-1146-2-35 19144200 31. Csapo R Alegre LM Effects of kinesio® taping on skeletal muscle strength—a meta-analysis of current evidence J Sci Med Sport 2015 18 4 450 456 10.1016/j.jsams.2014.06.014 25027771	0	PMC9734430	NO-CC CODE	2022-12-14 23:28:28	no	NeuroTransmitter. 2022 Dec 9; 33(12):22-25	latin-1	null	null	null	oa_other
==== Front Pharm Chem J Pharm Chem J Pharmaceutical Chemistry Journal 0091-150X 1573-9031 Springer US New York 2778 10.1007/s11094-022-02778-w Article Drug Repurposing: a Shortcut to New Biological Entities Rao Nutan [email protected] 1 Poojari Tushar 2 Poojary Charvi 2 Sande Ruksar 2 Sawant Sonal 2 1 Vivekanand Education Society’s College of Pharmacy, Chembur, Mumbai, 400074 India 2 Oriental College of Pharmacy, Sector 2, Behind Sanpada Railway Station, Sanpada West, Navi Mumbai, Maharashtra 400705 India 7 12 2022 112 20 4 2021 © 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. Drug repurposing has proved to be an efficient alternative to drug discovery owing to the facts that it is economical and risk factors being much lower or even negligible as the drug has already been approved for having safe use in humans. The contrast of drug discovery from drug repurposing, its advantages and the challenges faced during the process are the important factors to be considered in drug repurposing. The approaches in drug discovery include three methods namely computational, biological and mixed. Moreover, the recent advancement in application of drugs for COVID-19 proved drug repurposing is a vital strategy in medical science for the upcoming years. Keywords drug discovery drug repurposing strategies approaches computational biological COVID-19 advantages methodologies repurposed drugs ==== Body pmc 1. Introduction Drug repurposing, also known by the names of drug repositioning, drug retasking, drug re-profiling, drug recycling, drug redirection and therapeutic switching, can be defined as a process of recognition of new pharmacological indications from old/existing/failed/investigational already marketed/FDA approved drugs/prodrugs and the application of newly developed drugs to the treatment of diseases other than the drug’s original/intended therapeutic use [1]. It is well known that the cost of development of a new drug is extremely high and runs in 800 million–1.5 billion US dollars. Besides this, it takes anywhere between eight to ten years, to discover a new drug. After starting with 100,000 New Chemical Entities (NCEs), one ends up with two to four molecules, which can be called as new drugs to put on the market. The failure rate is very high in clinical trial of new drug; out of which nearly 50% drug fails in phase 3 out of the total cost of 800 million US dollar, nearly 400 million dollar goes in clinical trial and other process of new drug development [2]. Drug repurposing, also generally regarded as drug repositioning or drug rescue, emerged fundamentally in the early 1990s as a feasible alternative to the conventional drug discovery process. Repositioning depends on two prime scientific bases: (1) The discovery, through the human genome elucidation, that few diseases share sometimes common biological targets, and (2) The concept of pleiotropic drugs [3]. Repurposers have an advantage because they are working with compounds that have been approved or at least put through millions of dollars worth of preclinical and early clinical testing. As a result, repurposing can get drugs to the market cheaper and faster than the lengthy new drug research and development. Moreover, whereas 10% of new molecular entities are able to make it to the market from Phase II clinical trials and 50% new entities from Phase III. The rates for repurposed compounds are 25 and 65% for Phases II and III, respectively [4]. 2. Traditional Drug Discovery versus Drug Repurposing 2.1. Traditional Drug Discovery Steps: (i) Discovery and preclinical study (average 6.5 years). Research for a new drug starts in the laboratory and animal testing to answer basic questions about safety. (ii) Safety review (average 30 days). Review to ensure animal testing and assure safety. (iii) Clinical research phase 1 (average 1.5 years). This step includes the first trial in humans and is primarily concerned with the safety, pharmacokinetic and pharmacodynamic profiles of the drug. The drug is initially given as single, very low dose, then in gradually increasing amounts and subsequently, in multiple doses. Phase 1 trials only involve a small number of subjects and are generally conducted in normal volunteers. (iv) Clinical research phase 2 (average 2 years). Studies commonly referred to as phase 2 are relatively small-scale studies involving patients with the target disease or condition. (v) Clinical research phase 3 (average 3.5 years). If the drug in question proves safe in second phase, then phase three is carried out. This means that large numbers of patients are studied in medical centers throughout the country. (vi) Clinical research phase 4 (average 1.5 years). In this step, the drug is already approved by FDA. It is frequently termed as post marketing surveillance. These tests are generally large-scale and designed primarily to investigate the incidence of rare adverse reactions and to check if drug recall is required [5]. The typical pathway for a traditional drug discovery and development process or the Life Cycle of a pharmaceutical product is depicted in Fig. 1.Fig. 1. The Life Cycle of pharmaceutical products [6]. The pharmaceutical ecosystem can be stated as the convergence of networks of cross-interacting subsystems across the drug product pipeline, which has a stake in the efficiency of drug development and access to marketed drug. The tools for streamlining the utility of drug during pre- and post-marketing stages include the strategic issues, regulatory organizations and reforms, local and national cultures, politics, federal laws, economic and reimbursement policies, intellectual property and patent policies, product related factors. There are networks of multichannel interactions at various levels around the development of pharmaceutical product [7]. The scheme of cross-functional interactions in pharmaceutical ecosystem is shown in Fig. 2.Fig. 2. Cross-functional interaction in pharmaceutical ecosystem [7]. 2.2. Drug Repurposing Steps: (i) Compound identification (1.2 years). Compound identification is to select candidate drug for a particular drug target in the human organism. (ii) Compound acquisition (0 – 2years). Getting licenses for the new candidate drug. (iii) Development (1 – 5years). This step may start at preclinical, phase 1 or phase 2 drug research. To make sure that drugs are safe and effective, analysis of existing data is necessary. (iv) FDA post market safety survey. FDA monitors all drug and device safety once products are available for use by the public [8]. Figure 3 compares the timelines for steps involved in Traditional Drug Discovery process and Drug Repurposing process.Fig. 3. Comparative timelines of traditional drug discovery and drug repurposing [9]. 2.3. Advantages of Repurposing Drugs From the prior deliberation, it is evident that the drug repurposing approach is beneficial from the fact that approved drugs and several discarded compounds have already been tested in humans and elaborative data is available on their pharmacology, dose, possible toxicity and formulation. Drug repurposing has numerous advantages over conventional drug discovery approaches, including: (i) It considerably cuts research and development (R&D) expenditure. (ii) It decreases the drug development timeline, as several existing compounds have already demonstrated as safe in humans, and hence does not require Phase 1 clinical trials. (iii) Potential for reuse of drug molecules in spite of adverse effects and failed efficacy in some indications [10]. 2.4. Barriers of Drug Repurposing In spite of several advantages associated with drug repurposing, the process suffers from certain hurdles. These mainly include: (i) Lack of clear regulatory pathways. Pharmaceutical companies focus mainly on the development of new medicines and there is a lack of regulatory pathways to facilitate drug repurposing. (ii) Lack of financial incentives and research findings. Pharmaceutical industry, including the generic sector, has hardly any incentive to invest in the research necessary to gain regulatory approval for a drug that is no longer under patent. This is because there is no return on investment anticipated, given the lack of intellectual property protection and low prices of generic formulation [11]. However, the drug repurposing process is a low-risk, high-rewarding strategy for developing drugs (Fig. 4).Fig. 4. Risk and reward of different drug development strategies [8]. 2.5. Challenges Associated with Drug Repositioning The foremost challenge encountered by scientists lie in the relatively weak intellectual property protection allotted to such medicinal products, which can reduce their return on investment and discourage companies from developing them [12]. As the concerned drug has earlier been patented as a new chemical entity, succeeding medicines containing the same entity can only be protected by a new application patent, possibly supported by a novel formulation procedure. Patents on applications are essentially limited rather than those for a new chemical entity in terms of the therapeutic uses they offer. Repositioned drug, still has the same chemical entity are not strong in the front of a potential legal challenge on the basis that the new indication was predictable from data in the scientific literature [11]. Another challenge is to convince the physicians that an already existing drug can be used for a completely new pharmacological indication. The comparatively weaker protection provided by these patents may nevertheless be offset by certain advantages granted to companies repositioning drugs for the treatment of orphan diseases (defined as diseases which are very rare) such as fee reductions and a guaranteed period of market exclusivity. 2.6. Strategies of Drug Repurposing There are two main strategies for drug repurposing: (A) On-target drug repurposing; (B) Off-target drug repurposing. These strategies for drug repositioning are schematically depicted in Fig. 5.Fig. 5. On-target and off-target strategies for drug repositioning [12]. On-target drug repurposing. In the on-target strategy, we are investigating new indication of drug acting through the originally known target. The known pharmacological mechanism of a drug entity is applicable to a new therapeutic indication. In this strategy, the biological target of the drug molecule remains same, but the disease is different. Appealing aspect is that it is likely to be compatible with dosing of the original drug. For example, an on-target strategy is observed in the repositioning of Minoxidil (Rogaine) as the drug acts on the same target and produces two different therapeutic effects. Minoxidil was repurposed from an antihypertensive vasodilator to an anti-hair loss drug. As an antihypertensive vasodilator, Minoxidil has the characteristic of widening blood vessels and opening potassium channels, which permits more oxygen, blood, and nutrients to the hair follicles and thus show its pharmacological action. Off-target drug repurposing. In the off-target strategy, we discover new uses of a drug acting through an unanticipated target. The mechanism of action is not known. Drugs act on new targets, away from the original scope, for novel therapeutic indications. Therefore, the target on which the drug acts and the indication for which it is used, both are novel. Drug is primarily not optimized for the target, so one needs to be careful about the dosing. Methods such as docking and fingerprinting can be implemented. Aspirin (Colsprin) is an example of the off-target repurposing strategy, as it has been conventionally used as NSAID in the treatment of various pain and inflammatory disorders. Later it was discovered that it also suppresses blood coagulation (clot formation) by inhibiting the normal functioning of platelets, i.e., it acts as an antiplatelet drug. Hence, it is used in the treatment of heart attacks and strokes as well as in the treatment of prostate cancer has also been reported [12]. Refer Table 1 for indications, targets and novelty of the on- and off-target repurposing strategies.Table 1. Indication, Target, and Novelty of On- and Off-Target Repurposing [12] Reformulation, Line Extension On-Target Repurposing Off-Target Repurposing Indication Same Different Same Different Target Same Same Different Different Novelty Low Same High Highest 2.7. Approaches of Drug Repositioning Detection of novel drug–disease relationships being the main issue in drug repositioning, a number of approaches have been developed including:Computational approaches; Biological experimental approaches; Mixed approaches [12]. 3. Computational Drug Repositioning Approaches Majority of the existing computational approaches are based on the gene expression response of cell lines after treatment or merging several types of information about disease–drug relationships. They are divided into three categories, including:Network based approaches; Text mining approaches; Semantic approaches. 3.1. Network Based Drug Repositioning Network based computational approaches are broadly utilized in drug repositioning due to associated ability to integrate multiple data sources, which are further classified as follows. Network based propagation approaches employ prior data propagate from the source hubs to all network hubs and some sub-network hubs. According to different propagation ways, these approaches can be partitioned into two types. The local propagation approaches only take the limited information into consideration and may fail to make accurate predictions. The global propagation approaches employ data from the entire network enabling them to perform better than local approaches, which is why most current researchers concentrate on global approaches to achieve outstanding results. For example, Kohle, et al. [Am. J. Hum. Genet., 82(4), 949 – 958 (2008)] developed a network propagation based on the global information of a network to find novel disease gene interaction, which included three phases: (a) extracting drug disease relationship and constructing a disease gene network, (b) obtaining the global information of the network using random walk propagation algorithm in the network, and (c) defining global metrics to predict novel disease gene relationships, which performed better than other approaches. Network based cluster approaches have been put forward to discover the relationships between novel drug disease or drug target on the basis of the fact that biologic entities such as disease, drug, protein, etc. in the same module of biological networks share similar characteristics. These approaches intend to find several modules also known as clusters, groups or cliques using cluster algorithms in accordance with the topology structures of networks. The modules employ various (e.g., drug–drug or drug—target) relationships [13]. Characteristics of network based computational approaches are presented in Table 2.Table 2. Network Based Drug Repositioning [14] Name Method Network Description Key Findings Advantage Disadvantage RNSA Cluster PPI A global network algorithm to identify protein clusters on PPI network Some complex proteins This method considers both local and global information from networks. Overlap clusters can be detected as well. Some information may be dropped as the cluster size is small RRW Cluster PPI An effective network cluster approach to identify protein clusters on a PPI network Some complex proteins This is a general method with high prediction accuracy. It is a time costly and memory costly method that cannot detect overlap clusters ClusterONE Cluster PPI A global network algorithm to identify node clusters on network Some complex proteins This approach outperformed the other approaches including MCI, RRW, etc. both on weighted and unweighted PPI networks There is not a standard gold value to evaluate clusters Cluster Drug protein disease A variant clusterONE algorithm to cluster nodes on heterogenous networks (iloperidone, schizophrenia) Hypertension This is an efficient cluster approach that integrates multiple databases It Is difficult to distinguish between positive associations and negative associations on the network Cluster Drug target disease An algorithm to detect clusters on the network (Vismodegib, Basal cell carcinoma) gorlin syndrome This is a genera and highly robust approach This approach loses weakly associate genes o diseases and drugs MBiRW Cluster Drug disease A bi random walk based algorithm Levodopa, Parkinson disorder > Alzeihmer Predictions of this approach are reliable This approach needs to adopt more biological alternatives 3.2. Text Mining Based Drug Repositioning The extraction of novel and biological entity relationships from the literature has been a challenge. Text mining techniques have been widely used to tackle this problem and have been developed to mine new information from scientific literature as well as identify relationship between biological concepts or biological entities. The fundamental pipeline of biological text mining comprises four stages: (i) Information retrieval (IR), whereby relevant documents are extracted from the literature, further filtered to eliminate irrelevant concepts in document; (ii) Biological name entity recognition (BNER), whereby valuable biological concepts are identified with controlled vocabularies; (iii) Biological information extraction (BIE); (iv) Biological knowledge discovery (BKD). In the BIE and BKD steps, useful information is extracted to discover knowledge about biological concepts to build a knowledge graph, at the same time potential associations between knowledge such as drug- disease and drug- target relationships can also be detected [13]. Text mining based computational approaches are presented in Table 3.Table 3. Text Mining Tools for Drug Repositioning [15] Name Class Input Output Description Biovista Static Biological knowledge Gene protein relationships A mining framework to extract gene protein relationships Biowisdom Static Ontology Drug disease, drug target relationships A platform to discover novel biological entity relationships FACTA+ Static Tekst Abstracts and linked concepts A system to find associated concepts based on a user query EDGAR Static UMLS term Drug gene relationships A system to extract relationships between drugs an genes involved in cancer using syntactic analysis. Polysearch Dynamic Document Knowledge A web based text mining and natural language processing platform Extract 2 Static Bio entities Entity relationships A text mining based tool to map biological entities to ontology / taxonomy entries Anni 2.0 Static Bio entities Linked concepts An ontology interface of a text mining tool to extract DrugQuest Static Drugs Drug drug relations A knowledge discovery tool to detect drug drug relationships MaNER Dynamic Medical document Revelant entities A rule based system to mine relevant entities in medical documents BEST Dynamic Biomedical literature Relevant bioentities A knowledge discovery system to extract relevant bioentities 3.3. Semantics Based Drug Repositioning Semantics based approaches are widely used in information retrieval and image retrieval and recently have been applied to drug repositioning. The work flow in these methods mainly includes three steps. (i) Biological entity relationships are extracted from prior data in huge medical databases to construct the semantic network; (ii) Semantics network based on existing ontology networks are constructed by adding the prior information obtained in previous step; (iii) Finally, mining algorithms are designed to predict novel relationships in the semantic network [15]. The workflow of semantic network interference is illustrated in Fig. 6.Fig. 6. The workflow of semantic network interference [8]. 4. Biological Experimental Drug Repositioning Approaches The experiment based approach is also known as activity based repositioning which refers to the screening of original drugs for new pharmacological indications based on experimental assays. It involves protein target based and cell organism based screens in in vitro and/or in vivo disease models without requiring any structural information of target proteins. Several approaches of experimental repositioning involve target screening approach, cell assay approach, animal model approach and clinical approach [12]. 5. Mixed Approaches In recent years, numerous scientists have combined computational approaches and experimental approaches to find new indications for drugs, called “Mixed Approaches”, wherein the result of computational methods was validated by biological experiments and clinical tests. Mixed approaches provide opportunities to researchers for developing repositioned drug effectively and rapidly [12]. Refer Fig. 7 for an overview of drug repositioning approaches.Fig. 7. Approaches of drug repurposing [12]. 6. Methodologies of Drug Repurposing The methodologies adopted in drug repurposing are extensively divided into three categories depending on the quantity and quality of the pharmacological, toxicological and biological activity-related information available: (i) Drug Oriented Methodology: The structural characteristics of the dug molecules, biological activities, adverse effects and toxicities are evaluated, meant for identifying molecules with biological effects based on cell/ animal assays. This type is based on traditional pharmacology and drug discovery where studies are usually conducted to determine the biological efficacy of drug molecules without having an estimation about biological targets. (ii) Target Oriented Methodology: It comprises in silico screening or virtual high throughput screening of drugs or compounds from drug libraries/ compound databases such as ligand based screening or molecular docking followed by in vitro and in vivo throughput and/or high content screening of drugs against a selective protein molecule or a biomarker of interest. There is a prominent success rate in drug discovery as compared to drug oriented method as most biological targets represent the disease pathways/mechanisms. (iii) Disease/Therapy Oriented Methodology: It is relevant when there is more information on disease model available, as drug repurposing can be guided by the disease and/or treatment based upon availability of information concerning the disease process provided by (i) Proteomics, data on disease specific target proteins; (ii) Genomics, data on disease specific genetics; (iii) Metabolomics, disease specific metabolic pathways/profiles; (iv) Phenotypic data, on off-target mechanisms, pharmacological targets, disease pathways, pathological conditions, adverse and side effects [12]. Methodologies and steps of drug repositioning are summarized in Fig. 8.Fig. 8. Methodologies and steps in drug repositioning [12]. 7. Drug Repurposing in COVID-19 7.1. History The first case of novel coronavirus was identified in the end of December 2019 in Wuhan, China, where 27 cases of atypical pneumonia were recorded [16]. Since then, coronavirus disease has spread worldwide and on 11th March 2020 it was declared as pandemic by World Health Organization [17]. 7.2. COVID-19 / SARS-CoV COVID-19 is acute respiratory syndrome which affects respiratory system and causes pneumonia leading to acute respiratory distress syndrome (ARDS), multi organ failure and death [18]. COVID-19 also known as coronavirus disease is an infectious disorder caused by coronaviruses (CoV). The name “coronavirus” refers to the crown-like projections on the surface of pathogens. Fever, dry cough, difficulty in breathing, chest pain and other respiratory illnesses are symptoms of COVID-19. The coronavirus spreads through respiratory droplets produced when an infected person coughs, sneezes or talks. Spread is growing when people are in close contact within 6 feet [19]. 7.3. Mechanism of SARS-CoV Infection For any viral infection to occur binding of viruses to a host cell through target receptor is necessary. The invasion of CoV into the human cells is a complex process [20]. To enter into a cell, SARS-CoV and SARS-Cov-2 both requires interaction of spike glycoprotein with angiotensin converting enzyme-2 (ACE-2). The ACE-2 protein provides an easy entry for SARS-CoV [21]. The viral S glycoprotein consists of two subunits, S1 and S2, S1 is responsible for the virus attachment to the host cell surface though the receptor-binding domain (RBD), whereas S2 is needed for the fusion of the viral and cellular membranes. SARS-CoV and SARS-CoV2 mainly infect airways and alveolar epithelial cells, macrophages. The affinity of CoV is dependent on the ability of S protein to interact with the receptor of host cell [18]. Binding of S spike glycoprotein to the human ACE2 receptor by epithelial respiratory cells, vascular endothelial cells are essential for Human infection of COVID-19 [21]. Mechanisms of action of SARS CoV are illustrated in Fig. 9.Fig. 9. Mechanisms of action of SARS-CoV [21]. 7.4. Drug Repurposing Approaches in COVID-19 There are three types of approaches computational approaches, biological experimental approaches, and mixed approaches. Computational methods are new and useful in the drug repositioning. This method gives brief information about interaction between SARS-CoV and human host cell, protein-protein interaction, drug target in human, this information helps in identification of repurposed drug [4]. According to WHO Report, the antiretroviral drugs like Favipiravir and Remdesivir, antimalarial drugs Chloroquine and Hydroxychloroquine, and HIV drugs like Lopinavir and Ritonavir are mostly repurposed in the treatment of COVID-19/SARS-CoV [22]. 7.5. Drug Repurposing Strategies for COVID-19 The drug-repurposing work process is organized distinctively from conventional drug development. In drug repurposing, there are lesser steps and different parameters to follow, namely compound identification, compound acquisition, development and FDA post-marketing surveillance. Computational drug-repositioning approaches implemented on COVID-19 can be widely categorized as (i) network-based models, (ii) structure-based approaches, or (iii) machine/deep learning approaches. There are some literature works that used hybrid approaches, and it is classified, for example, a method consisting of both network and clustering as network based if network modeling is considered to be prevalent over machine learning [23]. 7.6. Examples of Repurposed Antiviral Drugs in SARS-CoV19 Remdesivir, an antiviral agent, is the monophosphoramidate prodrug nucleoside analog. The drug was developed to treat infections caused by Ebola viruses but was found to be effective against multiple RNA viruses including parainfluenza type 3 virus, measles viruses, nipah virus. Remdesivir is considered as the most promising repurposed drug against COVID-19 infection as it efficiently inhibits SARS-CoV2 infection in human. In Italy, Remdesivir is prescribed to the patients 200 mg per 12 hours as a loading dose followed by 100 mg per 12 h via intramuscular route for 10 days in the treatment of COVID-19 with other palliative medications while in Washington USA, Remdesivir was given intravenously for 7 days to the patients who were hospitalized for the treatment of COVID19. The FDA approved Remdesivir in emergency use for the treatment of COVID-19 patients in critical conditions. Japan has also approved the use of Remdesivir in the management of COVID-19 [18, 22]. Favipiravir is a potential antiviral agent developed by Toyama Chemical Co. Ltd., Japan. Favipiravir is a nucleic acid purine base analog, 6-fluoro-3-hydroxy-2-pyrazinecarboxamide approved for treatment of influenza in Japan. Mechanism of action of this drug is inhibition of RNA dependent RNA polymerase. Favipiravir undergoes metabolism in the liver mainly by aldehyde oxidase. Favipiravirribofuranosyl-5’-triphosphate is an active metabolite of Favipiravir that is responsible for its pharmacological effect. As per clinical data, recovery rate of COVID-19 patient is increased in co-morbidity free patients. Favipiravir has been approved for treatment of COVID-19 in China on 15th February, 2020 as well as in Russia for the treatment of hospitalized COVID-19 patients [20, 22]. 8. Examples of Repurposed Drugs for Various Diseases It is an oldest example of drug repurposing. Aspirin is an acetylsalicylic acid having analgesic used non-steroidal anti-inflammatory drug (NSAID) and anti-inflammatory effect. It inhibits the activity of the enzyme called cyclooxygenase (COX) which is responsible to the formation of prostaglandins (PGs) that in turn lead to inflammation, swelling, pain and fever. Initially marketed by Bayer company in 1899 as an analgesic, Aspirin at low doses is repurposed as an anti-platelet aggregation drug. It may also be repositioned in the area of oncology in the treatment of prostate cancer [24, 25]. Thalidomide was introduced as a safe antiemetic and hypnotic. It became popular for treating nausea and vomiting in early pregnancy. Unfortunately, drug caused teratogenicity and major birth defects. WHO banned Thalidomide in the year 1962. In 1964, Dr. Jacob Sheskin from Hadassah University, demonstrated its efficacy against erythema nodosum leprosum. However, in 1998 Celgene repositioned this drug as an orphan drug for complications of leprosy, but it strictly contraindicated during pregnancy. In 2006, Thalidomide was repurposed as a first-line treatment for multiple myeloma [24, 25]. Sildenafil is a potential antihypertensive drug that causes vasodilation. It inhibits phosphodiesterase enzyme eventually leading to the inhibition of platelet aggregation. Because of these properties, it was earlier used as a promising treatment for angina, but later on, unexpected side effects like penile erections were observed during its clinical trial. Owing to this side effect, Pfizer company marketed sildenafil as a drug for erectile dysfunction in the year 1998. In 2005, Pfizer repurposed Sildenafil in the treatment of pulmonary arterial hypertension [24, 25]. Dimethyl fumarate was first synthesized in 1819 as a mould inhibitor to protect leather. It was banned in Europe in 1988 due to the allergic skin reactions. It is commonly used in Germany to treat psoriasis, due to its anti-inflammatory activity under the brand name Fumaderm. Anti-inflammatory activity was regulated by increased expression of NRF2-dependent antioxidative genes. At higher doses, Dimethyl Fumarate was repurposed in the treatment of multiple sclerosis, under the brand name Tecfidera. It is less cardiotoxic and hepatotoxic than the other drugs used in multiple sclerosis. Dimethyl Fumarate was repositioned in multiple sclerosis based on similarity between the molecular profiles of psoriasis and multiple sclerosis [24, 25]. Minoxidil is an antihypertensive agent. It is direct acting peripheral vasodilator which decreases blood pressure by decreasing peripheral vascular resistance. The metabolite of minoxidil, minoxidil sulfate, is responsible for its antihypertensive effect. Minoxidil is repurposed in the treatment of alopecia (hair loss). Topical application of Minoxidil promotes hair growth. It is also recommended to patients with alopecia areata and alopecia totalis [26]. For more examples of repurposed drugs refer Table 4. Repurposed drugs approved for both common and orphan diseases are presented in Table 5.Table 4. Examples of Repurposed Drugs [27] Drug Name Class/Action Initial Use Repurposed Use Approval Status Amantadine Nmda Receptor Antagonist Influenza Parkinson’s Disease Approved Aspirin Nsaids/Salicylates Analgesic Antiplatelet Aggregation Approved Amphotericin-B Antifungal Antibiotic Antifungal Visceral Leishmaniasis Approved Doxycycline Tetracycline Antibiotic Antibacterial Malaria Approved for prevention of Malaria Galantamine Cholinesterase Inhibitor Polio, Paralysis Alzeihmer’s Disease Approved Minoxidil Antihypertensive Hypertension Hair Loss Approved Zidovudine Nrti Anticancer Antiviral Approved Dimethyl Fumarate Methyl Ester of Fumaric Acid Treat Allergies Multiple Sclerosis Approved for treating symptoms of multiple sclerosis Thalidomide Immunomodulator Antiemetic in pregnancy First Line Treatment Multiple Myeloma Banned in pregnancy due to teratogenicity but approved in treatment of multiple myeloma Sildenafil Antihypertensive Angina Erectile Dysfunction/Pulmonary Arterial Hypertension Approved Nelfinavir Hiv Protease Inhibitor Aids Clinical Trial For Multiple Cancer Not yet approved, under clinical trials Sunitinib Tyrosine Kinase Inhibitor Renal Cell Carcinoma Pancreatic Neuroendocrine Tumours Approved Table 5. Repurposed Drugs Approved for Both Common and Orphan Diseases [4] Compound Common Disease Orphan Disease Approval Status Azathioprine Rheumatoid arthritis Renal transplant Approved Bleomycin Various cancers Pleural effusion Approved Colchicine Gout Mediterranean fever Approved Cycloserine Urinary tract infection Tuberculosis Approved Eflornithine Unwanted facial hair Sleeping sickness Approved Cyclosporine Rheumatoid arthritis Psoriasis Transplant rejection Approved Everolimus Renal cancer Renal transplant Approved Histrelin Prostate cancer Precocious puberty Approved Interferon alpha Hepatitis B and C Various cancers Approved Rituximab Rheumatoid arthritis Various cancers Approved Infliximab Ulcerative colitis Rheumatoid arthritis Psoriasis Crohn’s disease Approved 9. Conclusion The traditional tedious drug discovery process has an embarked new way in the development of new therapies based upon existing/ approved medicine, better known as drug repositioning, a more strategic and rational approach. It has offered significant reduction in R&D cost, higher probabilities of success, shorter research time and much less investments. In pandemic like COVID-19 which hit the world badly, and urgent medicine was required in a short time, drug repurposing strategy worked quite well. Better understanding of the existing drug molecules, their structure, activity and structure activity relationship is required so as to ensure higher success rates in drug repositioning. However, this strategy can be effectively utilized in the discovery and development of new drugs with novel and efficient therapeutic indications for human use. Authors’ Contributions Dr. (Mrs.) Nutan Rao designed the study and managed the work done. Ms. Ruksar Sande, Ms. Charvi Poojary, Mr. Tushar Poojari and Ms. Sonal Sawant drafted the manuscript with collective efforts. Conflicts of Interest The authors declare that they have no conflicts of interest. Funding No funds, grants or any other support was received. ==== Refs References 1. Ashburn TT Thor KB Nat. Rev. Drug Discov. 2004 3 8 673 683 10.1038/nrd1468 15286734 2. H. P. Tipnis, A. Bajaj, Clinical Pharmacy, Third edition, Career Publ. (2006), p. 371. 3. H. S. Gns, S. Gr, M. Murhari, and M. Krishnamurthy, Biomedicine & Pharmacother., 700 – 716 (2018). 4. Thayer AM Chem. Eng. News 2012 90 40 15 25 10.1021/cen-09040-cover 5. H. P. Tipnis and A. Bajaj, Clinical Pharmacy, Third edition, Career Publ. (2006), pp. 377 – 379. 6. O. Osakwe and S. A. A. Rizvi, Pharmaceutical Industry, Society and Governance, in: Social Aspects of Drug Discovery, Development and Commercialization: From Laboratory to Clinic, 1st Edn. (2016), p. 28. 7. O. Osakwe and Syed A. A. Rizvi, Introduction, the Pharmaceutical Ecosystem, in: Social Aspects of Drug Discovery, Development and Commercialization: From Laboratory to Clinic, 1st Edn. (2016), p. 20. 8. Xue H Li H Wang Y Int. J. Biol. Sci. 2018 14 1232 1244 10.7150/ijbs.24612 30123072 9. Chakraborty A Trivedi V Austin J. Biotechnol. 2015 2 3 1047 10. Talevi A Bellera C Expert Opin. Drug Discoery. 2020 15 4 397 401 10.1080/17460441.2020.1704729 11. Verbaanderd C Meheus L Huys I Trends in Cancer 2017 3 8 543 546 10.1016/j.trecan.2017.06.007 28780930 12. M. Rudrapal, S. Khairnar and A. Jadhav, Drug Repurposing (DR): An Emerging Approach in Drug Discovery, in: Drug Repurposing, Intechopen Publ. (2020). 13. Li J Zhu X Chen J PLoS Comput. Biol. 2009 5 7 10.1371/annotation/5cc0d918-83b8-44e4-9778-b96a249d4099 14. Shahreza ML Ghadiri N Mossavi SR Briefings Bioinf. 2018 19 5 878 892 10.1093/bib/bbx017 15. Lekka E Defteros SN Persidis A Drug Discovery Today: Ther. Strategies 2011 8 3–4 103 108 16. Sohrabi C Alsafi Z O’Neill N Int. J. Surg. 2020 76 71 76 10.1016/j.ijsu.2020.02.034 32112977 17. COVID-19, World Health Organization, WHO Director-General’s Opening Remarks at the Media Briefing (11 March 2020). 18. Dehelean CA Lazureanu V Coricovac D J. Clin. Med. 2020 9 7 2084 10.3390/jcm9072084 32630746 19. Singhal T Ind. J. Pediatrics 2020 87 4 281 286 10.1007/s12098-020-03263-6 20. Lisia L Lacalb P Barbacciac M Grazianic G Biochem. Pharmacol. 2020 180 114169 10.1016/j.bcp.2020.114169 32710969 21. Wang Q Zhang Y Wu L Cell 2020 181 4 894 904 10.1016/j.cell.2020.03.045 32275855 22. Singh TU Parida S Lingaraju MC Pharmacol. Rep. 2020 72 6 1479 1508 10.1007/s43440-020-00155-6 32889701 23. Dotolo S Marabotti A Facchiano A Tagliaferri R Briefings Bioinf. 2021 22 2 726 741 10.1093/bib/bbaa288 24. Pushpakom S Iorio F Eyers P Nat. Rev. Drug Discov. 2019 18 41 58 10.1038/nrd.2018.168 30310233 25. Jourdan J Bureau R Rochais C Dallemagne P J. Pharm. Pharmacol. 2020 72 9 1144 1151 10.1111/jphp.13273 26. T. Badri, T. Nessel, D. Kumar, Minoxidil, 2nd Edn., Treasure Island: Statpearls Publ. (2021). 27. Li YY Jones SJM Genome Med. 2012 4 3 27 10.1186/gm326 22494857	0	PMC9734433	NO-CC CODE	2022-12-14 23:28:28	no	Pharm Chem J. 2022 Dec 7; 56(9):1203-1214	utf-8	Pharm Chem J	2,022	10.1007/s11094-022-02778-w	oa_other
==== Front Glob J Flex Syst Manag Global Journal of Flexible Systems Management 0972-2696 0974-0198 Springer India New Delhi 324 10.1007/s40171-022-00324-x Original Research Demand and Supply Disruptions During the Covid-19 Crisis on Firm Productivity http://orcid.org/0000-0002-5406-9906 Hasan Fakhrul [email protected] 1Fakhrul Hasan is a Lecturer in Finance and Accounting at the Newcastle Businesss School, Northumbria University, UK. Prior to that, he worked at the Liverpool Hope University, Liverpool, UK, De Montfort University, Leicester, UK and University of Cumbria, UK. He obtained his PhD in Finance from the Keele University in 2018. His main research interests include behavioural finance, dividend policy, capitalism, sustainability reporting, corporate governance, and stock markets. Some of his papers published in ABS high ranking journals, including Technology Forecasting and Social Changes, Journal of Applied Accounting Research and The Journal of Prediction Markets. Bellenstedt Mary Fiona Ross [email protected] 2Mary Fiona Ross Bellenstedt Fiona started her career in Mauritius in the banking sector ten years ago. She held various positions in the retail and private banking, where she had the opportunity to serve high net worth clients. In 2016, she flew to Germany where she completed a Bachelor of Business Administration at the University of Hertforshire. In 2019, Fiona decided to continue her studies, from where she did an MSc in Finance, Leadership & Management at the University of York. She brilliantly passed her Master's degree and received a University Award—Recognition of Excellence. In February 2022, she joined Deustche Bank as a Rating & Credit Analyst, where she manages a portfolio of 100 French mid-large cap companies and North African financial institutions. http://orcid.org/0000-0002-5392-9759 Islam Mohammad Raijul [email protected] 3Mohammad Raijul Islam is an Associate Lecturer at the Manchester Metropolitan University. His primary research area covers- Business Innovation, Strategy, HRM, Financial Management, Entrepreneurship and susainability. 1 grid.42629.3b 0000000121965555 Newcastle Business School, Northumbria University, Newcastle, UK 2 grid.5685.e 0000 0004 1936 9668 York Management School, York University, York, UK 3 grid.25627.34 0000 0001 0790 5329 Manchester Metropolitan University Business School, Manchester Metropolitan University, Manchester, UK 5 12 2022 119 7 6 2022 13 11 2022 © The Author(s) under exclusive licence to Global Institute of Flexible Systems Management 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 explores the supply chain (SC) disruption impacts to the performance outcomes of a semiconductor company during the Covid-19 pandemic and proposes appropriate risk mitigation strategies to overcome the crisis. The research uses a single case study methodology and 24 SC employees from Belgium and Germany who take part in the survey. To measure the effect of SC disruptions to the firm’s financial performance, some quarterly financial statement data are used from 2018 to 2021. The regression analysis results show that there is no significant impact of SC disruptions to the firm’s productivity and non-financial performance. The paired samples t-test suggests that there is no significant change in the firm’s financial performance before and during Covid-19 either due to the market’s political and economic stability or the semiconductor company develops effective SC risk management strategies. Keywords Covid-19 Demand and supply disruptions risk Firm productivity Firm financial performance Non-financial performance Semiconductor company ==== Body pmcIntroduction As the ongoing Covid-19 pandemic has created a new era, businesses continue to understand the pandemic’s crippling effects on various aspects of their daily operations (Do et al., 2021). This virus reveals that events characterised by unprecedented uncertainty impinge on normal demand and supply patterns, causing a significant disruption in the supply chain (SC) system (Kumar & Abdin, 2021; Masudin et al., 2021; Parast & Subramanian, 2021; Sarker et al., 2021). Over 1000 companies’ more than 94% fortunes are impacted by SC disruptions due to Covid-19 pandemic (Butt, 2021a). A survey by the Institute for Supply Management (2021) finds that 97% of organisations encounter global disruptions in supply availability, production capacity, lead times and transportation of goods following Covid-19 till March 2020. As SCs embrace all activities related to the flow and processing of goods from raw materials to the ultimate finished goods to the customer (Chen, 2018), understanding of the companies can handle the disruptions and developing emergency plans has become a critical field of research in SC risk management (SCRM) (Azadegan et al., 2020; Skipper & Hanna, 2009; Tummala & Schoenherr, 2011). Furthermore, the most significant Covid-19 global demand rise has been in the medical articles business, basic foodstuffs (like pasta) and other products (like toilet paper) (Hobbs, 2020; Paul & Chowdhury, 2020) as it exceeds the prevailing domestic production rate, thereby, generating higher demand and soaring prices (McKibbin & Fernando, 2021). For instance, a study by Sheth (2020) examines the impact of this pandemic on consumer behaviour. She finds that consumers engage in stockpiling behaviours due to the unpredictability of the future supply of commodities for basic needs while postponing unnecessary purchases and embracing digitalisation. As the technology industry faces unprecedented challenges due to the lockdown measures, it stimulates the demand for telecommunications (Deloitte, 2020a). The semiconductor industry being a critical element of any technology has been greatly affected by the Coronavirus due to its globalised SC. With most of its production taking place abroad, semiconductor corporations have had to address and adjust to rising expenses, lead times and supply shortages while meeting customer demand (Accenture, 2020). Supply chain risk management (SCRM) stems from two key elements: supply and demand (Blos et al., 2009). The literature on SCRM characterises panic buying as a demand risk and the supply plant shutdown which is called a classical supply risk. They both can negatively impact the business performance at numerous levels along with the customers and suppliers (Remko, 2020). Corporations have started addressing the need for investment in SCRM to minimise the consequences of disruptions (Dubey et al., 2019; Wallin et al., 2021) and the Covid-19 event has amplified this need. Although some studies address as to how disruption affects organisational and SC performance (Chen, 2018; Parast & Subramanian, 2021; Pérez Vergara et al., 2021; Wang, 2018), there is limited research examining the effects of SC risk on firm’s productivity and performance within a semiconductor company, particularly in Covid-19 perspective. This study has four aims. Firstly, it aims to provide an SC risk sources’ thorough conceptualisation. To achieve this objective, a range of relevant SC disruptions will be identified in the literature along with the Covid-19 context alignment. Secondly, it aims to assess the SC risk impacts on firm productivity and performance following Covid-19, using the empirical evidence, surveys, and financial reports of the semiconductor company. Given that semiconductor manufacturers are highly exposed to SC risk (Vakil & Linton, 2021), it facilitates an appropriate arena for studying SC disruptions. According to Zsidisin (2003), this evaluation is critical since the first component of SCRM is to assess the impact of an event or malfunction in SC operations and financial performance. Thirdly, it aims to contribute to SCRM literature with Covid-19 pandemic perspective. Fourthly, this paper aims to provide disruption mitigation strategies as to how organisations can develop dynamic capabilities to make their SC resilient. As SC disruptions are assumed to have a detrimental impact on the organisational performance, it is essential to determine the severity of these effects and to develop organisational capabilities. As such, understanding of the SC risk in a SC network has significant practical implications (Parast & Subramanian, 2021). This research is structured as follows: Sect. 2 is dedicated to the literature review and theory. In Sect. 3, we discuss the research methodology (survey development, data collection, measurements, and analysis). Section 4 discusses the obtained results. In the final section, we conclude this paper. Literature Review and Hypothesis Development Theoretical Background Supply Chain Risks In an era of increasing globalisation and dependence on suppliers, the likelihood of not achieving the expected SC performance is high which thus exposes organisations to SC risk. Wagner and Bode (2008) define SC risk as the possible deviation from the expected value of an SC performance measure. Tummala and Schoenherr (2011) view SC risk as an event that negatively impacts SC operations and desired performance metrics, such as service delivery levels and chain agility along with cost. They associate SC risk with undesired loss and uncertainty. This SC risk conception is principally grounded on the variance-based approach (Miller, 1983). In classical decision theory, risk is described as the fluctuation in the potential results spread, their expectations and subjective values (March & Shapira, 1987). Thus, wide variations make performance unforeseeable and raise the degree of risk. The literature highlights two main categories of SC risk, i.e. operational risk and disruption risk (Chen et al., 2013; Knemeyer et al., 2009; Tang, 2006). Manuj and Mentzer (2008) define operational risk as the spread of outcomes which is associated with the unfavorable occurrences within the company. It impacts the company’s production capacity, quality and speed, and/or profitability. Operational risk is mostly related to the coordination of supply & demand and arises from flawed or defective processes, systems and people. In contrast, disruption risks are usually natural disasters (e.g. tsunami, floods) and man-made risks (e.g. war or economic crisis). In principle, disruption risks are uncontrollable and have a far more negative impact on businesses in relation to operational risks (Chen et al., 2013; Ho et al., 2015). Supply Chain Operation Risks The SC variation encompasses all disruptions that impair performance at the SC and company level. It impacts the flow of information, materials, services and the demand- supply alignment (Jüttner, 2005). In an SC, variations come primarily from three points: upstream, i.e. suppliers’ performance; downstream, i.e. customer demands and inward, i.e. company’s manufacturing operations (Davis, 1993; Germain et al., 2008). Accordingly, Chen et al. (2013) define the SC operational risk is involved with three types of risks, namely- supply risk, demand risk and process risk. In addition, the Theory of Swift and Even Flow are the techniques to analyse the SC risk impact on the organisational performance. According to this theory, the faster and smoother materials flow through the process makes the process more efficient (Schmenner & Swink, 1998). Swift, and Even Flow theory have two elements: (1) ‘the reduction of variation’ which can be measured in terms of standard, quantities and timing (2) ‘the reduction of processing time’, which is the time taken to manufacture a product or to provide a service (Schmenner, 2014). Based on this theory, variance-based SC risk (supply, demand or process risk) will shape the SC performance. Risk Sources To distinguish SC risks from other business risks, Christopher and Peck (2004) present a ‘conceptual approach of disruption’ from an SC perspective. In their model, there are five SC disruption risks classified into three classes: disruption risks within the company (process and control); disruption risks beyond the company’s control but internal to the SC (demand and supply) and disruption risks outside the SC (environmental). Given the impact of the latest disruptions (i.e. Covid-19) on SCs (illustrated in Fig. 1), this research focuses on supply and demand disruptions of the COVID-19 reality of the SC.Fig. 1 Supply chain disruption risk drivers. Source: Christopher and Peck (2004) Supply Disruption Organisations face various disruptions related to the upstream part of their SCs (Wagner & Bode, 2008). Supply disruption is the possible deviation of incoming supply regarding time, quality and quantity which may lead to unfilled orders (Aldrighetti et al., 2019; Kumar et al., 2010). Lack of consistency in suppliers’ practices will affect their performance and increase supply disruptions (Chen et al., 2013). They are triggered by many forces including supply market production capacity constraints, delivery delays, changes in product technology and design, poor supplier service, lack of supplier involvement or supplier bankruptcy etc. (Wagner & Bode, 2008; Zsidisin et al., 2000). These are likely to have immediate or delayed adverse impacts on the performance of the purchasing company in the short and/or long-run, depending on the magnitude of the disruption and the purchasing company’s ability to recover (Sheffi & Rice, 2005). In addition to the above-mentioned supply risks, empirical studies demonstrate that improving supplier quality leads to superior customer service, firm’s performance and long-term competitive advantage (Hartley et al., 2002; Shin et al., 2000). As Giunipero and Eltantawy (2004) consider the likelihood of the suppliers’ products quality enhancement and uncertainties reduction, businesses tend to minimise supply risk. Tse and Tan (2012) demonstrate that product and service quality significantly reduces supply disruptions especially in the complex or multi-layered SC. Furthermore, suppliers should possess the competence to cope with the changing market demands (e.g. customer preferences) and should sustain their competitiveness through innovative product development (He et al., 2020; Zsidisin & Ellram, 2003). Similarly, suppliers’ inability to provide the requested product will adversely affect the SC effectiveness in its core purpose (Chen et al., 2013). According to Porter’s (1985) value chain model, success is based on the unbroken links between various activities in the chain, namely inbound and outbound logistics, which will eventually affect firms’ productivity and performance. Demand Disruption SC demand disruption is associated with the product demand (Diabat et al., 2012). Demand disruption arises from the failures emerging from downstream of SC operations (Jüttner, 2005). It could be caused by the product's distribution disruptions due to transportation constraints (McKinnon, 2006) and by the volatility and unpredictability of the customer demand (Nagurney et al., 2005). Fluctuating demand can be driven by inbound shocks like economic downturn, customers’ high bargaining power, seasonality, fashion volatility, new product introductions, or short product life-cycles (Diabat, et al., 2012; Johnson, 2001). They can also be ‘vendor-induced’, i.e. some marketing activities like sales promotion and order bundling which will enhance demand fluctuations (Paul & Chowdhury, 2021; Taylor & Fearne, 2006). Moreover, one of the SC primary objectives is to match the supply to demand (Cohen & Kunreuther, 2007). Unanticipated changes in demand reduce the accuracy of supply estimates and also make it challenging for manufacturers to meet this objective. The potential gap between projected and actual demand, along with poor SC coordination, will harm SC performance and its reliability. If the projection is above the actual demand, it can lead to overstocking, obsolescence, inefficient capacity utilisation or lower prices. If the projection is below the current demand, it can create costly shortages and an inability to meet customer needs (Chen et al., 2013; Wagner & Bode, 2008). Demand amplification or the bullwhip effect is not a recent phenomenon in SC dynamics (Geary et al., 2006). It was first presented by Forrester (1958, 1961) which is the operations and SC management discipline basis. It is directly associated with the demand risks and how they are likely to disrupt the entire SC (Butt, 2021b). The bullwhip effect occurs due to the increasing order variability as one moves upstream in the SC from retailer to manufacturer (Sucky, 2009). Although consumer sales indicate a relatively constant demand, the demand/order placed by the retailer to a wholesaler is greater than the actual demand seen by that retailer. The order placed by the wholesaler to the manufacturer and the order from the manufacturer to the supplier vary even more (Meiryani et al., 2022; Paik & Bagchi, 2007). According to Sucky (2009), the bullwhip effect usually results in excessive inventory investments throughout the SC which causes significant inefficiencies as the participants involved must hedge against demand variations. Figure 2 demonstrates the variation in orders at each phase of the SC.Fig. 2 The bullwhip effect. Source: Paik and Bagchi (2007) Conceptual Framework Covid-19 and Supply Chain Disruption Epidemics represent a particular phenomenon of SC risks due to their long-term nature, high risk and ripple effects propagation (Ivanov & Dolgui, 2020). Over time, the likelihood of supply disruptions leads to a growing interest in researching the impact of epidemics on SCs, highlighting the magnitude of their threats to business continuity (Baz & Ruel, 2021; Guan et al., 2020; Natarajarathinam et al. 2009; Udofia et al., 2021). Recently, the Covid-19 outbreak has disrupted the availability of many global SCs, affected the world economy and crippled many industries (Araz et al., 2020). It is apparent that this pandemic is an unseen event compared to other types of SC disruptions, and therefore, its impact on SC is considerably greater than anything witnessed in the past (Butt, 2021a). According to Walmart, several sectors and categories, such as disinfectant, toilet paper, and hair-colorant have experienced massive panic buying due to the coronavirus (Wallace, 2020). World’s 1,000 biggest corporations had more than 12,000 facilities, stores and operations in quarantine zones as of early March 2020 (Remko, 2020). Many UK retailers’ websites have collapsed due to the overflow of online shopping (Shaw, 2020). On eBay, a bundle of 20 facemasks was sold for more than $100 (Lufkin, 2020). As for the production and services, Covid-19 impacts have prompted drastic measures such as the trade barriers enforcement and export restrictions which harmed worldwide merchandise trade (UNs, 2020; WTO, 2020). Containment measures taken and enforced by numerous governments worldwide as part of a health strategy to curb the spread of the virus have disrupted the production plants’ operations (Butt, 2021b; Ivanov, 2020). Some industries such as tourism and aviation were the most affected ones due to lockdowns and reduced public movements (Moosavi et al., 2022; Obayelu et al., 2021). According to Handfield et al., (2020), Covid-19 impacted the SC materials flow both upstream and downstream. They also added that this pandemic has caused a bullwhip effect in the manufacturing industry on an unseen magnitude. Ivanov (2020) considers this disaster to be the worst in the last decade as it has dismantled global SCs. Furthermore, research has revealed that Covid-19 is a catalyst for companies to have short-term initiatives to address or lessen upcoming challenges and to reevaluate their existing SC strategies (Handfield et al., 2020; Mollenkopf et al., 2021). Supply and Demand Disruptions During Covid-19 On the supply side, Covid-19 impact is primarily mirrored in economic terms of tradeable sections, people and workforce (Butt, 2021b; Handfield et al., 2020). For example, Rio-Chanona et al., (2020) argued that the amount of labor that is withdrawn due to social distancing, travel restrictions, self-isolation measures, illness or mortality are likely to be the main challenges impacting on countries’ supply capacity. Similarly, Lemieux et al., (2020) examined the early effects of Covid-19 on the Canadian labor market. Their analyses found that Covid-19 caused a 32% decline in total weekly working hours along with a 15% rise in unemployment—the majority of which are public facing-jobs. On the demand side, the most significant Covid-19 impact was the sharp increase in global demand and stockpiling of medical supplies, causing unexpected demand shocks and stock-outs (Friday et al., 2021; Hasan & Shahbaz, 2021). Additionally, increased export restrictions imposed by some countries experiencing shortages have also led to higher prices. Butt (2021b) highlighted that the pandemic may impact through numerous transmission channels on the demand-side, for instance reduced household expenditure with rising business uncertainty about future demand. McKibbin and Fernando (2020) observed a decline in aggregate consumer demand, especially a consumption pattern distortion and consequent market anomalies due to panic buying and customers’ preferences shift. Coibion et al., (2020) investigated the determinant effects of lockdowns on consumer spending and employment. They found that they have an adverse impact. Finally, Chronopoulos et al. (2020) explored the evolution of household spending in the UK. They found that discretionary consumption has fallen. Based on the above argument, we develop the following hypothesis: H1 Manufacturers make optimal SCRM decisions to handle demand and supply disruptions caused by the Covid-19 outbreak. The Semiconductor Industry and Its Supply Chains The demand for semiconductors has grown rapidly over the last twenty years due to the ongoing development of new applications for integrated circuits (ICs) (Mönch et al., 2018). Semiconductors allow the functioning of most electronic devices. They are an integral part of all computers, video game consoles, smartphones and associated processors etc. (Kempf et al., 2021). Moreover, big data analytics, technological developments such as 5G, the Fourth Industrial Revolution and artificial intelligence are generating a growing demand for semiconductors (Chang & Wu, 2021). According to the Semiconductor Industry Associate (SIA) (2021), chip sales are expanding dynamically and are forecast to increase by 8.8% in 2022 globally. The semiconductor SCs are directly influenced by their products and manufacturing processes (Mönch et al., 2018). Firstly, it demands high-quality capabilities from its suppliers to meet the international industry standards (ISO 9000 certification) (Briscoe et al., 2004). Secondly, semiconductor firms are particularly exposed to the bullwhip effect as they are located well upstream in the overall SC (de Kok et al., 2005; Hasan et al., 2022; Mönch et al., 2018). Thirdly, the industry has undergone substantial value chain challenges with the requirement for technical specialisation, skilled engineers, heavy capital expenditures and a rapid pace of development (Hickey & Kozlovski, 2020). Fourthly, many semiconductor organisations nowadays operate globally seeking low-cost production sites, government support and tax benefits etc., which increase their SCs complexity (Mönch et al., 2018). Fifthly, semiconductor SC is threatened by geopolitical factors between the US and China due to the struggle for technological dominance which limits the flow of materials (Crawford et al., 2020; Hasan et al., 2021). Finally, their demand cycles are relatively volatile due to the misconception about the industry regularly delivering original, advanced and cheaper microchips. This has led to a shortening product life cycle and increased fragmentation of the SC (Macher et al., 2002). Therefore, this can already result in delays and disruptions for upstream producers. The Semiconductor Industry and Covid-19 KPMG (2020) study demonstrated that many global semiconductor firms have been experiencing SC bottlenecks which impact their sales and financial performance. Likewise, the Financial Times (2021) reported that due to strict lockdowns and increase in Coronavirus cases, numerous Asian semiconductor factories and their SCs have been affected, resulting in global chip shortages. That said, the world’s semiconductor producers are largely dependent on the Asian producers (specifically Japan, Taiwan and China) for their cheap materials (Deloitte, 2020b). Therefore, whenever the Asian production is disrupted, global manufacturing SCs are severely affected. Despite the impact on the SC, global semiconductor revenue reached USD 442 billion in 2020, up 5.4% from 2019 (International Data Corporation, 2021). The pandemic has driven the demand for healthcare items such as ventilators to treat critically ill-patients (SIA). Afterwards, the second burst occurred when people rushed to buy PCs, monitors and other devices to work or study remotely. This trend was followed by an increased demand for consumer electronics such as gaming devices, TVs and smartphones etc. Another notable fact is the significant change in demand for key semiconductor components which continue to outstrip supply. For example, Samsung indicated that there is actually a severe imbalance between demand and supply which affects the TV and home-appliances productions (Shed, 2021). Volkswagen stated that they are facing microchip shortages due to the chip-suppliers reserving their supplies for tech-companies producing tablets, smartphones and gaming devices (Keohane, 2021). In fact, production is higher than demand to match supply and demand (Sparkes, 2021). Organisational Productivity Organisational productivity represents the efficiency with which resources are transformed into finished goods (Kopelman et al., 1990) and without which organisational objectives are not feasible to achieve (Ali et al., 2011). The manufacturing assembly of products is an essential component of the SC, and it must be closely monitored and continuously improved. Slack et al., (1995) stated that among the various characteristics of production performance, capacity utilisation primarily affects the speed of response to customer demand such as its impact on flexibility, lead time and delivery capability. According to Gunasekaran et al. (2004), scheduling is critical to production performance. It represents the time/date at which operations are to be conducted. This alignment defines how resources flow through an operating system and how its efficiency has a major impact on SC performance. Since scheduling is highly dependent on customer demands, scheduling techniques must be considered within this context. Rushton and Oxley (1989) asserted that the focal point of an SC has an immediate effect on the delivery of the products & services to the customers. It is a key factor in customer satisfaction. Delivery, by its nature, occurs in a dynamic environment. This makes it difficult to predict as to how changes in any key element of the distribution structure will impact the overall system. Based on the above argument, we develop the following hypothesis: H2 Demand and supply disruptions have a negative impact on Excelsior’s productivity during Covid-19. Organisational Performance With the advancement of SC management (SCM) and the increased demand for quality, timely delivery, measuring organisational performance has become a necessary concern (Beamon, 1999). Organisational performance relates to the way in which a company achieves its market and financial objectives (Yamin et al., 1999). The short-term goal of SCM is mainly to enhance productivity, diminish inventory and cycle-time while the long-term goal is to have a strong cash-flow, gain market share and revenues (Simon et al., 2015; Tan et al., 1998). Financial metrics are used to evaluate and compare organisations in terms of profitability, financial stability, growth and organisation’s behaviour over time (Holmberg, 2000). Hence, any organisational approach including SCM aims to improve organisation’s performance. Some previous studies have used a mixture of financial and non-financial indicators to measure organisational performance and to assess SC performance. For example, some studies have considered ROI and ROE as financial indicators (Galankashi & Rafiei, 2021). Wu et al. (2006) used ROI and cash-flow from profitability and operations to measure SC performance. Li et al. (2006) assessed companies’ performance and SC by four types of achievements, i.e. ROI, profit margin, sales and growth of market share. On the other hand, a range of non-financial indicators have been identified in SCM research such as customer satisfaction/retention, product/service quality, lead-time, accuracy, flexibility, responsiveness, innovation, partnership and quality etc. (Gawankar et al., 2020; Gunasekaran et al., 2004; Tangen, 2004). In view of the above literature, Table 1 illustrates the financial and non-financial measures used in this study. Based on the above argument, we develop the following hypothesis:Table 1 Financial and non-financial measures on organisational performance Financial indicators Non-financial indicators Profit margin (PM) Product quality Return on assets (ROA) Customer satisfaction Return on equity (ROE) Market share H3 Demand and supply disruptions have a negative impact on Excelsior’s (a pseudonym) of non-financial and financial performance during Covid-19. Methodology Sample and Data Collection In this research, as the Excelsior SC department is selected, it yields a homogenous sample. Access is given to two sites in Belgium and Germany. The sample participants are assumed to have experience in SCM and the research problem knowledge. As here we use two different countries’ data, it makes this research more comparable. As this research is designed to collect information related to the experience and perceptions of SC risk over the Covid-19 course, a survey is the best method for data collection. Houser (2008) describes surveys to be among the most effective and commonly used techniques for capturing primary data. A survey questionnaire via Google form is chosen not to interfere with employees’ work; to avoid any physical contact and to increase the data accuracy amid the limited research conduct time. To administer the survey, all names are provided by the HR department. An email is sent to 26 participants at their office email address with complete information about the purpose and objectives of the study. The mailings and two reminders generated 24 responses with the response rate of about 92.3%. Regarding the length of service at the company, a period of service of > 10 years constitutes 12.5% of all employees surveyed, followed by 6–10 years (20.8%), 3–5 years (41.7%) and 2–5 years comprising the remaining 25%. In terms of job position, most respondents are in operations management (45.8%), with the next positions including logistic/SCM (29.2%), sales/distribution/service (8.3%), senior positions (12.5%) and not reported (4.2%). A detailed breakdown is shown in Fig. 3.Fig. 3 Descriptive statistics of the participants. Source: Primary data As for the second part of the first RQ, quarterly figures (i.e. [Q3, Q4 year 2018, Q1–Q4 year 2019—pre-Covid-19] and [Q1–Q4 year 2020, Q1, Q2 year 2021—during Covid-19] of Excelsior’s balance sheets and income statements are collected from their website to calculate financial ratios which are used to compare the company’s performance before and during Covid-19 pandemic. Survey Questionnaire Development & Structure A multi-stage process is undertaken to develop and validate the questionnaire. Initially, an in-depth examination of operations and SCRM literature is conducted to determine relevant concepts, operation definitions, and survey measurement metrics. Approved measures from previous surveys are adjusted to match our investigation for demand and supply disruption during the pandemic. The demand risk measure captures the risk arising from changes in customer demand and market volatility during Covid-19 (Chen et al., 2013; Ho et al., 2015; Parast & Subramanian, 2021; Wagner & Bode, 2008). Similarly, SC risk encompasses risk from Covid-19 and its upstream SC actors such as the suppliers’ performance (Chen et al., 2013; Ho et al., 2015; Parast & Subramanian, 2021; Wagner & Bode, 2008). Organisational productivity measures are taken from Udofia et al. (2020). Finally, organisational non-financial and financial performance measures are taken from (Chen, 2018; Simon et al., 2014). Table 2 lists the constructs, their details, and sources.Table 2 Constructs, details and sources Construct Details Sources Supply disruption Measured on four elements based on the upstream side of the company's SC, including activities such as purchasing and supplier quality Wagner and Bode (2008) Chen et al. (2013) Ho et al. (2015) Parast and Subramanian (2021) Demand disruption Measured on four elements that assess internal risks associated with changing demand, market uncertainty, mismatch between actual and projected demand and bullwhip effect Wagner and Bode (2008) Chen et al. (2013) Ho et al. (2015) Parast and Subramanian (2021) Firm productivity Assessed by production delays, change in technology employed and production capacity Udofia et al. (2021) Firm performance Measured by product quality, overall competitive position, customer satisfaction, PM, ROA, ROE (non-financial and financial performance) Simon et al. (2014) Chen (2018) As illustrated in appendices 2–3, the questionnaire comprises a consent form and multiple-choice questions. According to Vinten (1995), this type of interrogation requires participants’ minimum time and effort. The initial questionnaire is in English and is translated into German for Germany. It is felt that respondents are more likely to respond if it was written in their own language. A five-point Likert scale, ranging from strongly disagree (represented by value ‘1’) to strongly agree (represented by value ‘5’) is found suitable as it is a common format for evaluating respondents’ opinions in terms of degree of agreement with positive and negative items (Wang et al., 2015). The questionnaire is divided into three parts with definitions to guide participants. Part 1 explores the sources of disruption during Covid-19. Part 2 investigates the relationship between disruption sources and performance outcomes whereas part 3 profiles the overall SC employees. Measurements Non-financial Performance Respondents’ answers to SC disruptions are used as the unit of analysis. Eight measurement items, along with demand and supply disruptions are treated as independent variables. Six measurement items, firm productivity and performance are addressed as dependent variables. Financial Performance The profit margin (PM), ROA, ROE are used to evaluate Excelsior’s financial performance. The grounds are simple: PM reflects the entire operational performance of a company (Salancik & Pfeffer, 1980), ROA measures how effectively a company’s management produces revenue from its assets/resources (Mahajan & Singh, 2013) and lastly ROE indicates the efficiency with which businesses generate income from shareholders’ capital (Lau & Sholihin, 2005). The following formulas are used to calculate PM, ROA, ROE:1 Profitmargin=NetIncomeSales 2 Returnonassets=NetIncomeTotalAssets 3 Returnonequity=NetIncomeShareholders′equity Findings and Discussions Non-financial Performance Scale Reliability and Validity Prior to data analysis, the questionnaire is tested for its reliability and validity. Reliability is an estimate of the level of consistency between various measures of a variable. Validity is a major characteristic that reflects the degree of reliability of measurements (Wang et al., 2015). Content, convergent and discriminant validity are also examined. With respect to content validity, it is assumed that each of the concepts is precisely articulated and has a clear meaning (Yan et al., 2014). Convergent validity determines the level of correlation between two measures of the same construct. A correlation of > 0.7 is required for convergent validity. Conversely, discriminant validity is the level at which two similar concepts are different (Hair, 2010). The measurement model is presented in Table 3. It includes the factor loadings, average variance extracted (AVE) and composite reliabilities (CRs) of each construct (Appendix 5 describes the calculations). One factor loading (Fper3) is < 0.5 and therefore, it is removed. All AVE values are above the minimum, i.e. 0.40 (Namagembe et al., 2019), with values ranging from 0.4703 to 0.6879 which support the discriminant validity (Parast & Subramanian, 2021). The CRs range between 0.6353 and 0.8686 which supports convergent validity (Parast & Subramanian, 2021). Therefore, based on these results, there is a strong level of reliability. So, we can tell that our first hypothesis is true.Table 3 Constructs in the SCR non-financial model Variables Indicators Factor loading Composite reliability Average variance extracted Demand disruption DR1 0.7232 0.8269 0.5472 DR2 0.6157 DR3 0.8414 DR4 0.7607 Supply disruption SR1 0.6431 0.8268 0.5518 SR2 0.5676 SR3 0.8788 SR4 0.8362 Firm productivity Fpro1 0.8229 0.8686 0.6879 Fpro2 0.8525 Fpro3 0.8123 Firm performance Fper1 0.5843 0.6353 0.4703 Fper2 0.7741 Fper3 – Source: Primary data Descriptive Statistics The means, standard deviations, skewness and kurtosis are presented in Table 4.Table 4 Descriptive statistics (non-financial performance) Variables Min. statistic Max. statistic Mean SD Skewness statistic Kurtosis statistic SE SE Demand risk 3.75 5.00 4.4792 0.4418 − 0.275 0.472 − 1.430 0.918 Supply risk 2.75 5.00 4.2188 0.6355 − 0.679 0.472 − 0.199 0.918 Firm productivity 2.33 4.33 3.2639 0.6058 0.148 0.472 − 0.822 0.918 Firm performance 1.00 3.00 1.9583 0.6064 0.491 0.472 − 0.545 0.918 Source: Primary data (SPSS version 27) With a mean of 4.4792 and a low standard deviation of 0.4418, it is indicated that Excelsior experiences a very high demand disruption. Similarly, Excelsior experiences a very high supply disruption with a mean of 4.2188 and a low standard deviation of 0.6355. With a mean of 3.2639, Excelsior faces a moderate impact on its productivity with respect to demand and supply disruption. Finally, there is a low impact on the company’s non-financial performance with a mean of 1.9583 and standard deviation of 0.6064. The data are normally distributed with skewness values < 2, ranging from − 0.679 to 0.491, and kurtosis values < 7, ranging from − 0.199 to − 1.430. According to Namagembe and et al., (2019), the absolute value of univariate skewness must be < 2 and that of univariate kurtosis be < 7, to denote the existence of a normal distribution. Pearson’s Correlation Analysis Correlation analysis is essentially applied to measure the strength of relationships between variables (Lin, 2021). Presented in Table 5, the Pearson’s correlation coefficient is used to assess whether the correlation between the variables was statistically significant using the p-value. If its value is small (i.e. p < 0.01), it indicates that there is a statistically significant relationship between the variables presented (Kaawaase et al., 2021). Results indicate a significant and positive relationship between demand and supply disruptions (r=0.520,p<0.01). However, both demand and supply disruptions are not significantly associated with firm productivity (r=0.238,p>0.01 and r=0.201,p>0.01). Furthermore, there is a non-significant negative correlation between demand risk and firm’s non-financial performance (r=-0.227,p>0.01) whereas supply disruptions are insignificantly associated with the firm’s non-financial performance (r=0.067,p>0.01).Table 5 Pearson’s correlation analysis (primary data) Variables Demand disruptions Supply disruptions Firm productivity Firm performance Demand disruptions 1.000 Supply disruptions 0.520a 1.000 Firm productivity 0.238 0.201 1.000 Firm performance − 0.227 0.067 0.406b 1.000 aCorrelation is significant at the 0.01 level (2-tailed) bCorrelation is significant at the 0.05 level (2-tailed). (SPSS version 27) Regression Analysis Regression analysis is an effective statistical technique for examining the relationship between variables (Gray, 2002). This research uses a simple linear regression analysis. The formula below is utilised to demonstrate the linear relationship between the variables:4 y=∝+β∗x+μ where y and x represent the values of dependent and independent variables, respectively, α represents the intercept, β denotes the slope of the regression line and μ is the residual error term (Ambrosius, 2007). Table 6 summarises the results of the linear regression analysis. The beta coefficient between demand disruption and firm productivity is equal to 0.238 (a moderate positive correlation) and firm performance is equal to − 0.227 (a negative correlation). The low R2 values of 0.057 and 0.051 state that demand disruptions account for 5.7% and 5.1%, respectively, of total variance in firm’s productivity and non-financial performance. The F-statistic is an assessment of how well the regression has optimised the prediction of the outcome relative to the accuracy level of the model (Field, 2005). With values of 1.321 and 1.190, it indicates that the findings are not robust. The p-values are equal to 0.263 and 0.287 which are greater than all significant levels i.e. 1%, 5% and 10%. Therefore, demand disruptions have no significant impact on Excelsior’s productivity and non-financial performance.Table 6 Regression analysis (primary data) Regression weights Beta coefficient R2 Adjusted R2 F p-value DR → Firm pro 0.238 0.057 0.014 1.321 0.263 DR → Firm per − 0.227 0.051 0.008 1.190 0.287 SR → Firm pro 0.201 0.040 − 0.003 0.928 0.346 SR → Firm per 0.067 0.004 − 0.041 0.099 0.756 Source: Primary data (SPSS version 27) According to Table 7, there is a low beta coefficient of 0.201 between supply disruptions and firm productivity. The beta coefficient of 0.067 between supply disruptions and non-financial performance indicates a weak relationship. The low R2 values of 0.040 and 0.004 explain that supply disruptions account for 4% and 0.4% of the total variance in firm productivity and non-financial performance. Finally, the F-statistic is equal to 0.928 and 0.099. The p-values are equal to 0.346 and 0.756 which are greater than all significant levels. It demonstrates that there is a statistically insignificant relationship between supply disruptions and Excelsior’s productivity and non-financial performance. Based on these results we can accept our hypothesis three (H3).Table 7 Descriptive statistics (financial performance) N Mean SD % increase PM_before Covid-19 6 15.02 4.1109 PM_during Covid-19 6 15.35 4.0489 2.12 ROA_before Covid-19 6 4.57 1.7261 ROA_during Covid-19 6 4.81 1.8379 5.00 ROE_before Covid-19 6 6.00 2.0946 ROE_during Covid-19 6 6.30 2.2645 4.83 Source: Secondary data (Figures extracted from Excelsior’s financial reports. Quarterly figures 2018–2021) Financial Performance Descriptive Statistics The descriptive analysis results in Table 8 indicate that there is an increase-trend in average PM, ROA and ROE at Excelsior during Covid-19 opposed to pre-pandemic time. The average percentage increase is 2.12%, 5.00%, 4.83%, respectively, which indicates that Excelsior experiences an amelioration in its financial performance. Based on these results, we can accept our second hypothesis (H2).Table 8 The paired samples T-test of PM, ROA and ROE samples Paired samples test Paired differences t df Sig. (2-tailed) Mean SD SE mean 95% confidence interval of the difference Lower Upper PM_before Covid- 19 PM_during Covid-19 − 0.32667 7.20291 2.94057 − 7.88565 7.23232 − 0.111 5 0.916 ROA_before Covid-19 ROA_during Covid-19 − 0.24167 3.09042 1.26166 − 3.48486 3.00153 − 0.192 5 0.856 ROE_before Covid-19 ROE_during Covid-19 − 0.30333 3.69970 1.51039 − 4.18593 3.57926 − 0.201 5 0.849 Source: Secondary data (Figures extracted from Excelsior financial reports. Quarterly figures 2018–2021). (SPSS version 27) Paired Samples t-Test A normality test is first performed before running the test. As shown in Appendix 6, the scores show that all constructs meet the requirements of normality. The results from the paired samples t-test in Table 8 show that there is no significant difference in the overall financial performance of Excelsior between pre-Covid-19 and during-Covid-19 as all p-values are equal to 0.916, 0.856, 0.849 (p > 0.05). Theoretical Contributions This study analyses the impact of SC disruptions on firm productivity, non-financial and financial performance caused by Covid-19. This research contributes to literature in three essential aspects. Firstly, by mapping different types of disruptions in the literature review, the 24 participants from Excelsior’s SC department provide evidence of the multiple forces’ presence during the pandemic. This in-depth examination validates the typical retrospective descriptions of disruptions such as the bullwhip effect (Forrester, 1958). This paper empirically also tests the conceptual model of SC disruptions developed by Christopher and Peck (2004). So as far as disruptions are concerned, Excelsior’s operating environment is uncertain and evolving. Secondly in answering to the H1, this research provides empirical evidence that there is no profound/significant impact between SC disruptions and firm non-financial performance. These results are contrary to our expectations for two reasons– the general insight in the reviewed literature and worldwide scale of SC disruptions due to the pandemic. As it was presumed that the majority of the global semiconductor industry’s SC did not withstand Covid-19, many existing suppliers are unable to respond to the high demand (Accenture, 2020). Our findings need to be contrasted with the conclusions of Parast and Subramanian (2021). They showed that both demand and supply risks have an extreme impact on firm performance. However, this contradiction may appear due to the different sample, distinct country, and the number of respondents. Thirdly in answering to the H2, the results provide empirical evidence that there is no impact of SC disruptions on Excelsior productivity. These conclusions are conflicting with the extant literature (Butt, 2021b; Udofia et al., 2020). However, four explanations are possible (1) there is a perception gap between how academics and practitioners think SCR affects firm productivity vs how SCR variables really affect firm productivity. The constructs retained for firm productivity are expected to be experienced by Excelsior during Covid-19. Therefore, this judgment may have resulted in an insignificant impact on the company’s productivity. (2) Excelsior is engaged in agile production. Studies have shown that organisations during this pandemic crisis have dealt with these disruptions by adopting agile manufacturing. For instance, Butt (2021b) demonstrated that companies have prioritised some production when they expected a shortage of direct materials and inventories. (3) Excelsior may have focused on tier 1 supplier risk and get visibility on their inventory, production process and fulfillment status. Prior studies confirmed that businesses have identified their first-tier suppliers to battle Covid-19 (Butt, 2021b). (4) the impacts of Covid-19 on production and value chains have differed greatly across products and countries. The contextual settings in developed countries such as Germany and Belgium may not fully affect production. Production is highly automated, with inherent social distancing and institutional stability which may not require changes in some production practices (Swinnen & Vos, 2021). These benefits could have had little impact on Excelsior’s productivity. Nevertheless, additional research needs to be conducted to shed light on the unexpected results regarding SC disruptions risk on firm productivity, non-financial and financial performance. Practical Implications for Managers Results from the present study show that demand and supply disruptions are highly prevalent in times of crisis. Therefore, it is critical that SC managers prepare themselves and implement key strategies for a prosperous post-Covid-19 world. To answer RQ3, some insights for manufacturing companies are proposed in a bid to address SC disruption risks. Firstly, SC managers can foster agility by accumulating resources that act as ‘shock buffers’, i.e. inflating inventory, having flexible production methods, locating secondary suppliers and having a product design (Bode et al., 2011). Accordingly, Butt (2021b) shows that during Covid-19, managers of buying companies closely monitor their suppliers’ functions i.e. their production schedules, inventory positions and shipment status in order to forecast any sudden supplier shortfalls. However, taking such proactive measures require high upfront investments. Secondly, SC collaboration with internal and external partners reduces SC risk (Chen et al., 2013). Collaboration and information sharing across the SC enables SC partners to share knowledge about plans, needs and progress, thereby, improving SC performance and minimising uncertainty. Milliken (1987) argues that uncertainty arises from a lack of adequate information to predict precisely. Consequently, with accurate visibility into upstream and downstream movements, SC managers would be confident about order cycle times, demand forecasts, suppliers’ ability to deliver etc. Thus, investing in visibility are sound agility strategies that avoid double guessing and provide businesses with resources to respond to SC disruptions (Gunessee & Subramanian, 2020). Thirdly, for unforeseen turbulence, leveraging technology and extensive data mining, such as artificial intelligence, internet of things, blockchains, machine learning, control system, automated production, 3D printing etc. are significant in this pandemic (Sharma et al., 2020). Studies show that enterprises with more advanced digital capabilities are quicker to overcome SC disruptions during Covid-19 (Sajjad, 2021). Moreover, Cai and Luo (2020) demonstrate that these technologies assist in the production of high-demand products, speed up the delivery system and recalibrate and optimise SC planning during the pandemic. Thus, SC managers can better capture high-quality data along the value chain, increase SC visibility and take necessary corrective actions based on early warning signals. Covid-19 has undoubtedly brought attention to the circular economy which leads to resilience (Khan et al., 2021; Nandi et al., 2021). It focuses on the efficient use of resources and reduction of waste throughout the entire value chain. In times of high uncertainty, resources are precious and need to be used efficiently by integrating the ‘reduce-redesign-reuse’ approach. Fourthly, to get through a crisis, a more effective strategy is required to leverage existing relationships to stabilise the effect of the shock (Runfola et al., 2021). Companies need to know the production recovery status of other SC partners and assist them as it can be costly and difficult to replace them. Therefore, offering financial and non-financial incentives can contribute to a smooth SC recovery (Cai & Luo, 2020). Fifthly, firms need to have a transition from their traditional linear SC approach by adopting a more modern and holistic system such as the digital supply network (DSN) (Kilpatrick & Barter, 2020). DSN offers suppliers, producers and customers to work collaboratively through a dynamic data-sharing platform powered by real-time data (Sajjad, 2021). This empowers businesses to optimally align and connect with their SC network partners which subsequently enhance a firm’s agility and overall competitiveness. Conclusion SC disruptions caused by Covid-19 crisis highlight that pandemics have destructive effects on both demand and supply. This research presents an analysis of the impact of SC disruptions on firm productivity, financial and non-financial performance during Covid-19, in a semiconductor company. The results indicate that SC disruption risk does not necessarily have a significant impact on business performance outcomes. Moreover, the results have fueled the growth of SCRM literature. The ‘Triple-A’ mitigation measures are proposed namely, Agility, Adaptability and Alignment. They cover the importance of flexibility, collaboration with all SC partners, SCs digitalisation & localisation and adoption of a ‘reduce-redesign-reuse’ approach. Limitations and Future Research Avenues This research has some limitations. Firstly, the data are collected from branches based in Germany and Belgium. Therefore, the results can be verified in countries with comparable political, economic and geographical environments. This research offers the possibility to replicate it in other non-European countries with different economic and political contexts in order to ameliorate the external visibility of the results. Secondly, the data are obtained from a rather small sample size of 24 participants and in a single semiconductor company. Accordingly, generalisation of these results to a larger population should be done with care. Our results should be tested with various semiconductor manufacturers with the involvement of more participants. Thirdly, in measuring Excelsior’s financial performance during Covid-19 period, the lack of observations made it challenging to gain a thorough understanding of its actual impact. Future research involving longitudinal data could assist in addressing this limitation. Fourthly, this paper considers three financial measures, namely PM, ROA and ROE to assess the impact of SC disruptions on firm financial performance. However, they do not fully represent the overall picture of Excelsior’s financial performance. Other financial indicators like working capital or operating cash flow can be used for future research to understand whether businesses have rushed towards cash and liquidity to maintain operations. Finally, due to the uniqueness of Covid-19 and its impact on global SCs, it would have been more valuable to evaluate the effect of Covid-19 using a mixed-methods methodology. It is felt that the use of both interviews and surveys would have resulted in deeper insights and richer data. Key Questions Manufacturers make optimal SCRM decisions to handle demand and supply disruptions caused by the Covid-19 outbreak. Demand and supply disruptions have a negative impact on Excelsior’s productivity during Covid-19. Demand and supply disruptions have a negative impact on Excelsior’s (a pseudonym) of non-financial and financial performance during Covid-19. Appendix 1. Factor Loadings, Average Variance Extracted and Composite Reliabilities The steps to calculate the average variance extracted and composite reliabilities using SPSS version 27 are demonstrated below: Step 1: I computed the factor loadings first.analyse→dimensionreduction→factor→movethevariablestotherightrotation→varimax→continue→options→suppresssmallcoefficients→absolutevaluebelow0.5→continue→ok Step 2: Copy the table to an excel sheet.Insert the factor loading as λ Calculate the λ2 (e.g. DR 0.7232^2 = 0.5231) To calculate the ε = 1 − λ2. (e.g. 1 − 0.5231 = 0.4769) Calculate the sum of λ, λ2, ε N represents the number of factor loadings (number of observations) To calculate the average variance extracted is equal to the sum of λ2/N (e.g. = 2.1889/4 = 0.5472) To calculate the composite reliabilities is equal to sum λ^2/(λ^2 + ε) (e.g. = 2.941^2/(2.941^2 + 1.1811) Appendix 2. Descriptive Statistics (Financial Performance) Variables Min. statistic Max. statistic Mean SD Skewness statistic Kurtosis statistic SE SE PM_Before 11.97 20.52 15.0200 4.11094 0.942 0.845 − 1.850 1.741 PM_During 9.64 21.06 15.3467 4.04894 − 0.056 0.845 − 0.373 1.741 ROA_Before 3.22 6.92 4.5683 1.72612 0.944 0.845 − 1.810 1.741 ROA_During 2.18 7.20 4.8100 1.83786 − 0.296 0.845 − 0.844 1.741 ROE_Before 4.10 8.73 6.0033 2.09458 0.850 0.845 − 1.850 1.741 ROE_During 2.94 9.03 6.3067 2.26449 − 0.565 0.845 − 0.845 1.741 Source: Secondary data (SPSS version 27). Funding Not appropriate. Declarations Conflict of interest No conflict of interest. Ethical Approval Not appropriate. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Accenture (2020). Semiconductor companies: Business resilience in the wake of the Covid-19. Available at: https://www.accenture.com/_acnmedia/PDF-126/Accenture-High-Tech-Covid-19-SEMICONDUCTOR-Final.pdf Aldrighetti R Zennaro I Finco S Battini D 3Healthcare Supply Chain Simulation with Disruption Considerations: A Case Study from Northern Italy Global Journal of Flexible Systems Management 2019 20 S1 81 102 10.1007/s40171-019-00223-8 Ali SI Yousof J Khan MR Massod SA Evaluation of performance in manufacturing organisation through productivity and quality African Journal of Business Management 2011 5 6 2211 2219 Ambrosius WT Topics in Biostatistics 2007 Humana Press 143 164 Araz OM Choi T Olson DL Salman FS Data Analytics for Operational Risk Management A Journal of Decision Sciences Institute 2020 51 6 1316 1319 10.1111/deci.12443 Azadegan A Parast MM Lucianetti L Nishant R Blackhurst J Supply chain disruptions and business continuity: An empirical assessment Decision Sciences 2020 51 1 38 73 10.1111/deci.12395 Baz JE Ruel S Can supply chain risk management practices mitigate the disruption impacts on supply chains’ resilience and robustness? Evidence from an empirical survey in a COVID-19 outbreak era International Journal of Production Economics 2021 233 107972 10.1016/j.ijpe.2020.107972 Beamon BM Measuring supply chain performance International Journal of Operations and Production Management 1999 19 3 275 10.1108/01443579910249714 Blos MF Quaddus M Wee HM Watanabe K Supply chain risk management (SCRM): A case study on the automotive and electronic industries in Brazil Supply Chain Management 2009 14 4 247 252 10.1108/13598540910970072 Bode C Wagner S Petersen K Ellram L Understanding responses to supply chain disruptions: Insights from information processing and resource dependence perspectives Academy of Management Journal 2011 54 4 833 856 10.5465/amj.2011.64870145 Briscoe J Nels Lee T Fawcett SE Benchmarking challenges to supply-chain integration: Managing quality upstream in the semiconductor industry Benchmarking: an International Journal 2004 11 2 143 155 10.1108/14635770410532589 Butt AS Understanding the implications of pandemic outbreaks on supply chains: An exploratory study of the effects caused by the COVID-19 across four South Asian countries and steps taken by firms to address the disruptions International Journal of Physical Distribution & Logistics Management 2021 52 4 370 392 10.1108/IJPDLM-08-2020-0281 Butt, A. S. (2021b). Strategies to mitigate the impact of COVID-19 on supply chain disruptions: a multiple case analysis of buyers and distributors. The International Journal of Logistics Management, Vol. ahead-of-print. Cai M Luo J Influence of COVID-19 on manufacturing industry and corresponding countermeasures from supply chain perspective Journal of Shanghai Jiaotong University (science) 2020 25 409 416 10.1007/s12204-020-2206-z 32834699 Chang BC Wu KS The nonlinear relationship between financial flexibility and enterprise risk-taking during the COVID-19 pandemic in Taiwan’s semiconductor industry Oeconomia Copernicana 2021 12 2 307 333 10.24136/oc.2021.011 Chen HL Supply chain risk’s impact on corporate financial performance International Journal of Operations & Production Management 2018 38 3 713 731 10.1108/IJOPM-02-2016-0060 Chen J Sohal AS Prajogo DI Supply chain operational risk mitigation: A collaborative approach International Journal of Production Research 2013 51 7 2186 2199 10.1080/00207543.2012.727490 Christopher M Peck H Building the resilient supply chain The International Journal of Logistics Management 2004 15 2 1 13 10.1108/09574090410700275 Chronopoulos, D. K., Lukas, M., & Wilson, J. O. S. (2020). Consumer spending responses to the COVID-19 pandemic: An assessment of Great Britain. SSNR, Working Paper, University of St. Andrews. Clark, L., KMPG. (2020). The impact of Covid-19 on the semiconductor industry. https://advisory.kpmg.us/articles/2020/impact-of-covid-19-on-semiconductor-industry.html Cohen MA Kunreuther H Operations risk management: Overview of Paul Kleindorfer’s contributions Production & Operations Management 2007 16 5 525 541 10.1111/j.1937-5956.2007.tb00278.x Coibion, O., Gorodnichenko, Y., & Weber, M. (2020). The cost of the Covid-19 crisis: Lockdowns, macroeconomic expectations, and consumer spending. National Bureau of Economic Research, Working paper 27141, University of Texas at Austin. Crawford, A., Wu, D., Murphy, C., King, I., & Bloomberg. (2020). The US-China conflict over chips is about to get uglier. https://www.bloomberg.com/news/articles/2020-10-21/the-u-s-china-conflict-over-chips-is-about-to-get-even-uglier Davis T Effective supply chain management Sloan Management Review 1993 34 4 35 46 de Kok T Janssen F Van Doremalen J Van Wachem E Clerkx M Peeters W Phillips electronics synchronizes its supply chain to end the bullwhip effect Interfaces 2005 33 1 37 48 10.1287/inte.1040.0116 Deloitte. (2020a). Understanding the Covid-19’s impact on the telecom sector. file:///C:/Users/BYM/Downloads/gx-understand-sector-impact-telecom.pdf Deloitte. (2020b). COVID-19: A black swan event for the semiconductor industry?. https://www2.deloitte.com/global/en/pages/about-deloitte/articles/a-black-swan-event-for-the-semiconductor-industry-covid-19.html Diabat A Govindan K Panicker VV Supply chain risk management and its mitigation in a food industry International Journal of Production Research 2012 50 11 3039 3050 10.1080/00207543.2011.588619 Do QN Mishra N Wulandhari NBI Ramudhin A Sivarajah U Milligan G Supply chain agility responding to unprecedented changes: Empirical evidence from the UK food supply chain during COVID-19 crisis Supply Chain Management 2021 26 6 737 752 10.1108/SCM-09-2020-0470 Dubey R Gunasekaran A Childe SJ Papadopoulos T Blome C Luo Z Antecedents of resilient supply chains: An empirical study IEEE Transactions on Engineering Management 2019 66 1 8 19 10.1109/TEM.2017.2723042 Feizabadi J Maloni M Gligor D Benchmarking the triple-A supply chain: orchestrating agility, adaptability, and alignment Benchmarking: an International Journal 2019 26 1 271 295 10.1108/BIJ-03-2018-0059 Forrester JW Industrial dynamics: A major breakthrough for decision makers Harvard Business Review 1958 36 37 66 Forrester JW Industrial dynamics 1961 MIT Press Friday D Savage DA Melnyk SA Harrison N Ryan S Wechtler H A collaborative approach to maintaining optimal inventory and mitigating stock out risks during a pandemic: Capabilities for enabling health-care supply chain resilience Journal of Humanitarian Logistics and Supply Chain Management 2021 11 2 248 271 10.1108/JHLSCM-07-2020-0061 Galankashi RM Rafiei MF Financial performance measurement of supply chains: A review International Journal of Productivity and Performance Management 2021 71 5 1674 1707 10.1108/IJPPM-11-2019-0533 Gawankar SA Gunasekaran A Kamble S A study on investments in the big data-driven supply chain, performance measures and organisational performance in Indian retail 4.0 context International Journal of Production Research 2020 58 5 1574 1593 10.1080/00207543.2019.1668070 Geary S Disney SM Towill DR On bullwhip in supply chains—historical review, present practice and expected future impact International Journal of Production of Economics 2006 101 1 2 18 10.1016/j.ijpe.2005.05.009 Germain R Claycomb C Droege C Supply chain variability, organisational structure, and performance: The moderating effect of demand unpredictability Journal of Operations Management 2008 26 5 557 570 10.1016/j.jom.2007.10.002 Giunipero LC Aly Eltantawy R Securing the upstream supply chain: A risk management approach International Journal of Physical Distribution & Logistics Management 2004 34 9 698 713 10.1108/09600030410567478 Gray JB Introduction to linear regression analysis Technometrics 2002 44 2 191 192 10.1198/tech.2002.s714 Gunasekaran A Patel C McGaughey RE A framework for supply chain performance measurement International of Production of Economics 2004 87 3 333 347 10.1016/j.ijpe.2003.08.003 Gunessee S Subramanian N Ambiguity and its coping mechanisms in supply chains: Lessons from Covid-19 pandemic and natural disasters International Journal of Operations and Production Management 2020 40 7/8 1201 1223 10.1108/IJOPM-07-2019-0530 Guan D Wang D Hallegatte S Global supply-chain effects of COVID-19 control measures Nature Human Behaviour 2020 4 577 587 10.1038/s41562-020-0896-8 Hair JF Multivariate data analysis 2010 7 Prentice Hall Handfield RB Graham G Burns L Coronavirus, tariffs, trade wars and supply chain evolutionary design International Journal of Operations & Production Management 2020 40 10 1649 1660 10.1108/IJOPM-03-2020-0171 Hasan F Shahbaz AA COVID-19 and two different stock markets: An event study analysis The Journal of Prediction Markets 2021 15 2 3 18 10.5750/jpm.v15i2.1850 Hasan F Hossen R Shahbaz AA How steel industry has been affected due to trade war between the USA and China International Journal of Critical Accounting 2021 12 4 348 368 10.1504/IJCA.2021.118328 Hasan F Islam MR Ishrat F COVID-19 pandemic impact on the supply chains of the UK-based multinational manufacturing companies Business Ethics and Leadership. 2022 6 2 44 67 10.21272/bel.6(2).44-67.2022 Hartley J Greer B Park S Chrysler leverages its suppliers' improvement suggestions Interfaces 2002 32 4 20 27 10.1287/inte.32.4.20.46 He Y Li S Xu H Chunming Shi C An in-depth analysis of contingent sourcing strategy for handling supply disruptions IEEE Transactions on Engineering Management 2020 67 1 201 219 10.1109/TEM.2018.2868716 Hickey P Kozlovski E E-strategies for aftermarket facilitation in the global semiconductor manufacturing industry Journal of Enterprise Information Management 2020 33 3 457 481 10.1108/JEIM-05-2019-0124 Ho W Zheng T Yildiz H Talluri S Supply chain risk management: A literature review International Journal of Production Research 2015 53 16 5031 5069 10.1080/00207543.2015.1030467 Hobbs JE Food supply chains during the COVID-19 pandemic Canadian Journal of Agricultural Economics 2020 68 2 171 176 10.1111/cjag.12237 Holmberg S A systems perspective on supply chain measurements International Journal of Physical Distribution and Logistics Management 2000 30 10 847 868 10.1108/09600030010351246 Houser J Nursing research: Reading, using, and creating evidence 2008 Jones and Bartlett Publishers Institute for Supply Management. (2021). Covid-19’s Global impact on supply chains. https://www.ismworld.org/globalassets/pub/research-and-surveys/white-papers/white_paper_coronavirus_round4_research.pdf. International Data Corporation (IDC). (2021). Worldwide semiconductor revenue grew 5.4% in 2020 despite COVID-19 and further growth is forecast in 2021, According to IDC. https://www.idc.com/getdoc.jsp?containerId=prUS47424221. Ivanov D Predicting the impacts of epidemic outbreaks on global supply chains: A simulation-based analysis on the coronavirus outbreak (Covid-19/SARS-CoV-2) case Transportation Research Part e: Logistics and Transportation Review 2020 136 101922 10.1016/j.tre.2020.101922 32288597 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 Johnson ME Learning from toys: Lessons in managing supply chain risk from the toy industry California Management Review 2001 43 3 106 124 10.2307/41166091 Jüttner U Supply chain risk management: Understanding the business requirements from a practitioner perspective The International Journal of Logistics Management 2005 16 1 120 141 10.1108/09574090510617385 Kaawaase TK Nairuba C Akankunda B Bananuka J Corporate governance, internal audit quality and financial reporting quality of financial institutions Asian Journal of Accounting Research 2021 6 3 348 366 10.1108/AJAR-11-2020-0117 Kempf T Bobek V Horvat T The impacts of the American–Chinese trade war and Covid-19 pandemic on Taiwan’s sales in semiconductor industry International Journal of Economics and Finance 2021 13 4 62 76 10.5539/ijef.v13n4p62 Keohane, D. (2021). Car manufacturing hit by global semiconductor shortage. https://www.ft.com/content/e264fd41-7ee9-4fba-be3c-21446298efd9. Khan S Haleem A Deshmukh SG Javaid M Exploring the impact of Covid-19 pandemic on medical supply chain disruption Journal of Industrial Integration and Management 2021 6 2 235 255 10.1142/S2424862221500147 Kilpatrick, J., Barter, L., & Deloitte. (2020). COVID-19: Managing supply chain risk and disruption. https://www2.deloitte.com/content/dam/Deloitte/ca/Documents/finance/Supply-Chain_POV_EN_FINAL-AODA.pdf. Knemeyer AM Zinn W Eroglu C Proactive planning for catastrophic events in supply chains Journal of Operations Management 2009 27 2 141 153 10.1016/j.jom.2008.06.002 Kopelman RE Brief AP Guzzo RA Benjamin S The role of climate and culture in productivity Organisational climate and culture 1990 Jossey-Bass Kumar R Abdin MS Impact of epidemics and pandemics on consumption pattern: Evidence from Covid-19 pandemic in rural-urban India Asian Journal of Economics and Banking 2021 5 1 2 14 10.1108/AJEB-12-2020-0109 Kumar SK Tiwari MK Babiceanu RF Minimisation of supply chain cost with embedded risk using computational intelligence approaches International Journal of Production Research 2010 48 13 3717 3739 10.1080/00207540902893425 Lau CM Sholihin M Financial and nonfinancial performance measures: How do they affect job satisfaction? The British Accounting Review 2005 37 4 389 413 10.1016/j.bar.2005.06.002 Lemieux T Milligan K Schirle T Skuterud M Initial impacts of the Covid-19 pandemic on the Canadian labour market UTP Journal Press 2020 46 1 55 65 Li S Ragu-Nathan B Ragu-Nathan TS Rao SS The impact of supply chain management practices on competitive advantage and organizational performance The International Journal of Management Science 2006 34 2 107 124 Lin Y A study on the relationship between project management competency, job performance and job motivation in e-commerce industry Measuring Business Excellence 2021 25 1 24 57 10.1108/MBE-10-2020-0144 Lufkin, B., & BBC. (2020). Amid the coronavirus outbreak, people are flocking to supermarkets worldwide—but are they simply preparing, or irrationally panicking?. https://www.bbc.com/worklife/article/20200304-coronavirus-covid-19-update-why-people-are-stockpiling. Macher JT Mowery DC Simcoe TS E-business and disintegration of the semiconductor industry value chain Industry and Innovation 2002 9 3 155 181 10.1080/1366271022000034444 Mahajan P Singh F How do pre-slowdown financial characteristics impact the firms’ relative financial performance during economic recession? An empirical investigation Asia-Pacific Journal of Management Research and Innovation 2013 9 4 369 378 10.1177/2319510X14523105 Manuj I Mentzer JT Global supply chain risk management strategies International Journal of Physical Distribution & Logistics Management 2008 38 3 192 223 10.1108/09600030810866986 March JG Shapira Z Managerial perspectives on risk and risk taking Management Science 1987 33 11 1404 1418 10.1287/mnsc.33.11.1404 Masterson, V., & World Economic Forum. (2020). How a circular economy could aid the COVID-19” recovery. https://www.weforum.org/agenda/2020/11/circular-economy-covid-19-recovery-warwick/. Masudin I Ramadhani A Restuputri DP Amallynda I The Effect of Traceability System and Managerial Initiative on Indonesian Food Cold Chain Performance: A Covid-19 Pandemic Perspective Global Journal of Flexible Systems Management 2021 22 4 331 356 10.1007/s40171-021-00281-x McKibbin W Fernando R The Global macroeconomic impacts of COVID-19: Seven scenarios Asian Economic Papers 2021 20 6 1 30 10.1162/asep_a_00796 McKinnon A Life without trucks: The impact of a temporary disruption road freight transport on a national economy Journal of Business Logistics 2006 27 2 227 250 10.1002/j.2158-1592.2006.tb00224.x Meiryani N Koh Y Soepriyanto G Aljuaid M Hasan F The effect of transformational leadership and remote working on employee performance during COVID-19 pandemic Frontiers in Psychology 2022 13 919631 10.3389/fpsyg.2022.919631 36033029 Miller DC Handbook of research design and social measurement 1983 4 Long- Man Milliken FJ Three types of perceived uncertainty about the environment: State, effect and response uncertainty Academy of Management Review 1987 12 1 133 143 10.2307/257999 Mollenkopf DA Ozanne LK Stolze HJ A transformative supply chain response to COVID-19 Journal of Service Management 2021 32 2 190 202 10.1108/JOSM-05-2020-0143 Moosavi J Fathollahi-Fard AM Dulebenets MA Supply chain disruption during the COVID-19 pandemic: Recognizing potential disruption management strategies International Journal of Disaster Risk Reduction. 2022 75 102983 10.1016/j.ijdrr.2022.102983 35475018 Mönch M Uzsoy R Fowler JW A survey of semiconductor supply chain models part I: Semiconductor supply chains, strategic network design, and supply chain simulation International Journal of Production Research 2018 56 13 4524 4545 10.1080/00207543.2017.1401233 Nagurney A Cruz J Dong J Zhang D Supply chain networks, electronic commerce and supply side and demand side risk European Journal of Operational Research 2005 164 1 120 142 10.1016/j.ejor.2003.11.007 Namagembe S Ryan S Sridharan R Green supply chain practice adoption and firm performance: Manufacturing SMEs in Uganda Management of Environmental Quality 2019 30 1 5 35 10.1108/MEQ-10-2017-0119 Nandi S Sarkis J Hervani A Helms M Do blockchain and circular economy practices improve post COVID-19 supply chains? A resource-based and resource dependence perspective Industrial Management & Data Systems 2021 121 2 333 363 10.1108/IMDS-09-2020-0560 Natarajarathinam M Capar I Narayanan A Managing supply chains in times of crisis: A review of literature and insights International Journal of Physical Distribution & Logistics Management 2009 39 7 535 573 10.1108/09600030910996251 Obayelu AE Edewor SE Ogbe AO Trade effects, policy responses and opportunities of COVID-19 outbreak in Africa Journal of Chinese Economic and Foreign Trade Studies 2021 14 1 44 59 10.1108/JCEFTS-08-2020-0050 Paik S Bagchi PK Understanding the causes of the bullwhip effect in a supply chain International Journal of Retail & Distribution Management 2007 35 4 308 324 10.1108/09590550710736229 Parast MM Subramanian N An examination of the effect of supply chain disruption risk drivers on organizational performance: Evidence from Chinese supply chains Supply Chain Management 2021 26 4 548 562 10.1108/SCM-07-2020-0313 Paul SK Chowdhury P Strategies for Managing the Impacts of Disruptions During COVID-19: An Example of Toilet Paper Global Journal of Flexible Systems Management 2020 21 3 283 293 10.1007/s40171-020-00248-4 Paul SK Chowdhury P A production recovery plan in manufacturing supply chains for a high-demand item during COVID-19 International Journal of Physical Distribution & Logistics Management 2021 51 2 104 125 10.1108/IJPDLM-04-2020-0127 Pérez Vergara IG López Gómez MC Lopes Martínez I Vargas Hernández J Strategies for the Preservation of Service Levels in the Inventory Management During COVID-19: A Case Study in a Company of Biosafety Products Global Journal of Flexible Systems Management 2021 22 S1 65 80 10.1007/s40171-021-00271-z Porter MA Competitive advantage: Creating and sustaining superior performance 1985 Free Press Remko VH Research opportunities for a more resilient post-COVID-19 supply chain—closing the gap between research findings and industry practice International Journal of Operations & Production Management 2020 40 4 341 355 10.1108/IJOPM-03-2020-0165 Rio-Chanona RMD Mealy P Pichler A Lafond F Supply and demand shocks in the Covid-19 pandemic: An industry and occupation perspective Oxford Review of Economics Policy 2020 36 1 94 137 10.1093/oxrep/graa033 Runfola A Milanesi M Guercini S Rethinking interaction in social distancing times: Implications for business-to-business companies Journal of Business & Industrial Marketing 2021 36 13 105 115 10.1108/JBIM-05-2020-0242 Rushton A Oxley J Handbook of logistics and distribution management 1989 Kogan Page Ltd Sajjad A The COVID-19 pandemic, social sustainability and global supply chain resilience: A review Corporate Governance 2021 21 6 1142 1154 10.1108/CG-12-2020-0554 Salancik GR Pfeffer J The effects of ownership and performance on executive tenure in U.S. corporations Academy of Management Journal 1980 23 4 653 664 10.2307/255554 Sarker M Moktadir M Santibanez-Gonzalez ED Social Sustainability Challenges Towards Flexible Supply Chain Management: Post-COVID-19 Perspective Global Journal of Flexible Systems Management 2021 22 S2 199 218 10.1007/s40171-021-00289-3 Schmenner RW The pursuit of productivity Production and Operations Management 2014 24 2 341 350 10.1111/poms.12230 Schmenner RW Swink ML On theory in operations management Journal of Operations Management 1998 17 1 97 113 10.1016/S0272-6963(98)00028-X Semiconductor Industry Association (SIA). (2021). Global semiconductor sales increase 1.9% month-to-month in April; annual sales projected to increase 19.7% in 2021, 8.8% in 2022”. https://www.semiconductors.org/global-semiconductor-sales-increase-1-9-month-to-month-in-april-annual-sales-projected-to-increase-19-7-in-2021-8-8-in-2021/. Semiconductor Industry Association (SIA). (2020). From microchips to medical devices. Semiconductors as an essential industry during the Covid-19 pandemic. https://www.semiconductors.org/wp-content/uploads/2020/10/From-Microchips-to-Medical-Devices-SIA-White-Paper.pdf. Sharma A Adhikary A Borah SB Covid-19′s impact on supply chain decisions: Strategic insights from NASDAQ 100 firms using Twitter data Journal of Business Research 2020 117 443 449 10.1016/j.jbusres.2020.05.035 32834209 Shaw, N., & Wales Online. (2020). Supermarket sites collapse as people rush to book deliveries. https://www.walesonline.co.uk/news/uk-news/supermarket-sites-collapse-people-rush-17968808. Shed, S., & CNBC. (2021). The global chip shortage is starting to have major real-world consequences. https://www.cnbc.com/2021/05/07/chip-shortage-is-starting-to-have-major-real-world-consequences.html. Sheffi Y Rice J Jr A supply chain view of the resilient enterprise MIT Sloan Management Review 2005 47 1 41 48 Sheth J Impact of Covid-19 on consumer bahaviour: Will the old habits return or die? Journal of Business Research 2020 117 280 283 10.1016/j.jbusres.2020.05.059 32536735 Shin H Collier DA Wilson DD Supply management orientation and supplier/buyer performance Journal of Operations Management 2000 18 317 333 10.1016/S0272-6963(99)00031-5 Simon A Bartle C Stockport G Smith B Klobas JE Sohal A Business leaders’ views on the importance of strategic and dynamic capabilities for successful financial and non-financial business performance International Journal of Productivity and Performance Management 2015 64 7 908 931 10.1108/IJPPM-05-2014-0078 Skipper JB Hanna JB Minimizing supply chain disruption risk through enhanced flexibility International Journal of Physical Distribution & Logistics Management 2009 39 5 404 427 10.1108/09600030910973742 Slack N Chambers S Harland C Harrison A Johnston R Operations management 1995 Pitman Publishing Sparkes M., & New Scientist. (2021). There's a global shortage of computer chips—what's causing it?. https://www.newscientist.com/article/2271918-theres-a-global-shortage-of-computer-chips-whats-causing-it/. Sucky E The bullwhip effect in supply chains—an overestimated problem? International Journal of Production of Economics 2009 118 1 311 322 10.1016/j.ijpe.2008.08.035 Swinnen J Vos R COVID-19 and impacts on global food systems and household welfare: Introduction to a special issue Agricultural Economics 2021 52 3 365 374 10.1111/agec.12623 34149127 Tan KC Kannan VR Handfield RB Supply chain management: Supplier performance and firm performance International Journal of Purchasing and Materials Management 1998 34 3 2 9 Tang CS Perspectives in supply chain risk management International Journal of Production of Economics 2006 103 2 451 488 10.1016/j.ijpe.2005.12.006 Tangen S Performance measurement: From philosophy to practice International Journal of Productivity and Performance Management 2004 53 8 726 737 10.1108/17410400410569134 Taylor DH Fearne A Towards a framework for improvement in the management of demand in agri-food supply chains Supply Chain Management 2006 11 5 379 384 10.1108/13598540610682381 Tse YK Tan KH Managing product quality risk and visibility in multi-layer supply chain International Journal of Production Economics 2012 139 1 49 57 10.1016/j.ijpe.2011.10.031 Tummala R Schoenherr T Assessing and managing risks using the supply chain risk management process (SCRMP) Supply Chain Management 2011 16 6 474 483 10.1108/13598541111171165 Udofia EE Adejare BO Olaore GO Udofia EE Supply disruption in the wake of COVID-19 crisis and organisational performance: Mediated by organisational productivity and customer satisfaction Journal of Humanities and Applied Social Sciences 2021 3 5 319 338 10.1108/JHASS-08-2020-0138 United Nations. (2020). The impact of the Covid-19 pandemic on trade and development, transitioning to a new normal. https://unctad.org/system/files/official-document/osg2020d1_en.pdf. Vakil, B., Linton, T., & Havard Business Review. (2021). Why We’re in the midst of a global semiconductor shortage. https://hbr.org/2021/02/why-were-in-the-midst-of-a-global-semiconductor-shortage. Vinten G Open versus closed questions—an open issue? Management Decisions 1995 33 4 27 31 10.1108/00251749510084653 Wagner SM Bode C An empirical examination of supply chain performance along several dimensions of risk Journal of Business Logistics 2008 29 1 307 325 10.1002/j.2158-1592.2008.tb00081.x Wallace, A., & CNN Business. (2020). Walmart CEO says we’re in the ‘hair color’ phase of panic buying. https://edition.cnn.com/2020/04/11/business/panic-buying-walmart-hair-color-coronavirus/index.html. Wallin A Pihlajamaa M Malmelin N How do large corporations manage disruption? The perspective of manufacturing executives in Finland European Journal of Innovation Management 2021 25 6 19 43 10.1108/EJIM-09-2020-0383 Wang M Impacts of supply chain uncertainty and risk on the logistics performance Asia Pacific Journal of Marketing and Logistics 2018 30 3 689 704 10.1108/APJML-04-2017-0065 Wang M Jie F Abareshi A Evaluating logistics capability for mitigation of supply chain uncertainty and risk in the Australian courier firms Asia Pacific Journal of Marketing and Logistics 2015 27 3 486 498 10.1108/APJML-11-2014-0157 World Trade Organisation. (2020). Trade shows sign of rebound from Covid-19, recovery still uncertain. https://www.wto.org/english/news_e/pres20_e/pr862_e.htm. Wu F Yeniyurt S Kim D Cavusgil ST The impact of information technology on supply chain capabilities and firm performance: A resource-based view Industrial Marketing Management 2006 35 4 493 504 10.1016/j.indmarman.2005.05.003 Yamin S Gunasekruan A Mavondo FT Relationship between generic strategy, competitive advantage and firm performance: An empirical analysis Technovation 1999 19 8 507 518 10.1016/S0166-4972(99)00024-3 Yan Y Zha X Yan M Exploring employee perceptions of Web 2.0 virtual communities from the perspective of knowledge sharing Aslib Journal of Information Management 2014 66 4 381 400 10.1108/AJIM-08-2013-0070 Yusuf YY Gunasekaran A Musa A Dauda M El-Berishy NM Cang S A relational study of supply chain agility, competitiveness and business performance in the oil and gas industry International Journal of Production Economics 2014 147 531 543 10.1016/j.ijpe.2012.10.009 Zimon G Babenko V Sadowska B Chudy-Laskowska K Gosik B Inventory management in SMEs operating in polish group purchasing organizations during the COVID-19 pandemic Risks (basel) 2021 9 4 63 10.3390/risks9040063 Zsidisin GF Managerial perceptions of supply risk Journal of Supply Chain Management 2003 39 4 14 26 10.1111/j.1745-493X.2003.tb00146.x Zsidisin GA Panelli A Upton R Purchasing organization involvement in risk assessments, contingency plans, and risk management: An exploratory study Supply Chain Management 2000 5 4 187 198 10.1108/13598540010347307 Zsidisin GA Ellram LM An agency theory investigation of supply risk management The Journal of Supply Chain Management 2003 39 2 15 27 10.1111/j.1745-493X.2003.tb00156.x	0	PMC9734435	NO-CC CODE	2022-12-14 23:28:28	no	Glob J Flex Syst Manag. 2022 Dec 5;:1-19	utf-8	null	null	null	oa_other
==== Front Inf Technol Tourism Information Technology & Tourism 1098-3058 1943-4294 Springer Berlin Heidelberg Berlin/Heidelberg 241 10.1007/s40558-022-00241-w Review Enhancing sustainable development through tourism digitalisation: a systematic literature review http://orcid.org/0000-0002-0266-6354 Rodrigues Vitor [email protected] http://orcid.org/0000-0002-2220-5483 Eusébio Celeste http://orcid.org/0000-0002-5882-063X Breda Zélia grid.7311.4 0000000123236065 GOVCOPP - Research Unit in Governance, Competitiveness and Public Policy, Department of Economics, Management, Industrial Engineering and Tourism (DEGEIT), University of Aveiro, Aveiro, Portugal 5 12 2022 133 11 7 2022 22 10 2022 21 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. The world’s economic structure is increasingly moving towards a digital framework, boosted by the fourth industrial revolution. As a versatile sector, tourism is also embedded within this digital transformation process, albeit at a slower pace due to the uncountable challenges and uncertainties surrounding it. Nevertheless, the most recent implications of the pandemic crisis warned both managers and politicians of the urgent need for new development paths aligned with sustainability, particularly with the United Nations’ sustainable development goals. Furthermore, direct issues related to tourism activity, such as overtourism, pollution, and economic dependency, call for alternative and balanced approaches. Smart and digital solutions might play a key role in this process, but little is known concerning their potential. Aiming to extend knowledge concerning these potentials, a systematic literature review was conducted to examine the state-of-the-art about the implications of digital transformation in tourism as a catalyst for sustainable development, identifying gaps and providing directions for future research. From the analysis of 38 manuscripts, visitor experience, destination management, business solutions, and smart sustainable destinations emerged as the most common topics. However, inconsistencies were identified concerning the management narratives and the actual implementation of smart approaches. Additionally, the novelty of the concepts gravitating around smart tourism promotes some theoretical inconsistencies, which also need to be remedied. Keywords Sustainable development Sustainability Tourism 4.0 Smart tourism Tourism digitalisation Systematic literature review http://dx.doi.org/10.13039/501100001871 Fundação para a Ciência e a Tecnologia UI/BD/152274/2021 Rodrigues Vitor ==== Body pmcIntroduction Technology has accompanied tourism since its early stages, and new challenges are arising. The sector’s digital transformation is now an ongoing process, enhanced by a set of disruptive innovations that changed the industrial panorama and are starting to be transferred into the tourism sector. This trend was triggered by the emergence of Industry 4.0 (I4.0), where operations occur in a ‘phygital’ world, meaning that both digital and physical spheres converge into a unique system (Posada et al. 2015), enabling smartness to embrace the production process (Xu et al. 2018). The effectiveness of this process is guaranteed by the interconnectivity of a set of advanced information and communication technologies (ICTs) (e.g. artificial intelligence (AI), the internet of things (IoT), blockchain, cloud computing, big data), creating an intelligent value chain where data is constantly processed and exchanged through autonomous and independent methods (Posada et al. 2015; Xu et al. 2018). This ‘revolutionary’ trend was rapidly adopted by the tourism sector, creating space for the emergence of a new concept that some hastily called tourism 4.0 (T4.0). T4.0 can be simply understood as an extension of I4.0 principles, meaning that technologies, such as blockchain, IoT, AI, augmented reality (AR), virtual reality (VR) and others, are key elements in the operationalisation of tourism activity (e.g., Buhalis et al. 2019; Jeong and Shin 2020; Stankov and Gretzel 2020). Even so, very few studies attempted to conceptualise T4.0 (e.g., Pencarelli 2020; Stankov and Gretzel 2020), which promoted a growing, but decontextualised, use of the concept by governments and academics, making it difficult to clarify its true meaning. Alternatively, development processes that are driven by ICTs are commonly described as smart (Gajdosik and Orelová 2020; Gretzel et al. 2015a). For instance, ICTs represent a vital element of the smart tourism concept, aiming to create innovative processes by maximising and optimising the contribution of all the stakeholders (Buhalis 2020). However, the emphasis is on the technological potentialities and not on the technologies as a tangible dimension (Li et al. 2017). Within a smart context, the aim is also to improve sustainability through the efficient use of technologies and existing resources, which consequently enhance competitiveness or, in the words of Crouch and Ritchie (1999), “sustainable competitiveness”. Therefore, smart tourism involves innovative forms of collaboration and value creation based on the treatment of data from all the actors involved, facilitated by the implementation of new technologies. It aims to create value propositions based on efficiency and sustainability (Gretzel et al. 2015b), suggesting a clear alignment between technological solutions and the United Nations’ sustainable development goals (SDGs) (Sachs et al. 2019). Thus, the question that should be addressed within tourism management and planning fora is whether it is possible to maintain profitable tourism growth without disregarding the sustainable paradigm. Or, from a more ambitious perspective, how can the tourism industry be restructured to guarantee appropriate sustainability levels by adopting technologies and smart-oriented strategies? Due to their disruptive characteristics, tourism managers are increasingly aware of the benefits and added value of T4.0 and smart technologies (Buhalis et al. 2019). Technologies in services (and tourism inherently) are highly linked to the interactions between the service provider and customers and can be helpful in the co-creation of value (Buhalis et al. 2019; Pencarelli 2020). ICTs have also been supporting the tourism industry in developing effective marketing strategies to attract visitors and provide unique experiences at the destination (Buhalis et al. 2019; Jeong and Shin 2020). However, some of these solutions were primarily aimed at tourism companies and were not adapted for visitors, resulting in negative experiences (Stankov and Gretzel 2020). At the same time, smart technologies might arouse visitors’ behavioural intentions (Koo et al. 2016), influencing their motivations, recommendation intentions, and interest in a specific attraction or destination (Koo et al. 2016). Thus, it seems mandatory to thoroughly understand the extent to which visitors are willing to deal with new ICTs in a tourism context. At the same time, it is also relevant to evaluate tourism companies’ ability to effectively implement these ICTs in their favour and add value to the customer. Within a context where more and more voices emerge calling for a rearrangement of tourism development models, the role of ICTs and new emerging concepts, such as T4.0, needs to be questioned. An opportunity seems to exist for genuine sustainable tourism development, guiding the sector toward a more ecological path centred on the well-being of local communities (Niewiadomski 2020). The latest developments and proliferation of technologies are viewed as an opportunity to manage sustainability challenges within a tourism destination (Choudhary et al. 2020; Romão and Neuts 2017). For instance, the impacts caused by the latest pandemic crisis call for innovative ways of providing services and recovery strategies (Fennell 2021). This might be achieved through innovation processes and the gradual implementation of new technologies (Gössling 2020). In this sense, some discourses are already advocating disruptive approaches towards a sustainable reboot and rethink of tourism structure anchored on technological innovations (Mohanty et al. 2020), designing the path towards a new era of economic growth. Sustainable development implies that the exploitation of resources, allocation of investments, technological development, and institutional change are aligned and in harmony with the needs of the present and future generations (Brundtland 1987; Rogers et al. 2012). Thus, it must be adapted to the current panorama to guarantee that all the actors involved will be consistent and act in line with sustainability (Brundtland 1987; Rogers et al. 2012). However, an incorrect interpretation of the concepts, in both theoretical and managerial fields, might lead to a decontextualised integration in the perspective of strategic planning (Williams et al. 2020), distorting the discourses of policymakers and ‘emptying’ the concepts of their true content and meaning. Today’s society is driven by a digital setting in constant transformation. This scenario creates daily challenges for the sustainable development of both individuals and territories, making it vital to understand the influence of technologies (Popkova et al. 2022). In line with this, two main pillars sustain the rationale of the present study. First, despite it being argued that smart tourism is sustainability-oriented and that the implementation of technologies will enhance smart sustainable models, there is insufficient evidence supporting this (Gomis-López and González-Reverté 2020; González-Reverté 2019). Secondly, the relevance of new ICT solutions is increasing amongst tourism destinations due to the emergence of digital visitors and the need to promote fully sustainable strategical approaches (Gomez-Oliva et al. 2019). Thus, the following research question is addressed in this paper: To what extent can smart and technological approaches foster sustainable tourism development strategies? Therefore, based on a systematic literature review, this study aims to identify, summarise, and critically review what has been published concerning T4.0, smart tourism, and sustainable development. It should be added that, to date, to the best of our knowledge, no works have adopted a similar methodological approach, except for a systematic review related to the development of a model for sustainable smart tourism destinations (e.g., Shafiee et al. 2019). This literature review is expected to expand the discussion and understanding concerning the transition towards digital tourism and how the process relates to sustainable development. Additionally, this study aims to identify the main gaps in the literature and contribute to the emergence of new research concerning this theme. The paper is structured in five sections. The following section outlines the methods used to select and analyse the retrieved papers. Section 3 addresses the quantitative findings, particularly by describing and examining the number of studies published within a period range, main sources, academic fields, number of citations, keywords network analysis, and the geographical basis. In the fourth section, first the main research methodologies are identified, which is then followed by a content analysis organised by categories highlighting and discussing the key findings and topics. The paper closes with a summary of the main conclusions, underlining the implications of technology towards a more sustainable path within the tourism industry, beyond reporting the most prominent research gaps and suggestions for future research paths. Methodology Records selection A literature review is a versatile procedure aiming to indicate innovative directions to investigate a specific field (Davis et al. 2014; Snyder 2019). Different approaches can be implemented to conduct an effective literature review, namely systematic, semi-systematic, and integrative literature reviews (Pickering and Byrne 2013; Snyder 2019). This work draws on a systematic literature review to investigate the relationship between smart tourism and sustainable development, aiming to identify the methodologies applied, research subjects, and the main research gaps within the topic that might contribute to future research. By doing so, this method portrays emergent research themes and the fields demanding further theoretical support (Tölkes 2018). To accomplish this aim, the present study followed the PRISMA protocol encompassing four different stages adapted from Liberati et al. (2009). This protocol was chosen due to its reliability and use in various fields of study and its potential to improve consistency across reviews (Liberati et al. 2009). The research protocol was previously defined to select the most appropriate documents, identifying the excluding criteria. The procedure used to decide on the works to be analysed is detailed in Fig. 1. Fig. 1 Systematic review process based on the PRISMA diagram The Scopus database was used to identify the studies within the field in analysis. Scopus was chosen because it is the most comprehensive database covering scientific data and literature subject to peer review, and due to the availability of metrics and analytical tools supporting the analysis process (Elsevier 2021). To define the most appropriate search terms, the first step was the analysis of the existing and relevant literature on the topics, as suggested by Hausberg et al. (2019) and Siddaway et al. (2019). Thus, the analysis of seminal literature (e.g., Boes et al. 2016; Gretzel et al. 2015a,b; Pencarelli 2020; Rogers et al. 2012) and the review article by Shafiee et al. (2019) supported the design of the theoretical background, enabling the identification of several buzzwords (Hausberg et al. 2019). After a brainstorming and discussion process, different combinations of search terms were tested (Hausberg et al. 2019; Mehraliyev et al. 2019; Siddaway et al. 2019), allowing the researchers to determine the final search string: “tourism 4.0” or “smart tourism” or “digital tourism” AND “sustainable development goals” or “sustainab” or “sustainable development”. As the purpose was to be as holistic as possible concerning the sustainability side, it was decided not to focus on specific dimensions, e.g., environmental, social, or economic. Moreover, the asterisk wildcard character () was employed to search for term variations. The identification phase was accomplished in January 2022, according to the following search string filtered by title, abstract and keywords: “tourism 4.0” or “smart tourism” or “digital tourism” AND “sustainable development goals” or “sustainab*” or “sustainable development”. A total of 166 records were obtained. Then, the screening phase covered three stages. The records were firstly filtered by document type, excluding conference reviews, editorials, erratum, and reviews. Due to the novelty of the topics in discussion, this first screening opted to include both conference papers and book chapters to enlarge the sample and the probability of finding relevant insights that go along with the purpose of the present work. Thus, the number of documents was reduced to 149 records. Secondly, duplicated records were excluded. This second step excluded two publications. The third step encompassed an individual analysis (by title, abstract, and keywords) of each article to ensure their reliability for the study. If this process was inconclusive, the article was thoroughly analysed. Both theoretical and empirical studies were considered valid for the analysis if focusing on issues addressing the role of T4.0 and/or smart approaches towards sustainable tourism. Conversely, papers disregarding one of these dimensions or approaching the topic in a meaningless way, thus providing limited insights, were excluded from the literature review process. Concluded this screening phase, a total of 38 records was considered eligible and included for analysis. Records analysis This review was divided into two main categories of analysis: bibliometric and content analysis. Bibliometric analysis of the reviewed records was carried out through the review of (i) year-wise distribution of studies, (ii) journals by disciplinary field and the number of citations, and (iii) keyword co-occurrence. This last step was accomplished using the VOSviewer software to generate a keyword co-occurrence network. Secondly, the content analysis was mainly focused on the identification of the (i) research methods, and (ii) thematic fields (e.g. demand, supply, public policies) and discussion of the results concerning the implications of smart tourism and sustainability in each of the fields identified. Bibliometric analysis Year-wise distribution of studies A total of 38 records were retrieved for analysis. The number of studies analysing sustainability and smart tourism has been increasing since the first publication in 2014. The first article was published by Graziano (2014) and addressed the potential of smart tourism in Italian smart cities. The author performed a SWOT analysis examining the level of smartness and the future challenges, further proposing a territorial planning instrument looking towards both social and economic sustainable development. The author concluded that smart tourism would only be plausible if the decision-makers perceive the territorial context and dynamics, particularly to guarantee the sustainability of development strategies and plans. Most publications were registered between 2019 and 2021, which makes up 76.3% of the total, meaning that these subjects are relevant and gaining increasing prominence in the tourism field. The peak registered between 2019 and 2020 might be due to the increasing use of technologies in the tourism industry and the emergence of pro-sustainability segments within tourism demand. Moreover, particularly in 2020, the number of publications might be justified by the pandemic crisis that boosted the digitalisation process within the tourism sector, along with the need to design more sustainable development models (Fig. 2). As mentioned earlier in the introduction, the proliferation of smart and I4.0 technologies within the tourism sector is expected to generate greater awareness towards the necessity of investigating its implications. Moreover, several questions related to tourist behaviour towards smartness and companies’ adaptation to technological progress remain to be solved (Buhalis et al. 2019). Additionally, the potential of technologies in the sustainability field is still on a theoretical basis (Gössling 2020), and thus an increase in publication numbers is expected in upcoming years. Fig. 2 Number of records published by year Journals, disciplinary field, and citations Among the 38 publications under analysis, 94.7% were found to be articles, while only two were conference papers. These studies were published in 19 distinct journals or proceedings, representing a wide diversity of disciplinary fields (Table 1). In particular, the journal Sustainability is the most prolific, with 19 publications, while the remainder registered only one record. There were also two papers published in the e-Review of Tourism Research. Ten disciplinary fields were identified. The wide range of disciplinary backgrounds demonstrates the multi-disciplinarity of the research subjects. It should be mentioned that this diversity might also be justified by the fact that journals are associated with more than one field of study, as demonstrated in Table 1. There is a considerable prevalence of studies published in “geography, planning, and development” (63.2%), due to the high number of publications in the journal Sustainability. There is also a significant contribution of publications in the field of “tourism, leisure, and hospitality management” (18.4%), followed by the areas of “computer science” (13.2%), and “business, management and accounting” (10.5%). Perhaps one major gap is the fact that tourism and hospitality journals have no more than one or two publications. These results suggest that the aggregation of smart and digital tourism with sustainability topics is not embraced exclusively by tourism and hospitality journals. Instead, they are spread through distinct areas, despite the prevalence of managerial, planning, and development disciplinary fields. Table 1 Journals by disciplinary field Journal or proceeding Nr. of publications Tourism, leisure, and hospitality management Computer science Information systems Sociology and political science Social sciences Business, management and accouting Geography, planning and development Development Urban studies Ecology Austrian Journal of South-East Asian Studies 1 x Companion Proceedings of the 2019 World Wide Web Conference 1 x e-Review of Tourism Research 2 x x EuroMed Journal of Business 1 x European Journal of Geography 1 x European Journal of Tourism Research 1 x x GeoJournal of Tourism and Geosites 1 x IEEE Access 1 x Information Technology and Tourism 1 x x x International Journal of Recent Technology and Engineering 1 x International Journal of Information Management 1 x x x x Journal of Destination Marketing and Management 1 x x Journal of Regional Research 1 x x x Journal of Urban Technology 1 x Land 1 x Sustainability 19 x Proceedings of the Ninth International Symposium on Information and Communication Technology 1 x Tourism Management Perspectives 1 x Tourism Review International 1 x Total number of papers per disciplinary field 7 (18.4%) 5 (13.2%) 2 (5.3%) 1 (2.6%) 2 (5.3%) 4 (10.5%) 24 (63.2%) 1 (2.6%) 2 (5.3%) 1 (2.6%) A citation analysis was also employed based on Scopus metrics. Analysing the most cited articles might be useful for researchers to identify seminal literature, which is argued to support the development of an effective theoretical framework (Cavalcante et al. 2021). According to this rationale, the study by Sun et al. (2016) appears as the most influential, with a total of 584 citations, followed by Shafiee et al. (2019) and Pencarelli (2020), both with 80 citations. Only one article was found to have no citations (Liu et al. 2019), and 44% have less than ten citations. One possible reason justifying the low number of citations might be attributed to the fact that most of the publications were published predominantly since 2020. A second reason might be associated with the topic under analysis by the authors, e.g. circular economy, accessible tourism, and participatory planning, which might still be poorly related to the smart and/or sustainability subjects. The citation analysis network, indicating the number of times each author cites another, is presented in Fig. 3. The network shows a minimum number of one document and one citation per author. Limited results were generated. This is justified by the fact that the most prominent links are formed only by two clusters composed of 26 authors citing each other, from a total of 118. This suggests that these nodes have similar theoretical approaches to the topic (Hausberg et al. 2019). Among these, the studies by Pasquale Del Vecchio and Antonio J. Jara appear as the most mentioned within this network. Fig. 3 Keywords network by co-occurrence Keyword analysis An analysis of the keywords network by co-occurrence was carried out (Fig. 4). The VOSviewer software was used to create the network map. The node size represents the keywords’ weight, measured by the number of occurrences, while the lines indicate the link between the terms. From the 38 documents, a total of 131 keywords were identified. From these, 106 appeared only once, equivalent to 81% prevalence, while only 25 had two or more co-occurrences. For that reason, and for a more accurate visualisation of the network, all the keywords with a minimum number of one occurrence were included. This lower prevalence of keywords with two or more co-occurrences might be partially explained by the limited number of documents in analysis, but mainly by the fact that some terms (e.g., “smart”) are used in bulk and applied to a wide range of constructs, supporting the observation of Gretzel et al. (2015a), who criticise the arbitrary implementation of this concept. In this case, the term smart is presented in 16 keywords (e.g., smart communities, smart governance, smart frameworks, smart tools, smart tourism city, smart tourists), showing the lack of conceptual background around the topic. “Smart tourism” has the majority of occurrences and the highest link strength, along with “sustainability”, with nine occurrences and a total link strength of 53, and “smart city”, appearing in nine articles and with a link strength of 42. This was somewhat of an expected outcome as the first two keywords composed the search code. Then, some other keywords are worth mentioning, specifically “smart tourism destination” and “sustainable tourism”, composing the remaining top five positions. Fig. 4 Citation network analysis From this analysis, 16 clusters were identified in total. The first cluster combines studies around the topic of sustainable development. The second cluster is more focused on technology, involving issues such as innovation, value co-creation, and smart tourism plans. A third cluster is formed around sustainability, involving studies focused on tourist destinations’ solutions, such as circular economy practices or open innovation issues. The fourth cluster relates to studies mainly centred on the IoT and connecting terms such as information and communication technologies, near-field communication, or tourist attractions. The central topic in the fifth cluster is cultural tourism, related to heritage sites, world heritage, senior tourists, or Web 2.0. Cluster 6 groups words around the smart tourism destination topic, emphasising sustainable tourism. The seventh cluster is formed around the word smart tourism, including, among others, major terms such as tourism innovation and smart destination. Cluster 8 has the smart city topic as central, gathering other issues such as big data and social networks. Mobile application is the central topic compounding cluster 9. Following this, clusters 10, 11, and 12 evolve around themes such as tourist experience, case study, and governance, respectively. Finally, clusters 13 to 16 are based only on one article, each analysing a wide range of topics such as smart destination governance, neural networks, accessible tourism, and smart tools. One interesting fact is the modest relevance of the term “technology”, despite being a key element within both smart and digital paradigms. Nevertheless, and as in Schimperna et al. (2021), this is not representative of the slight importance of technological tools in the studies, even because several technologies are visible in the network, e.g., the IoT, mobile applications, big data, information and communication technologies, blockchain. On the other hand, this might indicate future research paths, perhaps highlighting the role of specific technologies or establishing a stronger link between the concepts. Finally, the relevant data retrieved from this network analysis, in addition highlighting the key research constructs and streams, supported the identification and structuring of the content analysis section (Niñerola et al. 2019; Schimperna et al. 2021). These results highlight the potential role of the smart paradigm within cities and tourism destinations, also addressing other key subjects, such as sustainable development and technologies, particularly in tourist-oriented and tourism destination management studies. Geographical context Contrary to other systematic literature review studies (e.g., Schimperna et al. 2021), the present geographical context analysis was based on the territorial area where the study took place. This enables a more profound understanding of the territorial basis where the case studies are being implemented. In this case, it gains additional relevance as it might allow differences to be identified between developed and developing countries. Among the studies under analysis, Europe is the predominant geographical focus (42.1%), followed by Asia (21.1%), and the Americas (5.3%) (Fig. 5). Within the European setting, almost all studies were conducted in the South, mostly in Spain (50%) and Italy (25%). This suggests that southern countries might be a step ahead concerning the implementation of technological solutions towards sustainability, probably aware of the short- and long-run impacts resulting from climate change (Hein et al. 2009; Pintassilgo et al. 2016), to which they are most exposed, and the urgent need to employ alternative development models. In the Asian context, China and South Korea stand out, both with two studies, followed by Hong Kong, Indonesia, Japan, and Vietnam. Although not significant in terms of prevalence, it is worth mentioning the studies conducted in Indonesia and Vietnam, two developing countries, where solutions such as IoT and a smart service centre are implemented to manage visitors. It is particularly interesting to understand that these countries are also aware and able to design technological facilities, despite the disparities with developed countries. Three studies used a multiple-country setting, particularly centred in European and Asian countries, while one used a global approach. Nine studies were not based on a geographical background, consisting primarily of theoretical discussions. Fig. 5 Distribution of studies by geographical context Content analysis Content analysis of the articles encompassed the assessment of the research methods and the research context of the studies. Concerning the latter, the goal was to identify the main fields in which the topics of smart tourism and sustainability were embraced. In this case, only the papers adopting an empirical research method were included in the analysis. Research methods Table 2 synthesises the methodologies used in the studies considered in this analysis. The majority are empirical studies (65.4%), while the remaining represent theoretical ones. Concerning the empirical studies, ten adopted a quantitative methodology, eight applied qualitative methods, six implemented mixed methods, and only one followed a systematic review approach. Furthermore, within each research approach, distinct data collection methods were employed. This might be justified by the novelty of the topics that call for the need to implement exploratory research, usually associated with qualitative research methods. Moreover, since the digital transformation and the transference of technologies to the tourism context is still in an early stage, there are only shreds of evidence about its implications for sustainable tourism. Table 2 Research methods Research design Methods Authors Theoretical n.a. Errichiello and Micera (2021), Graziano (2014), Gretzel and Scarpino-Johns (2018), Lee et al. (2020), Panagiotopoulou et al. (2020), Pencarelli (2020), Perles-Ribes and Baidal (2018), Perles-Ribes and Ramón-Rodríguez (2019), Tyan et al. (2020), Stephenson and Dobson (2020), Sun et al. (2016) and Vu et al. (2018) Quantitative Questionnaires Chung et al. (2019), Ivars-Baidal et al. (2021), Shen et al. (2020) and Vizuete et al. (2021) Secondary data analysis (e.g., Business analytics of Big Data) Encalada et al. (2017), Kim et al. (2019), Moustaka et al. (2019), Saltos et al. (2021) and Zubiaga et al. (2019) Prediction method Crivellari and Beinat (2020) Qualitative Delphi Gomis-López and González-Reverté (2020) and Ortega and Malcolm (2020) Ethnography Huang and Lau (2020), Idris et al. (2021) and Lim et al. (2017) Focus group Idris et al. (2021) Interviews Huang and Lau (2020), Idris et al. (2021), Lim et al. (2017), Polese et al. (2018) and Zeng et al. (2020) Secondary data analysis (e.g., institutional documents) Gomis-López and González-Reverté (2020), González-Reverté (2019) and Križaj et al. (2021) Mixed Del Vecchio et al. (2018), Del Vecchio et al. (2021), Gomez-Oliva et al. (2019), Liu et al. (2019), Ramos-Soler et al. (2019) and Slavec et al. (2021) Review n.a. Shafiee et al. (2019) Questionnaires and secondary data sources (e.g., business analytics) were prevalent in the quantitative studies. These methods were particularly applied to research on spatial distribution and visitor flow monitoring (e.g., Encalada et al. 2017; Kim et al. 2019; Zubiaga et al. 2019), eco-technologies (e.g., Chung et al. 2019), smart tourism tools (e.g., Vizuete et al. 2021), smart tourism destinations (e.g. Ivars-Baidal et al. 2021; Ortega and Malcom 2020; Saltos et al. 2021), and smart tourists and sustainable behaviours (e.g., Shen et al. 2020). In turn, qualitative approaches used the Delphi method, focus groups, ethnography, interviews, and secondary sources (e.g., strategic plans, institutional reports) to retrieve data. These studies were more dispersed regarding their focus of analysis. Specifically, topics were found related to gamification and accessible tourism (e.g., Huang and Lau 2020), smart and sustainable tourism systems (e.g., Lim et al. 2017; Moustaka et al. 2019), value co-creation and smart service ecosystem (e.g., Polese et al. 2018), promotional strategies (e.g., Idris et al. 2021), and smart tourism sustainability (e.g., Gomis-López and González-Reverté 2020; González-Reverté 2019; Križaj et al. 2021; Zeng et al. 2020). Questionnaires and interviews (e.g., Gomez-Oliva et al. 2019; Liu et al. 2019), interviews and secondary data provided by business analytics (e.g., Del Vecchio et al. 2018; Del Vecchio et al. 2022), or questionnaires and focus groups (e.g., Ramos-Soler et al. 2019) were the most observed methods among mixed methods approaches. Mixed- methods were particularly adopted by studies analysing topics concerning big data (e.g., Del Vecchio et al. 2018; Del Vecchio et al. 2022), cultural heritage and visitors’ experience (e.g., Gomez-Oliva et al. 2019; Ramos-Soler et al. 2019; Slavec et al. 2021), and smart tourism destinations (e.g., Gomez-Oliva et al. 2019; Liu et al. 2019). Finally, the paper by Shafiee et al. (2019) is the only one implementing a systematic review. However, it diverges from the aims of the present paper, since its main purpose was to present a model for smart tourism destinations, positioning sustainability as a central issue, and identifying the main elements supporting the model. Specifically, the review article by Shafiee et al. (2019) looks forward to the clarification of both smart tourism and smart tourism destination concepts, addressing that the digital transformation efforts patronised by governmental authorities are mostly technology-based, a path that the authors argue as insufficient for an effective smart transformation. Aiming to contribute to theory development through the design of a sustainable smart tourism destination model, the literature review process was based on the grounded theory method. Using the smart tourism destination as the main research field, the authors added smart city and smart tourism due to their complementarity. Additionally, and since sustainability is a main pillar within the smart ecosystem (Gretzel et al. 2015b), the authors add the concepts of ‘smart sustainable city’ and ‘sustainable cities’ to their search string. Although recognising some similarities, the present study is not established under the smart tourism destination concept. The novelty of this study relies on the identification and discussion of smart approaches within distinct tourism areas, particularly by addressing their potential contribution to enhance sustainable development and sustainability in its fullness. Research topics After a methodical full-text analysis of each paper, four main topics emerged (Table 3): visitor experience, destination management, business solutions, and smart sustainable destinations. Visitor experience concerns the analysis of how smart solutions influence or impact visitors’ on-site experience and, consequently, the implications for a destination’s sustainability. In its turn, destination management encompasses studies examining the contribution of smartness toward effective planning and management of a tourism destination. The topic of business solutions focuses on studies analysing the implementation of smart technologies to improve a business’s operationalisation, while smart sustainable destinations emerge from research centred on the ability of tourism destinations to foster sustainability through the adoption of smart approaches. Table 3 Research topics (empirical papers only) Research topic Authors Visitor experience Chung et al. (2019), Gomez-Oliva et al. (2019), Huang and Lau (2020), Ramos-Soler et al. (2019) and Shen et al. (2020) Destination management Crivellari and Beinat (2020), Del Vecchio et al. (2018), Encalada et al. (2017), Idris et al. (2021), Kim et al. (2019), Saltos et al. (2021), Slavec et al. (2021), Vizuete et al. (2021) and Zubiaga et al. (2019) Business solutions Del Vecchio et al. (2021), Lim et al. (2017), Moustaka et al. (2019) and Polese et al. (2018) Smart sustainable destinations Gomis-López and González-Reverté (2020), González-Reverté (2019), Ivars-Baidal et al. (2021), Križaj et al. (2021), Liu et al. (2019), Ortega and Malcolm (2020) and Zeng et al. (2020) Visitor experience Visitors already use smart technologies at tourism destinations, on their own or encouraged by the destination environment, contributing to memorable travel experiences (Jeong and Shin 2020). The rise of smart technologies promotes great changes in traditional tourists, shaping their consumption and behavioural patterns (Sigala 2018). This necessarily entails several implications regarding sustainability issues. Different perspectives have been addressed to analyse the interaction between visitors and technology and how it influences sustainability practices or contributes to the sustainable development of tourism destinations. The studies analysed in this section (Chung, Tyan and Lee 2019; Gomez-Oliva et al. 2019; Huang and Lau 2020; Ramos-Soler et al. 2019; Shen, Sotiriadis and Zhou 2020) do not follow the same rationale, as different technologies have been analysed (e.g., near-field communication, beacons, mobile applications, social networks, IoT) in distinct contexts (e.g., museums, accessible tourism, responsible behaviour, cultural heritage). Moreover, some studies follow a win-win approach, showing how interaction with technology can contribute to visitors’ experience while guaranteeing the destination’s sustainability (e.g., Gomez-Oliva et al. 2019; Ramos-Soler et al. 2019; Shen et al. 2020). In their turn, other studies investigate how adopting and disseminating eco-friendly technological tools by supply stakeholders impacts visitors’ perceptions (Chung et al.. 2019). There are also authors more focused on the capacity of technologies to promote inclusiveness within marginalised segments (Huang and Lau 2020). The role of ICTs in improving cultural tourism experiences was addressed by Ramos-Soler et al. (2019). Accordingly, ICTs (e.g., mobile apps) positively contribute to the overall experience among senior citizens, particularly during the pre-travel and on-site stages. However, when analysing the impact of specific tourist apps, the authors found that visitors did not use them, mainly due to a lack of knowledge about their existence. Still, it is recognised that embracing the senior market with tourist apps would improve the sustainability of world heritage sites, due to their propensity to use online review platforms and the associated potential of ICTs to promote cultural heritage. In turn, the study of Shen, Sotiriadis and Zhou (2020) analysed the influence of social networks on improving visitors’ sustainable behaviours during the overall experience. Accordingly, smart technologies can promote and drive visitors’ pro-sustainability behaviours in all three stages of a tourism trip, particularly during the pre-trip and on-site moments, also contributing to the sustainable management of tourism resources by the destination’s managers. Chung, Tyan and Lee’s (2019) research shows that the adoption of green technologies (e.g., near field communication, beacons) in museums positively influences visitors’ perceptions concerning corporate social responsibility practices, particularly environmental, social and economic ones, which, in turn, influence the way visitors perceive and evaluate the quality of tourism attractions. Thus, this will enhance sustainable practices by both supply and demand sides, once both parties’ interests are reinforced in the process. Supporting the implementation of technological tools to ensure a positive experience for the visitor, Gomez-Oliva et al.’s (2019) experiment demonstrated that the dissemination of technological solutions (e.g., IoT devices) through specific points of a tourism destination would enable autonomous sharing of tourists' content, co-created by residents (e.g., storytelling), based on a web-app that contributes to economic, environmental and socio-cultural sustainability of tourism destinations. This contribution is materialised through attracting digital tourists, collecting data regarding the environmental impact of tourist activities, and promoting local communities’ participation in the tourism development process. More precisely, the creation of smart areas within the destination, resulting in an alternative communication channel, allows the dissemination of cultural offers to both visitors and residents, revitalising the destination’s heritage and promoting the digitalisation transformation process. In their turn, Huang and Lau (2020) propose a gamification approach, namely an app to enhance the tourism experience of people with visual impairments. In addition to establishing an emotional bond with the destination, through a gamified approach that also enables visitors to increase their knowledge about it, this study contributes to the social inclusion of people with visual disabilities in the tourism context, providing the opportunity for tourism managers to develop more accurate and accessible travel apps, according to the needs expressed by these travel segments. The study strengthens the observation of Buhalis et al. (2019) regarding the added value of technologies for the inclusiveness of visitors with disabilities. Furthermore, the results demonstrate that these technological solutions enrich visitors’ experience and personal quality of life, mainly due to the sense of enjoyment and autonomy provided to citizens with visual constraints. Despite the different approaches, digitalisation seems to enable tourism attractions and destinations to enrich their bond with and knowledge about target markets, improving visitors’ experience and supporting sustainability. Destination management Smartness can have a great impact on tourism destination management (Boes et al. 2016; Gretzel et al. 2015a). Taking advantage of smart technological tools, destination managers are able to collect, analyse, and share a vast amount of data concerning the destination ecosystem (Sun et al. 2016). For instance, smart technologies can be implemented to track visitors’ movements and consumption patterns (Gretzel et al. 2015a), analyse online user-generated content (Encalada et al. 2017; Schimperna et al. 2021), or connect and interact with visitors (Mirzaalian and Halpenny 2019; Neuhofer et al. 2015). These potentialities offer new management model opportunities (Ivars-Baidal et al. 2019; Zubiaga et al. 2019), where technology and data are central actors (Ivars-Baidal et al. 2019). The main applications of smart technology in this field concern spatial distribution and monitoring visitors’ flows (Crivellari and Beinat 2020; Del Vecchio et al. 2018; Encalada et al. 2017; Kim et al. 2019; Zubiaga et al. 2019), sustainable tourism (Saltos et al. 2021; Vizuete et al. 2021), tourism promotion (Idris et al. 2021), and heritage preservation (Slavec et al. 2021). The phenomenon of overtourism can be a real issue for a tourism destination if managed improperly. The mass of visitors at specific attractions/areas might harm the destination’s overall environment (e.g., businesses’ life-cycle, local communities’ well-being, ecological equilibrium) and jeopardise the visitor experience as well (Zubiaga et al. 2019). In that way, effective management approaches are mandatory to guarantee the functionality of tourism destinations. Within it, big data analytics arise as a smart tool with considerable usefulness in tourism management, particularly to ensure truly sustainable development. For instance, in Encalada et al.’s (2017) study, the spatial distribution of visitors in Lisbon is analysed through geotagged photos published on social networks (Panoramio). By doing so, the authors identified the main tourism hotspots of Lisbon. More than that, they also found marginalised sites that were disregarded by tourism managers but valued by visitors, showing great potential for tourism purposes. These insights prove that by using technological tools, such as big and open data, tourism managers can adequately cope with sites that are under pressure, particularly by reallocating visitors to alternative points of interest. This will revitalise specific areas, create new business opportunities, and optimise the visitor experience, thus contributing to the sustainable development of the destination. Similarly, Del Vecchio et al. (2018) claim that through this type of analysis, managers can identify specific patterns regarding the destination (e.g., critical points, areas needing intervention, opportunities for development) and the demand (e.g., satisfaction, expectations, needs). Accordingly, there is great potential associated with the analysis of visitor-generated content on social media, particularly the creation of knowledge, allowing the destination to improve its performance on critical issues (e.g., accessibility, price, waste management) and to identify market segments (assessing visitors’ personal information). Moreover, social networks are also an important marketing channel. Through big data analytics, tourism managers can directly involve visitors in this process, contributing to a more personalised offer and guaranteeing a constant interaction between the provider and the demand, thus supporting the premiss of value co-creation (Buhalis et al. 2019; Pencarelli 2020). By doing so, tourism managers are a step ahead of sustainable tourism, as the results support the fact that big data analytics enable new opportunities for destinations based on non-conventional natural and cultural settings, which, in turn, have considerable positive implications for economic, environmental, and societal development. Implementing a network of IoT-related technologies in tourism sites enables tourism planners to efficiently manage visitor flows (Gretzel et al. 2015a). In this field, Zubiaga et al.’s (2019) study demonstrates how integrating IoT and Geographic Information Systems (GIS) within a management model can contribute to monitoring visitor flows and, consequently, improve destination management towards sustainability. To do so, the authors designed a monitoring system to collect and share data about visitors’ mobility patterns (e.g., occupation level, most-visited sites) within the historic centre of Ávila. The results demonstrate that the system supports decision-makers in developing strategies to avoid overcrowding situations and lighten the pressure on specific attractions/sites. Despite this primary outcome, it also provides the opportunity to design new attractions, improve less-visited places, and define new visitor routes, particularly in the surrounding areas of the destination. The added value of this solution is in the alarm method that notifies managers in overcrowded situations, allowing them to put into practice measures to control visitor flows (e.g., activating barriers) and to notify visitors through a mobile app, suggesting alternative activities or attractions to visit. Supporting the potential efficiency of IoT technologies within these contexts, a pilot project aiming to develop a visitor counting system using ultrasonic sensors and Bluetooth modules was implemented on Jeju Island (Indonesia) to deal with the gradual increase of visitors and the consequent environmental impacts on the destination (Kim et al. 2019). This solution allows authorities to properly design strategies toward limiting the number of visitors, thus preventing environmental constraints and associated costs. Moreover, the device also measures and provides environmental data (e.g., air pollution, wind speed, humidity, temperature) to visitors. Therefore, the system helps to support control and restrictive policies and, simultaneously, provides environmental information to visitors, allowing them to manage their experience effectively. The study by Crivellari and Beinat (2020) proposes a model that captures visitors’ mobility patterns to identify and predict future behaviour during visitation. Individual spatial choices are determined and predicted based on a long short-term memory neural network. In other words, the model is expected to define visitors’ movements within a destination. Accordingly, this solution might be helpful in developing location-based services, crowd control, and, in a broader context, destination management and planning. The model also paves the way to optimise the visitor experience, particularly by offering the possibility of informing visitors about crowded areas, providing personalised information and recommendations, or highlighting specific attractions, in line with the predicted trajectory. Through this prediction method, tourism planners might be able to properly comprehend the future spatial distribution of visitors, allowing them to reallocate facilities and services along with the trajectory, and predict crowded areas, thus contributing to more informed and balanced decisions. A distinct approach was chosen by Vizuete et al. (2021) and Saltos et al. (2021), although looking toward improving destination management conditions. By measuring the importance of 38 smart tourism tools based on visitors’ opinions, the study by Vizuete et al. (2021) concluded that visitors tend to value technologies related to safety, mobile payment, and e-tools (e.g., websites and blogs, recommendation systems, mobile technologies), meaning that the authorities should concentrate on implementing these technological facilities within the territory. As “data is the new oil”, the proliferation of technologies will enable both tourism suppliers and managers to understand and manage the needs of the demand side, customising services and adapting the territory. Consequently, new businesses will emerge, as well as direct investments, thus contributing to economic and social development. Moreover, by stimulating the stakeholders’ direct participation in selecting the most appropriate technologies, the path towards sustainable development is created, as the needs and worries of distinct actors that actively contribute to the destination’s growth can thus be collected together. In turn, based on the premise that social networks are valuable assets in a destination management context, Saltos et al. (2021) examined how Spanish destinations were using it to improve their management capacity. Despite finding that almost 90% of the analysed destinations were present on three or more social networks, no consistency was shown concerning their dynamism (e.g., daily publications), proactivity (e.g., engagement with visitors), or interaction (e.g., collaboration with websites from public administration, companies). This means that destination managers still use social networks for their traditional purpose, mainly promotion or to obtain information about visitors, disregarding active communication with visitors and other stakeholders. Moreover, the authors found no differences between smart and non-smart destinations, suggesting that the former were not benefiting from their technological advancements. Although focusing exclusively on the role of social networks, these findings seem to create a certain doubt concerning the added value of the smart destination paradigm, while demonstrating that there might be a lack of technological capabilities among tourism managers due to their apathy in engaging with the remaining stakeholders. Identifying major issues in the promotional information strategy of Madura Island, Idris et al. (2021) designed an android-based tourism information system using VR. The application was elaborated in an integrative way, by listening to a wide range of stakeholders, encompassing tourism companies, tourism managers, tourists, and local communities. Although recognising the added value of involving both managerial and societal sides in the design process, concerning the technological side, it seems that VR was poorly implemented, as the application only provided 360º panoramic images of each available feature (e.g., attractions, accommodation, amenities). Nonetheless, the application proved to be an effective solution for sharing tourism information about Madura Island and became a promotional tool contributing to the development of the tourism sector. Finally, Slavec et al. (2021) aimed to comprehend the extent to which it was feasible to involve visitors in the monitorisation of cultural heritage using technologies, such as smartphone cameras, visitors apps, and location-based games, for instance, by reporting damage to local authorities. Based on a focus group approach, the study suggests that preservation of cultural heritage might be improved through smartphone travel apps and location-based games, ensuring a greater interaction between visitors and cultural attractions. This solution might contribute to enhancing the destination’s sustainability, firstly by involving visitors in the preservation process as main actors, increasing their motivation towards sustainability issues, and, secondly, by ensuring that the authorities are aware of the heritage that needs urgent intervention. Smart technologies arise as valuable resources to collect, share, and analyse the massive amount of data resulting from the ‘interaction’ between visitors and destinations. Furthermore, their effective implementation is required for balanced and profitable management strategies to enhance destinations’ competitiveness. Business solutions Smart technologies play a critical role in the competitiveness of tourism businesses (Neuhofer et al. 2015). Specifically, these tools arise as strategic resources for the success or revitalisation of a tourism company. For instance, the COVID-19 pandemic seriously affected the global economic structure with tremendous negative impacts on the tourism industry, particularly the shutdown of several businesses. Tourism players must rethink their business models to prevent similar consequences in the future and guarantee their sustainability. With the economic ecosystem nowadays being ruled by disruptive businesses where sharing and circular models are emerging approaches, reconfiguration and flexibilisation are necessary to remain competitive (Buhalis et al. 2019). With this in mind, the study by Del Vecchio et al. (2022) regarding the development of an eco-friendly accommodation network is discussed. Further, Polese et al. (2018) propose an integrative smart service ecosystem based on value co-creation and innovation, in line with the observation of Buhalis et al. (2019) about the impacts of technologies and the smart paradigm on service management. The literature reviewed highlights some experiments used to illustrate how businesses could be improved through the implementation of digital solutions without compromising sustainable development. In particular, the study by Lim, Mostafa and Park (2017) designs a mobile application based on the preservation of cultural heritage to improve business operations and, consequently, visitors’ experience. The optimisation of transportation services through smart technologies is then discussed by Moustaka et al. (2019). Aiming to enhance Japanese cultural traditions (Omotenashi culture) and narrow the existing gap between service providers and foreign visitors due to language and communication barriers, the study conducted by Lim, Mostafa and Park (2017) proposes the design and implementation of a technological-based service-assisting system – Eatjoy – supporting this dysfunctional communication process. According to the results, the system enhanced visitors’ sustainable values and shone a light on cultural elements of Japanese history, positioning smart approaches as key factors towards tourism performance and competitiveness. The study demonstrates how technology can improve tourism services to support business growth without disregarding important assets of nations’ cultural traditions, which, in this case, constitute the basis of high-quality services in the tourism sector. Centred on an environmental perspective, the study of Del Vecchio et al. (2022) analyses the case of a network-based booking company promoting eco-friendly accommodation through a circular economy business model – ‘Ecobnb’. The results demonstrate that as well as promoting responsible travel behaviour among potential guests, the company offers the possibility for new or already existing accommodation establishments, to engage in a more ecological path.Support is provided towards the transition to a sustainable business model, which is the requirement to become part of the online network. Moreover, through big data analysis, the company can sustain and optimise its offer according to the expectation and needs of the target segments, by obtaining insights that improve decision-making and allowing companies involved in the network to design competitive strategies concerning ecological sustainability. Although focused on different dimensions of sustainability, both studies demonstrate how digitalisation supports the growth of tourism companies through an integrative process that is genuinely balanced and not exclusively driven by motivations of profit. By preserving elementary cultural aspects, such as Eatjoy, or operating in an ecological-oriented niche market, such as Ecobnb, these solutions prove that tourism businesses have the margin to develop more responsible growth models supported by ICTs solutions. The study by Moustaka et al. (2019) proposes a framework – TOMI – based on data analytics to improve the service of road passenger transport operators in Thessaloniki, Greece. Several technologies support the development of this framework, particularly big data analysis (e.g., bookings, routes, itineraries), IoT devices to monitor each vehicle’s data (e.g., engine status, fuel consumption, failures), and social network analysis to manage user-generated content. The aggregation of this data in one database allows different stakeholders (e.g., tour operators, travellers, local authorities) to benefit, first by optimising the service operation of tour companies (e.g., reducing costs, adopting fare strategies); secondly, by enabling the development of customised and economical services for travellers; and thirdly, through improving a destination’s competitiveness, by attracting new businesses, revitalising attractions, supporting decision-making, and diminishing environmental pollution resulting from tour operations. In turn, Polese et al.’s (2018) study discusses the perspective of Italian bed and breakfast businesses by analysing how smart service ecosystems contribute to value co-creation and innovation. Four components of the ecosystem were identified, particularly the stakeholders, the resources, the type of technologies, and the communication strategies. Through interviews with bed and breakfast owners, the authors understood that technologies operate as management tools, increasing the links between all the stakeholders within the ecosystem and, consequently, exchanging considerable amounts of data that can be used to improve business models. Particular attention was given to the relationship with guests. The process of continuous information exchange and data analysis (provided by guests’ opinions or recommendations through online platforms) ensures that businesses can detect areas that need improvement to increase guests’ satisfaction. Therefore, technologies ensure that all the stakeholders within the ecosystem interact and contribute to value co-creation, increasing the amount of data exchanged among them (Buhalis et al. 2019). Consequently, economic, societal, and environmental benefits are spread to all the actors and the territory, thus granting companies innovation in services, strengthening relationships among stakeholders, and establishing networks that contribute to territorial development. Smart sustainable destinations Tourism sustainability implies an effective balance between social, economic, and environmental pillars. However, the traditional development models, seem to emphasise the economic side, perhaps because most stakeholders consider that the benefits for the society stem from the economic dynamic of the tourism destination (Serra et al. 2017). On the other side, the environmental dimension, despite being referred to as a key feature in most strategic plans, is hardly operationalised (Križaj et al. 2021). As emphasised earlier, different visions lead to misinterpretations of both sustainable and sustainability constructs (Williams et al. 2020). The solution for effective sustainable tourism strategies might be through innovation dynamics where smartness and new technologies have a central role (Choudhary et al. 2020; Gössling 2020; Romão and Neuts 2017). However, there is no space for the ‘one-size-fits-all’ logic in smart and sustainable strategies. To argue about these assumptions, several studies discussing how both sustainability and smart issues complement each other within development policies in different geographical contexts and levels are presented. To investigate cities’ capacity to foster sustainability through smart approaches, González-Reverté (2019) conducted a content analysis of strategic documents related to urban sustainability in Spain. The results reveal the absence of action plans centred on implementing technological solutions as a catalyst for urban sustainability. The author identified a lack of coherence and coordination within those combining both dimensions, particularly concerning a shared sustainable vision. This promotes a disconnected, diversified, and rhetorical approach to urban sustainability in smart destinations’ plans, meaning that these strategies may not represent an efficient policy tool to achieve sustainable patterns within tourism growth. Gomis-López and González-Reverté (2020) examined how mature beach destinations incorporate smart and sustainable measures to prevent decline and improve competitiveness. The results of this study are in line with the conclusions of González-Reverté (2019). For instance, smart tourism plans aiming to improve sustainability were underrepresented, with technological solutions (e.g., apps, Wi-Fi, presence of digital platforms) being adopted on a very small scale and with very limited implications for sustainability. This limited technological scope, along with the restricted relevance of technological solutions towards urban sustainability, led the authors to suggest that smart strategies are understood as plans to revitalise the competitiveness of declining destinations, almost disregarding sustainability, which might constitute an issue in the mid or long term. In turn, Križaj et al. (2021) evaluated the smartness level of European smart tourism projects by analysing the adoption and implementation of technological solutions, as well as their sustainability orientation. Of the eligible projects (352), only 10% were truly smart, proving the commercialisation of the smartness paradigm (Gretzel et al. 2015a) and emptying the concept of its real meaning. Among the smart projects, emphasis was given to technologies such as big data, sensors, mobile apps, and IoT, particularly in the fields of transportation, air pollution, and social innovation. Concerning the sustainability orientation, most of the projects were focused on the sustainable dimension, while few highlighted the social or environmental dimensions. However, little information was provided about the meaning of the sustainable dimension and it was inferred that it might be related to all the dimensions of sustainability (e.g., economic, environmental, social). Through this approach, Križaj et al. (2021) propose a model that is expected to contribute to an effective classification of emerging ‘smart’ projects. Despite these findings, the study by Liu et al. (2019), based on an evaluation matrix for the construction of a smart tourism city, addresses sustainable development (e.g., mobility, accessibility, renewable energy, biodiversity) as a key feature. Nevertheless, based on the insights of the previous studies (e.g., Gomis-López and González-Reverté 2020; González-Reverté 2019), this indicates the existence of a gap between stakeholders’ narratives and actions in practice, compromising the real meaning and purpose of smart approaches. The truth is that it is not feasible to separate sustainability from smartness and vice-versa. As Ortega and Malcolm (2020) reported, based on the perceptions of tourism stakeholders about smart tourism destinations in Mexico, sustainability is the nuclear and strategic factor inthe effective implementation of a smart tourism destination. However, the concept of a smart destination still lacks a conceptual definition, which might lead to incoherent and uncoordinated development of smart approaches, as reported by the previous studies in this analysis. Nevertheless, smart tourism is also perceived as a strategy for promoting equity among tourists and locals, enhancing the development of new business models and enriching the overall tourism experience. Ivars-Baidal et al.’s (2021) study examined how tourism destinations evaluate the smart paradigm by analysing a system encompassing specific indicators, such as sustainability. The results demonstrate that sustainability was centred on the environmental field. Within it, it was concluded that governments showed a certain commitment to sustainability goals. Nonetheless, environmental certificates and indicators to measure environmental performance were scarce among companies, as were actions to raise awareness and plans to prevent climate change. Thus, despite integrating an indicator evaluating the sustainable performance of smart destinations, the focus is on the environmental component, disregarding the social and economic dimensions. This might misstate the real meaning and purpose of the smart tourism concept, calling once more for clarification of the concept, as suggested by Ortega and Malcolm (2020). Zeng et al. (2020) investigated the role of big data analytics as a trigger for the transformation of Qinhuangdao, China, into a smart tourism destination. Supported by qualitative interviews and under the affordance theory, a three-stage model was developed, identifying how big data analytics technologies (e.g., stream analytics, predictive analytics, visualisation applications) were used in different stages of the process. Great potential was associated with the implementation of this technological solution, particularly an informed transformation according to the main requirements and opportunities identified, the development of customised tourism services in line with visitors’ needs (e.g., personalised recommendations), and the creation of sustainable ecosystem involving different stakeholders in coordinated efforts towards effective development strategies. Conclusion This study conducted a literature review aiming to thoroughly investigate how sustainable development is being addressed in the literature along with the emerging concepts of smart and digital tourism. The rationale behind this study relied on the increasing need to adopt alternative strategies concerning tourism development models toward sustainability. Specifically, the transition to a digital economy, where new information and communication technologies associated with the 4.0 paradigm are central players, seems to arise as a feasible path. Thus, the content of 25 publications was analysed to identify the methods, successful approaches, and best practices in distinct contexts. Valuable insights for academics and managers result from this study. From the theoretical side, this is the first attempt to explore the combination of these topics, apart from the review study of Shafiee et al. (2019), detailed earlier. This study contributes to the smart and sustainable tourism literature by listing several topics in the context of smart technology use towards sustainability. Particularly, four main research themes - visitor experience, destination management, business solutions, and smart sustainable destinations - seem to be of significant relevance in the documents examined. Particular attention has been paid to the potential role of technologies in improving visitors’ overall tourism experience and how this interaction could also benefit the sustainability of a tourism destination. The influence of smart tourism on the inclusiveness of specific market segments (e.g., seniors, disabled) was also discussed, contributing to the dissemination of tourism experiences for all, under the sustainable development goals principles. Smart approaches were also associated with destination management. Particularly, the added value regarding the control of visitors’ flows was highlighted, providing the opportunity for the identification and promotion of new and alternative tourism products, reducing the pressure on the main hotspots and ensuring the well-being of visitors and local communities, as well as the preservation of tourism attractions/sites. Digital transformation of businesses was also emphasised as a factor contributing to the resolution of business constraints or as a key factor to the development of new models based on the co-creation of value where economic, societal, and environmental aspects are embraced. However, further studies are needed on implementing smart technologies in tourism firms and how those innovation strategies contribute to internal and external sustainability practices. This can be done through research on smart tourism, sharing economy and circular economy (Del Vecchio et al. 2022) and the adoption of case studies. Finally, the results showed the proliferation of smart and sustainable strategies in tourism destinations, although some dissimilarities were revealed. On one side, there is a gap between policy strategies’ narratives and the implementation of those policies in practice. On the other, sustainability is advocated as a key feature of the success of smart tourism destinations. This calls for a more accurate and holistic approach to smart and sustainable tourism strategies, namely, to support a concept that still lacks conceptual and empirical clarification. Throughout the discussion, the main technologies were also pinpointed. In this case, there is a clear concentration on social network sites and mobile applications, which opens space for future studies to explore the implications of other smart tourism technologies, using reference studies (e.g., Buhalis et al. 2019; Gretzel et al. 2015b; Jeong and Shin 2020) as background for the identification of the most prominent ones. This study also demonstrates the existence of several world regions with scarce or non-existent publications in this field. This is particularly true for North America, Central Europe, and Africa. Although recognising disparities between developed and developing countries concerning the adoption of the smart paradigm, its great potential has been addressed throughout this study and should be acknowledged with further research in distinct geographical contexts. From the managerial standpoint, this study has shown practical examples of how businesses and tourism destinations use smartness in their favour to support sustainability practices or to attain sustainable development. Greater interactive practices between stakeholders and effective data collection and analysis methods might assist playmakers in improving their business models and/or managing their destinations effectively. The study also shows that smartness contributes to value co-creation, enabling visitors and local communities to be part of the tourism experience and the managerial process. However, the extent to which companies and institutional organisations are prepared to deal with the ongoing digital transformation and yet be sustainable or contribute to the ecosystem’s sustainability should be thoroughly explored. Moreover, some existing smart tourism strategies seem to use both smart and sustainability concepts in a marginal ,propagandist way, more similar to a marketing point of view. This might be due to theoretical inconsistencies, which need to be addressed in future studies, defining a reliable basis for fully understanding the smart approach in tourism. Since the pandemic crisis has boosted the digitalisation transformation, it is expected that the number of publications encompassing these topics will grow in the short term. Thus, one recommendation for further research, in the systematic review field, concerns the analysis of the role of technologies within this panorama. In that way, researchers would be able to divide their findings by type of technology and, consequently, understand those with major relevance in the tourism context, since 4.0 technologies cannot be simply transferred to the tourism context, particularly because they were primarily designed as industrial solutions, while tourism is mainly characterised by services. Future studies should also concentrate on the comparison of digital dynamics before and after the COVID-19 pandemic as an increase in technology use is expected during and after the pandemic crisis. Another recommendation for future studies concerns the investigation of the role of smart approaches in crisis management. It is relevant to understand whether technological approaches arise as real solutions to prevent the total shutdown of the tourism industry in crisis scenarios (e.g., economic collapse, pandemics), saving millions of jobs and businesses, and providing alternative development models that might constitute the future for a more sustainable sector. Despite some studies focused on the impact of smart tourism on the visitor experience and its influence on sustainability, further evidence is needed to comprehend the implications of smartness and technologies on visitors. More precisely, more evidence should be provided concerning the extent to which visitors accept the increasing digitalisation of the sector, their digital literacy levels, and their contribution to a sustainable path. Despite the relevant contributions of this study, some limitations can be presented. As it was a preliminary approach to a systematic literature review, only articles and conference papers were analysed in the present work. The continuation of this work aims to integrate different document types, namely book chapters and additional ‘grey literature’ that might be relevant to extend the literature review around the topic in discussion. Furthermore, it is also intended to analyse the references of each document encompassed in this process to enlarge the sample and provide additional insights. A second limitation concerns the exclusive use of the Scopus database as the main source to access documents. Despite being one of the most extensive databases, cross-analysis with other databases, such as ISI Web of Science and the Directory of Open Access Journals (DOAJ), which also integrate peer-reviewed publications, would be a complementary way to broaden the sample. Additionally, the search query could be expanded to include keywords such as “smart technology”, “information and communication technology”, and “technology”. This would allow for an in-depth understanding of the phenomenon in analysis, particularly by identifying the leading technologies being implemented within this context. Even though sustainability has received the attention of several studies in the past, it is also true that the tourism industry still fails in the integration of the overall dimensions of the concept. Moreover, despite being argued that technologies might play a valuable role as management tools by bridging gaps and bringing people and places together, even if in a virtual framework, their incorporation in tourism destinations and businesses is still in a preliminary stage. This is particularly the case because tourism is mainly composed by micro, small and medium enterprises, with several constraints regarding human, financial, and investment resources that make it difficult to engage in the digital transformation process. In a certain way, this might also justify the slight empirical evidence regarding the relationship between these topics. Nonetheless, this study provided a thorough understanding of the relationship between smart and digital tourism and sustainability, as well as the basis for future research. Acknowledgements This work was financially supported by the research unit on Governance, Competitiveness and Public Policy (UIDB/04058/2020)+(UIDP/04058/2020), funded by national funds through FCT - Fundação para a Ciência e a Tecnologia, under the PhD Grant UI/BD/152274/2021. Data availability No datasets were analysed or generated during the development of the current study. Declarations Conflict of interest The authors have no relevant financial or non-financial interests to disclose. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Buhalis D Harwood T Bogicevic V Viglia G Beldona S Hofacker C Technological disruptions in services: lessons from tourism and hospitality J Serv Manag 2019 30 4 484 506 10.1108/JOSM-12-2018-0398 Brundtland G (1987) Report of the World Commission on Environment and Development: Our Common Future. United Nations General Assembly document A/42/427 Cavalcante WQ D F, Coelho A Bairrada CM Sustainability and tourism marketing: a bibliometric analysis of publications between 1997 and 2020 using vosviewer software Sustainability 2021 13 9 4987 10.3390/su13094987 Choudhary SA Khan MA Sheikh AZ Jabor MK bin Nordin MS Nassani AA Alotaibi SM Abro MMQ Vo XV Zaman K Role of information and communication technologies on the war against terrorism and on the development of tourism: evidence from a panel of 28 countries Technol Soc 2020 62 101296 10.1016/j.techsoc.2020.101296 Chung N Tyan I Lee SJ Eco-innovative museums and visitors’ perceptions of corporate social responsibility Sustainability 2019 11 20 5744 10.3390/su11205744 Crivellari A Beinat E LSTM-based deep learning model for predicting individual mobility traces of short-term foreign tourists Sustainability 2020 12 1 349 10.3390/su12010349 Crouch GI Ritchie JB Tourism, competitiveness, and societal prosperity J Bus Res 1999 44 3 137 152 10.1016/S0148-2963(97)00196-3 Davis J Mengersen K Bennett S Mazerolle L Viewing systematic reviews and meta-analysis in social research through different lenses SpringerPlus 2014 3 1 1 9 10.1186/2193-1801-3-511 24422185 Del Vecchio P Malandugno C Passiante G Sakka G Circular economy business model for smart tourism: the case of Ecobnb Euro Med J Bus 2022 17 1 88 104 10.1108/EMJB-09-2020-0098 Del Vecchio P Mele G Ndou V Secundo G Open innovation and social big data for sustainability: evidence from the tourism industry Sustainability 2018 10 9 3215 10.3390/su10093215 Elsevier (2021) Scopus – data. https://www.elsevier.com/solutions/scopus Encalada L Boavida-Portugal I Cardoso Ferreira C Rocha J Identifying tourist places of interest based on digital imprints: towards a sustainable smart city Sustainability 2017 9 12 2317 10.3390/su9122317 Errichiello L Micera R A process-based perspective of smart tourism destination governance Eur J Tour Res 2021 29 2909 10.54055/ejtr.v29i.2436 Fennell DA Technology and the sustainable tourist in the new age of disruption J Sust Tour 2021 29 5 767 773 10.1080/09669582.2020.1769639 Gajdošík T Orelová A Silhavy R Smart technologies for smart tourism development Artificial intelligence and bioinspired computational methods 2020 Switzerland Springer International Publishing 333 343 Gomez-Oliva A Alvarado-Uribe J Parra-Meroño MC Jara AJ Transforming communication channels to the co-creation and diffusion of intangible heritage in smart tourism destination: creation and testing in Ceutí (Spain) Sustainability 2019 11 14 3848 10.3390/su11143848 Gomis-López JM González-Reverté F Smart tourism sustainability narratives in mature beach destinations: contrasting the collective imaginary with reality Sustainability 2020 12 12 5083 10.3390/su12125083 González-Reverté F Building sustainable smart destinations: an approach based on the development of spanish smart tourism plans Sustainability 2019 11 23 6874 10.3390/su11236874 Gössling S Technology, ICT and tourism: from big data to the big picture J Sust Tour 2020 29 5 849 858 10.1080/09669582.2020.1865387 Graziano T Boosting innovation and development: the italian smart tourism: a critical perspective Eur J Geogr 2014 5 4 6 18 Gretzel U Scarpino-Johns M Destination resilience and smart tourism destinations Tour Rev Int 2018 22 3–4 263 276 10.3727/154427218X15369305779065 Gretzel U Sigala M Xiang Z Koo C Smart tourism: foundations and developments Electron Mark 2015 25 3 179 188 10.1007/s12525-015-0196-8 Gretzel U Werthner H Koo C Lamsfus C Conceptual foundations for understanding smart tourism ecosystems Comput Hum Behav 2015 50 558 563 10.1016/j.chb.2015.03.043 Hausberg JP Liere-Netheler K Packmohr S Pakura S Vogelsang K Research streams on digital transformation from a holistic business perspective: a systematic literature review and citation network analysis J Bus Econ Manag 2019 89 8 931 963 Hein L Metzger MJ Moreno A Potential impacts of climate change on tourism; a case study for Spain Curr Opin Environ Sustain 2009 1 2 170 178 10.1016/j.cosust.2009.10.011 Huang L Lau N Enhancing the smart tourism experience for people with visual impairments by gamified application approach through needs analysis in Hong Kong Sustainability 2020 12 15 6213 10.3390/su12156213 Idris I Adi KR Firmansyah R Nadhianty A Mobaroq MH Putri PG Pratama AS Wahono ER Developing smart tourism using virtual reality as a tourism promotion strategy in Indonesia Geo J Tour Geosites 2021 35 2 332 337 10.30892/gtg.35210-656 Ivars-Baidal JA Celdrán-Bernabeu MA Femenia-Serra F Perles-Ribes JF Giner-Sánchez D Measuring the progress of smart destinations: the use of indicators as a management tool J Dest Mkt Manag 2021 19 100531 10.1016/j.jdmm.2020.100531 Ivars-Baidal JA Celdrán-Bernabeu MA Mazón JN Perles-Ivars ÁF Smart destinations and the evolution of ICTs: a new scenario for destination management? Curr Issues Tour 2019 22 13 1581 1600 10.1080/13683500.2017.1388771 Jeong M Shin HH Tourists’ experiences with smart tourism technology at smart destinations and their behavior intentions J Travel Res 2020 59 8 1464 1477 10.1177/0047287519883 Kim JS Kang HJ Jwa JW Smart tourism monitoring device based on IoT technologies Int J Recent Technol Eng 2019 7 6s5 1281 1285 Koo C Joun Y Han H Chung N A structural model for destination travel intention as a media exposure Int J Contemp Hosp 2016 28 7 1338 1360 10.1108/IJCHM-07-2014-0354 Križaj D Bratec M Kopić P Rogelja T A technology-based innovation adoption and implementation analysis of european smart tourism projects: towards a smart actionable classification model (SACM) Sustainability 2021 13 18 10279 10.3390/su131810279 Lee P Hunter WC Chung N Smart tourism city: developments and transformations Sustainability 2020 12 10 3958 10.3390/su12103958 Li G Hou Y Wu A Fourth Industrial Revolution: technological drivers, impacts and coping methods Chin Geogr Sci 2017 27 4 626 637 10.1007/s11769-017-0890-x 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 J Clin Epidemiol 2009 62 10 1 34 10.1016/j.jclinepi.2009.06.006 Lim C Mostafa N Park J Digital Omotenashi: toward a smart tourism design systems Sustainability 2017 9 12 2175 10.3390/su9122175 Liu L Li Y Zhang H Terzibasoglu E Government’s perception of constructing smart tourism cities e-Rev Tour Res 2019 17 2 157 168 Mehraliyev F Choi Y Köseoglu MA Progress on smart tourism research J Hosp Tour Technol 2019 10 4 522 538 10.1108/JHTT-08-2018-0076 Mohanty P Hassan A Ekis E Augmented reality for relaunching tourism post-COVID-19: socially distant, virtually connected Worldw Hosp Tour Themes 2020 12 6 753 760 Moustaka V, Vakali A, Zikos N, Tsirakidis T, Anthopoulos G (2019), May L TOMI: A framework for smart tourism on the move innovation. In: Companion Proceedings of the 2019 World Wide Web Conference, pp. 123–129. 10.1145/3308560.3317051 Niewiadomski P COVID-19: from temporary deglobalisation to a re-discovery of tourism? Tour Geo 2020 22 3 651 656 10.1080/14616688.2020.1757749 Niñerola A Sánchez-Rebull MV Hernández-Lara AB Tourism research on sustainability: a bibliometric analysis Sustainability 2019 11 5 1377 10.3390/su11051377 Ortega JLC Malcolm CD Touristic stakeholders’ perceptions about the smart tourism destination concept in Puerto Vallarta Jalisco Mexico Sustainability 2020 12 5 1741 10.3390/su12051741 Panagiotopoulou M Somarakis G Stratigea A Smartening up participatory cultural tourism planning in historical city centers J Urban Technol 2020 27 4 3 26 10.1080/10630732.2018.1528540 Pencarelli T The digital revolution in the travel and tourism industry Inf Technol Tour 2020 22 3 455 476 10.1007/s40558-019-00160-3 Perles Ribes JF Ivars-Baidal J Smart sustainability: a new perspective in the sustainable tourism debate J Reg Res 2018 42 151 170 Perles Ribes JF Ramón-Rodríguez AB Obliquity in tourism economics: Smart and sustainable tourist destinations e-Rev Tour Res 2018 16 1 45 55 Pickering C Byrne J The benefits of publishing systematic quantitative literature reviews for PhD candidates and other early-career researchers High Edu Res Develop 2014 33 3 534 548 10.1080/07294360.2013.841651 Pintassilgo P Rossello J Santana-Gallego M Valle E The economic dimension of climate change impacts on tourism: the case of Portugal Tour Econ 2016 22 4 685 698 10.1177/1354816616654242 Polese F Botti A Grimaldi M Monda A Vesci M Social innovation in smart tourism ecosystems: how technology and institutions shape sustainable value co-creation Sustainability 2018 10 1 140 10.3390/su10010140 Popkova EG De Bernardi P Tyurina YG Sergi BS A theory of digital technology advancement to address the grand challenges of sustainable development Technol Soc 2022 68 101831 10.1016/j.techsoc.2021.101831 Posada J Toro C Barandiaran I Oyarzun D Stricker D de Amicis R Pinto EB Eiser P Dollner H Vallarino I Visual computing as a key enabling technology for industries 4.0 and industrial internet IEEE Comp Graphics App 2015 35 2 26 40 10.1109/mcg.2015.45 Ramos-Soler I Martínez-Sala AM Campillo-Alhama C ICT and the sustainability of World Heritage Sites. Analysis of senior citizens’ use of tourism apps Sustainability 2019 11 11 3203 10.3390/su11113203 Rogers PP Jalal KF Boyd JA An introduction to sustainable development 2012 Abingdon Routledge Romão J Neuts B Territorial capital, smart tourism specialisation and sustainable regional development: experiences from Europe Habitat Int 2017 68 64 74 10.1016/j.habitatint.2017.04.006 Sachs JD Schmidt-Traub G Mazzucato M Messner D Nakicenovic N Rockström J Six transformations to achieve the sustainable development goals Nat Sustain 2019 2 9 805 814 10.1038/s41893-019-0352-9 Saltos AE Flores-Ruiz D Barroso González M D L O Applying Social Networks in the management of Sustainable Tourist Destinations: an analysis of Spanish Tourist Destinations Land 2021 10 11 1142 10.3390/land10111142 Schimperna F Lombardi R Belyaeva Z Technological transformation, culinary tourism and stakeholder engagement: emerging trends from a systematic literature review J Place Manage Develop 2021 14 1 66 80 10.1108/JPMD-03-2020-0028 Serra J Font X Ivanova M Creating shared value in destination management organisations: the case of Turisme de Barcelona J Dest Mark Manag 2017 6 4 385 395 10.1016/j.jdmm.2016.06.005 Shafiee S Ghatari AR Hasanzadeh A Jahanyan S Developing a model for sustainable smart tourism destinations: a systematic review Tour Manag Persp 2019 31 287 300 10.1016/j.tmp.2019.06.002 Shen S Sotiriadis M Zhou Q Could smart tourists be sustainable and responsible as well? The contribution of social networking sites to improving their sustainable and responsible behavior Sustainability 2020 12 4 1470 10.3390/su12041470 Siddaway AP Wood AM Hedges LV How to do a systematic review: a best practice guide for conducting and reporting narrative reviews, Meta-analyses, and Meta-Syntheses Annu Rev Psychol 2019 70 747 770 10.1146/annurev-psych-010418-102803 30089228 Sigala M New technologies in tourism: from multi-disciplinary to anti-disciplinary advances and trajectories Tour Manag Persp 2018 25 151 155 Slavec A Sajinčič N Starman V Use of smartphone cameras and other applications while traveling to sustain outdoor cultural heritage Sustainability 2021 13 13 7312 10.3390/su13137312 Snyder H Literature review as a research methodology: an overview and guidelines J Bus Res 2019 104 333 339 10.1016/j.jbusres.2019.07.039 Stankov U Gretzel U Tourism 4.0 technologies and tourist experiences: a human-centered design perspective Inf Technol Tour 2020 22 3 477 488 10.1007/s40558-020-00186-y Stephenson ML Dobson GJ Deciphering the development of smart and sustainable tourism cities in Southeast Asia: a call for research ASEAS-Austrian J South-East Asian Stud 2020 13 1 143 153 10.14764/10.ASEAS-0033 Sun Y Song H Jara AJ Bie R Internet of things and big data analytics for smart and connected communities IEEE Access 2016 4 766 773 10.1109/ACCESS.2016.2529723 Tölkes C Sustainability communication in tourism: a literature review Tour Manag Persp 2018 27 10 21 10.1016/j.tmp.2018.04.002 Tyan I Yagüe MI Guevara-Plaza A Blockchain technology for smart tourism destinations Sustainability 2020 12 22 9715 10.3390/su12229715 Vizuete DD Gavilanes Montoya AV Muñoz Jácome EA Chávez Velásquez CR Borz SA An evaluation of the importance of smart tourism tools in the Riobamba Canton Ecuador Sustain 2021 13 16 9436 10.3390/su13169436 Vu TH, Dai NHB, ThiKhank NH, Quang DNV (2018) Overall structural system solution for supporting services and tourists management oriented on smart city in Vietnam. In: ACM International Conference Proceeding Series, Da Nang, Vietnam, pp 430–435 Williams AM Rodriguez I Makkonen T Innovation and smart destinations: critical insights Ann Tour Res 2020 83 102930 10.1016/j.annals.2020.102930 Xu LD Xu EL Li L Industry 4.0: state of the art and future trends Int J Prod Res 2018 56 8 2941 2962 10.1080/00207543.2018.1444806 Zeng D Tim Y Yu J Liu W Actualising big data analytics for smart cities: a cascading affordance study Int J Inf Manag 2020 54 102156 10.1016/j.ijinfomgt.2020.102156 Zhou K, Liu T, Zhou L (2015) Industry 4.0: Towards future industrial opportunities and challenges. In: 2015 12th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD), pp. 2147–2152. 10.1109/fskd.2015.7382284 Zubiaga M Izkara JL Gandini A Alonso I Saralegui U Towards smarter management of overtourism in historic centres through visitor-flow monitoring Sustainability 2019 11 24 7254 10.3390/su11247254	0	PMC9734436	NO-CC CODE	2022-12-14 23:28:28	no	Inf Technol Tourism. 2022 Dec 5;:1-33	utf-8	null	null	null	oa_other
==== Front SN Bus Econ SN Bus Econ Sn Business & Economics 2662-9399 Springer International Publishing Cham 377 10.1007/s43546-022-00377-1 Original Article ‘Cyclic syndrome’ of arrears and efficiency of Indian judiciary http://orcid.org/0000-0002-6319-0855 Mishra Sila [email protected] grid.417965.8 0000 0000 8702 0100 Department of Economic Sciences, Indian Institute of Technology, Kanpur, India 9 12 2022 2023 3 1 614 7 2022 18 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. One of the four pillars of democracy in India is the judiciary, which in the recent past has experienced the ‘cyclic syndrome’ of arrears. There are 3.5 crore cases pending in the Indian judicial system that has a bearing on contract enforcement. A burgeoning stream of literature has reported the role of the judiciary in economic growth and development. In the wake of a given potential economic multiplier of the judicial system, examining the factors affecting the performance of the judiciary should merit attention. The present study juxtaposes jurisprudence and production theory, not frequently examined in the same gust by employing Data Envelopment Analysis (DEA), Malmquist Productivity Index (MPI), Stochastic Frontier Analysis (SFA), and regression for High Courts and Subordinate Courts. Employing the dataset for the years 2014–19, we investigate the technical efficiency and productivity of the High Courts and their Subordinate Courts and examine the factors influencing the dissolved cases. Furthermore, we examine the impact of COVID-19 on the cases instituted and cases disposed of. To sum up, the paper, thus, touches upon two basic dimensions of justice for High Courts and Subordinate Courts in India: Timeliness in the disposal of cases and the proportionate use of the state’s resources. The study confirms the role of judges, judicial staff, and demand for justice on the supply of justice. Shreds of evidence point toward the need to introduce a “cocktail-based” approach instead of a “one-size-fits-all”. Supplementary Information The online version contains supplementary material available at 10.1007/s43546-022-00377-1. Keywords Judicial efficiency Data envelopment analysis (DEA) Stochastic frontier analysis (SFA) Malmquist productivity index (MPI) JEL Classification C23 D24 H11 H83 K40 K41 L11 issue-copyright-statement© Springer Nature Switzerland AG 2023 ==== Body pmcIntroduction “Inclusive economic institutions require secure property rights and economic opportunities not just for the elite but for a broad cross-section of society” (Acemoglu and Robinson 20121). A technically efficient judiciary subtly not just emboldens the faith of people in the justice delivery system but empirical studies have also confirmed its growth effects (Feld and Voigt 2006; Dam 2006; Chemin 2009). An efficient adjudication contributes to ease of doing business and with timely resolution of disputes, property rights are also safeguarded. The efficiency analysis of the judiciary is based on the utility-maximizing, rational behavior of judges (Posner 1993; Cooter 1983; Beenstock 2004). Where the judiciary maintains fiscal balance, the underperformance of the same jeopardizes the ease of doing business and shakes the confidence of citizens. Studies have also confirmed the labor-intensiveness of the judiciary, which often suffers from the bottlenecks caused by the number of judges (Beenstock and Haitovsky 2004). Ceteris paribus, judges have the tendency to improve the outcomes of courts (reduction in backlogs and dispute resolution rate); however, the physical and procedural constraints slow the work of judges (Djankov et al. 2003). Understanding the need for an impartial and efficient judiciary in fostering public trust and economic growth we build on the theoretical and preliminary findings to consider the case of the Indian judiciary. Indian judicial system is in the throes of a crisis and is currently experiencing a cyclic syndrome of delay2 of arrears and pendency (Ghosh 2018) as is evident from 3.53 crore cases pending among different levels of Courts in India (PIB 2019; Economic Survey of India 2019). Under Article 21 of the Indian constitution, a speedy trial is a fundamental constitutional right.3 Reviewing the arrears in the Indian Judiciary and the number of working days, the 230th Law Commission Report (2009) and Justice Malimath Committee (2000) recommended cutting down the holidays. Based on the data collected from the official website of the High Courts for the year 2020 (excluding the number of personal leaves taken by the judges), we found that on average the working days are 234 (Refer to Appendix Table 2 on Working days and vacancy for the year 2020). As per the EODB (Ease of Doing Business) report, in 2020 though India’s overall ranking improved to 63, in contract enforcement, it stands at 163rd position and 154th in property registration. Additionally, India ranks 98 in Civil Justice and 78 in Criminal Justice out of 128 countries (World Justice Project 2020).4 Circumstances have brought us to a long overdue moment of reckoning to deepen our understanding of the Indian judicial system and determine where the greatest opportunities for change lie. In Fig. 1, we present a schematic chart explaining the structure of the Indian Judiciary with the Supreme Court as the apex court followed by High Courts and their respective subordinate Courts. India follows a Common law system where laws are not codified in contrast with Civil law jurisdictions. To identify obsolete laws and examine the repealing of 1382 acts recommended in the year 1988, the Prime Minister Office formed Ramanujan Committee in 2014 which further recommended repealing 1741 such old statutes.5 Reminiscent of what Tacitus wrote in his book, “Corruptissima re publica plurimae leges” (greater the number of laws, the rampant is corruption), India is over-legislated and under-governed. The Law Commission of India in its 248th report recommended repealing 72 enactments. It is worth mentioning here that the Commission has in total recommended the repeal of 288 archaic laws to date through four of its interim Reports. In Appendix Table 1 we report some of the archaic laws still functional in India and their status across the world. To assess the performance of the judiciary, the India justice report, 2019 has employed an indicator methodology to rank the courts in India which is subject to criticism on the grounds of methodology because the rankings were constructed by quantifying indicators and then aggregating them by a weighted average method. It confuses and deludes policymakers (Akande et al. 2019). Nonetheless, the empirical evidence sheds no light on the kind of reform that should be adopted to bridge the inefficiency gap in the judicial system. Little research has been done in the Indian context that incorporates production theory to assess the performance of courts in India other than Gupta and Bolia (2020). Any kind of assessment should take into consideration multiple inputs and outputs. Thus, methods which come to our rescue are Data Envelopment Analysis (DEA), Stochastic Frontier Analysis (SFA), and Malmquist index. These three methods have been widely used to assess the performance of public administration. However, the application of the Malmquist index in the present context is limited to an analysis of Italian tax judiciary by Kittelesen and Forsund (1992) and Falavigna et al. (2018). The present study, thus, juxtaposes jurisprudence and production theory, not frequently examined in the same gust by utilizing DEA, SFA and MPI for High Courts and Subordinate Courts. The purpose is, with the assistance of economics, to throw new light on basic issues construing the efficiency of the Indian judicial system (Fig. 1).Fig. 1 Structure of the Indian Judiciary. Source: Author’s calculation There are some noticeable characteristics that discern the present study from the literature construing courts. A study done by Gupta and Bolia (2020) in the case of the Indian judiciary using Data Envelopment Analysis (DEA) noted that there is a gap in terms of analyzing the performance of different levels of the Indian judiciary. Their study is the first to use DEA to rank the performance of 24 High Courts for the years 2015–18. Our study has distinctive features over the earlier studies such as the usage of DEA, SFA, and MPI over the period of 2014–19 in the case of High Courts and their Subordinate Courts. Where Gupta and Bolia (2020) considered judges, staff, instituted cases, caseload, and cases disposed of as inputs and outputs of High courts our study in addition to these variables also incorporates the budget allocated to examine the technical efficiency of High Courts and Subordinate Courts. Moreover, this study delves into developing a different framework to compare the courts—both High Courts and Subordinate Courts in India which are here viewed as production units. Moreover, to contextualize our study, we have attempted to examine how the courts have performed in COVID-19. Taking a cue from the literature, we endeavor to address the following questions—how technically efficient are the High Courts and their Subordinate Courts of India? Has the productivity of the courts gone up in the last 5 years? What factors influence the dissolved cases in High Courts and Subordinate Courts? How has COVID-19 impacted the caseload and cases disposed of? To sum up, the paper, thus, touches upon two basic dimensions of justice for High Courts and Subordinate Courts in India: Timeliness in the disposal of cases and the proportionate use of the state’s resources. Literature review There is a substantial body of literature construing the efficiency of institutions. To gain deeper insights into the working of courts and improve their performance, scores of articles have used the concept of the ‘Judicial Production function’ (Voigt 2016). The clear rationale behind the present study is to explain how the efficiency of courts is impacted. Pursuing the set of explorations indicated earlier, we strive to construct the literature in three aspects—theoretical background, empirical studies, and research gap. Below, we review relevant studies and point out the gap in the literature. Theoretical background Noting the influence and the supporting role of law on the exchange, Stigler (1992) underscored two fundamentally different roles of economics in law. First, in offering expertise requested by lawyers and second, in the study of legal institutions and doctrine. Economic analysis provides an arduous framework for assessing the trade-offs policymakers face when restructuring legal institutions (Shavell 1999). Therefore, attention needs to be paid to the judiciary and its operation to expose its role in making the institutional ecosystem effective (Marciano et al. 2019). Posner (1985) considered the invocation of fairness and justice and due process of judicial decision-making as proxies for wealth maximization. However, a large part of economic theory assumes efficient courts that administer contracts both perfectly, fair, and freely (Chemin 2009; Botero et al. 2003). Judiciary like any other organization of society is an institution which sets the “rules of the game” and in turn impacts the process of economic development (North 2008). It might be tempting to preclude the poor performance of the criminal and civil justice system from any direct consequence on economic development; however, studies have shown that the trust of society is a form of informal institution that act as social capital, in addition to physical and human capital (Wang et al. 2019). It is this trust of society in rule of law that has the potential in reducing the cost of information collection (Zaheer et al. 1998) and is further deemed as a channel impacting economic growth. Drawing a distinction between developing and developed countries, Wang et al. (2019) reported that in a developing country where rule of law (considered as a proxy for the formal institution) is weaker, social trust (informal institutions) plays a significant role. Their evidence supported the new institutional economics theory of the prevalence of informal social norms when the formal legal rules fail new institutional economics (Williamson 2000). Judicial inefficiency severely impacts poor societies by impeding productive exchanges between private individuals (Botero et al. 2003). The Economic Survey of India (2018–19) in its chapter titled “Ending Matsyanyaya: How To Ramp Up Capacity In The Lower Judiciary” underscored that a culture of rule of law cannot be understated and should be pervasive as governance cannot be improved in isolation. Where the judiciary maintains fiscal balance, the underperformance of the same jeopardizes the ease of doing business and shakes the confidence of citizens. Marciano et al. (2019) reported that efficient rules will not operate effectively if not properly enforced. This ‘chicken or the egg dilemma’ as claimed by Posner (1988) states that without an established “justice” people might not have the means to afford a fast judiciary and without timely decisions, no judiciary can afford to deliver fair justice. A fast judiciary acts as a fundamental deterrent against economic agents’ willingness to deviate from previously signed contracts (Melcarne et al. 2021). Acemoglu and Robinson (2012) have discussed that the well-performing judiciary could also turn into an “extractive” institution following systematic violations of the rule of law. Building on this ambiguity Melcarne et al. (2021) investigated the renowned legal maxim “justice delayed is justice denied” empirically and suggested that countries that have fast judiciaries enjoy high levels of quality of justice (World Bank’s judicial quality index and the rule of law). Deyneli (2012) conveys that there is a dearth of studies in both legal and economic literature, that model and analyze the efficiency of courts, judges, and other judicial staff and noted that generally the literature has been limited to the demand side of justice service (Rosales-Lopez 2008). The supply side of justice service is dependent on the budget allocated for courts, the number of courts, judicial staff, working hours and technology, administration of the courts, management techniques, the system of legal education, the methods for licensing lawyers and selecting judges, education of the public about the legal system, the means of access to justice, the availability of alternative dispute resolution (ADR), judicial independence, and procedural reforms (Buscaglia and Ulen 1997). Drawing a caution on the expansion of courts beyond a point Posner (1988) stated that, unlike the way the economic system can add factories, courts cannot be added to meet the rising demand for justice, otherwise, it becomes difficult to synchronize the judgements of a large number of courts. Posner6 further argued that the crisis of overload in courts could be addressed with a simple but powerful tools of economic analysis like legislative reform and expansive interpretation of the nature and functioning of federal courts (Shapiro 1987). Staats et al. (2005) have provided five levels to measure the performance of the judiciary- independence, efficiency, accessibility, accountability and effectiveness. These parameters are key to the functioning of the judiciary and, thus, to the economic performance of any country. Both Efficiency and effectiveness which are used to gauge the performance of the judiciary are indistinctively used synonymously however, there is an underlying difference between the two. Efficiency or technical efficiency as defined by Marciano et al. (2019) draws a parallel with the concept of production theory of judicial activity implying the optimal utilization of resources/inputs to attain the output and places focus on the dispute resolution technology whereas, effectiveness (efficacy) measures the ability of the judiciary to deliver the demand of justice and inspects the equilibrium in the “justice market”. The effectiveness of an institution if considered as “humanly devised constraints” is directly impacted by de facto institutions that implement these rules (Marciano et al. 2019; Bubb 2013; Safavian and Sharma 2007; Williamson 2009; Williamson and Kerekes, 2011; Hodgson 2006; Voigt and Gutmann 2013). Empirical studies Espinosa et al. (2017) in their study reported that the Demand for litigation is affected by both, the distance and burden that the courts take on and the average productivity of courts falls with the transfer of judges. To deduce further, the courts are demand driven (Beldowski et al. 2020). Falavigna and Ippoliti (2020) investigated how judicial delays produce social costs which tend to affect the demand for justice. Talking about the role of judges literature is still ambiguous about the role of judges in improving court outcomes (reduction of backlogs and dispute resolution rate) where Falavigna et al. (2018) employing DEA confirmed the clear role of judges whereas, another study by Beenstock and Haitovsky (2004) and Dimitrova-Grajzl et al. (2012) by employing regression analysis proved how reveals that the disposition of cases does not depend on the number of judges. In line with the studies assessing the role of judges on judiciary Dayneli (2012) adopted a two-stage DEA to prove that there exists a positive and significant relationship between the salary of judges and efficiency scores. Schneider (2005) revealed using two-stage DEA that the qualification of judges and their career incentives impact the efficiency and the confirmation rate of courts. Additionally, employing SFA, Antonucci et al. (2014) showed that the efficiency of courts in Italy depends upon the vacancies and per capita expenditure. Moreover, considering the legislative reforms, a study in the context of Sweden by Agrell et al. (2020) provided evidence of how merging the district courts, on average, made the whole sector more efficient. The impact of judiciary on the firm in Indian context was probed by Chakraborty (2016) which reported how the judicial quality is a significant factor in determining firm’s performance (especially for contract-intensive) both international and domestic. From the methodological point of view, there have been various methods adopted to analyze the performance of courts like DEA and SFA. Metafrontier approach, Conditional DiD model, DEA (Output-Oriented with the assumption of VRS) with Bootstrap, Regression and Malmquist Productivity Index (MPI), FGLS (Feasible Generalized least square) among others (Beldowski, Dabros, and Wokciechowski 2020; Agrell, Mattson, and Mansson 2020; Espinosa et al., 2017; Falavigna et al. 2018; Antonucci et al. (2014); Deyneli 2012; Yeung and Azavedo 2011; Schneider 2005; Lewin et al. 1982). DEA is a linear programming and non-parametric approach to measuring the technical efficiency of firms. Technical efficiency is expressed as a ratio of actual to potential output and based on the technical efficiency scores one can compare and rank the firms in a sample. The studies comparing the courts by using DEA have either relied on output-oriented DEA with Variable Returns to Scale assumption (Falavigna et al. 2018; Schneider 2005) or DEA with Constant Returns to Scale assumption (Falavigna et al. 2018; Yeung and Azavedo 2011). The application of DEA with CRS assumption is not just a strong assumption but is also restrictive in nature in contrast with VRS. Additionally, there is a strand of literature that have performed two-stage DEA to examine the factors influencing the technical efficiency of Courts (Deyneli 2012). It is also worth mentioning here, that there is a burgeoning stream of literature that have ascertained not just the technical efficiency but also the productivity of courts like Kittelsen and Forsund (1992). Productivity is a descriptive measure expressed as a ratio of output to input, technical efficiency is a normative concept explaining the optimal conversion of inputs into output. Thus, a technically efficient decision-making unit (DMU) can still increase its productivity by making use of its scale economies (Coelli et al. 2005). Considering the deterministic nature of DEA there are studies that have adopted Bootstrap-DEA to obtain robust results like Falavigna et al. (2018) and SFA to overcome the limitation of DEA for being incapable to test hypothesis like the studies by Beldowski et al. (2020) and Antonucci et al. (2014). Apart from the methodology adopted to compare the performance of courts the selection of variables also deserves attention. The studies that have concentrated on the production function approach have broadly included Office areas, law clerk, other staff, Number of Judges, Demand for justice (pending and instituted cases) and in some cases expenditure allocated on judiciary as inputs; whereas, the choice of output in cases of courts has remained more of less the same i.e., cases resolved by courts in a year. From the literature reviewed so far, the category of courts that have been examined is Labor courts (Espinosa et al. 2017), Tax courts (Falavigna et al. 2015), magistrate courts, and district courts (Agrell et al. 2020; Beenstock and Haitovsky 2004). The studies examining the performance of based on technical efficiency of courts have been limited to developed countries like Poland (Beldowski, Dabros and Wokciechowski 2020), Sweden (Agrell, Mattson and Mansson 2020), France (Espinosa et al. 2017), Italy (Castro & Guccio, 2018; Falavigna et al. 2015; Antonucci et al. 2014), Brazil, Norway, Israel, Germany among others. Research gap Based on the literature reviewed so far, we found that though studies have been conducted with respect to advanced nations, to the best of our knowledge there is a dearth of studies in the context of developing economies (Ash et al. 2021). Additionally, DEA has been adopted by most of the studies to ascertain the technical efficiency but, its deterministic nature has been criticized and thus lately, the studies have corroborated the empirical findings with a stochastic approach (SFA). For the reasons stated in the above section on the limitation of non-parametric approaches, there are other robust parametric methods like two-step DEA-OLS analysis, panel least squares regression and SFA to examine the variables explaining the court performance (Beldowski et al. 2020). Moreover, the judiciary, like any other institution requires budget for its smooth functioning but, there are very less studies that have considered Expenditure on the judiciary as a key variable. The literature presents us with an immediate paradox on the role of judges as clearly explained in the Literature Review section. Their productivity increases when they are flooded with cases but, this hypothesis needs to be tested given the cyclic syndrome of arrears in the courts of developing countries. Additionally, the gap in the literature in the Indian context pertaining to both the levels of judiciary, i.e., High Courts and Subordinate Courts is also prominent. Methodology DEA model The requisite hallmark of a well-functioning judiciary is its ability to dispose of cases speedily. But, as far as measuring the performance of the court is concerned, due to the involvement of resources of multiple dimensions, it becomes a difficult task. DEA is one such method which can be used here. The origin of DEA dates to 1978 with the seminal work by Charnes and Cooper. DEA resembles the concept of production possibility frontier studied in economic theory. DMUs with efficiency scores of 1 form a frontier and all the other inefficient DMUs fall under that. DEA is a non-parametric data-driven method in which due care needs to be exercised while choosing appropriate returns to scale (RTS). CRS (Constant Returns to Scale) also called CCR (Charnes, Cooper, and Rhodes) and VRS (Variable Returns to Scale), also termed BCC (Banker–Charnes–Cooper) are the two types of models of RTS. In the CRS-type model, output changes by the same proportion as input and VRS both increasing and decreasing returns to scale are encompassed. This study uses the VRS model because of the reason that “in most sectors the true technology experiences variable returns to scale” (Berbegal-Mirabent, Lafuente and Sole 2013). Fare, Grosskopf and Lowell (1994) mentioned that O–O model is “very much in the spirit of classical production function defined as the maximum achievable output given input.” For the present study output-oriented model is found to be suitable because judges aim at maximizing the output (cases disposed of) given the inputs available. The output-oriented DEA7 model is formulated as follows. Assume that there are n DMUs (courts). Each court DMUj,j=1,2….n consumes m types of inputs and s types of outputs which in vector form can be expressed as xj=x1j,x2j,……,xmj and gj=g1j,g2j,…..gsj for court j. Following connotations are brought in for DEA and the related examination of efficiency:vj=v1,v2,….,vmTis the input weight vector, uj=u1,u,….,usTis the output weight vector, yj=y1j,y2j,y3j,….,xsjT>0,j=1,2,3…..,nthe output vector of DMUj, xj=x1j,x2j,x3j,….,xmjT>0,j=1,2,3…..,nthe input vector of DMUj. Computation of DEA comprises of three efficiencies: Technical efficiency (TE) is a measure of resource allocation efficiency ascertained in case of CCR model; Pure technical efficiency (PTE) is the indication of utilization level of inputs in BCC model; Scale efficiency (SE) is expressed as the ratio of TE and PTE. The dual and primal form of the output-oriented BCC model can be illustrated as: Basic form of BCC followed in the analysis.Dual DBCC0 Primal PBCC0 Maxz=z0 Min(wTx0+μ0)=z0 ∑j=1nxjλj≤x0 wTxj-μTyj+μ0≥0,j=1,2,3⋯..,n ∑j=1nyjλj≥zy0 μTy0=1 w≥1,μ≥0 ∑j=1nλj=1 λj=≥0,j=1,⋯.,n Source: Based on the review of literature Here, w0,μ0andμ00=optimalsolutionsofPBCC0. For our analysis n = 24, m = 4(3) and s = 1. Like statistics or any other empirical oriented methodology, sensitivity (stability or robustness) analysis is also important in DEA. Efficiency or inefficiency of DMU becomes questionable if the Degrees of freedom are inadequate. A general rule of thumb is (Cooper, Seiford and Tone 2007) n≥maxm×s,3m+s Malmquist index Malmquist index summarizes the change in productivity into technical efficiency change, pure technical efficiency change, scale efficiency changes and total factor productivity change (Tone 2004). MPI shows the intertemporal change in productivity at t based on productivity of previous period (t-1). Thus, for the jth court MPI at period t is:1 MPIjt=Djtxjt+1,yjt+1/Djtxjt,yjt. Djtxjt+1,yjt+1 and Djtxjt,yjt are the distance functions of jth court with level of period t as reference. MPIt,t+1 indicate the status quo between period t and t + 1. MPIj=1 is interpreted as the state of no change in total factor productivity,MPIj>1 implies gain in efficiency and MPIj<1 corresponds to the loss in efficiency. The general form of MPI is derived from the geometric mean of MPIjt and MPIjt+1 (Fare et al. 1994; Fare et al. 1998) which can be:MPIj=Djt+1(xjt+1,yjt+1)Djt(xjt,yjt)×Djt(xjt+1,yjt+1)Djt+1(xjt,yjt)Djt+1(xjt+1,yjt+1)Djt+1(xjt,yjt), MPIj=EFCH×TECH. In variable returns to scale (VRS), EFFCH=PECH×SEC,MPIj=Djt+1xjt+1,yjt+1Djtxjt,yjt×Djtxjt,yjtDjt+1xjt+1,yjt+1Djtxjt,yjtDjt+1xjt+1,yjt+1×Djtxjt+1,yjt+1Djt+1xjt,yjtDjt+1xjt+1,yjt+1Djt+1xjt,yjt. All the measures have been explained concerning movement from period t to t + 1. Table 1 provides a brief account of the decomposed measures of Malmquist Productivity Index (MPI). Finally, one needs to understand that MPIs are ascertained using the efficiency scores of DEA and frontiers are computed through DEA with output orientation. We have used DEAP version 2.18 for the analysis.Table 1 Summary of decomposed measures of MPI Decomposition Description (period t to t + 1) Cases (change in efficiency) EFFCH=Djt+1xjt+1,yjt+1Djtxjt,yjt Shows the change in technical efficiency on account of CRS. It indicates how well the production is close to PF EFFCH = 1: on the PF (invariable) EFFCH > 1: close to PF (efficiency gain) EFFCH < 1: away from PF (efficiency loss) TECH=Djtxjt+1,yjt+1Djt+1xjt,yjtDjt+1xjt+1,yjt+1Djt+1xjt,yjt It measures the magnitude of shift in PF for DMU TECH = 1: no change in technology TECH > 1: technical progress TECH < 1: technical backslide PECH=Djt+1xjt+1,yjt+1Djtxjt,yjt It accounts for technology level change in VRS PECH = 1: no change in technology PECH > 1: increasing technology level PECH < 1: decreasing technology level SECH=Djtxjt,yjtDjt+1xjt+1,yjt+1Djtxjt,yjtDjt+1xjt+1,yjt+1 It represents the role of scale change in change in productivity SECH = 1: scale change brings no productivity change SECH > 1: scale change promotes productivity change SECH < 1: scale change reduces productivity change PF production frontier, DMU decision-making unit, CRS constant returns to scale Source: Based on the review of literature Stochastic frontier analysis To measure the performance of DMUs, aimed at converting inputs into output(s), Aigner et al. (1977) and Meeusen and Van den Broeck (1977) put forward a parametric technique called the SFA model independently. The general form of which in panel data can be written as:lnqit=xit′β+vit-uit Orqit=expβ0+β1lnxit+vit-uit Orqit=expβ0+β1lnxit⏟Deterministiccomponent×expvit⏟Noise×exp-uit⏟inefficiency, Uit=Uiexp-ηt-T, Ui≥0∼iidtruncated at zero of the Nμ,σU2distribution, ηis the paramter to be estimated, where in the first equation qit is the number of disposed of cases of the i-th high court; xit is the vector of inputs such as judges, staffs, pending cases and expenditure allocated by i-th court; β stands for the vector of parameters to be estimated; vit∼iidN0,σv2 random errors, independent of the uit; uit≥0 are associated with technical inefficiency. TE of the ith country can be estimated as:TEi=exp-ui. SFA uses the MLE procedure to estimate the technical efficiency of the network readiness across countries. The variance parameter is:σU2=σu2+σv2, λ=σu2σv2. γ=σμ2σμ2+σν2,where0≤γ≤1. This method differs from other parametric techniques like OLS or IV-SLS in the sense that here, the theoretical value of a dependent variable, is treated as “expected maximum value” instead of treating it as average value. Apart from this, SFA residuals include iid random error and non-positive inefficiency terms (incapacitated to reach the frontier) rather than the assumption of zero mean and constant variance in OLS (Beldowski et al. 2020). The benefits of using panel data are the increase in the number of observations and more efficient estimators of the unknown parameters. Panel data come with some other added utilities like allowing the user to relax some strong distributional assumptions, to obtain consistent predictions of technical inefficiencies and estimate changes in technical efficiencies over time (Coelli 2005) Data and variable selection Identification of inputs and output is an integral part of DEA and SFA. The selection of variables is done through two mechanisms (1) by solving mixed-integer linear programming (MILP) primarily employed when there is no heuristic decision-making or expert judgement, proposed by (Peyrache et al. 2020) and (2) previous research on efficiency (Mattson and Tidana 2019). Courts are Labor-Intensive, and all the previous research has incorporated at least one measure of labor but has not considered the input on capital (Mattson and Tidana 2019). Given the context and constraints, to assess the efficiency of courts following factors should have been gauged in financial and Human Resources, diversity, infrastructure, and workload. The data for the present study on High Courts and Subordinate Courts has been extracted from Annual reports of the judiciary for the period 2015–20 on civil and criminal cases. This is the longest period for which the data are available in Annual Reports. Although data on civil and criminal cases pending, judges, judicial staff, and the budget can be extracted from the Annual Report Of Indian Judiciary, for the capital aspect of input in the judiciary, data on completely constructed court halls and residential accommodations are only provided for the year 2018–19 by “Nyaya Vikas” a portal and mobile app for monitoring of projects under the CSS for development of infrastructure facilities for the district and subordinate judiciary (Department of justice Handbook on Revised Guidelines 2018–19, GOI). It is worth mentioning here that the pending cases of a particular year have been derived by summing up the new cases instituted and the pending cases at the beginning of that year. Empirical results High Courts and Subordinate Courts To understand the status quo of 24 High Courts9 and their respective Subordinate Courts, in Appendix Table 3, we report the descriptive statistics in 5 dimensions for the period 2015–2020. From Fig. 2, we can see that on average the number of pending cases and disposed of cases in Subordinate Courts has been higher than in the High Courts which coincides with the observations from Fig. 3 where the number of judicial staff in subordinate courts seems to be higher than in the High Courts. Overall, there has been a surge in the workforce in both High Courts and Subordinate Courts. Fig. 2 Average number of pending and resolved cases in High Courts and Subordinate Courts. Source: Author’s calculation To be specific, judges in HC, other judicial staff in High Courts and the judicial officers in Subordinate Courts increased by 11%, 155 and 16%, respectively. Moreover, there has been a 60% surge in the expenditure allocated to High Courts in India (Refer to Fig. 4). Where on the one hand there has been an increase in the inputs employed in the courts, on the other hand there has been a growth of 18 and 17% in cases pending in the High Courts and Subordinate Courts.Fig. 3 Average Strength of workforce in High Courts and Subordinate Courts Source: Author’s calculation Fig. 4 Expenditure allocated to High Courts. Source: Author’s calculations Source: Author’s calculation Data envelopment analysis In the current study, we have employed output-oriented DEA with the assumption of Variable returns to Scale. The output-oriented model is configured to allow a DMU (in our case Courts) to become efficient by proportionately increasing the output given the inputs. It is expressed as the ratio of maximum output to actual output. In the software DEAP, the output-oriented technical efficiency ranges between 0 and 1 (inverse of an output-oriented theoretical model). A value of 1 means that the DMU is on the production frontier and is the benchmark for all the other DMUs, on the other hand, a value less than 1 indicates that the DMU is technically inefficient. Let us say a DMU’s Technical Efficiency score is 0.78 in output-oriented DEA, this would signify that it is technically inefficient and can become efficient by increasing 22% (1–0.78) of its output, and keeping the inputs fixed. As stated earlier, the non-parametric nature of DEA allows us to incorporate multiple inputs and outputs to obtain the technical efficiency score and rank the DMUs accordingly. For both levels of courts there is one output, namely Cases Disposed of, whereas the number of inputs for High Courts are Judges, Judicial Staff, cases pending and budgetary allocation, the inputs for lower courts are Judicial staff and cases pending. It is a pre-requisite to check for the isotonic relationship between inputs and outputs in DEA which is satisfied in our case (Refer to Appendix Table 4). It is worth mentioning at this juncture that we would be using the technical efficiency scores to rank the High Courts and Subordinate Courts and comment on the potential for change. Though the number of Technically efficient Subordinate Courts and High Courts went up between 2015 and 19, it went down in 2019–20. The results of Output-oriented DEA with VRS assumption are presented in Source: Based on author’s calculations Appendix Table 5 and Source: Based on author’s calculations Appendix Table 6 and the results are summarized in Figs. 5 and 6.Fig. 5 Ranking of High Courts based on the average VRSTE scores Source: Author’s calculation Fig. 6 Ranking of Subordinate Courts based on Average VRSTE. Note: The rankings are based on the average VRSTE scores over 2015–20 Source: Author’s calculation On average, it can be observed that other than Allahabad, Sikkim, and Meghalaya High Courts (Benchmark courts with perfect unity technical efficiency score), all the courts are technically inefficient and are required to increase the cases disposed of by keeping the inputs fixed. The lowest performing High Courts are UT of J&K and Ladakh, Bombay and Calcutta which need to increase the cases disposed of by more than 50% to reach the efficiency level. Similarly, the subordinate courts under the jurisdiction of Tripura, Sikkim and Allahabad High Courts set the benchmark and the remaining courts are technically inefficient. Subordinate Courts under the jurisdiction of Patna, Jharkhand, Orissa, and Meghalaya High Courts have been the poorest performing courts which need urgent policy measures to improve the case disposal rate to reach technical efficiency. DEAP also gives the Scale efficiency scores of the DMUs which is the ratio of output-oriented TE scores based on CRSTE and the scores from VRSTE. Based on the Scale Efficiency Scores of both, High Courts, and Subordinate Courts we can infer that over the years number of High Courts and Subordinate Courts witnessing DRS has shot up. This points toward an alarming situation where the courts are oversized, having exceeded their optimal size. Thus, to reduce their average input consumption, these High Courts must decrease their size. Practically, this could be done either by internal decay (i.e., producing less output) or by splitting the Courts into two separate Courts (Refer to Appendix Table 7). Malmquist productivity index In this section, we would analyze the DEA-based Malmquist productivity index which decomposes the change in productivity into technical efficiency change (catching-up effect) and Technology change (Frontier-shift). MPI primarily shows the intertemporal change in TFP and if it is observed to be greater than 1 then is a sign of improvement in the productivity of a particular DMU and vice versa. In Appendix Table 8 and Appendix Table 9, average decomposed MPI scores have been shown, where the product of EFFCH (technical efficiency change-second column) and TECHCH (technology change in the third column) gives TFP (total factor productivity change-fifth column). It can be observed that on average the productivity of both High Courts and Subordinate Courts has deteriorated by 6 and 9%, respectively, and the reason is the downfall in technical efficiency despite an improvement in technological change. We can infer that the courts in India need to have a proper balance between inputs and outputs. On average, the productivity of 8 High courts and Subordinate Courts under the legislature of 3 High Courts has gone up and remaining all courts have experienced plummeting productivity. The overall productivity of UT of J&K and Ladakh High Court has deteriorated (Refer to Appendix Table 10). The reason stems from the clear imbalance between inputs and output despite technological progress. A constant work in progress is required in the High Courts and the Subordinate Courts to move up from the status quo of deteriorating technical efficiency through transparent financial management, periodical review of the performance and maintenance of the supply of inputs like judges, staff, and budget. Additionally, the study suggests an innovation-centric approach and the introduction of new technology. These courts experience continuous struggles to move up the ladder of efficiency. The fact that the none of Subordinate Courts have experienced productivity gains barring the Subordinate Courts under Tripura High Courts (where the productivity has remained unchanged) underscores the need to change internal managerial conditions (referring to better use of resources by the court). To sum up, we can infer that the reason for the average productivity of courts is the lackluster internal management of resources (inability to approach the frontier) and imply significant policy interventions. After performing some non-parametric tests, we have found a statistically significant difference between the productivity of High Courts established before independence and the ones established after 1947 (Refer to Table 2).Table 2 Analysis of productivity of High Courts based on year of establishment Year of establishment Average Std. dev Max Min Two-sample t test Two-sample Wilcoxon Mann–Whitney rank-sum test (z-value). Kruskal–Wallis test Equality-of-populations rank (chi-square value). Two-sample Kolmogorov–Smirnov test For equality of distribution functions (D-value). Before independence 0.873 0.055 0.94 0.772 − 2.8* − 2.6* 6.7* 0.68* After independence 0.99 0.153 1.505 0.839 Source: Based on author’s calculation Stochastic frontier analysis In Table 3, we report the results of the true fixed-effects model of SFA in which column 2 shows the results when pending cases were considered as independent variables explaining the frontier and in column 3 we have considered the pending cases as an inefficiency-causing variable. In the case of High Courts, we have found a statistically significant positive impact of demand for justice, judges, judicial staff, and budgetary allocation on the cases dissolved. Furthermore, we have also found that there has been a decline in technical efficiency over the years (the negative coefficient of year). In a separate setup when we considered Pending cases as a technical inefficiency-causing variable, it was found to be statistically significant along with the statistically significant positive impact of other explanatory variables in our analysis. This explains that pending cases hinder the technical efficiency of High Courts. When a similar exercise was performed for Subordinate courts, judicial staff was found to be positively (and significantly) impacting the cases dissolved and the pending cases were found to be insignificant in explaining the technical inefficiency of Subordinate Courts.Table 3 Stochastic Frontier analysis with log of dissolved cases as a dependent variable High Courts Subordinate Courts (1) (2) (3) (4) log_pending 0.060* (0.0004) – 0.3814* (0.122) NA log_judges 0.483* (0.00001) 0.757* (0.119) 0.616** (0.285) 1.12* (0.668) log_staff 0.479* (0.00001) 0. 7747* (0.128) NA log_budget 0.020* (0.00002) 0.1761* (0.018) NA NA Year FE − 0.0420* (6.08E-06) − 0.057* (0.016) − 0.08* (0.019) 0.67* (0.067) MU (technical inefficiency) log_pending cases – 0.336* (0. 0934) NA −0.6896 (4.17) Usigma _cons 4.674* (1.47) – − 7.37 2.98 Log-likelihood (SFA)-UR 34.936 − 3.3699 − 7319 − 298.89 Log-likelihood (OLS)-R − 68.045 − 121.251 − 135.15 LR = − 2(LR(H0)-LR(H1)) 205.964* 235.762* 214.750*** − 327.48 All the variables were taken log of Values in parenthesis show standard error p < 0.05, p < 0.01, p < 0.001 Thus, based on SFA, we can infer that though demand for judges, Judges, judicial staff, and budgetary allocation improves case solvency, pending cases also cause technical inefficiency in High Courts. Thus, rising pendency is a matter of great concern for India’s High Courts and from the policy perspective, we emphasize not just judicious utilization of resources but also underscore the importance of infusing more resources (financial capital and manpower). Furthermore, we found no difference in the summary statistics of VRSTE scores of High Courts obtained from DEA and that from SFA (Refer to Appendix Table 11). Panel regression analysis To understand the average impact of demand for justice (pending cases), judges, judicial staff and expenditure with year and the court fixed effects dummy, we have performed panel regression analysis10, the results of which are reported in Table 4.Table 4 Panel LSDV (Robust) results for High Courts and Subordinate Courts Variables High Courts Subordinate Courts (1) (2) (3) (4) (5) (6) log_pending − 0.148 (0.395) 0.668* (0.0711) − 0.166 (0.387) .0739 (0.173) 0.751* (0.123) − 0.0444 (0.149) log_judges 0.580* (0.194) 0.211* (0.0722) 0.582* (0.173) 0.314 (0.405) 0.180 (0.195) 1.042 (0.402) log_staff 0.570* (0.212) 0.398* (0.111) 0.559* (0.188) NA NA NA log_budget − 0.0103 (0.158) − 0.192* (0.0644) 0.179 (0.158) NA NA NA Court FE Yes No Yes Yes No Yes Year FE No Yes Yes No Yes Yes Constant 7.504 (3.662) 3.635* (0.983) 3.610 (5.349) 13.77* (3.562) 1.874* (0.536) 7.896 (3.023) Observations 120 120 120 120 120 120 R-squared 0.980 0.944 0.982 0.980 0.897 0.986 Log of dissolved cases as dependent variable (All the variables were taken log of) Judges are judicial staff in cases of Subordinate Courts; Values in parenthesis show standard error p < 0.05, p < 0.01, p < 0.001. Source: Based on the author’s calculations It can be clearly seen that Judges on average have a significant positive impact on the cases disposed of in both, High Courts, and Subordinate Courts. The results refute the claim of the insensitivity of case disposition to the judicial staff revealed by Beenstock and Haitovsky (2004) and Dimitrova-Grajzl et al. (2012), and are in line with the emphasis placed on the recruitment of additional judges to clear backlogs in Economic Survey (2018–19). Furthermore, we find evidence of a direct positive impact of Cases pending on cases disposed of as was found in the case of Israeli courts by Beenstock (2004) and of Andhra Pradesh by Rabiyath and Rupakula (2010) confirming that the number of cases resolved depend on the demand for justice.The problem is not insurmountable, i.e., the disposal rate could be improved by filling up the vacancies and increasing the number of working hours (Refer to Appendix Table 2 on working hours in India’s High Courts in 2020). Interestingly, when we controlled for year-fixed effects, we found that Expenditure allocated to the judiciary has a negative impact manifesting the two dimensions of inequality.11 This signals a need to examine the budgetary allocation to the courts in India by the Finance Commission. There is a vicious cycle existing here in the sense that the judicial staff are occupied with disposing of cases, hence they devote very limited time to infrastructural deficits. Other aspects could be Red-Tapism and corruption. Mere allocation of funds does not guarantee an outcome, it needs to be at par with productivity. Thus, further research in this field could shed light as to why Budget has a statistically significant negative impact on cases disposed of. Impact of COVID-19 on instituted cases and disposed cases The problem of staggering pendency in Indian Courts was exacerbated by the unprecedented pandemic called COVID-19 which impacted each stratum of society. As India grappled with a health and economic crisis, the nationwide lockdown imposed on 24th March 2020, compelled the High Courts to function virtually. Consequently, the number of pending cases reached an all-time high but, in sheer contrast the number of fresh cases instituted, and cases disposed of plummeted (Refer to Fig. 7). In the Indian context, the prevailing literature is limited on the impact of COVID-19 (Rattan and Rattan 2021; Singh et al. 2022). To the best of our knowledge, the literature related to the pre-COVID-19 crisis period is scant and mostly pertains to the pre-COVID-19 crisis period and covers the government intervention with E-Courts and the process of digitization. Thus, to bring this study into the current context we also looked at the impact of COVID-19 and performed statistical tests the results of which are reported in Table 5.Fig. 7 Impact of COVID-19 on cases instituted and cases disposed of The first dotted vertical bar(gray) shows the first Case of COVID-19 which was found in Kerala and the second bar (Red) indicates the break due to nationwide lockdown on 24th March 2020. Source: Based on the author’s calculations Table 5 Statistical tests to understand the difference in cases instituted of and cases disposed of due to lockdown Variables Two-sample Kolmogorov–Smirnov test Kruskal–Wallis equality-of-populations rank test Two-sample Wilcoxon rank-sum (Mann–Whitney) test Two-sample t test with unequal variances Cases instituted 0.22* 31.95* 5.652* 5.68* Cases disposed of 0.233* 36.530* 6.044* 2.82*** Source: Based on Author’s calculation p < 0.05, p < 0.01, **p < 0.001 Thus, to understand the impact of the Lockdown on the number of fresh cases instituted and cases disposed of we collected monthly data of High Courts between January 2019 and July 2021 (National Judicial Data Grid-NJDG) and performed some non-parametric and classical tests. Based on the results reported in Table 5, we find a statistically significant difference in the number of fresh cases instituted in High Courts and the cases disposed of due to the lockdown. Conclusions and policy implications One of the four pillars of democracy is the judiciary. De facto, a technically efficient judiciary subtly not just emboldens the faith of people in the justice delivery system, but empirical studies have also confirmed its growth effects. The present study provides a methodological alternative to evaluate the performance of the judiciary. Following the attempt of Gupta and Bolia (2020), the current study presents evidence in the context of a developing country like India by juxtaposing jurisprudence with production theory, not frequently examined in the same gust by employing output-oriented DEA, SFA, Malmquist productivity index, and panel regression for both the levels of courts—High Courts and the Subordinate Courts. The current study does not enquire into probing the quality of verdicts but, somewhat it endeavors to examine the relative technical efficiency of courts in India. We revealed how the supply of justice (cases disposed of) is determined by the demand for justice, serving judges and judicial staff and the expenditure allocated in both, HCs and Subordinate Courts. Employing a unique dataset spanning over 2015–19 facilitated us to infer that the High Courts and Subordinate Courts are technically inefficient, and their productivity has declined over the analysis period. Furthermore, we were able to infer that on average the High Courts established post-independence fared better in technical efficiency than the ones established before independence. From the Malmquist productivity index, we were able to suggest that a constant work in progress is required in the High Courts and the Subordinate Courts to move up from the status quo of deteriorating technical efficiency through transparent financial management, periodical review of the performance and maintenance of the supply of inputs like judges, staff, and budget. We have also found that the issue of pendency is not insurmountable, i.e., the disposal rate could be improved by filling up the vacancies and increasing the number of working hours. However, the evidence does not suggest a “one-size-fits-all” approach to bring reform in the Indian Judicial system. Each court should examine and practice different ways to improve efficiency. In SFA we have also shown that pendency leads to technical inefficiency in High Courts. Interestingly, when we controlled for year-fixed effects, we found that Expenditure allocated to the judiciary has a negative impact manifesting the two dimensions of inequality in fiscal federalism. Analyzing the monthly database, we found a statistically significant difference in the number of fresh cases instituted and the cases disposed of in High Courts due to the nationwide lockdown in India. From the point of view of behavioral economics, the performance of the civil and criminal justice system has effects on human beings through a cultural spillover. Our study in the context of India’s judicial system by focusing on two levels—High Courts and Subordinate Courts could be a learning experience for lower-middle-income countries where the judiciary enjoys an important and powerful position and for the countries ranking low in the Rule of Law indices (World Justice Project). Our estimates confirm the role of judges, judicial staff, and pendency in the cases solved. Policy implications from the current study revolve around the increase in the number of judges and judicial staff, increasing the number of working days, and establishment of a separate professional body for the judicial administration (being prevalent in countries such as the UK, USA, and Canada) and technology deployment among others. These recommendations are pervasive for the global, South Asia regional peers, Lower Middle-income peers and poor countries that have fared low in the two sub-indices of the Rule of Law such as criminal and civil justice. The study is also not devoid of any limitations. First, data limitation on environmental factors like corruption, the budget allocated to Subordinate Courts, court halls, representation of women, technological factors, and Police which ideally could have been included in the analysis. Second, it is important to also analyze the efficiency of fast-track courts and Lok Adalat in India and construct a model that incorporates a simultaneously operating intertemporal model. Third, from the methodological point of view Bayesian approach, Network DEA and the application of Artificial Neural Networks could be what future research could focus on. Fourthly, from the global perspective, the availability of panel data on all the inputs and outputs could give a better picture. Fifthly, as stated the current study does not enquire into probing the quality of verdicts which could be a potential topic for future studies. Moreover, it would be interesting to look at the impact of COVID-19 on the Indian judiciary, which was compelled to work digitally for a longer period. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 49 KB) Acknowledgements I am grateful to Prof. Somesh Mathur of IIT-Kanpur for his valuable comments in the study. Author contributions This is a study conducted by a single author. Funding No funding was received for the completion of the study Data availability The data are available and could be provided when required. Declarations Conflict of interest There is no conflict of interest Ethical approval Ethics approval was not required for this study. Consent for publication I give consent for the publication of the manuscript. 1 Why Nations Fail: The Origins of Power, Prosperity, and Poverty. 2 Cyclic syndrome means repeated instances of backlogs. 3 In Hussainara Khatoon v. Home Secretary, State of Bihar, Patna (AIR 1979 SC 1369) the Supreme Court of India established that speedy trial is an indispensable element of Article 21. 4 According to the World Justice Report, India ranks 98 in “timely and effective criminal adjudication” and 88 in whether “ alternative dispute resolution mechanisms are accessible, impartial, and effective” out of 128 countries in 2020. 5 SOL2.pdf (https://www.niti.gov.in) accessed on 26th November 2021. 6 Posner, The Federal Courts, supra note 5, at 294. 7 For in-depth description of BCC output-oriented model in DEA, see also ‘Data envelopment analysis: A comprehensive text with models, applications, references and DEA-Solver software’ (2nd edition) by William W. Cooper, Lawrence M. Seiford and Kaoru Tone. 8 DEAP is software which is developed by CEPA to conduct DEA. It is a free software that can be downloaded from the University of Queensland’s home page. 9 Telangana High court has not included in the study. 10 SFA is different from OLS in the sense that in SFA models the theoretical value of the dependent variable is not its expected mean value but its expected maximum value (productivity frontier). Additionally, in SFA the residual term (i.e., the difference between the theoretical and the estimated values of the dependent variable) are composed of non-positive technical inefficiency (a productive unit’s incapability to reach its frontier output) and i.i.d. random error components; whereas, in OLS residuals is the random error component. 11 The horizontal imbalance between the States and the vertical inequality that exists within States and the Union. ==== Refs References Agrell PJ Mattsson P Månsson J Impacts on efficiency of merging the Swedish district courts Ann Oper Res 2020 288 2 653 679 10.1007/s10479-019-03304-0 Aigner D Lovell CK Schmidt P Formulation and estimation of stochastic frontier production function models J Econom 1977 6 1 21 37 10.1016/0304-4076(77)90052-5 Akande A Cabral P Gomes P Casteleyn S The Lisbon ranking for smart sustainable cities in Europe Sustain Cities Soc 2019 44 475 487 10.1016/j.scs.2018.10.009 Antonucci L Crocetta C d’Ovidio FD Evaluation of Italian judicial system Procedia Econom Finance 2014 17 121 130 10.1016/S2212-5671(14)00886-7 Ash E, Asher S, Bhowmick A, Chen DL, Devi T, Goessmann C, Siddiqi B (2021) Measuring gender and religious bias in the indian judiciary. Center for Law & Economics Working Paper Series, 2021(03) Beenstock M Haitovsky Y Does the appointment of judges increase the output of the judiciary? Int Rev Law Econ 2004 24 3 351 369 10.1016/j.irle.2004.10.006 Bełdowski J Dąbroś Ł Wojciechowski W Judges and court performance: a case study of district commercial courts in Poland Eur J Law Econ 2020 50 1 171 201 10.1007/s10657-020-09656-4 Berbegal-Mirabent J Lafuente E Solé F The pursuit of knowledge transfer activities: an efficiency analysis of Spanish universities J Bus Res 2013 66 10 2051 2059 10.1016/j.jbusres.2013.02.031 Botero JC La Porta R López-de-Silanes F Shleifer A Volokh A Judicial reform World Bank Res Obs 2003 18 1 61 88 10.1093/wbro/lkg005 Bubb R The evolution of property rights: state law or informal norms? J Law Econ 2013 56 3 555 594 10.1086/673208 Buscaglia E Ulen T A quantitative assessment of the efficiency of the judicial sector in Latin America Int Rev Law Econ 1997 17 2 275 291 10.1016/S0144-8188(97)00007-0 Castro MF Guccio C Measuring potential efficiency gains from mergers of Italian first instance courts through nonparametric model Public Finance Rev 2018 46 1 83 116 10.1177/1091142116652723 Chakraborty P Judicial quality and regional firm performance: the case of Indian states J Comp Econ 2016 44 4 902 918 10.1016/j.jce.2016.07.001 Chemin M The impact of the judiciary on entrepreneurship: evaluation of Pakistan’s “Access to Justice Programme” J Public Econ 2009 93 1–2 114 125 10.1016/j.jpubeco.2008.05.005 Coelli TJ, Prasada Rao DS, O’Donnell CJ, Battese GE (2005) Additional topics on data envelopment analysis. An introduction to efficiency and productivity analysis, 183–208 Cooper WW Seiford LM Tone K Data envelopment analysis: a comprehensive text with models applications, references and DEA-solver software 2007 New York Springer 489 Cooter RD (1983) The objectives of private and public judges. Public Choice, 107–132 Dam KW (2006) The judiciary and economic development. U Chicago Law & Economics, Olin Working Paper, (287) Deyneli F Analysis of relationship between efficiency of justice services and salaries of judges with two-stage DEA method Eur J Law Econ 2012 34 3 477 493 10.1007/s10657-011-9258-3 Dimitrova-Grajzl V Grajzl P Sustersic J Zajc K Court output, judicial staffing, and the demand for court services: evidence from slovenian courts of first instance Int Rev Law Econ 2012 32 1 19 29 10.1016/j.irle.2011.12.006 Djankov S La Porta R Lopez-de-Silanes F Shleifer A Courts Q J Econ 2003 118 2 453 517 10.1162/003355303321675437 Economic Survey of India (2019) Ending matsyanyaya: how to rampup capacity in the lower judiciary. https://www.indiabudget.gov.in/budget2019-20/economicsurvey/doc/vol1chapter/echap05_vol1.pdf (Accessed 28 Oct 2022) Espinosa R Desrieux C Wan H Fewer courts, less justice? Evidence from the 2008 French reform of labor courts Eur J Law Econ 2017 43 2 195 237 10.1007/s10657-015-9507-y Falavigna G Ippoliti R The impact of institutional performance on payment dynamics: evidence from the Italian manufacturing industry J Bus Econ Manag 2020 21 5 1285 1306 10.3846/jbem.2020.13195 Falavigna G Ippoliti R Manello A Ramello GB Judicial productivity, delay and efficiency: a directional distance function (DDF) approach Eur J Oper Res 2015 240 2 592 601 10.1016/j.ejor.2014.07.014 Falavigna G, Ippoliti R, Ramello GB (2018) DEA-based Malmquist productivity indexes for understanding courts Färe R Grosskopf S Lindgren B Roos P Productivity developments in Swedish hospitals: a Malmquist output index approach Data envelopment analysis: theory, methodology, and applications 1994 Dordrecht Springer 253 272 Färe R Grosskopf S Roos P Malmquist productivity indexes: a survey of theory and practice Index numbers: essays in honour of Sten Malmquist 1998 Dordrecht Springer 127 190 Fare R Färe R Fèare R Grosskopf S Lovell CK Production frontiers 1994 Cambridge University Press Feld LP, Voigt S (2006) Judicial independence and economic growth: some proposals regarding the judiciary. RD Congleton and Ghosh Y Indian judiciary: an analysis of the cyclic syndrome of delay, arrears and pendency Asian J Legal Educ 2018 5 1 21 39 10.1177/2322005817733566 Gupta M Bolia NB Efficiency measurement of Indian high courts using DEA: a policy perspective J Policy Modeling 2020 42 6 1372 1393 10.1016/j.jpolmod.2020.06.002 Hodgson J (2006) Conceptions of the trial in inquisitorial and adversarial procedure. In: Duff A, Farmer L, Marshall S, Tadros V (eds). Judgment and calling to account. pp 223–42 Kittelsen SA Førsund FR Efficiency analysis of Norwegian district courts J Prod Anal 1992 3 3 277 306 10.1007/BF00158357 Lewin AY Morey RC Cook TJ Evaluating the administrative efficiency of courts Omega 1982 10 4 401 411 10.1016/0305-0483(82)90019-6 Marciano A Melcarne A Ramello GB The economic importance of judicial institutions, their performance and the proper way to measure them J Inst Econ 2019 15 1 81 98 Mattsson P Tidanå C Potential efficiency effects of merging the Swedish district courts Socioecon Plann Sci 2019 67 58 68 10.1016/j.seps.2018.09.002 Meeusen W, van Den Broeck J (1977) Efficiency estimation from Cobb-Douglas production functions with composed error. Int Econ Rev, 435–444 Melcarne A Ramello GB Is justice delayed justice denied? An empirical approach Int Rev Law Econ 2021 65 105953 10.1016/j.irle.2020.105953 North DC (2008) Entitled, institutions, institutional change and economic performance (Cambridge: Cam-bridge University Press, 1990). The Limits of Rationality, 383 Peyrache A Rose C Sicilia G Variable selection in data envelopment analysis Eur J Oper Res 2020 282 2 644 659 10.1016/j.ejor.2019.09.028 PIB (2019) Economic Survey recommends Investment in Well-Functioning Legal System, given its Potential Economic and Social Multiplier Effect. https://pib.gov.in/Pressreleaseshare.aspx?PRID=1577016 (Accessed 28 Oct 2022) Posner RA Wealth maximization revisited Notre Dame JL Ethics & Pub Poly 1985 2 85 Posner RA On theory and practice: reply to richard Posner’s praxis Ohio St LJ 1988 49 1077 Posner RA What do judges and justices maximize? (The same thing everybody else does) Supreme Court Econ Rev 1993 3 1 41 10.1086/scer.3.1147064 Rabiyath S Rupakula RV Babu PG Thomas E Raja AV Hans-Bernd S Somashekar TS Disposal rates, pendency, and filing in Indian courts: an empirical study of the two states of Andhra Pradesh and Kerala Economic analysis of law in india: theory and application 2010 New Delhi Oxford University Press Rattan J Rattan V The COVID-19 crisis-the new challenges before the indian justice and court administration system IJCA 2021 12 1 10.36745/ijca.391 Robinson JA Acemoglu D Why nations fail: the origins of power, prosperity and poverty 2012 London Profile 45 47 Rosales-López V Economics of court performance: an empirical analysis Eur J Law Econ 2008 25 3 231 251 10.1007/s10657-008-9047-9 Safavian M Sharma S When do creditor rights work? J Comp Econ 2007 35 3 484 508 10.1016/j.jce.2007.05.004 Schneider F Shadow economies around the world: what do we really know? Eur J Polit Econ 2005 21 3 598 642 10.1016/j.ejpoleco.2004.10.002 Shapiro I Richard Posner’s praxis Ohio St LJ 1987 48 999 Shavell S (1999) The law and economics of judicial systems Singh M (2022) Challenges before the indian justice and court administration system in the COVID-19 Crisis. Resilience Transform Global Restruct, 12 Staats JL Bowler S Hiskey JT Measuring judicial performance in Latin America Latin American Politics Soc 2005 47 4 77 106 10.1111/j.1548-2456.2005.tb00329.x Stigler GJ Law or economics? J Law Econ 1992 35 2 455 468 10.1086/467262 Tone K Malmquist productivity index Handbook on data envelopment analysis 2004 Boston, MA Springer 203 227 Visaria S Legal reform and loan repayment: the microeconomic impact of debt recovery tribunals in India Am Econ J Appl Econ 2009 1 3 59 81 10.1257/app.1.3.59 Voigt S Determinants of judicial efficiency: a survey Eur J Law Econ 2016 42 2 183 208 10.1007/s10657-016-9531-6 Voigt S Gutmann J Turning cheap talk into economic growth: on the relationship between property rights and judicial independence J Comp Econ 2013 41 1 66 73 10.1016/j.jce.2013.01.007 Wang C Li Z Zhong T Social trust, rule of law, and economic exchange: evidence from China and its major trading partners Emerg Mark Financ Trade 2019 55 14 3134 3150 10.1080/1540496X.2019.1572505 Williamson CR Informal institutions rule: institutional arrangements and economic performance Public Choice 2009 139 3 371 387 10.1007/s11127-009-9399-x Williamson CR Kerekes CB Securing private property: formal versus informal institutions J Law Econ 2011 54 3 537 572 10.1086/658493 World Justice Project (2020). https://worldjusticeproject.org/rule-of-law-index/country/2020/India/Criminal%20Justice/ (Accessed 29 Oct 2022) Yeung LL Azevedo PF Measuring efficiency of Brazilian courts with data envelopment analysis (DEA) IMA J Manag Math 2011 22 4 343 356 10.1093/imaman/dpr002 Zaheer A McEvily B Perrone V Does trust matter? Exploring the effects of interorganizational and interpersonal trust on performance Organ Sci 1998 9 2 141 159 10.1287/orsc.9.2.141	0	PMC9734438	NO-CC CODE	2022-12-14 23:28:28	no	SN Bus Econ. 2023 Dec 9; 3(1):6	utf-8	SN Bus Econ	2,022	10.1007/s43546-022-00377-1	oa_other
==== Front Soc Netw Anal Min Soc Netw Anal Min Social Network Analysis and Mining 1869-5450 1869-5469 Springer Vienna Vienna 1005 10.1007/s13278-022-01005-4 Original Article Intelligent lead-based bidirectional long short term memory for COVID-19 sentiment analysis Kumari Santoshi [email protected] Pushphavathi T. P. [email protected] grid.464941.a Computer Science and Engineering, M S Ramaiah University of Applied Sciences, No. 470P, 4th Phase, Peenya Industrial Area, Bangalore, 560058 India 6 12 2022 2023 13 1 114 9 2022 12 11 2022 19 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. Social media is an online platform with millions of users and is utilized to spread news, information, world events, discuss ideas, etc. During the COVID-19 pandemic, information and ideas are shared by users both officially and by citizens. Here, the detection of useful content from social media is a challenging task. Hence, natural language processing (NLP) and deep learning are widely utilized for the analysis of the emotions of people during the COVID-19 pandemic. Hence, this research introduces a deep learning mechanism for identifying the sentiment of the people by considering the online Twitter data regarding COVID-19. The intelligent lead-based BiLSTM is utilized to analyze people's sentiments. Here, the loss of the classifier while learning the data is eliminated through the incorporation of the intelligent lead optimization. Hence, the loss is reduced, and a more accurate analysis is obtained. The intelligent lead optimization is devised by considering the role of the informer in identifying the enemy base to safeguard the territory from attack along with the Monarch's knowledge. The performance of the intelligent lead-based BiLSTM for the sentiment analysis is assessed using the metrics like accuracy, sensitivity, and specificity and obtained the values of 96.11, 99.22, and 95.35%, respectively, which are 14.24, 10.45, and 26.57% enhanced performance compared to the baseline KNN technique. Keywords Sentiment analysis Deep learning COVID-19 Optimization LSTM issue-copyright-statement© Springer-Verlag GmbH Austria, part of Springer Nature 2023 ==== Body pmcIntroduction The World Health Organization (WHO) proclaimed COVID-19 as a pandemic disease by considering the severity and the outrageous spreading that lost people's lives along with several diseases like respiratory failure, pneumonia, and various dangerous diseases. The coronavirus with severe respiratory syndrome is characterized as coronavirus 2 (SARS-CoV-2). As no appropriate treatment is available for COVID-19, several preventive measures like self-isolation, social distancing, hand hygiene, sanitizing, and several other public health practices are followed. The pandemic outbreak affects almost all nations that severely affect the economy globally. Therefore, the WHO devises several restrictions as corona affects severe damage to lives; hence, the Public Emergency of International Concern (PEIC) is declared. The restrictions lead to the closure of several businesses and people traveling. Thus, people started to share their views on social media regarding the happenings in the world because the pandemic makes social media more noticeable than before. Hence, the impact of the pandemic resulted in social media platforms as a big global data center, and people started to spend more time on these platforms and their applications. The people's reactions and conditions during the corona pandemic are analyzed using Twitter's social media platform. The people's emotions analysis by considering the products, disaster management, crime, election, stock market, and several other criteria are employed using Twitter data. Thus, the people's changes and reactions to COVID-19 are analyzed using Twitter data by evaluating the sentiment (Choudrie et al. 2021). The detection of people's sentiments by considering the Twitter data under three categories, neutral, positive, and negative, is termed sentiment analysis. Sentiment analysis is widely utilized in several application areas like health informatics, recommendation systems, opinion mining, etc. For example, while considering the business, profit enhancement is employed by providing the best services through strategic planning, which is offered by detecting consumer opinions through various sentiments. Likewise, the health status of the person, psychology, and behaviors are analyzed through the Tweets. In addition, the information acquisition in the text format is employed in several other ways, like the posts on Facebook social media for analyzing the sentiment. Here, the analysis of the sentiment from the textual data is employed through the natural language processing technique, in which the features are extracted through subjective knowledge. The information obtained from electronic devices is either consumer-generated or user-generated content for sentiment analysis. The view of each person varies for the information obtained from Twitter and other social media, hence the consolidation through evaluation and analysis for expanding the perceptions (Saifullah et al. 2021). Thus, the claims concerning the consumer's feelings regarding the products, negative or positive, are obtained by analyzing the specific subject based on evidence. However, analyzing the sentiment from the online data by considering the keyword is still challenging. Thus, the machine learning technique is utilized for extracting the sentiment from the online data for evaluating the different text analyses (Elgeldawi et al. 2021). NLP and deep learning are widely utilized for the analysis of the emotions of people during the COVID-19 pandemic (Kaur et al. 2021; Kumari et al. 2021). Text-based sentiment analysis is employed through emotion detection and recognition in the artificial intelligence (AI) arena (Chole and Gadicha 2022). The goal behind the sentiment analysis is to detect the person's feelings from the textual data through the detection and recognition stages for evaluating expressions like surprise, sadness, happiness, fear, disgust, and anger. The consumer's perception or happiness of purchasing the product is evaluated through the detection of emotion for business decision-making. The techniques like XG-boost, Random Forest, Support Vector Classifier, Decision Tree Classifier, Bernoulli, and K-Nearest Neighbor (K-NN) are some of the machine learning techniques (Chole and Gadicha 2020) utilized for emotion recognition. The significant attributes are fed as input to the classifier for processing the text data obtained online, in which the deep learning technique learns the features automatically by adjusting its parameters. The non-self-adjustable parameters are termed hyper-parameter and are adjusted through optimization-based tuning to obtain the desired output. Thus, the hyper-parameter decides the choice of the sentiment analysis output based on deep learning. The research aims to devise a novel sentiment analysis by considering the online Twitter data regarding COVID-19 for analysis of people's sentiments. Here, the optimized deep learning is designed for the sentiment analysis of the COVID-19 data is proposed. Furthermore, the novel intelligent lead optimization is designed for tuning the bidirectional long short-term memory (BiLSTM), in which the tunable hyper-parameters are tuned using the intelligent lead algorithm. Initially, from the online COVID-19 data, by using the essential keyword, the COVID-19 data are acquired from the input data for the reduction in computational complexity. Then, by using the distance-based similarity measure, the consumers with similar features are grouped using the graph-based strategy. Finally, using the BiLSTM, the sentiments like positive, negative, and neutral are employed. The contributions are:Proposed Intelligent lead optimization: The intelligent lead optimization is designed by considering the role of the informer in identifying the enemy base to safeguard the territory from attack along with the Monarch's knowledge. Thus, the proposed intelligent lead optimization based on the hybrid behavior improves the convergence rate. Proposed Intelligent lead-based BiLSTM: The intelligent lead-based BiLSTM is utilized for analyzing people's sentiments during COVID-19 by considering the online Twitter data. Here, the loss of the classifier while learning the data is eliminated through the incorporation of the intelligent lead optimization; hence, the loss is reduced, and a more accurate analysis is obtained. The organization of the intelligent lead-based sentiment analysis is organized as Sect. 2 details the conventional COVID-19-based sentiment analysis methods, and the detailed methodology is elucidated in Sect. 3. Finally, the analysis is presented in Sect. 4, and Sect. 5 concludes the work. Motivation The inaccurate identification of the sentiment and the complexities in finding the best solution are the most challenging issues in traditional sentimental analysis. Thus, this research introduces a novel technique to fulfill the challenges by reviewing the related works by the existing researchers, which are detailed in this section. Related works The sentimental analysis (SA) concerning the COVID-19 situation is detailed in this section and highlighted in Table 1. The SVM-based SA with the Twitter data was presented by Basiri et al. (2021) using the R programming technique, in which the keyword was utilized for identifying the sentiment based on the feature selection. Then, the score was assigned for the classified sentiment to analyze the polarity. The better performance was evaluated based on accuracy and obtained enhanced performance. The analysis depicts the positive sentiment sometimes and the negative sentiment daily basis. The SVM-based SA devised by Sontayasara et al. (2021) using the shuffle split cross-validation with the data processing acquired an accuracy of 71% using the linear kernel, which is higher than the sigmoid kernel. However, they failed to analyze the performance using large data. The NLP-based SA using Twitter data was suggested by Garcia and Berton (2021) and Hitesh et al. (2019) through sentence embedding and word embedding. Here, the summarization of the text was employed from the extracted features based on the topic modeling and obtained the negative sentiment during the pandemic's peak. This method failed to consider the keyword to retrieve the content which leads to the loss of some information, which is considered a drawback of the method. The KNN-based SA concerning the vaccination of COVID-19 was devised by Shamrat et al. (2021) for identifying three different sentiments using Twitter data (Chowdhury et al. 2021). In this, the data tokenization, normalization, and lemmatization were performed to remove noise, and the polarity was identified for the sentiment analysis using the KNN. The method analyzed the sentiment for three vaccines based on negative, neutral, and positive sentiments. The deep learning-based SA using the CNN was presented by Sitaula et al. (2021), in which the extraction of the informative attributes is employed using domain-specific, cross-domain, and fastText data, and the sentiment was analyzed separately using three various CNN. Finally, the decision taken by the three methods was combined to find the sentiment. The method obtained poor accuracy due to the failure to consider the essential attributes. SA based on deep learning was presented by Mishra et al. (2021) using long short-term memory (LSTM) by considering the rule-based sentiment estimation, in which the dictionary was utilized for estimating sentiment. In addition, the topic modeling helped to uncover the required topic from the corpus. It was evaluated based on accuracy, which helps to identify citizens' sentiments concerning tourism during COVID-19. However, while modeling the topic, the method failed to consider significant aspects like economy, transportation, and other factors to obtain a more accurate analysis. The SA with the topic modeling using the Latent Dirichlet Allocation (LDA) was suggested by Abdulaziz et al. (2021) and Monish et al. (2018) for obtaining the informative features by considering the large-scale data. Here, the SA was employed using the lexicon-based strategy, in which the polarity was evaluated by clustering the words into positive and negative groups. Then, the score was generated to identify the sentiment. Unfortunately, the method failed to evaluate the performance. A hybrid deep learning technique devised by combining five different deep learning techniques was presented by Basiri et al. (2021) using the Tweets obtained from social media. In this, the opinions of the people were identified using the metal learner, in which the decision regarding the sentiments for the tweets was fused for obtaining the stacked generalization capability, and the complexity was reduced using the iterative gradient boosting technique. As a result, the method obtained positive sentiment while the recovery rate is higher and evaluated higher negative sentiment during the high death charts that depict the method's accuracy. The method failure's drawback is selecting the keyword for analyzing the sentiment by considering individual countries, which may enhance the analysis accuracy further. Table 1 Related works S. no Author Method Advantage Disadvantage 1 Basiri et al. (2021) A hybrid deep learning technique Complexity was reduced Failed in selecting the keyword 2 Sontayasara et al. (2021) The SVM-based SA High accuracy in the data processing Failed to analyze the performance using large data 3 Garcia and Berton et al. (2021) NLP-based SA using the Twitter data Enhanced the performance Loss of some information 4 Shamrat et al. (2021) KNN-based SA Time efficient The prediction stage might be slow 5 Chowdhury et al. (2021) Data tokenization, normalization, and lemmatization are based on KNN Versatile to different calculations of proximity Inefficient and it's difficult to pick the correct value 6 Sitaula et al. (2021) Deep learning-based SA using the CNN Very High accuracy Failed to consider the essential attributes 7 Mishra et al. (2021) SA based on deep learning Helped to uncover the required topic from the corpus Failed to consider the significant aspects like economy, transportation 8 Abdulaziz et al. (2021) The topic modeling using the Latent Dirichlet Allocation (LDA) Minimized loss function Failed to evaluate the performance Challenges The challenges faced by the existing methods of sentimental analysis by considering the COVID-19 data are:The method devised by Garcia and Berton (2021) failed to consider the keyword to retrieve the content, which led to the loss of some information, which is considered a drawback of the method (Basiri et al. 2021). The poor performance was obtained by the method (Sitaula et al. 2021) due to the failure to consider the essential attributes that assist in elevating the accuracy of the method. The method (Sontayasara et al. 2021) failed to analyze the performance using large data. The insufficiency of labeled data and the inefficiency in evaluating complex sentences degrades the system's performance (Abdulaziz et al. 2021). Identifying the sentiment from the sentences with both positive and negative sentiment and detect as neutral is challenging (Abdulaziz et al. 2021). The higher computational complexity is due to the failure to extract the essential attributes before the sentimental analysis. Proposed methodology for COVID-19 sentiment analysis The prediction and the analysis of people's thinking are obtained using the social media platform as the issues concerning personal concerns, societal issues, and thoughts are shared by citizens. Thus, the prediction of the sentiments concerning COVID-19 among the people is analyzed using the Twitter social media data in the sentiment analysis method. Let the input data concerning the COVID-19 information be obtained from the Kaggle database to evaluate the sentimental analysis. The sentences on the COVID-19 information are obtained by searching the sentences with the significant keywords corresponding to COVID-19. Then, the graph-based sentence tracking is performed based on the distance measures with Euclidian and cosine similarity measures. Finally, using the tracked information, the sentiment analysis is performed using the BiLSTM, which is tuned using the intelligent lead algorithm. Here, the informer and the Monarch's role in safeguarding the territory by identifying the enemy base is considered for designing the Intelligent Lead algorithm. Figure 1 depicts the block diagram of the intelligent Lead based BiLSTM for the sentiment analysis of COVID-19 data.Fig. 1 Block diagram of intelligent lead-based BiLSTM for sentiment analysis using COVID-19 data Data acquisition The input data for the sentiment analysis are acquired from the Kaggle COVID-19 database for the sentiment analysis (COVID-192022; SemEval 2022)). Let, the database be denoted as C with data samples corresponding to the sentiments of COVID-19 and expressed as,1 C={Ca};(1≤a≤n) where the total sample in the database is notated as n, and ath sample in the database is notated as Ca. Keyword searching from the tweets The keywords are searched from the input text regarding the COVID-19 sentiment for selecting the sentences associated with COVID. Thus, the COVID-19-based sentiment analysis using the Intelligent Lead Optimization-based BiLSTM searches for the keywords in the data corpus as provided in Table 2.Table 2 Keywords utilized for searching the COVID-19-based Tweets from Kaggle Domain Keywords Treatments Hydroxychloroquine, plasma, treatment, trial, drug, hospital, patient, and vaccine Entertainment Show, love, video, like, and watch Politics Minister, state, Fauci, president, response, government, and Trump Proliferation care Reopen, spread, social, test, school, people, wear, stay at home, and mask Reports of case Confirm, update, record, patient, total, people, number, die, report, death, and case Economic aspects Pay, crisis, business, impact, and work Online events Host, tomorrow, virtual, impact, discuss, talk, live, webinar, and join Anti-racism War, right, die, death, American, black, kill, police, and protest Sports Sport, play, game, league, team, football, season, and player Charity Million, food, community, provide, donate, relief, fund, help, and support Thus, from the input data, the Tweets with essential attributes presented in Table 2 are selected for identifying the inter-word relationship for tracking COVID-19. COVID tracking using graph-based sentiment tracking COVID-19 is tracked by evaluating the sentiment tracking between the words based on the graph-based strategy. Instead of taking into consideration all the words, using graphs for keyword extraction offers a superior situation for finding the main group of terms. The Vector Space Model (VSM) and the Graph-based model are two well-liked methods for extracting keywords from the text. The given dataset is turned into a network for keyword extraction by treating each word as a node. Two words are connected by an edge to convey a connection if they co-occur in the dataset. Numerous centrality metrics may be used on graphs to identify the key phrases that should be utilized as the document's keywords. The following is a discussion of two of the common centrality measurements used in graphs cosine similarity and euclidean distance. Here, the distance between the words is estimated for the tracking of the sentiment based on the graph. The distances like the cosine similarity and the Euclidean distance are utilized for the evaluation of the distances and to form the graph for tracking COVID-19. Figure 2 depicts the graph representation for sentiment tracking based on distance measures. Based on the distance measures, the sentences with three different sentiments positive, negative, and neutral, are tracked (Wang et al. 2020).Fig. 2 Graph representation for tracking the sentiment Euclidean distance (Dis1) The distance between the two points is evaluated using the Euclidean measure, in which the line segment between the two points is obtained for tracking the sentiments. The Euclidean distance measure between the sentences is evaluated for tracking the sentences with the sentiments of COVID-19 and is expressed as,2 Dis1(u,v)=∑x=1n(ux-vx)1/2 where the two different sentences are referred to as u and v, and the Euclidean distance is notated as Dis1. Here, the total sentence is notated as n. Cosine similarity (Dis2) The cosine similarity between the sentences is measured between the two different sentences for tracking the sentiment regarding COVID-19 by considering the angle between the two vectors and is expressed as,3 Dis2(u,v)=u+vu+v where the cosine similarity is notated as Dis2. Thus, based on the two different distance measures, the similarity between the COVID-19 sentiments sentences are identified based on the graphical representation. Sentiment classification using intelligent lead optimization-based BiLSTM The sentiment classification by considering COVID-19 using the intelligent Lead based BiLSTM (Li et al. 2021) here the classifier is trained using the intelligent lead algorithm to reduce the loss that occurs while training the classifier for making generalizations while testing the unknown data to enhance the accuracy of the analysis. Computation efficiency and reliability can be impacted by a model’s design and choice of configuration parameters. With default parameter settings, the majority of models operate precisely and effectively, but some models perform better when the solver parameters are changed. When giving the keywords, knowledge of a model’s behavior can help you execute simulations more effectively. The optimization algorithm improves the proposed model's comprehension and updates the model's parameters to increase efficiency and precision. In order to retrieve only the data pertinent to the performance, it is necessary to filter out particular keywords from the data. By measuring the distance cosine similarity and euclidean distance, the text similarity-based method is used to visualize opinions in order to achieve that. The sentiments polarities like the neutral, positive and negative are evaluated using the intelligent Lead-based BiLSTM by considering the keywords. A detailed explanation is provided in the following subsections. System model of BiLSTM for sentiment classification The sentiment analysis using the recurrent neural network (RNN_-based technique provided enhanced results while considering the continuous data, but the vanishing gradient issue still exists that fails to update the weights, and hence the learning process can't be able to complete, which makes the performance degradation in the sentiment analysis process. Thus, for solving the issue regarding the vanishing gradient, long-term dependencies among the data are employed with the inclusion of a memory unit and are named as long short-term memory (LSTM). Besides, the time-series data classification for classifying the sentiment is obtained efficiently using the LSTM. In addition, the inclusion of the bidirectional layers named forward layer and backward layer in the LSTM constitutes the bidirectional LSTM (Bi-LSTM) for considering both the future and past context because the LSTM propagates the data from back to front. Still, it fails to process the data from front to back, which limits the learning capability. Thus, the BiLSTM is utilized for classifying the sentiment in the sentiment analysis technique. The deep learning methods comprise several layers like input, hidden, output, softmax, and activation function, in which the loss associated with the training limits the performance. Thus, the optimization is incorporated for tuning the weights and bias of the classifier optimally to reduce the loss while training the classifier. Here, the need for training is to learn the classifier with appropriate data to generalize to predict the sentiment more efficiently while providing the unknown test data. In this condition, when the loss occurs while training the classifier, the system's performance gets degraded. So to minimize the loss in training and to speed up the learning process optimization algorithm is utilized. In the sentiment classification, Intelligent Lead optimization is utilized for training the BiLSTM. The system model of the BiLSTM is depicted in Fig. 3.Fig. 3 System model of BiLSTM Let Ri,PiandKi be the output, input, and forget gates of the BiLSTM utilized for the sentiment analysis of COVID-19, which has modifiable parameters like N and W that depict the weight and bias. The multiplication process that is performed element-wise is notated as ∗ and the point-based multiplication is represented as ∙, respectively. The input gate processes the input and provides the output by considering the previous state information (ei-1) and the current input (Ri), which is represented as,4 Ri=sigmoidNR·ei-1,Yi+WR where the weight is notated as Y, and the bias is notated as W. The role of the forget gate is to discard or retain the information by considering the (ei-1) and present input and its output is represented as5 Pi=sigmoidNP·ei-1,Yi+WP The output of the memory cell utilizes the input gate and the forget gate's output and processes the output as Mi,6 Mi¯=tanhNM·ei-1,Yi+WM 7 Mi=Pi∗mi-1+Ri∗mi¯ The controlling function of the memory cell is employed by the output layer and hence the control function is represented as,8 Ki=sigmoidNK·ei-1,Yi+WK Finally, the sentiment analysis by considering the current input is presented as,9 ei=Ki∗tanhMi The sentiment analysis by considering the two additional layers, namely the forward and the backward layer constitute the BiLSTM, in which (ei←) and (ei→) refers to the backward and forward layer of the BiLSTM and are represented as,10 ei→=LSTMYi,ei-1→ 11 ei←=LSTMYi,ei-1← 12 Gi=aiei→+biei←+Wi The sentiment analysis by considering the COVID-19 data in the forward and the backward layer encoding provides the enhanced solution compared to the standard LSTM. The tunable weights and bias YandW are tuned using the Intelligent lead optimization algorithm. Proposed intelligent lead optimization algorithm In ancient days, the role of an informer is crucial in locating the enemy base and is updated to the king for safeguarding the country. Thus, the role of an informer is crucial for safeguarding the country and the quality of the information acquired by the informer should be higher to safeguard the country. Thus, the role of the informer (Pambudi and Kawamura 2022) along with the Monarch (Soradi-Zeid et al. 2020) is considered for designing the Intelligent Lead algorithm. Motivation The Monarch is considered sovereign over the territory and is the protector of the citizens of the country. The role of the Monarch is to protect the territory from the enemies, punish the misbehavers, and maintain the dharma. While considering the decision-making, Monarch discusses the particular topic with the ministers and provincial heads and gets their ideas, and implements the best idea for the betterment of citizens in the territory. In the intelligent lead algorithm, the role of safeguarding the territory is considered as it is considered as a crucial factor for living in a safer environment. Thus, for safeguarding the territory, Monarch appoints a group of informers (spies) under the head of a minister to locate the enemy base, who intrudes on the territory to capture the territory and plunder it. Here, the role of informers plays an essential role in safeguarding the country. The informer's moves to identify the enemy base, which is unknown initially based on three different movements. They are:Random search This is a non-cooperative movement, in which the informer explores more areas in the search space to obtain the global best solution, but this informer is a low-rank informer. SwingMove search This is also a non-cooperative movement, in which the informer exploits deeply inside the search space to locate the enemy base. Here, the informer moves with his location within a small perimeter and is a high-ranked informer with high-quality information concerning the enemy base. MoveToward search This is a cooperative movement, in which the movement of the informer is based on another informer and is a mid-ranked informer. In this movement, the informer initially explores more areas and after some iteration, he enters into the exploitation phase, when he is closer to the solution (enemy base). Here, for the optimization algorithm, the solution is referred to as an informer, the optimal best solution for solving the optimization issue is the detection of an unknown enemy base and the location of the informer is the solution vector that is evaluated based on the fitness function. The information obtained only by the high-rank and mid-rank informer makes the solution converge at a local solution. Thus, to enhance the search space, the role of low ranked informer based on the random search is utilized. In addition, safeguarding the territory is employed by the Monarch based on the information obtained from the informer. Thus, in the decision-making to safeguard the territory, the combined role of an informer and the Monarch are considered and hence the role of both are hybridized in the intelligent lead optimization to obtain the global best solution for tuning the parameters of the classifier. Mathematical modeling The intelligent lead algorithm is detailed in this section with its mathematical modeling. Initialization: The maximal number of iterations in the intelligent lead optimization is referred as imax, and the dimension of the search space is notated as S. Let the variable B is utilized for defining the objective function and is expressed as B=b1,b2,...bS, and the total number of informers is the same as the population size and is referred to as D. Here, BαandfBα refers to the position of the informers and the corresponding value, respectively, in which α=1,2,3,...,D. Fitness evaluation: The fitness function is utilized for guiding the optimization to find the global best solution by estimating the goodness of the position updated by the search agents in the search space. Here, in the intelligent lead optimization, the mean square error is evaluated to estimate the fitness and is expressed as,13 ILfit=1a∑i=1aPredictedvalue-Actualvalue2 where the fitness function of the intelligent lead optimization is notated as ILfit, and the total samples are notated as a. Here, the fitness is evaluated for all the informers in the search space and the informer with the least fitness is considered as an informer with the highest quality of information regarding the enemy-base identification. Thus, based on fitness all the informers are ranked, in which the position of the informer is not changed until the informer's position is referenced by another informer. Exploration phase: In the exploration phase, the informer with the lowest rank among all the informer searches in the search space randomly explores more areas to analyze the features in the feature space. Here, the random search movement is followed by the informer. Phase switching: In this phase, the move toward criteria is followed, in which the cooperative movement is employed, which means one informer follows the other informer to identify the enemy base. Initially, the informer explores more area in the search space and then identifies an informer Bm and move toward him to find the enemy base and the position updation is represented as,14 Bα(i+1)informer=Bα(i)+c[-1,1]Bm(i)-Bα(i) where the position of the αth informer is updated as Bα(i+1) for the iteration (i+1), Bm(i) refers to the randomly selected best informer for entering into the exploiting phase. Here, the position of the informer is enhanced by hybridizing the Monarch knowledge with the informer's best informer selection in a random manner and is expressed as,15 B(i+1)Monarch=Bbest+E∗Brand1-Brand2 16 B(i+1)-B(i)=Bbest+E∗Brand1-Brand2-B(i) where the best knowledge of the Monarch is notated as Bbest, the position of the Monarch is notated as B(i+1), and the two randomly selected ideas of the Monarch to choose the best informer randomly is notated as Brand1andBrand2, respectively. The term E refers to the coefficient vector utilized for updating the Monarch knowledge. Then, the fractional derivative concept is applied for the knowledge updation of the Monarch and is expressed as,17 B(i+1)-δB(i)-12B(i-1)=Bbest+E∗Brand1-Brand2-B(i) Then, the position updation of the informer with the Monarch's knowledge in selecting the random informer for the intelligent lead optimization is expressed based on the rule (Binu and Kariyappa 2020) as,18 B(i+1)IL=0.5B(i+1)Monarch+0.5B(i+1)Informer where the position updation based on the intelligent lead optimization is notated as B(i+1)IL, the position of the informer is notated as B(i+1)Informer, the knowledge of the Monarch is notated as B(i+1)Monarch.19 B(i+1)IL=0.5Bα(i)+c[-1,1]Bm(i)-Bα(i)+0.5B(i)δ-1+12δB(i-1)+Bbest+E∗Brand1-Brand2 20 B(i+1)IL=12B(i)δ-c(-1,1)+B(i)c(-1,1)+12δB(i-1)+Bbest+E∗Brand1-Brand2 Thus, using the intelligent lead optimization the feature space is explored randomly to obtain the global best solution among the search space in terms of detecting the enemy base. Here, the incorporation of the Monarch knowledge in identifying the enemy base with the informer's knowledge assists to identify the best solution. Exploitation phase: In the exploitation phase, the SwingMove is adapted by the informer for identifying the enemy base. Here, the position of the informer is updated as,21 Bα(i+1)=Bα(i)+c(-1,1)Qi The position of the informer is further strengthened by incorporating the knowledge of the King in decision-making, which is provided in Eq. (21). Thus, the enemy-base identification is safeguarding the territory using the intelligent lead optimization as per the rule (Binu and Kariyappa 2020) is expressed as,22 B(i+1)IL=0.5Bα(i)+c(-1,1)Qi+0.5B(i)δ-1+12δB(i-1)+Bbest+E∗Brand1-Brand2 The position updation of the informer using the intelligent lead optimization is notated as,23 B(i+1)IL=12δB(i)+c(-1,1)Qi+12δB(i-1)+Bbest+E∗Brand1-Brand2 Thus, the enemy-based identification in safeguarding the territory by the combined knowledge of Monarch and the informer ensures more accurate detection with a fast convergence rate. Besides, the in-depth analysis of the exploitation phase helps to identify the local best solution for solving the optimization issues. Here, in the intelligent lead optimization, the balanced exploration and exploitation phase in identifying the global best solution and the local best solution ensures the more accurate solution for solving the optimization issues in detecting the sentiment of the COVID-19 data. Fitness re-evaluation: After identifying the enemy base by the informer, the fitness is re-evaluated to find the correctness of the solution obtained by the current iteration. If the present solution is best, then the position is updated using the present global best solution, else remains in the old position. Stopping criterion: The above-mentioned steps are repeated until imax. The pseudo-code for the intelligent lead optimization is presented in Algorithm 1. Algorithm 1 Pseudo-code for the intelligent lead algorithm.Pseudo-code for the intelligent lead algorithm 1 Begin: 2 Initialize the parameters: swing factor Q, iteration imax, and population size D 3 Exploration: 4 Generate the initial solution I 5 Evaluate the fitness 6 Sort according to the ranks //*The highest-ranked informer is the one with minimal fitness value 7 for 8 { 9 i=1toimax 10 Low-ranked informer explores to identify the enemy base based on a random search I 11 Mid-ranked informer follows another informer by considering the Monarch knowledge to identify the enemy base using MoveToward movement 12 High-ranked informer tries to locate the enemy base based on the Monarch Knowledge using the SwingMove movement 13 Re-evaluate the fitness 14 Update the position of the low-ranked informer based on the exploration phase using Eq. (20) 15 Update the position of the mid-ranked informer based on the phase switching using Eq. (14) 16 Update the position of the high-ranked informer based on the exploitation phase using Eq. (23) 17 Rank based on the fitness 18 i=i++ 19 } 20 Return the best solution 21 end Results and discussion The MATLAB tool is utilized for the implementation of the sentiment analysis method using intelligent Lead-based BiLSTM using Windows 10 OS with 8 GB RAM. Description of dataset The dataset utilized for evaluating the performance is taken from Kaggle open source for sentimental analysis (COVID-192022). It comprises of 9.17 MB data with user descriptions like location, name, and ID. The data corresponding to the COVID-19 vaccine are obtained from the Tweets of the massive population across 180 countries that express the feelings of the people. Laptop and restaurant data: The data taken from the restaurant and laptop trail data are obtained from SemEval-2014 Task-4 dataset (2022). In this, the reviews concerning the laptops and restaurants with two different polarities are included and are utilized for the assessment of the proposed method. Assessment metrics The specificity, sensitivity, and accuracy are evaluated for assessing the performance of the intelligent Lead based BiLSTM technique. Specificity: For the sentiment analysis by the intelligent Lead-based BiLSTM, the capability of predicting the negative sentiments from the available data is termed specificity and is expressed as,24 SAspe=SAtnSAtn+SAfp Sensitivity: For the sentiment analysis by the intelligent Lead-based BiLSTM, the capability of predicting the positive sentiments from the available data is termed sensitivity and is expressed as,25 SAsen=SAtpSAtp+SAfn Accuracy: The closeness of the sentiments estimated by the intelligent Lead based BiLSTM with the target is termed accuracy and is expressed as,26 SAacc=SAtn+SAtpSAtn+SAtp+SAfp+SAfn where the true positive is notated as SAtp, the true negative is notated as SAtn, the false positive is notated as SAfp, and the false negative is notated as SAfn. Conventional methods The conventional methods utilized for the comparative assessment of the intelligent lead-based BiLSTM are KNN (Shamrat et al. 2021), stochastic gradient descent-based neural network (SGD-NN) (Akçay 2020), Random Forest (Neogi et al. 2021), TD-LSTM (Wang et al. 2016), ATAE-LSTM (Tang et al. 2015), BiLSTM (Li et al. 2021), Spy based BiLSTM (Pambudi and Kawamura 2022), and King based BiLSTM (Soradi-Zeid et al. 2020). Comparative assessment The comparative assessment of the intelligent Lead based BiLSTM for the sentiment analysis is employed by varying the training data. Assessment using twitter data Figure 4 portrays the assessment of the intelligent Lead-based BiLSTM using the Twitter data by varying the training data, in which the accuracy is presented in Fig. 4a, the sensitivity is presented in Fig. 4b and the specificity is presented in Fig. 4c. The accuracy acquired by the proposed method with 70% of training data is 92.40%, which is 14.73% enhanced performance compared to the 40% training data. Besides, the proposed method with the same training data is 16.74, 15.39, 15.59, 17.57, 12.71, 14.71, 13.64, and 12.08% superior to the KNN, SGD-NN, Random Forest, TD-LSTM, ATAE-LSTM, BiLSTM, Spy algorithm based BiLSTM, and King algorithm based BiLSTM conventional methods. Table 3 depicts the detailed analysis of Fig. 4.Fig. 4 Assessment using Twitter data based on a accuracy, b sensitivity, and c specificity Table 3 Assessment using Twitter data Methods/Training percentage KNN SGD-NN Random forest TD-LSTM ATAE-LSTM BiLSTM Spy algorithm based BiLSTM King algorithm-based BiLSTM Intelligent lead based BiLSTM Accuracy 40 74.51 76.33 75.42 72.98 78.15 77.23 77.27 78.60 78.79 50 75.53 77.01 76.73 74.05 78.78 77.33 77.96 79.48 80.01 60 76.70 77.97 77.72 74.75 79.38 78.23 79.00 80.75 91.77 70 76.93 78.18 78.00 76.17 80.66 78.80 79.80 81.24 92.40 80 77.63 78.44 78.20 76.47 82.20 79.17 79.88 85.51 95.51 Sensitivity 40 82.02 83.86 82.79 81.14 85.92 85.12 85.16 86.57 86.87 50 83.27 85.19 84.79 81.46 87.05 85.21 86.24 88.00 88.74 60 84.80 86.59 86.30 82.60 88.04 86.88 87.89 89.60 91.07 70 84.86 86.92 86.64 83.75 89.46 87.94 89.16 90.20 91.51 80 86.02 87.29 86.87 84.08 89.69 88.55 89.25 96.27 96.63 Specificity 40 67.74 69.56 68.81 65.54 71.16 70.11 70.15 71.43 71.50 50 68.55 69.60 69.44 67.39 71.31 70.23 70.46 71.76 72.08 60 69.37 70.12 69.91 67.64 71.52 70.36 70.90 72.71 93.39 70 69.77 70.23 70.13 69.35 72.66 70.45 71.24 73.09 94.22 80 70.02 70.38 70.31 69.62 75.53 70.59 71.32 75.61 95.35 Assessment using laptop data Figure 5 portrays the assessment of the intelligent Lead based BiLSTM using the laptop data by varying the training data, in which the accuracy is presented in Fig. 5a, the sensitivity is presented in Fig. 5b and the specificity is presented in Fig. 5c. The sensitivity acquired by the proposed method with 60% of training data is 87.06%, which is 1.87% enhanced performance compared to the 40% training data. Besides, the proposed method with the same training data is 5.25, 5.14, 7.20, 5.82, 2.32, 4.82, 1.73, and 0.74% superior than the KNN, SGD-NN, Random Forest, TD-LSTM, ATAE-LSTM, BiLSTM, Spy algorithm based BiLSTM, and King algorithm based BiLSTM conventional methods. Table 4 depicts the detailed analysis of Fig. 5.Fig. 5 Assessment using laptop data based on a accuracy, b sensitivity and c specificity Table 4 Assessment using laptop data Methods/Training percentage KNN SGD-NN Random forest TD-LSTM ATAE-LSTM BiLSTM Spy algorithm based BiLSTM King algorithm based BiLSTM Intelligent lead based BiLSTM Accuracy 40 77.34 77.57 73.75 77.06 79.24 78.09 80.31 80.48 90.13 50 78.57 86.40 76.97 77.78 80.74 79.02 81.62 82.33 90.99 60 78.77 79.01 77.44 78.32 81.25 79.43 81.90 82.47 91.56 70 84.20 80.23 77.88 78.83 82.21 81.26 82.41 83.82 92.53 80 79.65 80.69 78.49 78.95 83.40 82.83 83.77 92.17 95.32 Sensitivity 40 80.29 80.46 75.98 79.99 82.56 81.18 84.53 84.79 85.43 50 82.42 82.42 80.62 80.99 84.94 82.54 85.27 86.16 86.60 60 82.49 82.59 80.79 81.99 85.04 82.86 85.55 86.41 87.06 70 83.05 84.65 81.52 82.66 86.30 86.00 86.38 87.92 88.25 80 83.95 85.46 82.16 82.70 88.51 87.99 88.58 91.39 91.84 Specificity 40 70.94 71.22 68.26 70.69 72.37 71.52 72.44 72.52 91.22 50 71.18 71.81 69.85 71.08 72.89 71.94 74.30 74.79 91.73 60 71.50 71.88 70.61 71.11 73.80 72.42 74.56 74.80 92.39 70 71.64 72.15 70.72 71.44 74.40 72.81 74.72 75.93 93.09 80 71.74 72.23 71.28 71.64 74.46 73.88 75.14 89.06 94.92 Assessment using restaurant data Figure 6 portrays the assessment of the intelligent Lead based BiLSTM using the restaurant data by varying the training data, in which the accuracy is presented in Fig. 6a, the sensitivity is presented in Fig. 6b and the specificity is presented in Fig. 6c. The specificity acquired by the proposed method with 80% of training data is 95.32%, which is 5.06% enhanced performance compared to the 40% training data. Besides, the proposed method with the same training data is 19.41, 19.15, 18.74, 19.26, 16.62, 19.91, 17.52, and 15.99% superior than the KNN, SGD-NN, Random Forest, TD-LSTM, ATAE-LSTM, BiLSTM, Spy algorithm based BiLSTM, and King algorithm based BiLSTM conventional methods. Table 5 depicts the detailed analysis of Fig. 6.Fig. 6 Assessment using restaurant data based on a accuracy, b sensitivity and c specificity Table 5 Assessment using restaurant data Methods/Training percentage KNN SGD-NN Random forest TD-LSTM ATAE-LSTM BiLSTM Spy algorithm based BiLSTM King algorithm based BiLSTM Intelligent lead based BiLSTM Accuracy 40 76.98 83.08 83.60 80.07 84.74 70.00 83.98 86.07 91.91 50 79.60 83.20 83.85 81.75 85.05 75.25 84.45 86.44 91.95 60 79.81 83.85 84.21 82.70 85.99 75.71 85.13 88.49 92.23 70 80.82 84.48 85.03 83.37 86.52 76.43 85.87 89.83 94.23 80 82.43 84.89 85.32 83.93 86.36 81.67 86.11 91.50 96.11 Sensitivity 40 78.99 90.31 91.16 84.49 92.10 66.15 91.47 93.13 94.25 50 83.55 90.44 91.61 87.79 92.38 75.82 92.18 93.72 94.27 60 83.92 91.63 92.03 89.64 93.89 76.70 93.08 94.08 95.94 70 85.92 92.89 93.47 90.98 94.86 77.91 94.19 96.46 97.75 80 88.86 93.57 94.04 91.73 95.33 87.81 94.46 96.65 99.22 Specificity 40 75.74 76.69 76.87 76.45 78.22 74.44 77.32 78.22 90.49 50 76.44 76.80 76.93 76.54 78.58 75.44 77.56 79.22 91.22 60 76.50 76.91 77.24 76.58 78.95 75.48 78.04 79.56 91.78 70 76.53 76.92 77.44 76.60 79.06 75.71 78.40 79.87 93.82 80 76.82 77.06 77.46 76.96 79.48 76.34 78.62 80.08 95.32 Comparative discussion The assessment by considering the best measure acquired by the intelligent Lead based BiLSTM for the sentiment analysis is portrayed in Table 6. The maximal accuracy and sensitivity are acquired while analyzing using the restaurant data with the value of 96.11 and 99.22%. Likewise, the maximal specificity is acquired using the Twitter data with a value of 95.35%. The performance superiority of the proposed method in terms of accuracy is 14.24, 11.67, 11.22, 12.67, 10.14, 15.03, 10.41, and 4.80%, in terms of sensitivity is 10.45, 5.69, 5.22, 7.55, 3.92, 11.50, 4.80, and 2.59% and in terms of specificity is 26.57, 26.19, 26.26, 26.99, 20.79, 25.97, 25.20, and 20.70% for the traditional methods like KNN, SGD-NN, Random Forest, TD-LSTM, ATAE-LSTM, BiLSTM, Spy based BiLSTM, and King based BiLSTM, respectively.Table 6 Comparative assessment with best measures Methods/Training percentage KNN SGD-NN Random forest TD-LSTM ATAE-LSTM BiLSTM Spy algorithm based BiLSTM King algorithm based BiLSTM Intelligent lead based BiLSTM Accuracy 82.43 84.89 85.32 83.93 86.36 81.67 86.11 91.50 96.11 Sensitivity 88.86 93.57 94.04 91.73 95.33 87.81 94.46 96.65 99.22 Specificity 70.02 70.38 70.31 69.62 75.53 70.59 71.32 75.61 95.35 The intelligent lead-BiLSTM acquired superior performance compared to all the conventional sentimental analysis methods due to the incorporation of the intelligent lead optimization in training the BiLSTM for the minimization of the loss in data learning. Besides, the proposed method acquired fast convergence by integrating the informer's knowledge with the Monarch in decision-making regarding the detection of the enemy base for safeguarding the country. In addition, the balanced exploration and exploitation phase based on the informer's ranks helps to identify the global best solution for tuning the modifiable parameters of the sentiment analysis method. Thus, the enhanced performance is acquired by the intelligent Lead based BiLSTM. Conclusion This research introduces a sentiment analysis technique using the intelligent Lead based BiLSTM. Here, in the intelligent lead optimization, the role of the informer along with the Monarch knowledge in decision-making for locating the enemy-based is considered. The optimal best solution for solving the optimization issue is the detection of an unknown enemy-base and the location of the informer is the solution vector that is evaluated based on the fitness function. In addition, the loss of the classifier while learning the data is eliminated through the incorporation of the intelligent lead optimization hence the loss is reduced and more accurate analysis is obtained in terms of assessment metrics. Besides, the balanced exploration and exploitation phase helps to obtain the global best solution for solving the optimization issues in sentiment analysis by considering the COVID-19 data. The assessment based on the accuracy, sensitivity, and specificity acquired by the intelligent Lead based BiLSTM is 96.11, 99.22, and 95.35%, respectively. The accuracy of the detection still needs to be improved for further online applications. Hence, in the future, a novel deep learning technique will be devised with a graph-based technique for identifying the polarity of the sentiment to enhance the performance further. Author contributions All authors have made substantial contributions to conception and design, revising the manuscript, and the final approval of the version to be published. Also, all authors agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. 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 References Abdulaziz M Alotaibi A Alsolamy M Alabbas A Topic based sentiment analysis for COVID-19 tweets Int J Adv Comput Sci Appl 2021 10.14569/IJACSA.2021.0120172 Akçay MT Estimation of constant speed time for railway vehicles by stochastic gradient descent algorithm Sak Univ J Comp Inf Sci 2020 3 3 355 365 Basiri ME Nemati S Abdar M Asadi S Acharrya UR A novel fusion-based deep learning model for sentiment analysis of COVID-19 tweets Knowl Based Syst 2021 228 107242 10.1016/j.knosys.2021.107242 Binu D Kariyappa BS Rider-deep-LSTM network for hybrid distance score-based fault prediction in analog circuits IEEE Trans Ind Electron 2020 68 10 10097 10106 10.1109/TIE.2020.3028796 Chole V Gadicha V A review towards human intuition based chess playing system using AI & ML Turk J Comput Math Educ (TURCOMAT) 2020 11 2 792 797 Chole V Gadicha V Locust Mayfly optimization-tuned neural network for AI-based pruning in Chess game Int J Image Graph 2022 10.1142/S0219467823500286 Choudrie J Patil S Kotecha K Matta N Pappas I Applying and understanding an advanced, novel deep learning approach: a Covid 19, text based, emotions analysis study Inf Syst Front 2021 23 6 1431 1465 10.1007/s10796-021-10152-6 34188606 Chowdhury AA, Das A, Saha SK, Rahman M, and Hasan KT (2021) Sentiment analysis of COVID-19 vaccination from survey responses in Bangladesh COVID-19 vaccines tweets—sentiment analysis, https://www.kaggle.com/code/ariwansrisetya/covid-19-vaccines-tweets-sentiment-analysis/data?select=vaccination_all_tweets.csv. Accessed Aug 2022 Elgeldawi E Sayed A Galal AR Zaki AM Hyperparameter tuning for machine learning algorithms used for Arabic sentiment analysis Informatics 2021 8 4 79 10.3390/informatics8040079 Garcia K Berton L Topic detection and sentiment analysis in Twitter content related to COVID-19 from Brazil and the USA Appl Soft Comput 2021 101 107057 10.1016/j.asoc.2020.107057 33519326 Hitesh MSR, Vaibhav V, Kalki YA, Kamtam SH, Kumari S (2019) Real-time sentiment analysis of 2019 election tweets using word2vec and random forest model. In: 2019 2nd international conference on intelligent communication and computational techniques (ICCT), pp 146–151. IEEE Kaur H Ahsaan SU Alankar B Chang V A proposed sentiment analysis deep learning algorithm for analyzing COVID-19 tweets Inf Syst Front 2021 23 6 1417 1429 10.1007/s10796-021-10135-7 33897274 Kumari Santoshi Ranjith Ediga Gujjar Abhishek Narasimman Siranjeevi Aadil Sha Zeelani H S Comparative analysis of deep learning models for COVID-19 detection Global Transitions Proceedings 2021 2 2 559 565 10.1016/j.gltp.2021.08.030 Li X Li Z Tian Y Sentimental knowledge graph analysis of the COVID-19 pandemic based on the official account of Chinese universities Electronics 2021 10 23 2921 10.3390/electronics10232921 Mishra RK Urolagin S Jothi JA Neogi AS Nawaz N Deep learning-based sentiment analysis and topic modeling on tourism during COVID-19 pandemic Front Comput Sci 2021 10.3389/fcomp.2021.775368 Monish P, Kumari S, Babu CN (2018) Automated topic modeling and sentiment analysis of tweets on SparkR. In: 2018 9th international conference on computing, communication and networking technologies (ICCCNT), pp 1–7. IEEE Neogi AS Garg KA Mishra RK Dwivedi YK Sentiment analysis and classification of Indian farmers' protest using twitter data Int J Inf Manag Data Insights 2021 1 2 100019 Pambudi D Kawamura M Novel metaheuristic: spy algorithm IEICE Trans Inf Syst 2022 105 2 309 319 10.1587/transinf.2021EDP7092 Saifullah S, Fauziah Y, and Aribowo AS (2021) Comparison of machine learning for sentiment analysis in detecting anxiety based on social media data. arXiv preprint arXiv:2101.06353 SemEval-2014 Task 4. https://alt.qcri.org/semeval2014/task4/index.php?id=data-and-tools. Accessed on Aug 2022 Shamrat FMJM Chakraborty S Imran MM Muna JN Billah MM Das P Rahman OM Sentiment analysis on twitter tweets about COVID-19 vaccines using NLP and supervised KNN classification algorithm Indones J Electr Eng Comput Sci 2021 23 1 463 470 Sitaula C Basnet A Mainali A Shahi TB Deep learning-based methods for sentiment analysis on Nepali COVID-19-related tweets Comput Intell Neurosci 2021 2021 1 11 10.1155/2021/2158184 Sontayasara T Jariyapongpaiboon S Promjun A Seelpipat N Saengtabtim K Tang J Leelawat N Twitter sentiment analysis of Bangkok tourism during COVID-19 pandemic using support vector machine algorithm J Disaster Res 2021 16 1 24 30 10.20965/jdr.2021.p0024 Soradi-Zeid S Jahanshahi H Yousefpour A Bekiros S King algorithm: a novel optimization approach based on variable-order fractional calculus with application in chaotic financial systems Chaos Solitons Fractals 2020 132 109569 10.1016/j.chaos.2019.109569 Tang D, Qin B, Feng X, and Liu T (2015) Effective LSTMs for target-dependent sentiment classification. arXiv preprint arXiv:1512.01100 Wang Y, Huang M, Zhu X, and Zhao L (2016) Attention-based LSTM for aspect-level sentiment classification. In: Proceedings of the 2016 Conference on empirical methods in natural language processing, pp 606–615 Wang D, Liu P, Zheng Y, Qiu X, and Huang X (2020) Heterogeneous graph neural networks for extractive document summarization. arXiv preprint arXiv:2004.12393	0	PMC9734439	NO-CC CODE	2022-12-14 23:28:28	no	Soc Netw Anal Min. 2023 Dec 6; 13(1):1	utf-8	Soc Netw Anal Min	2,022	10.1007/s13278-022-01005-4	oa_other
==== Front Indian J Labour Econ Indian J Labour Econ The Indian Journal of Labour Economics 0971-7927 0019-5308 Springer India New Delhi 408 10.1007/s41027-022-00408-3 Article Specific Human Capital and Skills in Indian Manufacturing: Observed Wage and Tenure Relationships from a Worker Survey Singh Jaivir [email protected] 1 Das Deb Kusum [email protected] 2 Abhishek Kumar [email protected] 3 1 grid.10706.30 0000 0004 0498 924X Centre for the Study of Law and Governance, Jawaharlal Nehru University, New Delhi, India 2 grid.8195.5 0000 0001 2109 4999 Department of Economics, Ramjas College, University of Delhi, New Delhi, India 3 grid.10706.30 0000 0004 0498 924X Centre for Economic Studies and Planning, Jawaharlal Nehru University, New Delhi, India 5 12 2022 2022 65 4 10071028 2 11 2022 © The Author(s), under exclusive licence to Indian Society of Labour Economics 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. Successive Indian governments have attempted to increase the growth in employment alongside encouraging skill enhancement. Against this background, we empirically explore issues surrounding the investment in specific capital by workers. In particular, we try to discern the presence of specific human capital investment by investigating whether there is a link between tenure and wages and find that there is indeed such a link evident in India. This allows us to infer that it is valuable to have long-term relations between employers and their workers and therefore labour market institutions that support long-term employer–employee relationship need to be encouraged. Keywords Specific human capital Indian labour market policy Skills Wage and tenure relationship JEL Classification J24 J83 J41 http://dx.doi.org/10.13039/100004421 World Bank Group 502915-01 issue-copyright-statement© The Author(s), under exclusive licence to Indian Society of Labour Economics 2022 ==== Body pmcIntroduction Unfortunately, it is an overarching and persistent fact that job creation in the Indian manufacturing sector has fallen short of the growth in the workforce. Among the many attempts to increase employment, the current NDA government has pushed the Make in India policy—Starting in 2014, the critical components of this approach were to increase the growth of the manufacturing sector so that both employment in the sector and share of the sector in the national income goes up substantially. This has not quite fructified along envisioned lines with the growth of manufacturing averaging around 6.9% between 2014–2015 and 2019–2020 (as against the objective of 12 to 14% per annum), and the share of manufacturing dropped from 16.3% of GDP in 2014–2015 to 15.1% in 2019–2020 (as against an objective of 25% of GDP by 2022).1 The changes in employment are equally dismal with sectoral contribution to total employment being constant around 12% and fall in manufacturing employment by 9 million between 2011–2012 and 2017–2018 (Mehrotra and Parida 2019). Be this as may, as an essential component of the job creation policy, the NDA government has attempted a series of programs aimed at skill formation to enhance the quality of employment—a vital input into the Make in India endeavour so as "to transform India into a global design and manufacturing hub".2 In the face of the COVID-19 pandemic, the NDA government finds itself now speaking about Atmanirbhar Bharat Abhiyan, which aims to build capacities across sectors and promote local products, with the role of skill formation continuing to be a central concern. In pursuance of this overall agenda of skill enhancement, starting in 2015 a series of policy moves were made.3 The spirit behind these programs is to encourage public–private partnerships in the area of skilling, where the government funds entrepreneurs, who in turn skill workers in collaboration with employers and are refunded by the government based on their performance. This strategy to enhance skills does not attempt to strengthen labour market institutions that can guarantee long-term employment and real wage stability—instead as time has gone by, the labour protection regime has only been weakened, culminating with the NDA ruled states effectively suspending protective labour laws during a raging pandemic,4 and subsequently, rendering such changes more permanent by providing the legislative basis to strengthen moves by individual states to enact competitively weaker labour laws.5 The absence of any thinking in this regard implies that the policymakers have ignored a crucial point made by Becker in his seminal work on human capital—the distinction between general human capital and specific human capital (Becker 1975). General human capital is productive across employers, while specific human capital is associated with increased productivity of the worker only to a particular employer/firm or employee-job match. Specific investments are more valuable if the match continues, than if it is truncated. If employers have invested in specific skills, they will want workers to continue, and to the extent, workers have invested in gaining the specific skills they will want to ensure returns to their investment with wage stability and long-term employment. If workers feel that the employment opportunities associated with the specific skills that they have invested in will evaporate soon, they will be reluctant to invest in these specific skills. This becomes a problem, particularly if employers need these specific skills to compete in the international market (Estevez-Abe et al. 2001). Without some guarantee of long-term employment and real wage stability, these specific skills will be undersupplied. This is indeed an important concern in a labour abundant country seeking to gain a comparative advantage by becoming skill abundant. There is, by now, a good deal of empirical support for the proposition that labour market institutions affect workers incentives to acquire firm-specific skills on the job and thereby shape the export patterns of countries (Tang 2012). While the Make in India policy seems to aspire "to transform India into a global design and manufacturing hub", the skilling policy is devoid of any recognition of specific skills. In this context there is no real attempt to check for the presence of patterns of specific human capital in India—typically a discussion of human capital in the Indian context is confined to broad general human capital concerns [See (Singh et al. 2020; Sharma 2019; Chakravarty and Bedi 2019; Mitra and Verick 2013; Mehrotra et al. 2013; Kumar et al., 2013; Dev and Venkatanarayana 2011)]. Thus, over this paper, we are motivated to explore issues surrounding the investment in specific capital by workers, attempting to assess empirical patterns of specific human capital investment. In particular, we hope to discern basic displays of specific human capital investment by crucially investigating whether there is a link between tenure and wages—this has come to be an important investigation all over the world, typically undertaken to signify the presence of a specific capital investment in the employment relation. We go on to raise cognate questions as to what are the motivations for workers to gain skills by training themselves and if tenure is taken as an important incentive for worker investment in the job, what factors influence such tenure. These questions are important to ask, but often enough, this type of inquiry has been absent in India, mainly on account of the lack of data (as discussed later in the paper). In the face of this, we seek to use data from a special worker-oriented survey conducted in 2017. While this was a small sample cross-section survey, it is nevertheless very valuable in helping us open up some questions concerning the Indian labour market. We begin with Section 2, where we discuss the background literature related to specific human capital investment, providing an overall basis of our study. Next in Section 3 after a brief discussion on the lacunae in Indian labour data, we provide details regarding the survey, the type of information gathered and followed this by tabulating some noticeable patterns evident from the data. In Section 4, we describe the models we seek to estimate and define the variables used in the estimation. This is followed by Section 5, which presents the results of the estimations, and we conclude in Section 6. Background and Empirical Context The key analytical point associated with specific human capital is that it involves a series of ex-ante investment decisions by both employers and employees, which are subject to an ex-post risk of quasi-rent appropriation (Klein et al. 1978). When employers invest in specific human capital, workers can quit, putting to waste the fruits of training and recruitment costs incurred by the employer. At the other end of the relationship, there are concerns of the appropriation of quasi-rent as well, because workers who have invested in specific human capital on the job can be fired by opportunistic employers near retirement, disabusing them from enjoying the returns to investing in the job. Clearly, this leads up to hold up in investment, with both or one of the parties underinvesting. Such hold up can, of course, be mitigated by writing contingent contracts but it has been widely held that this is impossible–the incomplete contract argument. Since efficient contracts that condition relation-specific investment cannot be written, hold-up problems in human capital investment end up being governed by the legal/institutional regime, within which the relationships are embedded. A variety of theoretical discussions have discussed various responses including long-term contracts, wage rigidities, fixed-wage contracts and renegotiation [See (Hashimoto and Yu 1980; Macleod and Malcomson 1993; Grout 1984; Grossman and Hart 1986; Hart and Moore 1988 and Hermalin and Katz 1993)]. The empirical literature has also grown looking for turnover costs, forms of employment contracts and wage characteristics that are consistent with hold-up theories. Particularly prominent is the positive relationship between wage and tenure—the pioneering work of Becker, Mincer and Schultz [(Becker 1975), (Mincer 1962), Schultz (1961)] suggested that to avoid inefficient separations, costs and returns would be shared by both employer and worker and since on the job training increases with tenure, we should also see a rise in wages alongside tenure. To explore the veracity of this proposition, a series of studies were undertaken (mainly in the late 1980s and early 1990s) using panel data in the USA—some finding a more definite positive relation between wage and tenure than others [See (Altonji and Shakotko 1987; Brown 1989 and Topel 1991)]. Since then, empirical studies investigating this relationship can be found all over the world—for example, the presence of a positive wage tenure has been noted in Germany (Dustmaan and Meghir 2005), Italy (Sulis 2014) and China (Qu and Wang 2019), among others. Given the multiple Indian endeavours to impart skills to workers, it is crucial to get a sense of how skills improve on the job and whether such skills are transferable and more importantly what are the returns to experience and seniority in this context. Unfortunately, such questions are hard to pursue because of the lack of substantial data in this regard. However, nevertheless, we have relied on a small survey to initiate an inquiry into this genre of questions in the Indian context. Survey Description Labour Data in India Data pertaining to labour in India is strangely plentiful and simultaneously very scarce. There are several agencies and mechanisms involved in the collection of the data (Papola 2014). Labour laws produce a good bulk of the data—in a sense several labour laws decree that establishments covered by the law have to furnish returns providing information about the establishment. For example, the Annual Survey of Industries (ASI) used by scholars all over the world as the principal source of industrial statistics of India is produced by the combined requirements of the Collection of Statistics Act 1953 and the Factories Act, 1948. As per the Factories Act enterprises employing more than a certain threshold of workers (i.e. those factories employing ten or more workers using power; and those employing 20 or more workers without using power) have to submit details about their establishments, which form the basis for the data. This data and a good deal of other data associated with labour is typically gathered from the employer—and provides information only on a few labour characteristics—say, numbers employed and wages paid but little or no information on the many other worker characteristics such as educational attainment, skills, tenure or tasks undertaken in the workplace. Some of this lacuna was overcome by the data generated from surveys conducted by the National Sample Survey Office (NSSO). The NSSO carried out quinquennial surveys on employment and unemployment with the aim to capture the many associated characteristics that include age, education, gender, social group, level of living, industry and occupational category and ends up facilitating the creation of valuations for labour force participation rate, worker population ratio, unemployment rate, industry and occupational distribution of workers, the extent of underemployment, wages of employees to name just a few of the useful indices available to us on account of this data. In the 2004 (NSSO 60th Round) some data was collected on vocational training in the 66th Round details on education, and many aspects of work training were included. This was followed by the NSSO 68th Round undertaken in 2011–2012 where we again get some information on training and skill formation. While useful in giving us snapshots of the education and training currently gathered by sampled workers, this data tells us very little about the changes that take place over the lifespan of a worker. Furthermore, since the data is available only once in five years, one cannot do any in-depth analysis regarding trends or demand for skills. Thus, many questions asked of human capital accumulation cannot be asked in the Indian context because there is limited data. Recently the Employment and Unemployment Survey by NSSO has been replaced by the Periodic Labour Force Survey (PLFS), and this is now the primary source of labour market data at both the National and State level. The survey is oriented mainly towards collecting data on the employment status of workers but also includes questions that collection information on training. Seeking information on type of training, source of funding, duration of training and whether any training was undertaken over the last 365 days. Thus, some additional information on skill training is becoming available, but again information on many important characteristics of workers and the jobs they do, such as the length of tenure is still missing in PLFS. Furthermore, the institution of this new survey is not without criticism—some severe lacunae have been pointed out by scholars working on labour issues in India (Kapoor 2019). All this effectively means that it is fortuitous that we could work to manage a small survey which provides some vital information that allows us to empirically explore some of the specific human capital issues on hand. Survey Details Our specially commissioned survey was conducted over April–June 2017, supported by a World Bank-funded project 'Jobs and Development' 2014–2016 and undertaken by the Indian Council for Research on International Economic Relations (ICRIER). This survey is linked to an earlier survey supported by the same program of the World Bank, that aimed to look at issues associated with the employment of contract workers in the Indian manufacturing sector. The earlier survey gathered data from 500 firms, with these firms being chosen using a larger ASI frame set for the year 2013–2014 and was located in five states, namely Haryana, Tamil Nadu, Maharashtra, Gujarat and Karnataka. The survey covered eight industry divisions, viz. Manufacture of Food Products; Manufacture of Textiles; Manufacture of Wearing Apparel; Manufacture of Leather and Leather Products; Manufacture of Computer, Electronic and Optical Products; Manufacture of Electrical Equipment; Manufacture of Motor Vehicles, Trailers and Semi-Trailers; and Manufacture of Other Transport Equipment. Further details regarding the survey can be gathered from our work that used data from the survey to investigate the employment of contract labour, one study covering the entire sample (Singh et al. 2019a), and another a sub-sample confined to the state of Haryana (Singh et al. 2017), as well as a study looking at union activity in the manufacturing sector (Singh et al. 2019b). The worker-oriented study survey used in this paper aimed to ask questions regarding human capital accumulation from workers instead of employers, covering workers from the same set of industry groups covered in the earlier employer-oriented survey. The five industry groups located in districts of the State of Haryana, neighbouring Delhi, are Food Processing, Textile and Garments, Leather and Leather Products, Electronics and Computer Equipment and Auto Products, which can be mapped to eight Industry groups as per NIC-08 Classification as shown in Table 1 below.Table 1 Industry groups covered. NIC-2008 (2 Digit) Industry group (NIC-08) Industry group (Survey) 10 Manufacture of food products Food processing 13 + 14 Manufacture of Textiles + Manufacture of Wearing Apparel Textile and garments 15 Manufacture of Leather and Related Products Leather and leather Products 26 + 27 Manufacture of Computer, Electronic and Optical products + Manufacture of Electrical Equipment Electronics and computer equipment 29 + 30 Manufacture of Motor vehicles, trailers and semi-trailers + Manufacture of Other Transport Equipment Auto products (Source: Authors own compilation) Haryana is a state located in the northern part of India which contributed 3.63% to India's GDP in 2017–2018. The Industrial sector contributed 32% towards state's GSVA in 2017–2018 at constant (2011–2012) prices, and the industry has grown at a CAGR of 7.50% between 2011–2012 and 2017–2018. The total number of people engaged in organised manufacturing during 2017–2018 was 8.42 lakh, which is approximately 6% of the All-India total. Unlike other industrial states such as Tamil Nadu, Gujarat, and Maharashtra, the state has no coastal border or any major port to facilitate trade through the sea. The state is a leading state in terms of production and exports of automobile products such as passenger cars, two-wheelers, mobile cranes & tractors. Maruti Udyog Ltd., Hero MotoCorp Ltd, Yamaha Motor Pvt Ltd. and Escorts Group are some of the leading automobile companies based in Haryana. The Gurgaon–Manesar–Bawal belt is the auto hub of India. Apart from automobiles, the other major industry in the state is the textile and wearing apparel industry employing around 1.77 lakh people. Districts such as Panipat, Gurugram, Faridabad, Hisar and Sonipat are the textile centers in Haryana and engage in production and exports of primarily the cotton readymade garments. In Haryana, the proportion of workers employed as contract workers is one of the highest among the Indian states. In 2013–2014, out of every 10-worker engaged in manufacturing, about 5 of them were on a contract job. The two charts below provide trend and contrast Haryana with All India in terms of worker engagement. The state has witnessed significant worker unrest as well owing to this practice, the 2012 violence incidence at Maruti Suzuki factory in Manesar was one such event. Given the industrial nature and practice of employing workers on contract, Haryana is the ideal state in the northern region of India to study the issue of relationship-specific investment in India (Fig. 1).Fig. 1 Directly Employed v/s Contract Workers in Manufacturing: Haryana and All-India. Panel A: Harayana, Panel B: All-India. (Source: Various Issues, Annual Survey of Industries, Ministry of Statistics and Programme Implementation, Govt. of India.) The survey collected information on skills, tenure and wages of workers from a sample of 100 workers engaged in the organised sector. Ideally, we would have liked to work with a larger sample, but resources limited us to canvass a small sample. We were confronted with two questions—one, how many workers need to be surveyed from each industry and two, how to choose the worker to be interviewed? To keep some parity with the earlier employer-oriented survey, we used the ASI frame 2013–2014. We computed the total number of persons engaged in each of the Industry Groups as a ratio to the total number of persons employed across all five Industry Groups. This share then allowed us to decide how many workers to allocate to out of 100 to each industry group. (For example, the Food Processing Industry Group engaged 52,816 persons, which was about 8% of the total number of persons employed across all Industry Groups, which meant that eight workers belonging to the Food Processing Industry Group were canvassed during the survey.) To choose the specific worker, each worker employed in the industry was given a random number and five times the number of workers chosen to represent the industry were chosen randomly and located by identifying the firm employing them as per the ASI frame 2013–2014. We proceeded to interview the stipulated number of workers for each industry working down the list. (For example, in the Food Processing Industry we worked with a pool of 40 workers each associated with an identifiable firm, and out of this pool eight workers were interviewed working down the list—if a worker from a firm could not be found we moved to the next worker on the list.) Turning to a description of the questions canvassed—workers were asked to identify their status as a regular, contract or casual labour, wage levels, education details, how long they have been in the current job and other details regarding their past experience, details and attitudes regarding skilling and training as well as their views on links between skills learnt and job regularisation. Models and Explanatory Variables Our primary target over this paper is to see if we can detect some basic configurations of specific human capital investment, using the data we have on hand. Thus, we aim to see whether there is a link between tenure and wages, if tenure is taken as an important incentive for worker investment in the job what factors influence such tenure, and what are the motivations for workers to train or invest in skilling on the job. This is attempted by looking at three sets of relationships – (i) Wage–Tenure Relationship (ii) Determinants of Tenure (iii) Drivers of Worker Training. We proceed below to describe the empirical models that we use to look at these relationships. Wage–Tenure Relationship In the first exercise, we attempt to look at the wage–tenure relationship, suggesting that the underlying relationship, following the specifications in much of the literature, can be captured by1 Wi=fXi,Ti,Vi where Wi is the wage of worker i; the coefficient of Xi is the return to general human capital, gathered by gaining experience and captured empirically as the total market experience; Ti represents job-specific capital, empirically measured as the tenure with the employer; and Vi includes other characteristics, which may be person-specific or industry-specific. We include three variables as components of Vi in our empirical model, the Skill level of the worker, the nature of job—regular or contract worker and to capture the role of industry-specific factors in wage determination, the industry in which the worker is employed. To be able to adapt this model to the data on hand we have to, among other things, represent wages in a limited dependent variable format because the data on wages was collected in wage bands. Thus, the empirical model that was estimated is of the form62 Wi∗=β1Xi+β2Ti+β3Si+β4Ri+β5Ii+εiW=0W∗<6000W=0W∗<6000W=16000<W∗<9000W=29000<W∗<12000W=312000<W∗<15000W=415000<W∗ where Wi∗ and Wi are the latent and observed variables relating to wages received by worker i, respectively, and are believed to depend on Xi the general human capital of the worker (measured as the age of worker), Ti the specific human capital of the worker (measured as the years spent in the current job), Si the skill level of the worker, Ri whether the worker is a regular or contract worker and Ii the industry in which the worker is employed. The term εi is an error term and we assume that it is normally distributed—this assumption allows us to estimate the model as an Ordered Probit Model and we estimate the parameters of the model β1⋯.β5 using the Maximum Likelihood method. Determinants of Tenure As a follow-up from the previous model, it was thought it would be useful to see if we can identify some factors that may be influencing the tenure of workers. Thus, the equation we estimate is3 Ti=g0+g1Fi+g2Ei+g3Si+g4Ri+g5ETi+g6Hi+g7Xi+g8SBRi+ei where Ti is the tenure, dependent variable and the independent variables include worker related characteristics such as whether the worker stays with family (Fi), education levels of the worker (Ei), the skill level of the worker (Si), whether the worker is a regular or contract worker (Ri), whether the employer-provided training or not (ETi), hours worked by the worker (Hi), age of the worker (Xi) and the job regularisation policy of the employer was contingent on skills (SBRi). We assume the error term is independently, identically and normally distributed and we use the method of Ordinary Least Squares to estimate the parameters of the equation γ0 …γ8. Worker Training or Propensity to Train In pursuit of our attempts to see whether our data provides some evidence about the training of workers as a firm-specific human capital investment, we look at the factors that affect the propensity of workers to be trained. It is impossible from the data to look into the psychological profile of workers. However, we can infer inclinations from observed actions of reported participation in training endeavours—the fact that a worker reports training reflects his inclination. Thus, we look at factors that may be influencing the propensity of workers to undergo training on the job. The model characterising the propensity to be trained as follows:4 Y∗i=η0+η1Ei+η2Ti+η3Xi+η4Si+η5PTi+η6Ri+η7Li+uiYi=(Yi∗>0)Yi=0(Yi∗=0) where Yi∗ and Yi are the latent and observed variables related to the propensity to be trained. Here the observed variable takes note if the worker reported any training on the job whatsoever. The independent variables include Ei which represents the education level of the worker, Ti the tenure or the number of years the worker has worked in her current job, Xi the age of the worker, Si skills of the worker, PTi whether the worker says he would pay to be trained, Ri captures the type of worker-contract or regular and Li represents the nature of the industry in which the worker is employed-labour intensive or capital intensive. The error term ui is assumed to be normally distributed, and the parameters of the model η0 …η7 are estimated as a probit model using the Maximum Likelihood method. Description of Variables The empirical estimates of models described above involve a number of variables, and we proceed to describe the content of these variables. Age Worker (Xi) This variable consists of the reported age of the worker. Since we do not have any information regarding the number of years the worker has worked, which is the usual measure for general human capital, we use the age of the worker as a proxy for general human capital. Industry (Ii) This variable captures the industry in which the worker is employed viz. Auto Components, Food Processing, Electronic Appliances, Garments and Leather: As noted earlier, the survey covered workers working in five industries groups located in the State of Haryana namely Food Processing, Textile and Garments, Leather and Leather Products, Electronics and Computer Equipment and Auto Products. While estimating parameters of models that we have specified, these industries appear as dummy explanatory variables. Labour Intensive (Li) In the estimation of Eq. (4) that looks at the propensity of workers to train, we collapse the five industries mentioned above into labour-intensive industries (Garments, Leather, Food Processing) and capital-intensive industries (Auto Products and Electronics). This division is made based on the capital-labour ratio value of these industries between 2009 and 2014. The industries which have a capital-labour ratio value higher than the average value of the five industries combined are taken as capital-intensive industries, whereas the industries, having a lower average capital-labour ratio value than the overall average is taken as labour-intensive industries. Education Worker (Ei) The workers reported their level of education in the survey. Given the small size of the sample, it is difficult to use the information in a finely portioned manner, so education appears in the equations as a dummy variable taking the value of unity if the worker's education level lies between passing 10th Class and being a college graduate and taking the value zero if illiterate or passed a class till the 8th Standard. Hours Worked (Hi) The surveyed workers were asked whether they worked 8-h, 10-h or 12-h shifts and based on this information we construct a binary variable which takes the value of one if the worker works an eight-hour shift and zero if they work 'ten hours' or 'twelve-hour' shifts. Employed Present firm (Ti) We asked the workers surveyed, how long they had been working in the establishment where they were currently employed. This variable is taken as our measure of tenure and is our measure of specific human capital. Employer Training (ETi): We asked the workers, whether they were trained by their employers or not and based on this information the variable takes the value unity if the answer was in the affirmative otherwise it takes the value zero. Job training (Yi) The survey questioned workers, whether they received any training, whether initiated by the employer or otherwise, forming a broader set than the previous variable. We use this information to construct a binary dependent variable (Model 3) which captures the effects of a latent propensity to train by workers which can’t be observed directly. NSQF (Si) This is an attempt to measure the skill of a worker using information reported in the survey as to the tasks performed by the worker. Recently the Indian government has identified a National Skill Qualification Framework (NSQF), which categorises tasks/jobs based on complexity on an ascending scale of 1–10. We took the tasks reported by workers in our sample and slotted them in the categories put out by the NSQF. Our sample showed a range between 2 and 5, and we decided that we would label workers with a score of 4 & 5 as being more skilled than those with a score of 2 and 3. This is an interesting index to use because it captures the complexity of tasks performed by a worker and which is reflective of the innate ability of the worker and an overall association with investment in human capital linked to the worker. While estimating parameters, these skill levels appear as dummy explanatory variables. Paid Training (PTi) Questioning the workers expressing the willingness to train, whether they would pay for training, and some said they would while others said they would not. We use this information to create a binary variable which takes the value of unity if the worker says he will pay for training and zero otherwise. Skill Based Regularisation (SBRi) We questioned the workers if their employer had a regularisation policy. For those that said "Yes" (see the discussion on Patterns Evident from the survey above) were asked whether the skills learnt formed the basis on which their employer gives them a 'permanent' job and if they said 'Yes' the variable takes the value of the unit or zero otherwise. Worker Family (Fi) Over the survey, the workers were asked whether the worker stays with his family or not. This variable takes the value of unity if he stays with his family and a value of zero if he does not. This variable attempt to measure job-specific investment by the worker—if he lives with his family, then he has invested in establishing a home nearby. Worker Type (Ri) This variable takes the value unity if the worker reports that he has a regular job and zero if he said that he worked as a contract worker, giving us a sense as to which segment of the labour market he belongs. Wage Category (Wi) The workers surveyed were asked information about their monthly pay by identifying in which of the five categories listed in the survey questionnaire, their pay could be placed. These categories (Rupees per month) included (a) less than 6000, (b) 6000–9000, (c) 9000–12,000, (d) 12,000–15,000 and (e) more than 15,000. The collection of wage information in wage bands was done purposely since obtaining point estimates for wages could have induced measurement bias on account of reluctance in reporting precise amounts. Most of the above 14 variables are dummy variables except for the variables capturing the age and tenure of workers. The summary statistics of the variables can be seen in the accompanying Table 2. The overall sample comprised 96 workers as 4 of them had to be dropped due to reporting issues.Table 2 Regression Summary Statistics. Variable Mean No of observations Standard deviation Min. Max. Age worker (Xi) 29.20 96 7.01 19 52 Auto components (Ii) 0.31 96 0.47 0 1 Wage category (Wi) 0.62 95 0.73 0 3 Education worker (Ei) 0.76 95 0.43 0 1 Hours worked (Hi) 0.11 96 0.32 0 1 Electronic appliances (Ii) 0.06 96 0.24 0 1 Employed present firm (Ti) 3.00 96 3.05 0.08 16 Employer training (ETi) 0.28 96 0.45 0 1 Food processing (Ii) 0.08 96 0.28 0 1 Garments (Ii) 0.48 96 0.50 0 1 Job training (Yi) 0.29 93 0.46 0 1 Labour intensive (Li) 0.63 96 0.49 0 1 Leather (Ii) 0.06 96 0.24 0 1 NSQF (Si) 0.68 96 0.47 0 1 Paid training (PTi) 0.33 95 0.47 0 1 Skill Based Regularisation (SBRi) 0.77 94 0.43 0 1 Worker Family (Fi) 0.52 96 0.50 0 1 Worker Type (Ri) 0.40 96 0.49 0 1 (Source: ICRIER Worker Survey on Labour issues in Indian Manufacturing sector 2017) Empirical Results Over this section, we present the results from our estimations of Eqs. (2), (3) and (4) described in Sect. 4—reflecting the wage–tenure relationship, the determinants of tenure and the propensity of workers to train, respectively. The results appear statistically robust and prima facie indicate support for the view that specific human capital is quite important in the Indian manufacturing sector. Estimates of the Wage–Tenure Relationship The maximum likelihood estimates of the model endeavouring to capture the wage–tenure relationship can be seen in Table 3. As mentioned earlier, the age of the worker stands in as a measure of general human capital since the data was not able to give us a figure for the total work experience of the worker. The number of years that the worker had worked in the place of current employment is our measure of the specific human capital. The NSQF classification is a measure of the type of job pursued by the worker while simultaneously measuring skill. We also include the type of worker—whether hired as a regular worker or contract worker to see if there is a link between such categorisation and wages. Apart from this, the next set of variables seek to capture industry-level effects. This is captured by setting up dummy variables in relation to the leather industry, depending on whether the worker worked in the Garment, Auto Components, Electronic Appliances or Food Processing industries. By including these other variables, we have attempted to minimise problems of misspecification as can be seen in Table 6 displaying the diagnostic tests associated with the estimated model that the results of the link test in STATA indicate that there is no misspecification and other diagnostic values are well within reasonable bounds.Table 3 Maximum Likelihood Estimates of the Ordered Probit Model Wage– Tenure Relationship. Dependent variables: wages Explanatory variable Coefficients Age Worker (general K) (Xi) 0.044 (0.020) Employed present firm (specific K) (Ti) 0.135 (0.058) NSQF (Si) 1.348* (0.413) Garments (Ii) 2.081 (0.965) Auto components (Ii) 1.804* (0.923) Electronic appliances (Ii) 2.739* (1.043) Food processing (Ii) 2.698 (1.106) Worker type (R i) 0.540* (0.321) Log-likelihood − 68.561 Pseudo R2 0.279 (Source: ICRIER Worker Survey on Labour issues in Indian Manufacturing sector 2017) *Significant at 1% level, Significant at 5% level Significant at 10% level Values in parenthesis represent Standard Errors The results show that all the variables are significant and have a positive sign. Thus, the variables of interest namely the age of the worker—reflecting a dimension of general human capital and the number of years employed in the current firm—reflecting human capital specific to the job are positively and significantly related to wage levels. The positive and significant relationship with the NSQF further links specific human capital (the more skilled/complex job) with wages. The dimensions of the wage tenure relationship are explicitly evident when viewed graphically, as shown in Fig. 2. Here we see that the predicted probability of drawing a worker with a long tenure declines if she is in the lowest wage bracket (Rupees 6000 to Rupees 9000). In the next wage bracket (Rupees 9000–12,000) the probability rises, hitting a maximum of around 10 years but then tapers down. Over the next two higher wage brackets (Rupees 12,000–15,000 and More than Rupees 15,000) the probability of drawing a worker with long tenure is not as high as the previous bracket but nevertheless is increasing.Fig. 2 Tenure and Wages: Predicted Probabilities. (Source: Authors’ computation based on ICRIER Worker Survey on Labour issues in Indian Manufacturing sector 2017) While this model only uses a cross-section data where typically large extensive long-term data sets are used, it nevertheless provides reasonable and robust support for the proposition that there is a wage tenure relationship present in the Indian manufacturing sector. This in turn gives us attendant support that specific human capital is present and important in the Indian manufacturing sector. Determinants of Tenure: Estimates of the Model Given the information available within our data set, it is important to see if we can identify some of the factors that might be influencing the tenure of workers. In other words, we try to see if we can identify some of the characteristics of workers that have been with a firm for a relatively long time. Of course, the wage paid is not included due to obvious endogeneity problems. We linked tenure (as dependent variable) with whether the worker stayed with his/her family, education levels, the NSQF value of the job, whether the worker had a regular job or was a contract worker, whether trained by the employer, whether the worker worked an eight-hour shift or longer, the age of the worker and whether the worker reported skill-based regularisation by their employer. The OLS estimates of the model are shown in Table 4. As can be seen, neither the fact that the worker lives with his family nor education levels is significant. Apart from perhaps reflecting the point that the education variable is not very finely partitioned and that this may be contributing to the insignificant result, it could also suggest the fact that much of the specific capital associated with the job is learned on the job rather than through education. This is evident from the significance of the variable capturing whether an employer imparts training—this perhaps reflects the idea that the employer (and the worker) are investing in a long-term relationship. This is no doubt reflected in the strong significance of the relationship between tenure and whether the worker has been employed on regular terms or a contract worker. The higher complexity/skill of the job is also associated significantly (albeit at the 10% level of significance) with tenure. The fact that the relationship between those who report working more than the reasonable eight hours and tenure is negative indicates that vulnerable low skilled workers are pushed to short tenures. The variable asking workers their subjective opinion as to whether skills they have learnt enable more permanent jobs was not significant in explaining tenure.Table 4 OLS Estimates-Determinants of Tenure Dependent variable: tenure (experience on the current job) Explanatory variable Coefficient Worker family (Fi) − 0.0220 (0.706) Education worker (Ei) − 0.275 (0.881) NSQF (Si) 1.099 (0.510) Worker type (Ri) 1.600* (0.817) Employer training (ETi) 1.675* (0.634) Hours worked (Hi) − 2.031 (0.848) Age worker (Xi) 0.135* (0.051) Skill based regularisation (SBRi) − 0.921 (1.194) Cons − 1.742 (1.357) R-squared 0.357* (Source: ICRIER Worker Survey on Labour issues in Indian Manufacturing sector 2017) Significant at 1% level, Significant at 5% level, Significant at 10% level Values in parenthesis represent Robust Standard Errors As part of diagnostics, the Ramsey Reset Test has been undertaken, with P-value of 0.491 the null hypothesis of No Omitted variable bias is accepted. The mean VIF for the model estimated is 1.27, ruling out multicollinearity. Overall, the significant correlates with tenure support the view that longer tenure is associated with situations where specific capital is important—where the nature and complexity of the job demand it, and the employer sees virtue in training the worker on the job. Worker Training: Estimates of the Model The last model in this study attempts to capture the factors that influence the propensity of workers to train or in other words, gain specific human capital. While we cannot observe the latent variable namely the inclination of workers to train, but only observe as to whether they were trained or not, our model aims at a maximum likelihood estimate of a Probit model. Thus, in Table 5, we see the coefficients associated with a series of variables with respect to whether training is imparted to a worker or not. These variables include worker education levels, tenure, age, NSQF levels, whether the worker has a regular job or is a contract worker whether the worker is willing to pay for training or not and whether the worker works in a labour-intensive industry. The measures of goodness fit for the model are presented in Table 6. Table 5 Maximum Likelihood Estimates of the Probit Model-Worker Training Dependent variable: the propensity to skill Explanatory variable Coefficient Education worker (Ei) 1.084* (0.595) Employed present firm (Ti) 0.164(0.074) Age worker (Xi) − 0.016 (0.027) NSQF (Si) 0.487 (0.394) Paid training (PTi) − 0.579 (0.379) Worker type (Ri) 1.208* (0.384) Labour intensive (Li) − 0.869** (0.394) Cons − 1.748(1.034) Log-likelihood − 38.322 Pseudo R2 0.307 (Source: ICRIER Worker Survey on Labour issues in Indian Manufacturing sector 2017) Significant at 1% level, Significant at 5% level, *Significant at 10% level Values in parenthesis represent Standard Errors Table 6 Regression Diagnostics and Goodness of Fit (Source: Authors’ computation based on ICRIER Worker Survey on Labour issues in Indian Manufacturing sector 2017) Measures Wage tenure model Propensity to skill model Log-likelihood: model − 68.563 − 38.322 Log-likelihood: intercept-only − 95.175 − 55.332 Chi-square: deviance 137.125 (df = 84) 76.645 (df = 83) Chi-square: LR 53.225 (df = 8) 34.019 (df = 7) Chi-square: p-value 0.000 0.016 R-square: McFadden 0.280 0.307 R-square: McFadden(adjusted) 0.164 0.163 R-square: McKelvey & Zavoina 0.547 0.551 R-square: Cox-Snell/ML 0.429 0.312 R-square: Cragg-Uhler/Nagelkerke 0.496 0.443 R-square: Count 0.653 0.802 R-square: count(adjusted) 0.298 0.333 Information criterion: AIC 159.125 92.645 AIC divided by N 1.675 1.018 Information criterion: BIC 187.218 112.732 Variance: e 1 1 Variance: y-star 2.209 2.228 The age of the worker is insignificant, and so is the variable where workers state whether they are willing to pay for training or not. The NSQF value of the job is also insignificant, most probably reflecting the point that workers who perform already skilled tasks are not trained further. Turning to other variables, the propensity to train is significantly linked to education levels and so is the tenure level, both along expected lines. Also, significant but with a negative sign, it appears that workers associated with labour-intensive industries have less of a propensity to train than those in capital intensive industries. The most interesting and significant result is associated with the variable that captures whether the worker is a regular worker or a contract worker. This result tells us that regular workers have a propensity to get trained, but contract workers do not—in other words, contract workers may not have an incentive to invest in the job. This is indicative of the enormous segmentation in the Indian labour market and is very well illustrated in Fig. 3. Using the underlying estimates of parameters of the model it plots the probability of two types of workers–regular and contract, who has been employed for varying years in the present firm, of undergoing training. As can be seen, workers who are directly employed have a greater chance of undergoing training than their counterpart who is employed through a contractor. Fig. 3 Job Training and Tenure: Predicted Probabilities (Source: Authors’ computation based on ICRIER Worker Survey on Labour issues in Indian Manufacturing sector 2017) Summing up over the three models The predominant finding of our empirical investigation is that there is a reasonably strong link between wages and tenure, allowing us to infer that a value can be ascribed to the continuation of relations between employers and their workers rather than truncating the relationship. In other words, with due admission that we are working with a small sample, we have reasonably robust support to acknowledge the presence of relationship-specific human capital in the Indian manufacturing sector. As we turn to the linkages between tenure and the characteristics of workers—it appears that longer tenure is the device through which both employers and workers seem to overcome hold-up problems, evident in our findings that employers pay for training workers who have a longer tenure and that more skilled workers have longer tenures than those who are less skilled. However, the most interesting finding of our empirical investigation has been to see the link between tenure and the type of worker—clearly being a regular worker with more substantial labour rights gets her a longer tenure than a contract worker. In this the patterns that we gathered from the more subjective inputs of our survey showing that contract workers wait with some expectation to become regular workers may condition some of the learning on the job—but till this is fructified, such workers are probably not investing in the job on hand. To the extent we can see the propensity to train as a proxy for expressing a desire to invest in skilling for the job, our results show that regular workers have a greater propensity to train and gain skills rather than contract workers. It can be effectively gathered from this—since a good amount of employment in the manufacturing sector is in the form of contract employment (36% of the workers in the manufacturing sector are contract workers7)–there is a loss of specific human capital. The large-scale advent of contract labour in the Indian manufacturing sector can be traced to a Supreme Court of India judgment—after the Steel Authority judgment,8 Indian employers have been able to hire workers through labour contractors, paying such workers lower wages and effectively denying them any long-term claims on their job (Das et al. 2017). Several studies have made it apparent that contract labour allows employers to use the segmented labour market to bargain lower wages for regular workers [See (Singh et al. (2019a, b), (Kapoor and Krishnapriya 2019), (Sen and Maiti 2013)]. This benefit comes at a cost and a good portion of this cost comprises of the inadequate specific human capital gained by contract workers. In the new Code on Industrial Relations enacted recently to replace existing labour laws,9 it appears that contract labour may gain some rights but regular workers will lose many existing rights—thus, prima facie it looks like it that the effects of the new law on the formation of specific human capital are not very promising. Concluding Comments Invoking a much larger data source than we have used, estimates from 2017 to 2018 PLFS report show that nearly 90% of the population in the age group 15–59 years have not received any vocational training. Out of the remaining 10% who have received the training, only 2% point has received through the formal channel which is associated with a structured educational institution resulting in diploma/certificates and qualifications. The three industries that account for 40% of those receiving the formal training are Electrical, Power and Electronics, IT/ITes and Textiles and Handlooms Apparels. In terms of employment outcomes, 12.4% of those who were formally trained were unemployed. Most importantly it turns out that non-formal sources of training which include hereditary, self-learning, learning on the job and other non-formal training are the most prominent methods of skilling. Among these the learning on the job is the most popular source of non-formal skill formation. It is against this background that as we noted in the introduction, the Indian state is desirous of skilling workers sufficiently so that Indian manufacturing output and exports compete in the international market. This cannot be done without enhancing both general and specific human capital—without expanding both categories of human capital, it is hard to imagine up a sizeable skilled workforce. However, for investment in specific human capital to go up, the inherent hold-up problems must be mitigated, and that means having labour institutions in place that can prevent hold-ups on the part of the worker, which in turn implies more secure worker rights. While pre-existing laws may need reform but the current move in the face of the COVID 19 pandemic to suspend labour laws, followed by the enactment of new labour laws that weaken labour rights, counters the aspiration to have a skilled workforce. It is indeed challenging to think how the current Atmanirbhar Bharat Abhiyan policy of the Indian Government will fructify. Acknowledgements This paper is a part of the Jobs and Development project (2015–2018) at ICRIER supported by the World Bank. An earlier version of the document was presented at the Jobs and Development conference organized in Bogota, Colombia, in May 2018. The authors thank the conference participants for their comments on the paper. We would also like to thank K.V. Ramaswami and Homagni Chaudhury for their valuable comments on this paper. A big thank you to Prateek Kukreja, who assisted in conducting the survey that gathered the crucial data used in this paper. This paper would have been impossible to write were it not for Prof Deb Kusum Das, our co-author, who unfortunately passed away well before his time. Funding The paper is a part of the ‘Jobs and Development’ research project at ICRIER supported by the World Bank (Grant No.: 502915-01). Financial support from the World Bank is gratefully acknowledged; World Bank Group [502915-01]. Declarations Conflict of interest Opinions and recommendations in the paper are exclusive of the author(s) and not of any other individual or institution to which authors are associated. The authors declare that they have no relevant material or financial interests related to the research presented in this paper. 1 See key macroeconomic-indicators published by National Statistical Office, Ministry of Statistics and Programme Implementation (https://eaindustry.nic.in/Key_Economic_Indicators/Key_Macro_Economic_Indicators.pdf, last accessed 17th July 2020) and M. Suresh Babu, "Why 'Make in India' has failed". The Hindu. (https://www.thehindu.com/opinion/op-ed/why-make-in-india-has-failed/article30601269.ece, last accessed on 23rd July 2020). 2 See About Make in India (https://www.makeinindia.com/about,last accessed on 23rd July 2020). 3 One of the first steps in this regard was to marginally refurbish the Apprentice Act, 1961, the law which makes it obligatory for a set of employers to engage apprentices in designated trades and contribute towards setting up training institutes—Industrial Training Institutes (ITIs) and Industrial Training Centers. Those trained in these institutes are employed for a short duration by employers who participate in the apprenticeship endeavour. It needs to be noted that this arrangement is quite far removed from standard systems of apprenticeship (such as in Germany) where apprentices can look forward to long-term relationships with employers. In addition to this, a National Policy on Skill Development and Entrepreneurship was also declared in 2015, resulting in a new ministry, the Ministry for Skill Development and Entrepreneurship that is dedicated to various programs to skill the Indian workforce. Prominent among these are schemes such as the Pradhan Mantri Kaushal Vikas Yojana (PMKVY) along with other schemes like Deen Dayal Upadhyay Gramin Kaushalya Yojana (DDUGKY) and more recently the project associated with Skills Acquisition and Knowledge Awareness for Livelihood Promotion (SANKALP). 4 Uttar Pradesh, Madhya Pradesh, Gujarat, Rajasthan and Himachal Pradesh have initiated steps, details regarding which can be found at https://www.mondaq.com/india/employment-and-workforce-wellbeing/935398/suspension-of-labour-laws-amidst-covid-19. 5 See Somesh Jha Codes give more power to states to be flexible on labour laws Business Standard September 4 2020 https://www.business-standard.com/article/printer-friendly-version?article_id=120092401255_1 6 The numbers are in Indian Rupees and the intervals the ones used in the survey. 7 Based on estimates from Annual Survey of Industries (ASI) 2017–2018 report (http://www.csoisw.gov.in/CMS/UploadedFiles/VolumeI_2017_2018.pdf, last accessed on 23rd July 2020). 8 Steel Authority of India v. National Union Water Front Workers AIR 2001 SC 3527. 9 See Somesh Jha ‘How 3 labour codes aim to reform employment contract, lay-off, work safety’ Business Standard September 09 2020 https://www.business-standard.com/article/printer-friendly-version?article_id=120090900775_1. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Altonji JG Shakotko RA Do wages rise with job seniority? The Review of Economic Studies 1987 54 3 437 459 10.2307/2297568 Babu, M. Suresh. 2020. Why ‘Make In India’ has failed. The Hindu, 20 January 2020. Becker GS Human capital: A theoretical and empirical analysis, with special reference to education 1975 Chicago University of Chicago Press Brown JN Why do wages increase with tenure? On-the-Job Training and Life-Cycle Wage Growth Observed within Firms, the American Economic Review 1989 79 5 971 991 Chakravorty B Bedi AS Skills Training and Employment Outcomes in Rural Bihar Indian Journal of Labour Economics 2019 62 173 199 10.1007/s41027-019-00167-8 Das DK Singh J Choudhury H Hill Elizabeth Palit Amitendu Judicial production of labour market flexibility: Contract labour employment in Indian organised manufacturing Employment Policy in Emerging Economies: The Indian case 2017 London Taylor & Francis 42 63 Dev, S. M., and Venkatanarayana, M. (2011). Youth employment and unemployment in India, IGIDR Working Paper No. 2011–009. Mumbai: Indira Gandhi Institute of Development Research. Dustmann C Meghir C Wages, experience and seniority The Review of Economic Studies 2005 72 1 77 108 10.1111/0034-6527.00325 Estevez-Abe M Iversen T Soskice D Hall Peter A Soskice David Social protection and the formation of skills: A reinterpretation of the welfare state Varieties of capitalism: the institutional foundations of comparative advantage 2001 Oxford Oxford University Press 145 183 Grossman SJ Hart OD The costs and benefits of ownership: A theory of vertical and lateral integration Journal of Political Economy 1986 94 4 691 719 10.1086/261404 Grout PA Investment and wages in the absence of binding contracts: A Nash bargaining approach Econometrica 1984 52 2 449 460 10.2307/1911498 Hart O Moore J Incomplete contracts and renegotiation Econometrica 1988 56 4 755 785 10.2307/1912698 Hashimoto M Yu BT Specific capital, employment contracts, and wage rigidity The Bell Journal of Economics 1980 11 2 536 549 10.2307/3003378 Hermalin BE Katz ML Judicial modification of contracts between sophisticated parties: A more complete view of incomplete contracts and their breach Journal of Law Economics and Organization 1993 9 2 230 255 Jha, Somesh. (2020a) How 3 labour codes aim to reform employment contract, layoffs, work safety. Business Standard, 9 September 2020a. Jha, Somesh. (2020b) Codes give more power to states to be flexible on labour laws. Business Standard, 24 September 2020b. Kapoor, Radhicka, and Krishnapriya, P.P. (2019). Explaining the contractualisation of India’s workforce, ICRIER Working Paper No. 369.New Delhi: Indian Council for Research on International Economic Relations. Kapoor, Radhicka. (2019). An Employment Data Strategy for India. Paper presented at the India Policy Forum, NCAER, New Delhi, July 2019. Klein B Crawford RG Alchian AA Vertical integration, appropriable rents, and the competitive contracting process The Journal of Law and Economics 1978 21 2 297 326 10.1086/466922 Kumar R Mitra A Murayama M Toiling children in India: The gender dimension International Journal of Social Economics 2013 40 10 885 897 10.1108/IJSE-2012-0156 MacLeod WB Malcomson JM Investments, holdup, and the form of market contracts The American Economic Review 1993 83 4 811 837 Mehrotra, S., and Parida, J. (2019). The Employment Crisis: Rising Educational Levels and Falling Non-agricultural Job Growth. CSE Working Paper No. 23. Bangalore: Azim Premji University. Mehrotra S Gandhi A Sahoo BK Estimating India's Skill Gap: On a Realistic Basis for 2022 Economic and Political Weekly 2013 48 13 102 111 Mincer J On-the-job training: Costs, returns, and some implications Journal of Political Economy 1962 70 5 50 79 10.1086/258725 Mitra, A., and Verick, S. (2013). Youth employment and unemployment: an Indian perspective. ILO Asia- Pacific Working Paper. New Delhi: ILO DWT for South Asia and ILO Country Office for India. Obhan, A., and Bhalla, B. (2020). “India: Suspension of labour laws amidst Covid-19” Mondaq (blog), 18 May 2020. https://www.mondaq.com/india/employment-and-workforce-wellbeing/935398/suspension-of-labour-laws-amidst-covid-19. Papola, T. S. (2014). An assessment of the labour statistics system in India, New Delhi: International Labour Organization. Qu, D., Guo, S., & Wang, L. 2019. Experience, tenure, and gender wage difference: evidence from China. Economic research-Ekonomska istraživanja 32 (1): 1169–1184. Schultz TW Investment in human capital The American Economic Review 1961 51 1 1 17 Sen K Saha B Maiti D Trade, labour institutions and flexibility in Indian manufacturing Labour Economics 2013 24 180 195 10.1016/j.labeco.2013.08.008 Sharma S Skill building & employment in India: Interrogating an uneasy relationship Indian Journal of Industrial Relations 2019 55 2 205 216 Singh J Das DK Abhishek K Kukreja P Law, skills and the creation of jobs as ‘contract’ work in India: Exploring survey data to make inferences for labour law reform The Indian Journal of Labour Economics 2017 60 4 549 570 10.1007/s41027-018-0113-8 Singh J Das DK Abhishek K Kukreja P Factors influencing the decision to hire contract labour by Indian manufacturing firms Oxford Development Studies 2019 47 4 406 419 10.1080/13600818.2019.1624705 Singh S Parida JK Awasthi IC Employability and earning differentials among technically and vocationally trained youth in India The Indian Journal of Labour Economics 2020 63 363 386 10.1007/s41027-020-00222-9 Singh, J., D.K Das, K. Abhishek and P. Kukreja, (2019a). “Exploring the pattern of trade union activity in the Indian manufacturing sector. In K.R. Shyam Sundar (ed.) Globalization, Labour Market Institutions, Processes and Policies in India (pp. 87–107). Singapore: Palgrave Macmillan Sulis G Wage returns to experience and tenure for young men in Italy Scottish Journal of Political Economy 2014 61 5 559 588 10.1111/sjpe.12058 Tang H Labor market institutions, firm-specific skills, and trade patterns Journal of International Economics 2012 87 2 337 351 10.1016/j.jinteco.2012.01.001 Topel R Specific capital, mobility, and wages: Wages rise with job seniority Journal of Political Economy 1991 99 1 145 176 10.1086/261744	0	PMC9734441	NO-CC CODE	2022-12-15 23:15:20	no	Indian J Labour Econ. 2022 Dec 5; 65(4):1007-1028	utf-8	Indian J Labour Econ	2,022	10.1007/s41027-022-00408-3	oa_other
==== Front Mol Biol Rep Mol Biol Rep Molecular Biology Reports 0301-4851 1573-4978 Springer Netherlands Dordrecht 36478297 8087 10.1007/s11033-022-08087-5 Review The alternative renin-angiotensin-system (RAS) signalling pathway in prostate cancer and its link to the current COVID-19 pandemic Sehn Fabian [email protected] 12 Büttner Hartwig [email protected] 2 Godau Beate [email protected] 1 Müller Marten [email protected] 1 Sarcan Semih [email protected] 1 Offermann Anne [email protected] 3 Perner Sven [email protected] 34 Kramer Mario W. [email protected] 1 Merseburger Axel S. [email protected] 1 Roesch Marie C. [email protected] 1 1 grid.412468.d 0000 0004 0646 2097 Department of Urology, University Hospital Schleswig-Holstein, Campus Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany 2 Takeda Pharma Vertrieb GmbH und Co. KG, Jägerstrasse 27, 10117 Berlin, Germany 3 grid.412468.d 0000 0004 0646 2097 Institute of Pathology, University Hospital Schleswig-Holstein, Campus Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany 4 grid.418187.3 0000 0004 0493 9170 Research Center Borstel, Leibniz Lung Center, Pathology, Parkallee 1-40, 23845 Borstel, Germany 7 12 2022 18 20 5 2022 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. Background The renin-angiotensin system is known to maintain blood pressure and body fluids. However, it has been found to consist of at least two major constituents, the classic and the alternative pathway, balancing and supporting each other’s signalling in a very intricate way. Current research has shown that the renin-angiotensin system is involved in a broad range of biological processes and diseases, such as cancer and infectious diseases. Methods and results We conducted a literature review on the interaction of the renin-angiotensin system and prostate cancer and explored the research on the possible impact of the SARS-CoV-2 virus in this context. This review provides an update on contemporary knowledge into the alternative renin-angiotensin system, its role in cancer, specifically prostate cancer, and the implications of the current COVID-19 pandemic on cancer and cancer care. Conclusion In this work, we aim to demonstrate how shifting the RAS signalling pathway from the classic to the alternative axis seems to be a viable option in supporting treatment of specific cancers and at the same time demonstrating beneficial properties in supportive care. It however seems to be the case that the infection with SARS-CoV-2 and subsequent impairment of the renin-angiotensin-system could exhibit serious deleterious long-term effects even in oncology. Keywords Renin-angiotensin system Prostate cancer SARS-CoV-2 MAS1 Ang (1–7) ==== Body pmcIntroduction Often in physiology, seemingly simple enzymatic systems can regulate a host of pleiotropic and frequently contradictory actions. The renin-angiotensin system (RAS) for instance, has long time been studied for its important role in cardiovascular disease, and various pharmacological intervention principles have been identified to achieve the desired medical effects, such as lowering blood pressure. These include well-established treatment-modalities with angiotensin-converting-enzyme (ACE)-antagonists, selective angiotensin (AT) I antagonists, and diuretics [1]. The RAS itself can be subdivided into the classic and the alternative RAS (see Fig. 1). Fig. 1 Classic and alternative RAS signalling; ACE angiotensin converting enzyme, AGT angiotensinogen, Ang Angiotensin, AT1R angiotensin II type 1 receptor, AT2R angiotensin II type 2 receptor, Ala Alamandine, MAS1 Mas-receptor, MrgD G-protein-coupled receptor Mas, REN renin The alternative RAS represents a powerful antagonist to the classic RAS, counteracting its effects on blood pressure, fluid and electrolyte homeostasis. Besides these well-known roles, a growing body of evidence has linked the RAS with the pathogenesis of cancer. However, in the same pleiotropic manner as its role in controlling vascular functions, the RAS signalling cascade exerts its actions both pro- and antineoplastic [2]. Recent interest in this enzymatic system sparked with the discovery that several coronaviruses utilize one of its central constituents, the enzyme ACE II, as their entry point by binding to it via spike proteins [3]. The binding to and resulting depletion of ACE II is tilting the balance of this intricate system towards the classic RAS at the expense of the alternative pathway with potentially pathogenic consequences. This fostered research into the effects of established medications and repurposing medicines modulating this pathway [3]. To date, however, the data on this effort has been inconclusive. First shown after the SARS outbreak of 2003, survivors exhibit increased susceptibility to infections, cardiovascular complications as well as an increased occurrence of various tumour types [4]. As SARS-CoV-2 utilizes the same underlying principle, it is to be expected that similar observations can be made in the aftermath of the current pandemic. Not only does the virus and its management impose logistical burdens, but the aforementioned tilting of the RAS balance may have a serious impact on the occurrence and progression of cancer. Possible mechanistic pathways have been investigated and described by Saha and Anirvan [5]. Prostate cancer itself is a severe disease, being the second most common type of cancer in men worldwide and is among the five leading causes of cancer death [6]. Primary treatment consists of either radical prostatectomy or radiation therapy, however a relevant proportion of patients develop metastases. Many different treatment options exist in the metastatic situation. The choice of treatment is dependent on multiple factors such as molecular and pathologic features of the cancer, localisation and number of metastases, and patient’s choice. Nevertheless, androgen deprivation therapy still is the mainstay of management of advanced and metastatic prostate cancer. Because of the interplay between the RAS and the androgen receptor pathway [7], impacting the androgen receptor pathway should raise concerns about interfering with the RAS. A comprehensive picture on the mechanistic pathways linking cardiovascular disease and androgen deprivation has not yet been delineated, however, this seems to be a multi-facetted problem [8]. The possible interaction does not work one way only, as it has been shown that the mortality rate due to SARS-CoV-2 infection is significantly higher in prostate cancer (PCa) patients than in a cohort of male patients with any other malignancy [9]. On the other hand, patients receiving androgen depriving therapy exhibited a significantly lower risk of SARS-CoV-2 infection [10]. In this review, we present a summary of the contemporary knowledge of the role of RAS in PCa with a focus on the alternative RAS. The classic RAS signalling cascade Since the classic pathway of the RAS has been known for much longer, the bulk of research focussed on it, its constituents and their medicinal manipulation. In our opinion, current knowledge on the classic pathway has been summed up excellently by other authors [11, 12], so we quickly touch on this pathway only insofar as it relates to PCa. Angiotensin (Ang) II, besides being a major constituent of the vascular homeostasis, can also activate signal cascades that affect classic hallmarks of cancer, such as inflammation, proliferation and angiogenesis [13]. Experiments in vitro and in mouse xenografts have shown that the AT1 receptor (AT1R) transduces its signalling cascade through mitogen-activated protein kinase (MAPK) and signal transducer and activator of transcription 3 (STAT3), so that blockade of AT1R leads to inhibition of both androgen-dependent and independent PCa cells activated by the epidermal growth factor (EGF) [12]. Consequentially, administration of selective AT1R inhibitors lead to inhibited growth and vascularization of PCa xenografts [12, 14]. Inhibition of AT1R has also been shown to improve vascular perfusion of cancers and thereby possibly increases delivery of chemotherapy [15]. These effects appear to be complementary to those of dedicated VEGF-directed drugs and further evaluation clearly is warranted. It also seems like angiotensin II and III may modify the expression of their corresponding receptors in PCa cell lines, most notably angiotensin II type 2 receptor (AT2R) [14]. Further exploration showed that the AT1 receptor is upregulated in most human PCa tissues versus normal prostate tissues [12]. Additionally, high AT1R expression negatively correlated with survival in patients with ovarian cancer [16]. This naturally posed the question, whether medicinal interference with already approved substances might have a positive impact on prognosis. An analysis of the Finnish Cancer Registry revealed that post-diagnostic use of angiotensin receptor blockers decreased the risk for death from PCa irrespective of Gleason score, risk group or metastatic state [17]. However, the use of antihypertensive drugs such as ACE inhibitors, β-blockers and diuretics were associated with a slightly increased risk for the development of PCa in general [18]. Moreover, it has been shown that Ang II can boost the migration tendency of aggressive PCa cells [14], thus potentially speeding up metastasis. This is supported by the observation that in many cancers, interference with the Ang II –AT1R axis shows its beneficial effects especially in recurrence and distant metastases [19]. The interplay of Ang II and other pathways, such as relaxin 2, also seems crucial in remodelling both the tumour microenvironment as well as the cancer cells itself by facilitating the transition from the androgen-dependent to the androgen-independent phenotype via modulation of the expression of androgen receptors [19]. The alternative RAS signalling cascade The alternate signalling pathway, describing the actions of Ang (1–7) mediated through the G-protein coupled receptor MAS1 [20], has also been studied in the light of its cardiovascular effects. However, Ang (1–7) does not only bind to MAS1, interaction with constituents of the classic pathway, AT1R and AT2R, have also been shown, yet on a much smaller scale [21]. This is further emphasised by the fact that some of its cardiovascular effects can be nullified by both AT1R- and AT2R-inhibition [22]. In addition to MAS1, Ang (1–7) has been found to specifically bind to another member of the Mas-receptor family: MAS-related G protein-coupled receptor member D (MrgD) [23]. While MAS-deficient mice suffered from endothelial dysfunction under cardiovascular stress [24], they did not show a strong phenotype under physiological conditions [25], suggesting that signalling of Ang (1–7) may at least in part be compensated for by MrgD [23]. Ang (1–7) also, however, exhibits effects on cancer. While the role of the classic RAS seems to be well understood, the implications of Ang (1–7) are contradictory in different organ systems: Despite the fact it has been shown to exert anti-neoplastic properties in breast [26], colon [27] lung [28], pancreatic [29] and PCa [30], pro-neoplastic effects have been discovered in renal cell carcinoma [31, 32]. Similarly to the classic pathway, constituents of the alternative pathway are differentially expressed in neoplastic versus non-neoplastic tissues. ACE II expression has been shown to be downregulated in non-small-cell lung cancer (NSCLC) [28], hepatocellular carcinoma (HCC) [33], breast cancer [34], pancreatic cancer [35] and gallbladder cancer cells [36]. It is debatable whether the levels of ACE II expression can be utilized as prognostic marker, as it has been shown that these in fact correlate with outcome, at least in HCC and breast cancer [33, 37]. In addition to their cellular experiments, Zhang et al. also performed retrospective analyses on the expression of ACE II in human breast cancer samples [37]. They could show that ACE II expression in cancerous tissues was significantly decreased compared to normal tissue and patients with higher ACE II expression exhibited a better prognosis with regard to relapse free survival than those with lower expression (HR = 0.81). These findings have been confirmed in clear cell renal cell carcinoma (ccRCC), in addition to showing that Ang (1–7) might be the mediator of these effects [38]. Interestingly, treatment of ccRCC with small-molecule inhibitors of vascular endothelial growth factor receptor (VEGFR-TKI) decreased expression of ACE2. This was ameliorated by the combined administration of VEGFR-TKI and Ang (1–7), which resulted in further suppressed tumour growth and improved survival. Given the fact that TKI targeting the VEGFR are in widespread use e.g. in RCC management, this combination should be further evaluated. The receptor of Ang (1–7), MAS1, can be upregulated in colon cancer [27]. It has been discussed, however, that this is not a marker of oncogenic processes itself, but rather part of a general activation of the RAS in cancer [27]. Further data from breast and oesophageal cancer supports this notion, as upregulation of MAS1 has been shown to represent a positive prognostic marker [39]. However, siRNA mediated knockout of MAS1 increased proliferation of osteosarcoma cells [40]. Taken together, most of the evidence suggests that the classic RAS plays a pro-neoplastic role in the development of cancers that can (partly) be ameliorated by the alternative RAS pathway. Both for its effects on the vascular system and cancer itself, Ang (1–7) has been investigated as a medication in trials in cancer patients. It was found to exert positive effects on multilineage cytopenia as well as demonstrating clinical benefit and shrinking of some tumour types [41]. These results have also been observed in PCa, albeit in mouse models only [42]. Stimulation of MAS1 with a selective agonist has been found to exhibit favourable effects, too, such as delaying the development of cancers and mitigating cancer cachexia [43]. A summary of selected experiments on the role of the alternative RAS can be found in Table 1. Another receptor also capable of binding Ang (1–7), MrgD, was found upregulated in lung cancer tissue and its main ligand, beta-alanine, enhanced spheroid formation in vitro [44]. It is unclear, however, whether these affects are solely contributable to beta-alanine and whether Ang (1–7) results in a different signalling. The intracellular signal transduction cascade of MAS1 as shown below (see Fig. 2) follows both the MAPK/ERK and the AKT/mTOR signalling cascades. Fig. 2 Downstream signalling cascade from MAS1; AKT protein kinase B, ERK Extracellular signal-regulated kinase, MAPK mitogen-activated protein kinase, mTOR mammalian target of rapamycin, PKD1 protein kinase D1, PIP3 Phosphatidylinositol (3,4,5)-trisphosphate, Raf rapidly accelerated fibrosarcoma, Ras rat sarcoma kinase Table 1 Effects of the alternative RAS signalling pathway on cancer Cancer Model Manipulation Results References Breast In vitro ACE II overexpression ↓ Proliferation ↓ Vascularisation ↓ Migration ↓ Metastasis ↓ pVEGFR2 [37] Breast In vivo MAS1 analysis MAS1 low: ↑ Tumour growth ↑ Metastasis ↑ Grading [39] Pancreas Murine MAS1 stimulation ↓ Muscle atrophy ↓ Weight loss ↑ Locomotor activity ↓ Tumour development [43] Colon In vivo MAS1 analysis ↑ MAS1 expression in neoplastic tissue [27] Lung Murine Addition of Ang (1–7) ↓Tumour growth ↓ Vessel density ↓ VEGF [45] Lung In vitro Addition of Ang (1–7) ↓ Migration ↓ Invasion ↓ MMP-2 & -9 [46] Nasopharynx Murine Addition of Ang (1–7) ↓ Tumour growth ↓ Vessel density ↓ VEGF/↓ PlGF [47] Nasopharynx Murine Addition of Ang (1–7) ↑ Autophagy [48] Prostate Murine Addition of Ang (1–7) ↓ Metastasis ↓ VEGF ↓ Tumour growth ↓ Osteoclastogenesis [30] Prostate Murine Addition of Ang (1–7) ↓ Tumour growth ↓ Proliferation ↓ Intratumoral vessel density ↓ VEGF/↓ PlGF/↑ sFlt-1 [42] Prostate In vitro Addition of Ang (1–7) ↓ Proliferation ↓ Apoptosis ↑ AT2R / ↑MAS1 ↓ ESR1 / ↑ ESR2 Modulation of NF-kB Modulation of IKK Modulation of MMP-2 & -9 [49] Prostate In vitro Addition of Ang (3–7) Addition of Ang (1–9) ↑ Growth ↑ Mobility ↑ ESR1/↑ ESR2 ↓ Colony size ↑ AR expression (in PC3) [50] AR androgen receptor, ESR oestrogen receptor, IKK IκB kinase, MMP Matrix metallopeptidase, NF-kB Nuclear Factor kappa-light-chain-enhancer of activated B cells, PC3 human prostate cancer cell line, PlGF placental growth factor, pVEGFR2 phosphorylated receptor 2 of the vascular endothelial growth factor, sFlt-1 soluble fms-like tyrosine kinase-1, VEGF vascular endothelial growth factor Prostate cancer within the context of the current COVID-19 pandemic As it has recently been found that SARS-CoV-2 enters its host cells by ACE II in a TMPRSS2 (transmembrane protease serine subtype 2)-dependent manner [3], implications on the functioning of the RAS, especially the alternative RAS were expected. While the expression of TMPRSS2 is influenced by androgens, data on its manipulation are contradictory at best. Mechanistically one could argue that downregulation of the receptor ACE II induced by androgen-depriving therapy (ADT) for instance, may have a protective effect against the virus. While this was found to be the case in one study [10], another study demonstrated different, contrary results [51]. This could be in part due to population differences among the subjects observed, as well as the clinical settings and thus needs confirmation. Additionally, it was found that lower testosterone plasma levels are able to predict worse clinical outcome of the infection, suggesting a more favourable role of androgens during the course of the infection [52]. Many hypotheses have been put forward to explain the increased infection rates of males versus females [53], but most are speculative so far. The overall differential expression of TMPRSS2 between males and females for instance has been found to be very similar in lung tissue and one common driver of PCa, the fusion of TMPRSS2 and ERG has so far not been found to elicit any differential infection patterns or influence the severity of the infection. The influence of this fusion gene in prostate cancer therefore might warrant further exploration. The depletion of ACE II upon SARS-CoV-2 infection [54] is expected to mitigate the alternative RAS signalling, thereby skewing the resulting signal more in favour of the potentially deleterious classic RAS pathway [5]. The resulting effects have been studied in the aftermath of the 2003 epidemic of the very similar SARS-CoV with the consequential spike of incidences of various cancers and other inflammatory diseases [55]. Many open questions still remain and are worth pursuing. Summary The bulk of the evidence so far shows that the classic RAS exhibits pro-tumorigenic properties, while the alternative RAS mostly displays anti-tumorigenic ones. Even though some mechanistical questions remain, RAS has been identified as a feasible drug target not only in cardiovascular disease, but in cancer as well [56]. As aldosterone is the final player in the RAAS, regulating not only blood pressure and volume, but also triggering migratory stimuli on hormone-dependent PCa cells [57] it again demonstrates the importance of this pathway in the pathology of PCa. This conclusion is not new, as it has been discouraged to use drugs influencing the aldosterone axis in PCa before [58]. Even though Ang II signalling can trigger apoptosis [59] and signalling through MAS1 seems to prevent it [49], the deleterious effects of the classic RAS signalling cascade appear to outweigh those of the alternative pathway on cancer by far. Taken together, skewing the RAS signalling pathway from the classic to the alternative axis seems to be a viable option if not in preventing, but at least in treating specific cancers and at the same time demonstrating beneficial properties in supportive care (see Table 1; Fig. 3). Fig. 3 Interplay of AT1R, MAS1 and AT2R In the light of the current pandemic situation, cancer patients are among the most vulnerable and therefore careful consideration needs to be placed on not only how to protect them from the disease, the impact of their medication on and susceptibility to the virus but also on the possibility of Covid-19 convalescents later becoming cancer patients themselves [4, 5]. Further research on the influence of RAS on PCa and its modulation is warranted. This in fact can lead to an improved molecular understanding and novel, personalised treatment options. Lung cancer to date is at the forefront of targeted approaches to specific mutations, but it has been shown that PCa, too, yields various, sometimes already actionable mutations. Already, survival with PCa has been improved by the introduction of novel therapeutics [60], with a better molecular understanding of the pathogenesis and development of resistance, further improvement is to be expected. Conclusion While much has been investigated into the role of the classic RAS signalling pathway, data on the interplay between the alternative RAS and PCa still is fragmentary. Taken together, while a SARS-CoV-2 infection poses a clear and acute danger to cancer patients in general, possible short term and long-term effects due to the interplay of SARS-CoV-2 with the RAS might be a danger for cancer patients as well. We suggest to further explore the oncogenic role of the classic RAS and alternative RAS, in particular with a focus on the androgen signalling pathway in prostate cancer. In light of the current pandemic a better understanding of the classic RAS and alternative RAS will improve both SARS-CoV-2 management and oncologic care. Author contributions FS: Data collection and management, literature research, data analysis, manuscript writing and editing. HB, MM, SS, AO, SP, MWK: Manuscript editing. BG: Data collection. ASM: Idea, manuscript editing, critical revision. MCR: Idea, manuscript writing and editing, data collection. Funding The authors did not receive support from any organization for the submitted work. Declarations Conflict of interest Fabian Sehn, Hartwig Büttner: Part-time employee of Takeda; the present publication has been prepared independently and outside of the employment; the employer is not involved in any of subjects dealt with in this publication and did not provide any form of support. Sven Perner: Consultant: Ventana Medical Systems, Roche, Novartis, Astellas, Bristol-Myers Squibb, MetaSystems, Merck Serono, MSD. Research: Ventana Medical Systems, Roche, Bristol-Myers Squibb, MSD, Boehringer-Ingelheim. Lectures/Speaker/Honoraria: Ventana Medical Systems, Roche, Novartis, Astellas, Bristol-Myers Squibb, MetaSystems, Merck Serono, MSD. Mario W. Kramer: Honoraria/Consultation: Bayer, BMS, Eusai, Novartis, Merck, MSD, Pfizer, Roche. Travel Grants: Ipsen, Janssen, Merck, Novartis. Axel S. Merseburger: Lectures/Speaker/Honoraria: AstraZeneca, Bristol-Myers Squibb, Eisai, Ipsen, MSD, Merck Serono, Janssen, Takeda, TEVA, Astellas, Novartis, Pfizer und Roche. Consultant: AstraZeneca, Astellas, Bristol-Myers Squibb, Ipsen, Janssen, EUSAPharm, MSD, Merck Serono, Novartis, Takeda, Teva, Pfizer und Roche. Research and clinical trials: AstraZeneca, Astellas, Bristol-Myers Squibb, Ipsen, Janssen, EUSAPharm, MSD, Merck Serono, Novartis, Takeda, Teva, Pfizer und Roche. Marie C. Roesch: Lectures/Honoraria/Travel Grants: Ipsen, Novartis, Sanofi, Solution Akademie. Beate Godau, Marten Müller, Semih Sarcan , Anne Offermann declare that they have no conflict of interest. Research involving human and/or participants No human nor animal subjects have been involved in the presented publication. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Ballew JR Fink GD Characterization of the antihypertensive effect of a thiazide diuretic in angiotensin II-induced hypertension J Hypertens 2001 19 9 1601 1606 10.1097/00004872-200109000-00012 11564980 2. Pinter MJ Rakesh K Targeting the renin-angiotensin system to improve cancer treatment: Implications for immunotherapy Sci Transl Med 2017 9 410 5616 10.1126/scitranslmed.aan5616 3. Hoffmann M SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor Cell 2020 181 2 271 280e8 10.1016/j.cell.2020.02.052 32142651 4. Wu Q Altered lipid metabolism in recovered SARS patients twelve years after infection Sci Rep 2017 7 1 9110 10.1038/s41598-017-09536-z 28831119 5. Saha A Anirvan P Cancer progression in COVID-19: integrating the roles of renin angiotensin aldosterone system, angiopoietin-2, heat shock protein-27 and epithelial mesenchymal transition Ecancermedicalscience 2020 14 1099 10.3332/ecancer.2020.1099 33082849 6. Bray F Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries CA 2018 68 394 4246 30207593 7. Mishra JS Hankins GD Kumar S Testosterone downregulates angiotensin II type-2 receptor via androgen receptor-mediated ERK1/2 MAP kinase pathway in rat aorta J Renin Angiotensin Aldosterone Syst 2016 10.1177/1470320316674875 8. Melloni C Roe MT Androgen deprivation therapy and cardiovascular disease Urol Oncol 2020 38 2 45 52 10.1016/j.urolonc.2019.02.010 30879969 9. Chakravarty D Sex differences in SARS-CoV-2 infection rates and the potential link to prostate cancer Commun Biol 2020 3 1 374 10.1038/s42003-020-1088-9 32641750 10. Montopoli M Androgen-deprivation therapies for prostate cancer and risk of infection by SARS-CoV-2: a population-based study (N = 4532) Ann Oncol 2020 31 8 1040 1045 10.1016/j.annonc.2020.04.479 32387456 11. George AJ Thomas WG Hannan RD The renin–angiotensin system and cancer: old dog, new tricks Nat Rev Cancer 2010 10 11 745 759 10.1038/nrc2945 20966920 12. Uemura H Angiotensin II receptor blocker shows antiproliferative activity in prostate cancer cells: a possibility of tyrosine kinase inhibitor of growth factor Mol Cancer Ther 2003 2 11 1139 1147 14617787 13. Haendeler J Dimmeler S Unger T Schölkens BA Regulation of angiogenesis by angiotensin II 2004 Berlin Heidelberg: Berlin, Heidelberg Springer 99 109 14. Domińska K Similarities and differences between effects of angiotensin III and angiotensin II on human prostate cancer cell migration and proliferation Peptides 2012 37 2 200 206 10.1016/j.peptides.2012.07.022 22884921 15. Chauhan VP Angiotensin inhibition enhances drug delivery and potentiates chemotherapy by decompressing tumour blood vessels Nat Commun 2013 4 2516 10.1038/ncomms3516 24084631 16. Zhang Q Angiotensin II promotes ovarian cancer spheroid formation and metastasis by upregulation of lipid desaturation and suppression of endoplasmic reticulum stress J Exp Clin Cancer Res 2019 38 1 116 10.1186/s13046-019-1127-x 30845964 17. Siltari A Antihypertensive drug use and prostate cancer-specific mortality in Finnish men PLoS ONE 2020 15 6 e0234269 10.1371/journal.pone.0234269 32598349 18. Smith L Cross-sectional associations between angiotensin-converting enzyme inhibitor use and cancer diagnosis in US adults Clin Exp Med 2020 20 3 409 416 10.1007/s10238-020-00622-7 32219665 19. Domińska K Interaction between angiotensin II and relaxin 2 in the progress of growth and spread of prostate cancer cells Int J Oncol 2016 48 6 2619 2628 10.3892/ijo.2016.3458 27035428 20. Chappell MC Emerging evidence for a functional angiotensin-converting enzyme 2-angiotensin-(1–7)-MAS receptor axis: more than regulation of blood pressure? Hypertension 2007 50 4 596 599 10.1161/HYPERTENSIONAHA.106.076216 17785634 21. Bosnyak S Relative affinity of angiotensin peptides and novel ligands at AT1 and AT2 receptors Clin Sci (Lond) 2011 121 7 297 303 10.1042/CS20110036 21542804 22. Jiang F Angiotensin-converting enzyme 2 and angiotensin 1–7: novel therapeutic targets Nat Rev Cardiol 2014 11 7 413 426 10.1038/nrcardio.2014.59 24776703 23. Tetzner A et al (2016) G-protein-coupled receptor MrgD is a receptor for angiotensin-(1–7) involving adenylyl cyclase, cAMP, and phosphokinase. Hypertension, 68(1), 185 24. Peiró C Endothelial dysfunction through genetic deletion or inhibition of the G protein-coupled receptor mas: a new target to improve endothelial function J Hypertens 2007 25 12 2421 2425 10.1097/HJH.0b013e3282f0143c 17984663 25. Walther T Sustained long term potentiation and anxiety in mice lacking the mas protooncogene J Biol Chem 1998 273 19 11867 11873 10.1074/jbc.273.19.11867 9565612 26. Zhao Y Angiotensin II/angiotensin II type I receptor (AT1R) signaling promotes MCF-7 breast cancer cells survival via PI3-kinase/Akt pathway J Cell Physiol 2010 225 1 168 173 10.1002/jcp.22209 20458733 27. Bernardi S Characterization and significance of ACE2 and Mas receptor in human colon adenocarcinoma J Renin Angiotensin Aldosterone Syst 2012 13 1 202 209 10.1177/1470320311426023 22048948 28. Feng Y The angiotensin-converting enzyme 2 in tumor growth and tumor-associated angiogenesis in non-small cell lung cancer Oncol Rep 2010 23 4 941 948 20204277 29. Zhou L Angiotensin-converting enzyme 2 acts as a potential molecular target for pancreatic cancer therapy Cancer Lett 2011 307 1 18 25 10.1016/j.canlet.2011.03.011 21481527 30. Krishnan B Angiotensin-(1–7) attenuates metastatic prostate cancer and reduces osteoclastogenesis Prostate 2013 73 1 71 82 10.1002/pros.22542 22644942 31. Zheng S Ang-(1–7) promotes the migration and invasion of human renal cell carcinoma cells via Mas-mediated AKT signaling pathway Biochem Biophys Res Commun 2015 460 2 333 340 10.1016/j.bbrc.2015.03.035 25783053 32. Burns WC Angiotensin II mediates epithelial-to-mesenchymal transformation in tubular cells by ANG 1–7/MAS-1-dependent pathways Am J Physiol Renal Physiol 2010 299 3 F585 F593 10.1152/ajprenal.00538.2009 20554647 33. Ye G The association of renin-angiotensin system genes with the progression of hepatocellular carcinoma Biochem Biophys Res Commun 2015 459 1 18 23 10.1016/j.bbrc.2015.02.030 25701390 34. Yu C Downregulation of ACE2/Ang-(1–7)/Mas axis promotes breast cancer metastasis by enhancing store-operated calcium entry Cancer Lett 2016 376 2 268 277 10.1016/j.canlet.2016.04.006 27063099 35. Zhou L Decreased expression of angiotensin-converting enzyme 2 in pancreatic ductal adenocarcinoma is associated with tumor progression Tohoku J Exp Med 2009 217 2 123 131 10.1620/tjem.217.123 19212105 36. Zong H Loss of angiotensin-converting enzyme 2 promotes growth of gallbladder cancer Tumour Biol 2015 36 7 5171 5177 10.1007/s13277-015-3171-2 25663464 37. Zhang Q ACE2 inhibits breast cancer angiogenesis via suppressing the VEGFa/VEGFR2/ERK pathway J Experimental Clin Cancer Res 2019 38 1 173 10.1186/s13046-019-1156-5 38. Khanna P ACE2 abrogates tumor resistance to VEGFR inhibitors suggesting angiotensin-(1–7) as a therapy for clear cell renal cell carcinoma Sci Transl Med 2021 13 577 170 10.1126/scitranslmed.abc0170 39. Luo Y Expression of MAS1 in breast cancer Cancer Sci 2015 106 9 1240 1248 10.1111/cas.12719 26080617 40. Lässig F Expression of canonical transient receptor potential channels in U-2 OS and MNNG-HOS osteosarcoma cell lines Oncol Lett 2021 21 4 307 10.3892/ol.2021.12568 33732383 41. Gallagher PE Angiotensin-(1–7): a peptide hormone with anti-cancer activity Curr Med Chem 2014 21 21 2417 2423 10.2174/0929867321666140205133357 24524765 42. Krishnan B Angiotensin-(1–7) reduces proliferation and angiogenesis of human prostate cancer xenografts with a decrease in angiogenic factors and an increase in sFlt-1 Prostate 2013 73 1 60 70 10.1002/pros.22540 22644934 43. Murphy KT Mas receptor activation slows tumor growth and attenuates muscle wasting in cancer Cancer Res 2019 79 4 706 719 10.1158/0008-5472.CAN-18-1207 30420474 44. Nishimura S MRGD, a MAS-related G-protein coupled receptor, promotes tumorigenisis and is highly expressed in lung cancer PLoS ONE 2012 7 6 e38618 10.1371/journal.pone.0038618 22715397 45. Soto-Pantoja DR Angiotensin-(1–7) inhibits tumor angiogenesis in human lung cancer xenografts with a reduction in vascular endothelial growth factor Mol Cancer Ther 2009 8 6 1676 1683 10.1158/1535-7163.MCT-09-0161 19509262 46. Chen X AAV-Mediated angiotensin 1–7 overexpression inhibits tumor growth of lung cancer in vitro and in vivo Oncotarget 2017 8 1 354 363 10.18632/oncotarget.13396 27861149 47. Pei N Angiotensin-(1–7) decreases cell growth and angiogenesis of human nasopharyngeal carcinoma xenografts Mol Cancer Ther 2016 15 1 37 47 10.1158/1535-7163.MCT-14-0981 26671566 48. Lin YT Pre-treatment with angiotensin-(1–7) inhibits tumor growth via autophagy by downregulating PI3K/Akt/mTOR signaling in human nasopharyngeal carcinoma xenografts J Mol Med (Berl) 2018 96 12 1407 1418 10.1007/s00109-018-1704-z 30374682 49. Domińska K Influence and mechanism of Angiotensin 1–7 on biological properties of normal prostate epithelial cells Biochem Biophys Res Commun 2018 502 1 152 159 10.1016/j.bbrc.2018.05.138 29802847 50. Domińska K The impact of Ang-(1–9) and Ang-(3–7) on the biological properties of prostate cancer cells by modulation of inflammatory and steroidogenesis pathway genes Int J Mol Sci 2020 21 17 6227 10.3390/ijms21176227 32872192 51. Caffo O On the relationship between androgen-deprivation therapy for prostate cancer and risk of infection by SARS-CoV-2 Ann Oncol 2020 31 10 1415 1416 10.1016/j.annonc.2020.06.005 32562741 52. Rastrelli G Low testosterone levels predict clinical adverse outcomes in SARS-CoV-2 pneumonia patients Andrology 2020 9 88 98 10.1111/andr.12821 32436355 53. Mjaess G COVID-19 and the male susceptibility: the role of ACE2, TMPRSS2 and the androgen receptor Prog Urol 2020 30 10 484 487 10.1016/j.purol.2020.05.007 32620366 54. Banu N Protective role of ACE2 and its downregulation in SARS-CoV-2 infection leading to macrophage activation syndrome: therapeutic implications Life Sci 2020 256 117905 10.1016/j.lfs.2020.117905 32504757 55. Li W Angiotensin-converting enzyme 2 is a functional receptor for the SARS coronavirus Nature 2003 426 6965 450 454 10.1038/nature02145 14647384 56. Acconcia F The network of angiotensin receptors in breast cancer Cells 2020 9 6 1336 10.3390/cells9061336 32471115 57. Feldman R Aldosterone mediates a mineralocorticoid receptor-mediated increase in prostate cancer cell migration FASEB J 2020 34 S1 1 1 58. Dhondt B Abiraterone and spironolactone in prostate cancer: a combination to avoid Acta Clin Belg 2019 74 6 439 444 10.1080/17843286.2018.1543827 30477405 59. Ishida J Fukamizu A Angiotensin II and apoptosis Nihon Rinsho 1999 57 5 1117 1123 10361444 60. von Amsberg G Merseburger AS Treatment of metastatic, castration-resistant prostate cancer Urologe A 2020 59 6 673 679 32274540	36478297	PMC9734445	NO-CC CODE	2022-12-14 23:28:28	no	Mol Biol Rep. 2022 Dec 7;:1-8	utf-8	Mol Biol Rep	2,022	10.1007/s11033-022-08087-5	oa_other
==== Front J Child Adolesc Trauma J Child Adolesc Trauma Journal of Child & Adolescent Trauma 1936-1521 1936-153X Springer International Publishing Cham 503 10.1007/s40653-022-00503-z Original Article The Rising of the Shield hero. Development of the Post-Traumatic Symptom Questionnaire (PTSQ) and Assessment of the Protective Effect of self-esteem from trauma-related Anxiety and Depression http://orcid.org/0000-0001-7000-5999 Rossi Alessandro Alberto [email protected] 12 Panzeri Anna [email protected] 3 Taccini Federica [email protected] 24 Parola Anna [email protected] 5 Mannarini Stefania [email protected] 12 1 grid.5608.b 0000 0004 1757 3470 Department of Philosophy, Sociology, Education, and Applied Psychology, Section of Applied Psychology, University of Padova, Padova, Italy 2 grid.5608.b 0000 0004 1757 3470 Interdepartmental Center for Family Research, University of Padova, Padova, Italy 3 grid.5608.b 0000 0004 1757 3470 Department of General Psychology, University of Padova, Padova, Italy 4 grid.5608.b 0000 0004 1757 3470 Department of Developmental Psychology and Socialisation, University of Padova, Padova, Italy 5 grid.4691.a 0000 0001 0790 385X Department of Humanities, University of Naples Federico II, Naples, Italy 7 12 2022 119 18 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 Adverse life events such as life-threatening accidents, domestic and/or sexual violence, organic diseases (i.e., cancer), or COVID-19 can have a strong traumatic impact – generating reactions as intrusive thoughts, hyperarousal, and avoidance. Indeed, the traumatic impact of COVID-19 seems to lead individuals to experience anxiety and depression. However, the Anxiety-Buffer Hypothesis suggests that self-esteem could be considered a shield (buffer) against traumatic experiences and their outcomes (i.e., anxiety and depression). The present study has two objectives. First, to develop a measure of the impact of the traumatic event considering the aforementioned reactions. Second, to test the process – triggered by COVID19-related traumatic experience – in which self-esteem buffers the path that leads to anxiety and depression. Method In Study 1 (N = 353), the Post-Traumatic Symptom Questionnaire (PTSQ) was developed and a deep investigation of its psychometric properties was conducted. In Study 2 (N = 445), a structural equation model with latent variables was performed to assess the buffering effect of self-esteem. Results The PTSQ has excellent fit indices and psychometric properties. According to the ABH, results confirm the buffering effect of self-esteem in the relationships between traumatic symptoms and both anxiety and depression. Conclusion On the one hand, the PTSQ is a solid and reliable instrument. On the other hand, that self-esteem is a protective factor against anxiety and depression related to a traumatic experience – such as COVID-19. Targeted psychological interventions should be implemented to minimize the psychological burden of the illness while promoting adaptation and positive aspects of oneself. Supplementary Information The online version contains supplementary material available at 10.1007/s40653-022-00503-z. Keywords Trauma Anxiety buffer hypothesis terror management theory anxiety depression ==== Body pmcIntroduction Accidents endangering one’s own life, the tragic death of some of our loved ones, domestic and/or sexual violence, and serious organic diseases (e.g., cancer or COVID-19) are all potentially traumatic events that can have a dramatic psychological impact (Friedman et al., 2007; Rossi et al., 2022a; Thakur & Jain, 2020; Zhang et al., 2021). The impact of the traumatic event can be associated with a wide range of physical as well as psychological difficulties and disorders in adults, children, and adolescents (Briere & Spinazzola, 2005; Friedman et al., 2007; Kratovic et al., 2021). On the one hand, scientific literature showed that the impact of traumatic events can be so strong that can have a serious long-term influence on physical health (Sareen et al., 2007) in terms of poor quality of life (QOL), general health symptoms, general medical conditions, pain (e.g., musculoskeletal), cardio-respiratory symptoms, and gastrointestinal health (Brosschot et al., 2016; Yaribeygi et al., 2017; Zhang et al., 2021) – for a review (Pacella et al., 2013). On the other hand, the psychological impact of traumatic events can be so strong that can lead to developing post-traumatic stress symptoms (PTSS) and/or post-traumatic stress disorder (PTSD) – consisting in the late onset and persistence of mental disorders caused by experiencing, suffering or encountering one or more threats for themselves or others (Agaibi & Wilson, 2005) – even 10 years after its onset (Kessler et al., 1995). Moreover, individuals who have experienced a traumatic event are between 2 and 6 times more likely to present with psychiatric comorbidities, including anxiety, depression, suicidal ideation, and self-injury behaviors (Breslau & Davis, 1992; Breslau et al., 1991; Karatzias et al., 2019; Tessitore et al., 2022). Also, approximately 40% of individuals who have experienced a traumatic event continue to present relevant symptomatology (Djelantik et al., 2020; Kessler et al., 1995; Orcutt et al., 2002) – probably due to the presence of maladaptive psycho(physio-)logical mechanisms that contribute to the development and the maintenance of anxiety and depression symptoms. Indeed, the psychological impact of traumatic events often entails persistent reenactment of the event, avoidance symptoms, negative change in general reactivity, and increased arousal and reactivity (American Psychiatric Association, 2013). According to the terror management theory (TMT) (Greenberg et al., 1986), individuals’ awareness of mortality – elicited by a traumatic event (e.g., accidents, experienced violence, or severe organic disease) and the continuously dwelling – generates terrifying fears of death and thus anxiety and depression that constantly conflicts with humans’ intrinsic desire to live and their natural tendency to survive. (Rossi et al., 2020). Moreover, scientific literature has shown that adverse events can generate three main psycho(physio-)logical reactions/domains (Pacella et al., 2013; Yaribeygi et al., 2017) – intrusivity, hyperarousal, and avoidance – and all of them are associated with heightened anxiety reactions. Firstly, intrusivity refers to the intrusions of unwanted cognitive and/or sensorial stimuli recalling the traumatic event that re-expose the person to that feared event – and can thus generate an intense sense of anxiety as the original event did. Secondly, traumatic events can generate hyperarousal reactions, a state of chronic hypervigilance for potential threats and heightened arousal, that prompts the individual to amplify anxious reactions in response to (also neutral) stimuli, thus contributing to the development and the maintenance of anxiety as the anticipation of future (potential) threats (Barlow, 2002; Harding et al., 2008). Lastly, traumatic events can generate the need for avoidance – both cognitive and behavioral. Avoidance is a coping strategy that the individual implements to protect himself or herself from the possibility that the traumatic event may recur by avoiding all the trauma-related experiences – aiming at preventing further damage to an already weakened psychological structure (Bishop et al., 2018). However, even this coping strategy – when abused – often contributes to maintaining and amplifying dysfunctional anxiety states (Bishop et al., 2018). Consequently, these three main reactions/domains to traumatic events (overlapping with symptoms of PTSS and PTSD) seem to converge under the overarching factor related to the suffered trauma – the so-called ‘impact of the event’ – which makes the individual more prone to develop and maintain anxious symptomatology (Gagne et al., 2018). The negative/adverse event and the subsequent possible development of trauma can influence the way individuals think and behave – thus representing an important risk factor for mental health – making them more prone to develop anxious and depressive symptoms (Kessler et al., 2017). Furthermore, although traumatic events can lead to the development of depressive symptoms (as aforementioned), several research shows that also prolonged and chronic states of anxiety can lead to the development of depressive symptoms (Rossi et al., 2020) – by intensifying the negative expectations, negative repetitive thoughts (i.e., worry, rumination), pessimism, and negative feelings (Starr & Davila, 2012; Thompson et al., 2005). Such depressive symptomatology includes sadness, loss of positive feelings, and Beck’s negative triad (Beck, 1979) – consisting of a negative view of the self, the world, and the future (Rossi et al., 2021). Moreover, traumatic events-related depressive symptomatology should not be underrated due to its association with anticonservative ideation and attempts (Breslau & Davis, 1992; Breslau et al., 1991; Karatzias et al., 2019; McIntyre & Lee, 2020; Thakur & Jain, 2020). Therefore, the impact of a traumatic event (in addition to possible PTSS and PTDS) can lead the individual to develop conditions of intrusivity, hyperarousal, and avoidance which – in turn – contribute to the development and maintenance of anxiety states, which can lead to the development of depressive symptoms (Santini et al., 2020; Thakur & Jain, 2020). However, over the past 20 years, the literature has shown that certain psychological variables can play a key role in protecting – as a psychological shield – the individual from the negative consequences (i.e., anxiety and depression) of traumatic events (Benight, 2012). The anxiety-buffer hypothesis (ABH) (Greenberg et al., 1992) – from the TMT (Greenberg et al., 1986) – suggests that self-esteem has a shielding (buffering) effect on the relationship between the impact of traumatic events and both anxiety and depression (Benight, 2012; Rossi et al., 2020). More in detail, self-esteem – conceived as the beliefs and evaluations of individuals towards themselves, and the attitudes that derive from them – is based on personal values deeply rooted in the culture and social context of the individual from which one’s own personal value (Becker, 1971, 1973). Therefore, by respecting the standards of one’s own culture of belonging and worldview, self-esteem is strengthened both by the social validation of oneself and by the intimate and personal feeling of being an individual with a certain degree of value and who assumes a significant role in one’s society (Pyszczynski et al., 2004; Solomon et al., 2004). In this context, the ABH (Greenberg et al., 1992) hypothesizes that – by reconnecting the individual to an enlarged universe of purely individual and intimately personal meanings and values – self-esteem could act as a protective shield (buffer) against the damaging psychological effects of life threats and stressors. Consequently, self-esteem should buffer the negative consequences of traumatic events. Considering this background, a major traumatic event has hit the world: the COVID-19 pandemic (Brooks et al., 2020; Torales et al., 2020; Wang et al., 2020; Zhang et al., 2021). COVID-19 is a serious and potentially deadly infectious disease that threatened the entire world population, since-at the time of its outbreak-there was no preventive immunity or even a well-defined cure or vaccine (Baud et al., 2020); and it still causes deaths. According to scientific literature, COVID-19 represented a challenge to the physical and mental health of individuals and generated widespread post-traumatic reactions (Esterwood & Saeed, 2020; Fisher et al., 2020; Shevlin et al., 2021; Silver, 2020; Waseem et al., 2021). A recent review by Zhang and colleagues demonstrated that during the COVID-19 pandemic in all countries the general population developed PTSS to varying degrees, with a PTSD prevalence of 15% (Zhang et al., 2021). Moreover, it is important to highlight that even people not infected by COVID-19 still experienced high psychological stress (Zhang et al., 2021). In addition, scientific studies on the psychological impact of COVID-19 highlighted that specific categories of the population are more prone than others to develop psychological issues, in particular, individuals in the emerging adulthood phase also referred to as young adults (18–30 y.o.) (Parola, 2020; Parola et al., 2020). Indeed, in the last two years, scientific research has highlighted that in the emerging adulthood individuals present heightened levels of psychological difficulties, including symptoms of distress, anxiety, and depression (Panzeri et al., 2021; Parola et al., 2020). Importantly, emerging adulthood may expose to a higher risk of developing psychological issues because individuals in this part of the lifespan go through a process of individualization, structuring self-esteem, and personal growth (van den Berg et al., 2021) – in which they progressively rediscuss and strengthen their role in the society, relationships, and the professional field (Graupensperger et al., 2022). Thus, the trauma associated with the pandemic situation and its limitations may have disrupted this process, hence hindering the possibility to fulfill personal, professional, and relational wills (Cao et al., 2020; Cheng et al., 2014; Zhang et al., 2021). Taken together, all these evidence strongly advice to assessing and monitoring the psychological condition of individuals, in particular of those in the emerging adulthood phase. In light of this, the present study has two aims. First, to develop a self-report questionnaire measuring the impact of the traumatic event that specifically takes into account intrusivity, hyperarousal, and avoidance. Second, to test the hypothesis that if COVID-19 was an event that had a traumatic impact – especially on young adults (Parola, 2020; Parola et al., 2020) – that led to anxiety and depressive symptoms, then, according to the ABH, self-esteem (acting as a shield) should buffer the relationship from the impact of traumatic event to anxiety, which in turn should lead to depression. Study 1. Development of the Post-Traumatic Symptom Questionnaire (PTSQ) Materials and methods Sample Size Determination The subject-per-parameter ratio “n:q criterion” (subjects per free model parameter) was used to plan a priori the minimum number of subjects needed given the main statistical analyses of this study (see dedicated section). A ratio of 5 subjects per parameter was guaranteed (Brown, 2015; Hu & Bentler, 1999; Kline, 2016). Enrollment Procedure Using the snowball sampling method (Johnson, 2014), participants were recruited from the general population through advertisements on social media (e.g., Facebook, Twitter, etc.). Inclusion criteria were: (A) having experienced at least a traumatic event that the subject remembers well to this day, (B) being aged between 18 and 30 y.o., (C) being a native Italian speaker; (D) providing informed consent; (E) do not provide missing answers, (F) have no inability to complete the assessment procedure; and (G) do not complete the procedure in less than 4 or greater than 8 min. At the beginning of the survey, participants were asked if they had experienced a traumatic event that they remember well to this day. Based on the aforementioned inclusion/exclusion criteria, only those who had this characteristic went on to complete the survey and were enrolled in the study. Participants A sample of 412 subjects was eligible to complete the survey. However, 59 questionnaires were excluded due to missing data/answers (n = 36) and/or inappropriate completion times (< 4 or > 8 min; n = 23). The final sample included 353 participants: 111 males (31.4%) and 242 females (68.6%), aged from 18 to 30 y.o. (mean = 26.13, SD = 2.96) – all of them reported having experienced a traumatic event. Most of the sample was in a relationship (58.4%), 32.6% were single, 8.5% were cohabitants, and 0.6% were married. Considering the level of education, most of the sample had a bachelor/master’s degree (48.4%), 37.1% had a high school license, 9.9% had a master’s degree/higher specialization degree, and 4.5% had a middle school license. Lastly, considering employment status, 36.1% were dependent workers followed by students (36%) and student/workers 19.7% were students/workers; the remaining part of the sample was unemployed (8.2%). Each of the participants experienced a traumatic event. 22.1% reported to have experienced a ‘severe accident (e.g., car crash, domestic accident, etc.) that threatened his/her own life, followed by ‘severe organic illness, still ongoing (i.e. cancer)’ (21.8%); ‘threat to one’s life (e.g. being hit/hurt) from family members and/or partners and/or strangers’ (15.9%); and ‘having been sexually abused by family members and/or partner and/or strangers’ (12.5%). Moreover, 10.8% ‘loss of a member of one’s immediate family in a tragic way (e.g., serious and long physical illness, car accident, plane crash, suicide)’, followed by ‘witnessing traumatic event experienced by another person’ (9.1%) and ‘other’ (7.9%). The Development Procedure of the Questionnaire The Post-Traumatic Symptom Questionnaire (PTSQ) was created to fill some of the gaps in previous instruments for measuring post-traumatic symptoms. First, it was chosen to use a number of items that would allow for easy and rapid administration suitable in all types of contexts – both clinical and research settings – thus differentiating it from longer instruments such as the Impact of Event Scale-Revised (IES-R), which is much longer with its 22 items (Weiss, 2007; Weiss & Marmar, 1996). Second, unlike the original Impact of Event Scale (Horowitz et al., 1979), the PTSQ was created with the intention of also measuring the dimension of hyperarousal according to DSM-5-TR. In line with previous studies, the item pool for the Post-Traumatic Symptom Questionnaire (PTSQ) was developed using a three-step double-blind study procedure (Milavic et al., 2019; Pietrabissa, Rossi, Borrello, et al., 2020; Pietrabissa, Rossi, Simpson, et al., 2020; Rossi et al., 2021). The detailed procedure is reported in the supplementary materials. Measures A biographic information form collected general demographic information (e.g., sex, age, civil status, education level, and employment status). Moreover, participants were asked what type of traumatic event they had experienced. The Post-Traumatic Symptom Questionnaire (PTSQ) The Post-Traumatic Symptom Questionnaire (PTSQ) is a 12-items questionnaire measuring the three main reactions/domains of the impact of traumatic events: (A) intrusivity, (B) avoidance, and (C) hyperarousal. The first domain, (A) intrusivity (INTR), measures the extent to which the subject who has experienced a traumatic event reports having intrusive thoughts and images – as well as unpleasant emotions – that recall the traumatic event itself. The second domain, (B) avoidance (AV), measures the extent to which the subject tends to avoid things, situations, and people – as well as thoughts and behaviors – that may remind them of the traumatic event. Lastly, the third domain, (C) hyperarousal (HY.AR), measures the extent to which the subject reports having excessive reactions of hypervigilance, fear, and alertness in their daily lives after experiencing the traumatic event. Items are rated on a 5-point Likert-type scale, from 1 (= “not at all”) to 5 (= “extremely”). Higher scores indicate higher levels in that domain. Moreover, consistently with the proposed theoretical background, an overarching second-order general was assumed: post-traumatic symptoms (PTS). See Appendix ‘A’ and ‘B’ for the English and the Italian versions of the PTSQ, respectively. Statistical Analysis The R software was used with the following packages: lavaan (lavaan, 2012), psych (Revelle, 2018), and semPlot (Jorgensen et al., 2019). To test the factorial structure of the PTSQ, a confirmatory factor analysis (CFA) was performed. Considering the theoretical background as well as the semantic content of the items, a second-order factorial structure (Brown, 2015) was specified: each item loaded onto its specific first-order factors reflecting the main three reactions/domains of traumatic events – namely, (A) intrusivity, (B) avoidance, and (C) hyperarousal – and an overarching general factor so-called ‘post-traumatic symptom’ (PTS) (Fig. 1). Given the response scale of the PTSQ, the diagonal weighted least square (DWLS) estimator was used to perform the CFA (Brown, 2015; Forero et al., 2009; Li, 2016; Parola et al., 2022). Fig. 1 Study 1. Graphical representation of the PTSQ model Model fit was assessed using the Chi-square statistics (χ2), the Root-Mean Square Error of Approximation (RMSEA), the Comparative Fit Index (CFI), and the Standardized Root Mean Residual (SRMR) (Brown, 2015; Hoyle, 2012; Kline, 2016; van de Schoot et al., 2012). The following cut-off criteria were chosen to evaluate the goodness of fit: (A) statistically non-significance of the χ2, (B) an RMSEA lower than 0.08, (C) a CFI higher than 0.95, and (D) an SRMR lower than 0.08 (Brown, 2015; Hoyle, 2012; Hu & Bentler, 1999; van de Schoot et al., 2012). Additionally, the ability of the items to discriminate subjects with a low or high level of the measured construct was tested using the item discriminant power (IDP) (Chiorri, 2011; Ebel, 1965). Moreover, adjusted item-total correlation (rit−tot) was also computed (Tabachnick & Fidell, 2014). Given the problems and criticisms associated with Cronbach’s alpha (Barbaranelli et al., 2014; Raykov, 2011; Raykov, 2012; Raykov & Marcoulides, 2011), the internal consistency of each scale was also calculated with McDonald’s omega (McDonald, 1999). Also, correlations between items and domains were computed (Tabachnick & Fidell, 2014). Lastly, a multivariate analysis of variance (one-way MANOVA) was conducted to assess for possible statistical differences between the different traumatic experiences, on the PTSQ subscales simultaneously. For multiple comparisons, the strength of differences was interpreted using partial eta-squared (η2p) whereas, for pairwise comparisons, Hedge’s g (Hedges, 1981) was used. The strength of the differences was interpreted using Cohen’s benchmarks (Cohen, 1988): null (η2p < 0.010; g < 0.20), small (η2p from 0.011 to 0.059; g from 0.20 to 0.49), moderate (η2p from 0.060 to 0.139; g from 0.50 to 0.79), and large (η2p > 0.140; g > 0.80). Games-Howell post-hoc correction for multiple comparisons was applied (Howell, 2013). Results Structural Validity The PTSQ showed a good fit to the data. Despite that the Chi-square statistic was statistically significant [χ2 (51) = 133.686; p < .001], all other fit indices revealed a good fit to the data: RMSEA = 0.068; 90%CI[0.054, 0.082], CFI = 0.997, SRMR = 0.057. As reported in Table 1, all items’ loadings were statistically significant and ranged from 0.735 (item#10; HY.AR.) to 0.968 (item#3, INTR). Table 1 Study 1. Item descriptive statistics, item psychometric properties, and confirmatory factor analysis Descriptive statistics IDP r Adj CFA Mean SD SK K t d λ R 2 Item#1 3.159 1.169 -0.011 -0.778 -29.62 4.19 0.836 0.905 0.819 Item#2 3.031 1.193 0.021 -0.842 -30.17 4.25 0.832 0.899 0.807 Item#3 3.164 1.139 -0.036 -0.785 -34.08 4.81 0.883 0.968 0.936 Item#4 3.232 1.198 -0.097 -0.895 -28.96 4.12 0.782 0.839 0.703 Item#5 2.895 1.377 0.071 -1.186 -28.28 4.18 0.644 0.761 0.579 Item#6 2.813 1.392 0.115 -1.198 -22.85 3.37 0.645 0.796 0.634 Item#7 2.686 1.338 0.175 -1.150 -27.20 3.98 0.636 0.810 0.656 Item#8 2.776 1.307 0.183 -0.981 -28.88 4.25 0.761 0.855 0.731 Item#9 2.445 1.289 0.455 -0.905 -18.78 2.91 0.558 0.640 0.410 Item#10 2.705 1.301 0.225 -0.988 -22.50 3.48 0.611 0.735 0.541 Item#11 2.754 1.318 0.148 -1.052 -33.01 5.12 0.772 0.929 0.862 Item#12 2.646 1.302 0.275 -0.943 -24.46 3.80 0.661 0.808 0.653 INTR 12.586 4.263 0.036 -0.803 -21.93 3.34 0.615 0.815 0.664 AV 11.170 4.443 0.036 -0.955 -19.97 3.03 0.446 0.554 0.306 HY.AR 10.550 4.216 0.222 -0.691 -26.12 3.95 0.656 0.955 0.912 ITE 34.306 10.500 0.105 -0.486 Notes: IDP = Item discriminant power onto the specific factor (i.e., INTR, AV, or HY.AR), t = independent sample t-test, d = Cohen’s d (effect size). CFA = confirmatory factor analysis. In the CFA columns, absolute values of standardized factor loading (\|λ\|) are reported. λ = factor loading onto the specific factor (i.e., INTR, AV, or HY.AR); for INTR, AV, and HY.AR, λ = refers to factor loading of the first-order factors onto the general factor (i.e., ‘post-traumatic symptoms’). Psychometric Properties Considering the three domains, the IDP analysis showed that 12 items of the PTSQ discriminated well between subjects with a low or high level of the construct (Table 1). The discrimination parameter ti ranged from − 18.78\| (item#19 – HY.AR) to -34.08 (item#3 – INTR), with an associated effect size (Cohen’s d) ranging from 2.91 (large) to 4.81 (large), respectively. In addition, the item-total correlation (adjusted) revealed a moderate-to-strong association between each item and its PTSQ factor. Considering the general factor (PTS), the IDP analysis showed that three domains of the PTSQ discriminated well between subjects with a low or high level of impact of the traumatic event (Table 1). The discrimination parameter ti ranged from − 19.97 (AV) to -26.12 (HY.AR), with an associated effect size (Cohen’s d) ranging from 3.03 (large) to 3.95 (large), respectively. Also in this case, the item-total correlation (adjusted) revealed a moderate-to-strong association between each domain and the PTSQ general score. Considering internal consistency, the Cronbach’s alpha revealed that the PTSQ showed good internal consistency for each domain: INTR = 0.928, 95%CI[0.915, 0.940]; AV = 0.838, 95%CI[0.808, 0.864]; HY.AR = 0.824, 95%CI[0.792, 0.852]. Also the McDonald’s omega revealed that the PTSQ showed good internal consistency for each domain: INTR = 0.928, 95%CI[0.916, 0.941]; AV = 0.839, 95%CI[0.811, 0.866]; HY.AR = 0.831, 95%CI[0.802, 0.859]. Moderate-to-large correlations among the three domains and the general total score were found. A moderate correlation was found between the INTR domain and the AV domain (r = .383, p < .001) whereas a strong correlation was found between the INTR domain and the HY.AR domain (r = .665; p < .001). A moderate correlation was found between the AV domain and the HY.AR domain (r = .431, p < .001). Lastly, large correlations were found between the general factor (PTS) and the INTR domain (r = .835, p < .001), the AV domain (r = .752, p < .001), and the HY.AR domain (r = .854, p < .001). Correlations between items are reported in Supplementary material (Table S1). Differences Among Traumatic Experiences A statistically significant multivariate effect was found among type of traumatic experience on the three post-traumatic domains: Wilks’ Λ = 0.675; F = 8.071, p < .001; η2p = 0.123 – Fig. 2; Table 2. Fig. 2 Study 1. boxplot Table 2 Study 1. Item descriptive statistics, and post-hoc comparisons among different kind of traumatic events A B C D E F G Post-hoc comparisons M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) M (SD) INTR 13.32 (4.35) 13.88 (3.70) 14.18 (3.62) 13.58 (4.25) 11.29 (3.41) 9.50 (2.78) 8.00 (3.31) A > F; A > G; B > E; B > F; B > G; C > E; C > F; C > G; D > E; D > F; D > G; E > G AV 12.00 (4.39) 13.59 (3.67) 12.70 (3.25) 9.26 (4.60) 11.21 (4.34) 9.19 (4.34) 9.07 (3.68) A > D; A > F; A > G; B > D; B > F; B > G; C > D; C > F; C > G; HY.AR 10.95 (4.23) 11.46 (4.02) 11.34 (4.07) 11.83 (4.02) 9.79 (4.30) 7.91 (3.41) 6.89 (2.53) A > F; A > G*; B > F; B > G; C > F; C > G; D > F; D > G*; E > G; PTS § 36.27 (10.74) 38.93 (8.63) 38.23 (8.59) 34.68 (11.01) 32.29 (9.71) 26.59 (7.24) 23.96 (6.39) A > F; A > G; B > E; B > F; B > G; C > F; C > G; D > F; D > G; E > G; * p < .01; ** p < .001; redundant comparisons (e.g., B > E and E < B) were not reported in the table. Considering MANOVA’s assumption, for the general total score (composite score, PTS) a separate One-Way ANOVA was performed with Games-Howell post-hoc comparisons. Note: A = ‘severe accident (e.g., car crash, domestic accident, etc.) that threatened his/her own life’; B = ‘threat to one’s life (e.g. being hit/hurt) from family members and/or partners and/or strangers’; C = ‘having been sexually abused by family members and/or partner and/or strangers’; D = ‘severe organic illness, still ongoing (i.e. cancer)’; E = ‘loss of a member of one’s immediate family in a tragic way (e.g., serious and long physical illness, car accident, plane crash, suicide)’; F = ‘witnessing traumatic event experienced by another person’; G = ‘other’ Fig. 3 Study 2. Graphical representation of the structural equation model (N = 445).( Note. Model fit: χ2 (84) = 224.050; p < .001; RMSEA = 0.061; 90%CI: 0.052, 0.071; CFI = 0.972; SRMR = 0.035.) Considering ‘intrusivity’, a statistically significant univariate effect was found: F = 14.449, p < .001, η2p = 0.200. More in detail, statistically significant differences were found between (A) ‘severe accident’ and (F) ‘witnessing traumatic event’ (g = -0.964) and (G) ‘Other’ (g = -1.296). Moreover, statistically significant differences were found between (B) ‘threat to one’s life by another person’ and (E) ‘loss of a family member’ (g = -0.722), (F) witnessing a traumatic event (g = -1.289), and (G) Other (g = -1.644). Also, statistically significant differences were found between (C) ‘have been sexually abused’ and (E) ‘loss of a family member’ (g = -0.82), (F) witnessing a traumatic event (g = -1.421), and (G) Other (g = -1.764). In addition, statistically significant differences were found between (D) ‘severe chronic illness (still ongoing)’ and (E) ‘loss of a family member’ (g = -0.573), (F) witnessing traumatic event (g = -1.051), and (G) Other (g = -1.386). Lastly, statistically significant differences were found between (E) ‘loss of a family member’ and (G) Other (g = -0.977). Considering ‘avoidance’, a statistically significant univariate effect was found: F = 9.868, p < .001, η2p = .146. More in detail, statistically significant differences were found between (A) ‘severe accident’ and (D) severe chronic illness (still ongoing)’ (g = -0.609), (F) ‘witnessing traumatic event’ (g = -0.642) and (G) ‘Other’ (g = -0.695). Statistically significant differences were found between (B) ‘threat to one’s life by another person’ and (D) severe chronic illness (still ongoing)’ (g = -1.023), (F) witnessing traumatic event (g = -1.121), and (G) Other (g = -1.231). Also, statistically significant differences were found between (C) ‘have been sexually abused’ and (D) severe chronic illness (still ongoing)’ (g = -0.826), (F) witnessing traumatic event (g = -0.937), and (G) Other (g = -1-061). Considering ‘hyperarousal’, a statistically significant univariate effect was found: F = 8.859, p < .001, η2p = 0.133. More in detail, statistically significant differences were found between: (A) ‘severe accident’ and (F) ‘witnessing traumatic event’ (g = -0.758) and (G) ‘Other’ (g = -1.051). Moreover, statistically significant differences were found between (B) ‘threat to one’s life by another person’ and (F) witnessing a traumatic event (g = -0.931), and (G) Other (g = -1.27). Also, statistically significant differences were found between (C) ‘have been sexually abused’ and (F) witnessing a traumatic event (g = -0.901), and (G) Other (g = -1.251). In addition, statistically significant differences were found between (D) ‘severe chronic illness (still ongoing)’ and (F) witnessing a traumatic event (g = -1.017), and (G) Other (g = -1.339). Lastly, statistically significant differences were found between (E) ‘loss of a family member’ and (G) Other (g = -0.793). Considering the total score ‘post-traumatic symptom’, a statistically significant univariate effect was found: F = 13.218, p < .001, η2p = 0.186. More in detail, statistically significant differences were found between: (A) ‘severe accident’ and (F) ‘witnessing traumatic event’ (g = -0.981) and (G) ‘Other’ (g = -1.256). Moreover, statistically significant differences were found between (B) ‘threat to one’s life by another person’ and (E) ‘loss of a family member’ (g = -0.731), (F) witnessing a traumatic event (g = -1.513), and (G) Other (g = -1.88). Also, statistically significant differences were found between (C) ‘have been sexually abused’ and (F) witnessing a traumatic event (g = -1.446), and (G) Other (g = -1.826). In addition, statistically significant differences were found between (D) ‘severe chronic illness (still ongoing)’ and (F) witnessing a traumatic event (g = -0.759), and (G) Other (g = -1.026). Lastly, statistically significant differences were found between (E) ‘loss of a family member’ and (G) Other (g = -0.985). Study 2. The Anxiety Buffer Hypothesis: self-esteem as a Shield. A Structural Equation Model Approach Materials and methods Sample Size Determination In line with Study 1, the “n:q criterion” was still used to determine a priori the minimum sample to be enrolled. A ratio of 5 subjects per parameter was guaranteed (Brown, 2015; Hu & Bentler, 1999; Kline, 2016). Procedure In line with Study 1, the same snowball sampling method (Johnson, 2014) was used to enroll participants from the general population through personal invitations or materials advertised via social media platforms (e.g. Facebook, Twitter, etc.). Inclusion criteria were: (A) having experienced COVID-19 as a traumatic event, (B) being aged between 18 and 30 y.o., (C) being a native Italian speaker; (D) providing informed consent; (E) do not provide missing answers, (F) have no inability to complete the assessment procedure; and (G) do not complete the procedure in less than 8 or greater than 20 min. In line with previous studies (e.g. Rossi et al., 2020), data were entirely collected in a single week-interval to avoid confounding effects due to the pandemic fluctuations. In line with the Study 1, at the beginning of the survey, participants were asked whether they had experienced COVID-19 as a traumatic event. Based on the above inclusion/exclusion criteria, only those with this feature completed the survey and were enrolled in the study. Participants An initial sample of 510 individuals was eligible to complete the survey. However, 42 individuals were excluded due to the presence of missing values and 23 individuals did not complete the procedure within the given time frame. Thus, the final sample consisted of 445 participants: 76 males (17.1%) and 369 females (82.9%) aged from 18 to 30 years (mean = 26.18, SD = 2.89). All of the participants experienced COVID-19 as a traumatic event. Most of the participants were in a relationship (56.4%), 32.1% were single, 10.8% were cohabitants, and 0.7% were married. Considering the level of education, most of the sample had a bachelor/master’s degree (48.8%), 36.2% had a high school license, 10.3% had a master’s degree/higher specialization degree, and 4.7% had a middle school license. Lastly, considering employment status, 37.3% were dependent workers followed by students (36.4%) and student/workers 19.6% were students/workers; the remaining part of the sample was unemployed (6.5%). Measures The socio-demographic information form used in Study 1 and the Post-Traumatic Symptom Questionnaire (PTSQ) were administered. In this study, the PTSQ still provides good internal consistency with both Cronbach’s alpha (INTR = 0.781, 95%CI[0.746, 0.813]; AV = 0.838, 95%CI[0.812, 0.861]; HY.AR = 0.818, 95%CI[0.788, 0.844]) and McDonalds’ Omega (INTR = 0.786, 95%CI[0.753, 0.818]; AV = 0.840, 95%CI[0.815, 0.864]; HY.AR = 0.819, 95%CI[0.791, 0.846]). In addition, the following self-report measures were administered. Rosenberg Self-Esteem Scale (RSE) The RSE (Rosenberg, 1965) is one of the most used scales for assessing self-esteem. It consists of 10 statements assessing feelings about one’s self. Respondents express their degree of agreement with each statement on a 4-point Likert-type scale (ranging from 1 = “not at all” to 4 = “always”), and it provides a single-factor structure. Higher values indicate greater self-esteem. In the present study, the RSE showed a high internal consistency: Cronbach’s alpha = 0.860, 95%CI[0.841, 0.877]; McDonalds’ Omega = 0.871, 95%CI[0.854, 0.889]. Anxiety Subscale of the Symptom Checklist-90Revised (SCL-90R – ANX) The SCL-90R ANX subscale (Derogatis & Unger, 2010) is one of the most used worldwide scales for assessing anxiety symptoms. It is a 10-item measuring physical, cognitive, and psychological signs of anxiety within the past week. Respondents rate the severity of their symptoms on a 5-point Likert-type scale (ranging from 1 = “not at all” to 5 = “always”). The ANX provides a single-factor structure. Higher values indicate more severe anxiety symptoms. In the present study, the ANX subscale showed a high internal consistency: Cronbach’s alpha = 0.924, 95%CI[0.914, 0.934]; McDonalds’ Omega = 0.928, 95%CI[0.918, 0.938]. Depression Subscale of the Symptom Checklist-90Revised (SCL-90R – DEP) The SCL-90R DEP subscale (Derogatis & Unger, 2010) is one of the most used self-report scales for assessing depression symptoms. It is a 13-item measuring cognitive, emotive, and somatic signs of depression during the past week. Respondents rate the gravity of their symptoms on a 5-point Likert-type scale (ranging from 1 = “not at all” to 5 = “always”). The DEP provides a single-factor structure. Higher values indicate more severe depressive symptoms. In the present study, the DEP subscale showed a high internal consistency: Cronbach’s alpha = 0.912, 95%CI[0.900, 0.923]; McDonalds’ Omega = 0.917, 95%CI[0.905, 0.928]. Statistical Analyses The R software was used with the following packages: lavaan (lavaan, 2012) and psych (Revelle, 2018). Graphical representations were performed with GraphViz in DiagrammeR (Iannone, 2018). Preliminarily, multivariate multiple regression was performed to exclude the potential confounding effects of covariates on psychological variables. Also, the Pearson correlation coefficient (r) was computed to evaluate the relationships between variables (Tabachnick & Fidell, 2014). These analyses are reported in the supplementary material – Table S2 and S3. A structural equation modeling (SEM) approach with latent variables was used. The steps described below were followed. First. Before testing the hypothesized model, the structural validity of each scale used in this study was evaluated using CFAs, separately. The DWLS estimator was used. Model fit was assessed using the aforementioned fit indices (χ2, RMSEA, CFI, SRMR) and their cutoff values for good model fit (Brown, 2015; Hoyle, 2012; Kline, 2016; van de Schoot et al., 2012). Second. The ‘common method bias’ was checked using Harman’s single-factor test (Harman, 1976; Podsakoff et al., 2003; Brown 2015). First, a six correlated factors model was specified according to the measurement model of each questionnaire (PTSQ – three factors, RSE – single factor, ANX – single factor, and DEP – single factor). Second, a first-order single factor model was specified (all the items loaded onto a single latent dimension). Models were compared: a statistically significant chi-square difference (Δχ2; p < .050), a ΔRMSEA greater than .015, and a ΔCFI greater than .010 suggest the absence of the bias (Brown, 2015; Cheung & Rensvold, 2002; Millsap, 2012). Third. A partially disaggregated parcel approach was used: item parcels were used as indicators (i.e., items) of latent variables (Coffman & MacCallum, 2005; Little et al., 2002; Little et al., 2013). At least 3-item-parcel were created for each latent variable (each factor should be at least ‘just identified’) (Hoyle, 2012; Kline, 2016; Little et al., 2013) – statistics of item parcels are reported in the supplementary material. The ‘item-to-construct balance strategy’ (Little et al., 2013) was used to create parcels of the three unidimensional scales (RSE, ANX, and the DEP) – by inspecting factor loadings resulting from each measurement model (Little et al., 2002; Little et al., 2013). The ‘domain-representative strategy’ (Little et al., 2002; Little et al., 2013) was used to create parcels of the hierarchical second-order structure of the PTSQ – by aggregating together items of each dimension. Fourth. Item parcels’ descriptive statistics were examined: a normal distribution was found for the large majority of item parcels (see supplementary material). Consequently, the maximum likelihood (ML) estimator was used to test the hypothesized SEM (Hoyle, 2012; Kline, 2016). A 10,000 bootstrap resampling procedure was applied (MacKinnon, 2012). Fifth. The multiple mediation model with latent variables was tested using a three-step approach (Daniel et al., 2015; Hayes, 2022; MacKinnon, 2012; MacKinnon et al., 2007; VanderWeele & Vansteelandt, 2014). First, a predictor-only model was specified: the traumatic experience of COVID-19 (X) predicts depressive symptoms (Y). Second, a simple mediation model was specified by excluding the effect of self-esteem (buffering variable): the traumatic experience of COVID-19 (X) predicts depressive symptoms (Y) through anxiety symptoms (M). Statistical analyses of these intermediate models are reported in the supplementary material – Figure S1 and Figure S2. Lastly, a sequential multiple mediation model (full model) was specified (Fig. 3). The variance of each latent variable was fixed to unity. More in detail, the traumatic experience of COVID-19 (X) predicts depressive symptoms (Y) through anxiety symptoms (M2). However, according to the ABH, self-esteem (M1) should protect against the negative consequences – anxiety (M2) and depression (Y) – of the traumatic experience of COVID-19 (X). The goodness of the model was evaluated by using the abovementioned ‘goodness-of-fit’ indices (χ2, RMSEA, CFI, SRMR) and their cutoff values. Finally, all regression coefficients (β) reported in the text were unstandardized, whereas in Table 5 standardized regression coefficients (B) were also displayed. Results Preliminary Analysis The multivariate multiple regression analysis showed no statistically significant effects of potential confounding external variables. In addition, correlation analyses suggested small-to-large associations between the variables involved in the multiple mediation model – Supplementary material, Table S2 and Table S3. Structural Models The PTSQ showed adequate goodness-of-fit indices: χ2 (51) = 190.880; p < .001; RMSEA = 0.079, 90%CI[0.067, 0.091], p(RMSEA < 0.05) < .001; CFI = 0.989; SRMR = 0.061. Factor loadings of the first-order items ranged from 0.704 (item#4; INTR) to 0.836 (item#11 – HY.AR) (INTR: mean = 0.77, SD = 0.06; AV: mean = 0.82; SD = 0.06; HY.AR: mean = 0.79; SD = 0.04). Factor loadings of the second-order variable ranged from 0.682 (AV) to 0.974 (HY.AR) (mean = 0.80; SD = 0.13). Even the RSE revealed good fit indices: χ2 (35) = 102.124; p < .001; RMSEA = 0.066, 90%CI[0.051, 0.081], p(RMSEA < 0.05) = .039; CFI = 0.992; SRMR = 0.054. Factor loadings of the items ranged from 0.427 (item#9) to 0.821 (item#10) (mean = 0.69; SD = 0.14). Also the ANX showed good indices: χ2 (35) = 127.717; p < .001; RMSEA = 0.077, 90%CI[0.063, 0.092], p(RMSEA < 0.05) = .001; CFI = 0.995; SRMR = 0.046. Factor loadings of the items ranged from 0.718 (item#2) to 0.879 (item#3) (mean = 0.81; SD = 0.06). Lastly, the DEP revealed good fit indices: χ2 (65) = 161.861; p < .001; RMSEA = 0.058, 90%CI[0.047, 0.069], p(RMSEA < 0.05) = .117; CFI = 0.994,; SRMR = 0.056. Factor loadings of the items ranged from 0.461 (item#1) to 0.878 (item#8) (mean = 0.73; SD = 0.10). Harman’s Single-Factor Test The first CFA with six correlated factors provided good fit indices: χ2 (930) = 2075.199; p < .001; RMSEA = 0.053, 90%CI[0.050, 0.056], p(RMSEA < 0.05) = .074; CFI = 0.989; SRMR = 0.063. On the contrary, the CFA with a single latent factor provided poor fit indices: χ2 (945) = 8667.104; p < .001; RMSEA = 0.136, 90%CI[0.133, 0.138], p(RMSEA < 0.05) < .001; CFI = 0.923; SRMR = 0.120. The Harman’s single-factor test model comparison suggested the absence of the ‘common method bias’: Δχ2 (15) = 6591.9, p < .001; \|ΔRMSEA\| = 0.083, and \|ΔCFI\| = 0.065. Sequential Mediation Model The hypothesized model (Fig. 4, Table 3) provided adequate goodness-of-fit indices: χ2 (84) = 224.050; p < .001; RMSEA = 0.061, 90%CI[0.052, 0.071], p(RMSEA < 0.05) = .028; CFI = 0.972; SRMR = 0.035. All of the item-parcels showed a factor loading higher than 0.66 (Table S4, supplementary material). According to the ABH, traumatic experience of COVID-19 (X) was negatively associated with self-esteem (M1), path a1: β = -0.602 (SE = 0.075) [95%CI: -0.754; -0.464], z = -8.059, p < .001, and self-esteem – in turn – negatively predicted anxiety symptomatology (M2), path d: β = -0.162 (SE = 0.065) [95%CI: -0.292; -0.037], z = -2.500, p = .012 – thus showing the buffering effect of self-esteem. Lastly, anxiety symptomatology (M2) positively predicted depressive symptomatology (Y) path b2: β = 0.703 (SE = 0.096) [95%CI: 0.536; 0.910], z = 7.333, p < .001. In addition, still in line with the ABH, self-esteem (M1) was negatively associated with depressive symptomatology (Y), path b1: β = -0.731 (SE = 0.114) [95%CI: -0.977; -0.530], z = -6.395, p < .001 – further revealing the buffering effect of self-esteem also on traumatic-related depressive symptoms. Furthermore, traumatic experience of COVID-19 (X) was positively associated with both anxiety symptomatology (M2) [path a2: β = 0.601 (SE = 0.088) [95%CI: 0.434; 0.782], z = 6.769, p < .001] and depressive symptomatology (Y) [path c1: β = 0.544 (SE = 0.114) [95%CI: 0.344; 0.788], z = 4.794, p < .001] – suggesting a partially mediated model that highlighted the buffering effect of self-esteem. The total indirect effect (traumatic experience of COVID-19 → self-esteem → anxiety symptomatology → depressive symptomatology) was statistically significant: β = 0.069 (SE = 0.031) [95%CI: 0.015; 0.137], z = 2.179, p = .029. Also, the total model effect was statistically significant: β = 1.476 (SE = 0.169) [95%CI: 1.197; 1.856], z = 8.711, p < .001. The total explained variance (R2) was equal to 0.772. Table 3 Summary of Standardized Parameter Estimates (Beta) with 95% Confidence Intervals for Key Pathways Tested (Fig. 4 ) Path B β (SE) 95%CI [L - U] z-value p-value R2 Traumatic experience of COVID19 (X) → Self-esteem (M1) (a1) − 0.516 -0.602 (0.075) [-0.754; -0.464] -8.059 p < .001 0.266 Self-esteem (M1) → Anxiety (M2) (d) − 0.154 -0.162 (0.065) [-0.292; -0.037] -2.500 p = .012 0.339 Anxiety (M2) → Depression (Y) (b2) 0.413 0.703 (0.096) [0.536; 0.910] 7.333 p < .001 0.772 Traumatic experience of COVID19 (X) → Anxiety (M2) (a2) 0.488 0.601 (0.088) [0.434; 0.782] 6.796 p < .001 Self-esteem (M1) → Depression (Y) (b1) − 0.407 -0.731 (0.114) [-0.977; -0.530] -6.395 p < .001 Traumatic experience of COVID19 (X) → Depression (Y) (c1) 0.260 0.544 (0.114) [0.344; 0.788] 4.794 p < .001 Indirect effect of X on Y via M1 (a1b1) 0.210 0.440 (0.082) [0.305; 0.625] 5.359 p < .001 Indirect effect of X on Y via M2 (a2b2) 0.202 0.422 (0.075) [0.291; 0.586] 5.627 p < .001 Indirect effect of X1 on Y via M1 and M2 (a1db2) 0.033 0.069 (0.031) [0.015; 0.137] 2.179 p < .001 Total effect X1 on Y 0.705 1.476 (0.169) [1.197; 1.856] 8.711 p < .001 Note: B = standardized beta; β = unstandardized beta; 95%CI = 95% confidence intervals (lower and upper bound) for the unstandardized beta; R2 = explained variance Fig. 4 Study 2. Conceptual graphical representation of the structural equation model Discussion Scientific literature highlighted how traumatic events can have a negative impact on mental health (Benjet et al., 2016). Traumatic events represent a risk factor for the onset and worsening of anxious reactions, then followed by depressive symptoms. In this line, COVID-19 represented a traumatic event and triggered intense adverse psychological reactions such as anxiety and depression. The present study had two aims. First, to develop and evaluate the psychometric properties of a self-report questionnaire measuring the impact of the traumatic event that specifically takes into account intrusivity, hyperarousal, and avoidance. Second, to understand the psychological impact of traumatic events and emphasized how self-esteem can protect (buffer effect) from the negative outcomes (i.e., anxiety and depression) of traumatic events. Considering Study1, the PTSQ was developed using a solid and theoretically-driven methodology. It proved to be a reliable and psychometrically sound assessment tool to measure the psychological impact of traumatic events – specifically focused on the three main reactions/domains of intrusivity, hyperarousal, and avoidance. Moreover, it is important to note that although the sample was collected from the general population, the questionnaire was administered only to individuals who had actually reported experiencing a traumatic event such as a severe accident (e.g., car crash, domestic accident, etc.) that threatened his/her own life; or threat to one’s life (e.g. being hit/hurt) from family members and/or partners and/or strangers; or a severe organic illness (i.e. cancer). Moreover, the PTSQ showed a second-order (i.e., hierarchical) factorial structure with 3 well-separated (but reasonably correlated) first-order factors – clearly reflecting the three main reactions/domains of traumatic events – providing good fit indices. Also, all the items had excellent factorial loadings on the hypothesized factors. Moreover, item analysis showed the ability of the single indicator to discriminate between individuals with a low and high level of the measured construct. Furthermore, the PTSQ allowed observing how different traumatic experiences can have some aspects in common in the three components. Indeed, traumatic events that are experienced in first-person had a stronger psychological impact compared to traumatic events experienced in second-person (e.g., ‘witnessing traumatic event experienced by another person’). Considering Study 2, the structural equation model used to test the research hypotheses provided good results. The first model tested (Model 1, predictors only – supplementary material) showed that a state of post-traumatic symptoms might lead to the development of depressive symptoms (Santini et al., 2020; Thakur & Jain, 2020): a one-point increase in the severity of post-traumatic symptoms was associated with an increase of 0.981 points in the severity of the depressive symptomatology. Still, even when considering the mediation effect of anxiety symptoms (Model 2, simple mediation model – supplementary material), post-traumatic symptoms and depression exhibited a positive association in line with a consistent body of scientific literature (Rossi et al., 2020). Simultaneously, post-traumatic symptoms showed a strong and positive association with anxiety symptomatology (Friedman et al., 2007; Friedman et al., 2011) which in turn can lead to developing depressive symptoms (Bowman, 2001). This pattern suggests a partial mediation model from post-traumatic symptoms, to depression through anxiety – since the former are frequently characterized by intense experiences hyperarousal, startling, and concern about the past and the future (Gagne et al., 2018). Nevertheless, according to the research hypotheses, the final model undisclosed the buffering effect of self-esteem on the relationship between traumatic symptoms to anxiety. Notwithstanding the strong positive association between post-traumatic symptoms and anxiety (β = 0.547), self-esteem was able to hinder it. This result is perfectly in line with the TMT and the ABH (Greenberg et al., 1986; Greenberg et al., 1992). Indeed, self-esteem acts as a shield by protecting the individual from the self-reinforcing mechanisms that go between negative psychological constructs. In practice, from a theoretical point of view, at the moment when the impact of the traumatic event leads the individual to develop anxiety and depressive symptoms, self-esteem – by recovering the individual’s personal and social value and meaning – interferes with this concatenation of negative states (Greenberg et al., 1986; Greenberg et al., 1992; Rossi et al., 2020). Consequently, from a statistical point of view, self-esteem shows negative associations (negative β-values) with the psychological constructs of outcome of the traumatic event (i.e., anxiety and depression). However, it is important to note that since the relationship between posttraumatic symptoms, anxiety and depression was maintained even when their relationships were buffered by self-esteem, a partial mediation model is the one that best describes the psychological phenomenon inherent in TMT and ABH (Greenberg et al., 1986; Greenberg et al., 1992; Pyszczynski et al., 2004; Solomon et al., 2004). Moreover, these results (sequential partial mediation) suggest a possible explanation for the fact that self-esteem cannot – by itself – completely prevent the presence of anxiety and depression. Indeed, again considering that the relationship between the impact of the traumatic event and the negative outcomes are not totally mediated (buffered) by self-esteem, therefore, the relationship between the negative variables continues to hold – even though the shield (i.e., self-esteem) helps to dampen their strength (Salzman & Halloran, 2004). These findings showed that self-esteem can buffer the adverse effect of anxiety symptoms generated by traumatic symptoms. Consequently, these results provide additional support for the soundness of the ABH – that highlighted the buffering role of self-esteem on the relationship between post-traumatic symptoms, anxiety, and depression (Pyszczynski et al., 2004). Considering clinical implications, findings can suggest a potential line of intervention in order to offer psychological help for individuals facing the emerging adulthood phase and dealing with the adverse psychological outcomes of the impact of traumatic events such as the prolonged pandemic of COVID-19, with the aim to relieve it. As mentioned above, young adults are the category most at risk of traumatic consequences related to COVID-19 (Cao et al., 2020; Silva Junior et al., 2020). Although COVID-19 impacted individuals of all life ages, the consequences such as restrictions have especially impacted emerging adults’ opportunities for personal growth and the structuring of one’s personal identity and self-esteem (van den Berg et al., 2021). Specifically, the consequences of COVID-19 could exacerbate the vulnerabilities of emerging adulthood. Among the life span life, emerging adulthood is a crucial developmental period with significant changes in life roles (Arnett, 2006, 2016). Individuals are called to define and adapt self and identity (Arnett, 2016). This phase includes a large number of challenges and risks that can affect psychological adjustment (Berzin, 2010; Burt & Masten, 2010). To this extent, the ABH posits that self-esteem can protect against various stressors, which in turn can intensify the need for self-esteem to buffer psychological difficulties (Harmon-Jones et al., 1997). As a result, improved self-esteem is supposed to serve as a buffer against anxiety, weakening the negative psychological reactions to stressors and threats to the individuals’ health. Therefore, implanting and improving psychological strategies to specifically target self-esteem may represent an efficient approach to reducing the adverse psychological consequences of traumatic symptoms related to post-traumatic events such as COVID-19 – in particular among young adults who showed to be more vulnerable and prone to develop them (Liu et al., 2020; Panzeri et al., 2021). This study is not free of limitations. The observational/correlational research design did not allow for defining a causal relationship among variables, but only relationships of statistical predictions (Fiedler et al., 2011) – as in line with the aim of the study. Cross sectional research design has limitations compared to longitudinal designs but is still able to provide a photograph of the participants psychological conditions at a given time. Importantly, a considerable amount of scientific literature refers to ‘statistical mediation’ as ‘mediation’, but a desirable way to assess mediation would be through longitudinal studies. Moreover, the online survey consisting of self-reports may have been influenced by well-known biases, such as social desirability. Also, the sample presents a high prevalence of females (68.6%), despite no associated effects emerging from the preliminary analyses. Moreover, a multi-group analysis (moderated mediation) comparing the model across males and females was not performed because of the small presence of males – which would have not allowed to provide an accurate estimation of model parameters (Hoyle, 2012; Kline, 2016). Future research may try to overcome these limitations. Concluding, both theoretical and statistical reasons supported the choice of a mediation model rather than a moderation one. Theoretically, ABH and TMT (Greenberg et al., 1986; Greenberg et al., 1992) hypothesize self-esteem to be an intermediating shield (buffer) between stressors and anxiety (Pyszczynski et al., 2004), thus a mediation approach can better reflect this pattern. Indeed, self-esteem is both able to impact subjective anxiety and depression and it can also be modified by negative emotional traumatic experiences that can deeply (negatively) modify the evaluation of oneself (Greenberg et al., 1986; Heinrich & Gullone, 2006; Sowislo & Orth, 2013). Literature showed that, by activating negative cognitions and emotions, traumatic experiences can significantly worsen the self-evaluation and self-concept (Greenberg et al., 1986) (i.e., feeling a failure, feeling worthless) (Beck, 1979), also progressively reducing self-efficacy and self-esteem (Sowislo & Orth, 2013). For the abovementioned reasons, a moderation model would have not been in line with the theoretical background and could not allow considering the complexity of relationships among constructs. The present study also has some strengths. First, both Study 1 and Study 2 are grounded on well-established theoretical foundations relying on several experimental and longitudinal studies’ support (Brage & Meredith, 1994; Greenberg et al., 1992; Heinrich & Gullone, 2006; Pyszczynski et al., 2004). Second, Study 1 provides a brief but solid measure to assess post traumatic symptoms. In fact, the PTSQ is much shorter than other common scales for assessing trauma response such as the impact of event scale – revised (IES-R) which has 22 items (Weiss, 2007; Weiss & Marmar, 1996) but the PTSQ is equally robust, psychometrically grounded. In addition, it is worth noting that the 15-item version of the IES (Horowitz et al., 1979) is missing the hyperarousal symptom dimension according to the DSM – which is measured by the PTSQ instead: indeed, the PTSQ accurately measures the three components of intrusiveness, avoidance, and hyperarousal. Moreover, the solid psychometric foundation of the instrument, is also seen in comparison with drastically shorter scales – such as the 6-item Impact of Event Scale (Giorgi et al., 2015). In fact, that scale having only two items for each of the 3 dimensions, turns out to be under-identified in the latter. The PTSQ, having 4 items per latent dimension, turns out to be over-identified – which is the ideal condition for measurement scales (Brown, 2015; Hoyle, 2012; Kline, 2016). Because of its length, the PTSQ can be easily integrated into surveys or batteries of instruments and has – in addition – proven to be a useful, accurate and valid measure of psychological constructs in research settings (Rossi et al., 2022; Schipolowski et al., 2014). Third, the sample size from the general Italian population allowed utilizing robust statistical analysis and methodologies (MacKinnon, 2012; MacKinnon et al., 2007) to offer interesting results. Additionally, all the models hypothesized and tested provided a good fit. Moreover, findings from this research could be extended and generalized to people coping with the adverse impact of traumatic experiences, in particular health-related (e.g., smallpox and/or diagnosis of severe cancer) (Betancourt et al., 2016; Brown & Lees-Haley, 1992; Chew et al., 2020; Huremović, 2019). As an instance, these results could lead to useful applications to relieve the psychological consequences of traumatic reactions toward self-threats (Rossi Ferrario & Panzeri, 2020). For example, based on these findings, clinicians may develop interventions to improve people’s psychological health and well-being. Overall, this research provides a valuable contribution to the current stream of literature disentangling the psycho-social impact of COVID-19. Future longitudinal studies are needed to understand how psychological difficulties-progressing over time may be affected by self-esteem. Also, future studies may deepen the process through which other protective or risk factors could impact psychological outcomes. Conclusion In conclusion, this study provides additional evidence about the anxiety-buffer effect on self-esteem, once again confirming the effectiveness and usefulness of TMT in providing valuable research and clinical insights into understanding and handling trauma-related experiences. People dealing with traumatic-related events (and their impact) should be supported through targeted psychological interventions aiming at reducing the trauma’s psychological impact and favoring resilient psychological health outcomes. Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2 Supplementary Material 3 Author Contributions AAR: conceptualization, formal analysis, methodology, writing the original draft, review, and editing. A.Panzeri: data collection, writing the original draft, methodology, review, and editing. A.Parola: data collection, review, and editing. FT: contribution in developing the PTSQ, review, and editing. SM: review and editing. Funding Sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Data Availability The datasets presented in this article are not readily available because due to privacy restrictions, data were available from the corresponding author on a reasonable request. Declarations Ethical Approval The study was in accordance with the ethical standards of the Ethical Committee of the University of Padua. 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 Agaibi CE Wilson JP Trauma, PTSD, and resilience:a review of the literature Trauma Violence & Abuse 2005 6 3 195 216 10.1177/1524838005277438 American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). American Psychiatric Association. Arnett, J. J. (2006). Emerging Adulthood: understanding the New Way of coming of Age. Emerging adults in America: coming of age in the 21st century (pp. 3–19). American Psychological Association. 10.1037/11381-001. Arnett JJ College Students as emerging adults:the Developmental Implications of the College Context Emerging Adulthood 2016 4 3 219 222 10.1177/2167696815587422 Barbaranelli C Lee CS Vellone E Riegel B Dimensionality and reliability of the self-care of heart failure index scales: further evidence from confirmatory factor analysis Research In Nursing & Health 2014 37 6 524 537 10.1002/nur.21623 25324013 Barlow, D. H. (2002). In D. H. Barlow (Ed.), Anxiety and its Disorders: the nature and treatment of anxiety and panic (2nd ed.). The Guilford Press. Baud D Qi X Nielsen-Saines K Musso D Pomar L Favre G Real estimates of mortality following COVID-19 infection The Lancet Infectious Diseases 2020 20 7 773 10.1016/s1473-3099(20)30195-x Beck, A. T. (1979). In A. T. Beck (Ed.), Cognitive therapy of depression. The Guilford Press. Becker, E. (1971). Birth and death of meaning: an interdisciplinary perspective on the Problem of Man. Free Press. Becker, E. (1973). The denial of death. Simon & Schuster. Benight CC Understanding human adaptation to traumatic stress exposure: beyond the medical model Psychological Trauma: Theory Research Practice and Policy 2012 4 1 1 8 10.1037/a0026245 Benjet C Bromet E Karam EG Kessler RC McLaughlin KA Ruscio AM Shahly V Stein DJ Petukhova M Hill E Alonso J Atwoli L Bunting B Bruffaerts R Caldas-de-Almeida JM de Girolamo G Florescu S Gureje O Huang Y Lepine JP Kawakami N Kovess-Masfety V Medina-Mora ME Navarro-Mateu F Piazza M Posada-Villa J Scott KM Shalev A Slade T ten Have M Torres Y Viana MC Zarkov Z Koenen KC The epidemiology of traumatic event exposure worldwide: results from the World Mental Health Survey Consortium Psychological Medicine 2016 46 2 327 343 10.1017/s0033291715001981 26511595 Berzin SC Vulnerability in the transition to adulthood: defining risk based on youth profiles Children and Youth Services Review 2010 32 4 487 495 10.1016/j.childyouth.2009.11.001 Betancourt TS Brennan RT Vinck P VanderWeele TJ Spencer-Walters D Jeong J Akinsulure-Smith AM Pham P Associations between Mental Health and Ebola-Related Health Behaviors: a regionally Representative cross-sectional survey in post-conflict Sierra Leone PLOS Medicine 2016 13 8 e1002073 10.1371/journal.pmed.1002073 27505186 Bishop LS Ameral VE Palm Reed KM The impact of Experiential Avoidance and Event Centrality in Trauma-Related rumination and posttraumatic stress Behavior Modification 2018 42 6 815 837 10.1177/0145445517747287 29241356 Bowman GS Emotions and illness Journal of Advanced Nursing 2001 34 2 256 263 10.1046/j.1365-2648.2001.01752.x 11430288 Brage D Meredith W A causal model of adolescent depression The Journal of psychology 1994 128 4 455 468 10.1080/00223980.1994.9712752 7932297 Breslau N Davis GC Posttraumatic stress disorder in an urban population of young adults: risk factors for chronicity The American Journal of Psychiatry 1992 149 5 671 675 10.1176/ajp.149.5.671 1575259 Breslau N Davis GC Andreski P Peterson E Traumatic events and posttraumatic stress disorder in an Urban Population of Young adults Archives Of General Psychiatry 1991 48 3 216 222 10.1001/archpsyc.1991.01810270028003 1996917 Briere J Spinazzola J Phenomenology and psychological assessment of complex posttraumatic states Journal of Traumatic Stress 2005 18 5 401 412 10.1002/jts.20048 16281238 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 The Lancet 2020 395 10227 912 920 10.1016/s0140-6736(20)30460-8 Brosschot JF Verkuil B Thayer JF The default response to uncertainty and the importance of perceived safety in anxiety and stress: an evolution-theoretical perspective Journal of Anxiety Disorders 2016 41 22 34 10.1016/j.janxdis.2016.04.012 27259803 Brown RS Lees-Haley PR Fear of future illness, Chemical Aids, and Cancerphobia: a review Psychological Reports 1992 71 1 187 207 10.2466/pr0.1992.71.1.187 1529055 Brown, T. A. (2015). Confirmatory factor analysis for Applied Research (Second ed.). The Guilford Press. Burt, K. B., & Masten, A. S. (2010). Development in the transition to adulthood: vulnerabilities and opportunities. In J. E. Grant, & M. N. Potenza (Eds.), Young adult mental health (pp. 5–18). Oxford University press. Cao W Fang Z Hou G Han M Xu X Dong J Zheng J The psychological impact of the COVID-19 epidemic on college students in China Psychiatry Research 2020 287 112934 10.1016/j.psychres.2020.112934 32229390 Cheng C Jun H Baoyong L Psychological Health Diathesis Assessment System: a nationwide survey of resilient trait scale for chinese adults Studies of Psychology and Behavior 2014 12 6 735 742 Cheung GW Rensvold RB Evaluating goodness-of-fit indexes for testing measurement invariance Structural equation modeling 2002 9 233 255 10.1207/S15328007SEM0902_5 Chew Q Wei K Vasoo S Chua H Sim K Narrative synthesis of psychological and coping responses towards emerging infectious disease outbreaks in the general population: practical considerations for the COVID-19 pandemic Singapore Medical Journal 2020 61 7 350 356 10.11622/SMEDJ.2020046 32241071 Chiorri, C. (2011). Teoria e tecnica psicometrica. Costruire un test psicologico. McGraw-Hill Companies. Coffman DL MacCallum RC Using parcels to convert path analysis models into latent variable models Multivariate Behav Res 2005 40 2 235 259 10.1207/s15327906mbr4002_4 26760108 Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Lawrence Erlbaum Associates. Daniel RM De Stavola BL Cousens SN Vansteelandt S Causal mediation analysis with multiple mediators Biometrics 2015 71 1 1 14 10.1111/biom.12248 25351114 Derogatis, L. R., & Unger, R. (2010). Symptom Checklist-90-Revised. Corsini Encyclopedia of psychology. John Wiley & Sons, Inc. 10.1002/9780470479216.corpsy0970. Djelantik AAAMJ Robinaugh DJ Kleber RJ Smid GE Boelen PA Symptomatology following loss and trauma: latent class and network analyses of prolonged grief disorder, posttraumatic stress disorder, and depression in a treatment-seeking trauma‐exposed sample Depression and Anxiety 2020 37 26 34 10.1002/da.22880 30724427 Ebel, R. L. (1965). Measuring educational achievement. Prentice-Hall. Esterwood, E., & Saeed, S. A. (2020). Past Epidemics, Natural Disasters, COVID19, and Mental Health: Learning from History as we Deal with the Present and Prepare for the Future. Psychiatric Quarterly, 91(4), 1121-1133. https://doi.org/10.1007/s11126-020-09808-4 Fiedler K Schott M Meiser T What mediation analysis can (not) do Journal of Experimental Social Psychology 2011 47 6 1231 1236 10.1016/j.jesp.2011.05.007 Fisher EB Miller SM Evans M Luu SL Tang PY Valovcin D Castellano C COVID-19, stress, trauma, and peer support—observations from the field Translational Behavioral Medicine 2020 10 3 503 505 10.1093/tbm/ibaa056 32569372 Forero CG Maydeu-Olivares A Gallardo-Pujol D Factor analysis with Ordinal Indicators: a Monte Carlo Study comparing DWLS and ULS Estimation Structural Equation Modeling: A Multidisciplinary Journal 2009 16 4 625 641 10.1080/10705510903203573 Friedman, M. J., Keane, T. M., & Resick, P. A. (2007). In M. J. Friedman, T. M. Keane, & P. A. Resick (Eds.), Handbook of PTSD: Science and Practice (3rd ed.). The Guilford Press. Friedman MJ Resick PA Bryant RA Brewin CR Considering PTSD for DSM-5 Depression and Anxiety 2011 28 9 750 769 10.1002/da.20767 21910184 Gagne C Dayan P Bishop SJ When planning to survive goes wrong: predicting the future and replaying the past in anxiety and PTSD Curr Opin Behav Sci 2018 24 89 95 10.1016/j.cobeha.2018.03.013 Giorgi G Fiz Perez FS Castiello D’Antonio A Mucci N Ferrero C Cupelli V Arcangeli G Psychometric properties of the impact of event Scale-6 in a sample of victims of bank robbery Psychol Res Behav Manag 2015 8 99 104 10.2147/PRBM.S73901 25878514 Graupensperger S Calhoun BH Patrick ME Lee CM Longitudinal effects of COVID-19-related stressors on young adults’ mental health and wellbeing Applied Psychology: Health and Well-Being 2022 14 3 734 756 10.1111/aphw.12344 35102692 Greenberg, J., Pyszczynski, T., & Solomon, S. (1986). The Causes and Consequences of a Need for Self-Esteem: A Terror Management Theory. In R. F. Baumeister (Ed.), Public Self and Private Self (pp. 189–212). Springer New York. 10.1007/978-1-4613-9564-5_10 Greenberg J Solomon S Pyszczynski T Rosenblatt A Burling J Lyon D Simon L Pinel E Why do people need self-esteem? Converging evidence that self-esteem serves an anxiety-buffering function J Pers Soc Psychol 1992 63 6 913 922 10.1037/0022-3514.63.6.913 1460559 Harding KJ Skritskaya N Doherty E Fallon BA Advances in understanding illness anxiety Curr Psychiatry Rep 2008 10 4 311 317 10.1007/s11920-008-0050-1 18627669 Harmon-Jones E Simon L Greenberg J Pyszczynski T Solomon S McGregor H Terror management theory and self-esteem: evidence that increased self-esteem reduced mortality salience effects J Pers Soc Psychol 1997 72 1 24 36 10.1037/0022-3514.72.1.24 9008372 Hayes, A. F. (2022). Introduction to mediation, moderation, and conditional process analysis: a regression-based approach (3rd ed.). Guilford publications. Hedges LV Distribution theory for Glass’s Estimator of Effect size and related estimators Journal of Educational Statistics 1981 6 2 107 128 10.3102/10769986006002107 Heinrich LM Gullone E The clinical significance of loneliness: a literature review Clinical Psychology Review 2006 26 6 695 718 10.1016/j.cpr.2006.04.002 16952717 Horowitz, M., Wilner, N., & Alvarez, W. (1979). Impact of Event Scale: A Measure of Subjective Stress.Psychosomatic Medicine, 41(3). https://journals.lww.com/psychosomaticmedicine/Fulltext/1979/05000/Impact_of_Event_Scale__A_Measure_of_Subjective.4.aspx Howell, D. C. (2013). Statistical methods for psychology. Wadsworth, Cengage Learning. Hoyle, R. H. (2012). Handbook of Strucural equation modeling. The Guilford Press. Hu LT Bentler PM Cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives Structural equation modeling 1999 6 1 1 55 10.1080/10705519909540118 Huremović, D. (2019). Psychiatry of Pandemics. A Mental Health Response to Infection Outbreak (D. Huremović, Ed.). 10.1007/978-3-030-15346-5 Iannone, R. (2018). DiagrammeR: Graph/Network Visualization. In (Version 1.0.0) https://CRAN.R-project.org/package=DiagrammeR Johnson, T. P. (2014). Snowball Sampling: Introduction. In Wiley StatsRef: Statistics Reference Online. 10.1002/9781118445112.stat05720 Jorgensen, T. D., Pornprasertmanit, S., Schoemann, A. M., & Rosseel, Y. (2019). semTools: Useful tools for structural equation modeling. In (Version 0.5-2) https://CRAN.R-project.org/package=semTools Karatzias T Hyland P Bradley A Cloitre M Roberts NP Bisson JI Shevlin M Risk factors and comorbidity of ICD-11 PTSD and complex PTSD: findings from a trauma-exposed population based sample of adults in the United Kingdom Depression and Anxiety 2019 36 9 887 894 10.1002/da.22934 31268218 Kessler RC Aguilar-Gaxiola S Alonso J Benjet C Bromet EJ Cardoso G Degenhardt L de Girolamo G Dinolova RV Ferry F Florescu S Gureje O Haro JM Huang Y Karam EG Kawakami N Lee S Lepine JP Levinson D Navarro-Mateu F Pennell BE Piazza M Posada-Villa J Scott KM Stein DJ Ten Have M Torres Y Viana MC Petukhova MV Sampson NA Zaslavsky AM Koenen KC Trauma and PTSD in the WHO World Mental Health surveys European Journal of Psychotraumatology 2017 8 sup5 1353383 10.1080/20008198.2017.1353383 29075426 Kessler RC Sonnega A Bromet E Hughes M Nelson CB Posttraumatic stress disorder in the National Comorbidity Survey Archives Of General Psychiatry 1995 52 12 1048 1060 10.1001/archpsyc.1995.03950240066012 7492257 Kline, R. B. (2016). Principles and practice of structural equation modeling. The Guilford Press. Kratovic L Smith LJ Vujanovic AA PTSD symptoms, suicidal ideation, and suicide risk in University students: the role of distress tolerance Journal of Aggression Maltreatment & Trauma 2021 30 1 82 100 10.1080/10926771.2019.1709594 lavaan RY An R Package for Structural equation modeling Journal Of Statistical Software 2012 48 1 36 Li CH The performance of ML, DWLS, and ULS estimation with robust corrections in structural equation models with ordinal variables Psychological Methods 2016 21 3 369 387 10.1037/met0000093 27571021 Little TD Cunningham WA Shahar G Widaman KF To Parcel or not to Parcel: exploring the question, weighing the merits Structural Equation Modeling: A Multidisciplinary Journal 2002 9 2 151 173 10.1207/s15328007sem0902_1 Little TD Rhemtulla M Gibson K Schoemann AM Why the items versus parcels controversy needn’t be one Psychological Methods 2013 18 3 285 300 10.1037/a0033266 23834418 Liu CH Zhang E Wong GTF Hyun S Hahm HC Factors associated with depression, anxiety, and PTSD symptomatology during the COVID-19 pandemic: clinical implications for U.S. young adult mental health Psychiatry Research 2020 290 113172 10.1016/j.psychres.2020.113172 32512357 MacKinnon, D. P. (2012). Introduction to statistical mediation analysis. Routledge. MacKinnon DP Fairchild AJ Fritz MS Mediation analysis Annual Review of Psychology 2007 58 1 593 614 10.1146/annurev.psych.58.110405.085542 McDonald, R. P. (1999). Test theory: a unified treatment. Lawrence Erlbaum Associates. McIntyre RS Lee Y Preventing suicide in the context of the COVID-19 pandemic World Psychiatry 2020 19 2 250 251 10.1002/wps.20767 32394579 Milavic, B., Padulo, J., Grgantov, Z., Milić, M., Mannarini, S., Manzoni, G. M., Ardigò, L. P., & Rossi, A. (2019). Development and factorial validity of the psychological skills inventory for Sports, Youth Version – Short Form: Assessment of the psychometric properties [Article]. PLoS One, 14(8), 10.1371/journal.pone.0220930. Article e0220930. Millsap, R. E. (2012). Statistical approaches to measurement invariance [Book]. Routledge. 10.4324/9780203821961. Orcutt HK Erickson DJ Wolfe J A prospective analysis of Trauma exposure: the mediating role of PTSD Symptomatology Journal of Traumatic Stress 2002 15 3 259 266 10.1023/a:1015215630493 12092919 Pacella ML Hruska B Delahanty DL The physical health consequences of PTSD and PTSD symptoms: a meta-analytic review Journal of Anxiety Disorders 2013 27 1 33 46 10.1016/j.janxdis.2012.08.004 23247200 Panzeri, A., Bertamini, M., Butter, S., Levita, L., Gibson-Miller, J., Vidotto, G., Bentall, R. P., & Bennett, K. M. (2021). Factors impacting resilience as a result of exposure to COVID-19: the ecological resilience model [Article]. PLoS One, 16(8 August), 10.1371/journal.pone.0256041. Article e0256041. Parola, A. (2020). Novel coronavirus outbreak and Career Development: a narrative approach into the meaning of italian university graduates. Frontiers In Psychology, 11(2255), 10.3389/fpsyg.2020.02255. Parola A Rossi A Tessitore F Troisi G Mannarini S Mental Health through the COVID-19 Quarantine: a growth curve analysis on italian young adults [Article] Frontiers In Psychology 2020 11 567484 10.3389/fpsyg.2020.567484 33132973 Parola, A., Fusco, L., & Marcionetti, J. (2022). The parental career-related behaviors questionnaire (PCB): Psychometric properties in adolescents and young adults in the Italian context. Current Psychology. 10.1007/s12144-022-02764-1. Pietrabissa G Rossi A Borrello M Manzoni GM Mannarini S Castelnuovo G Molinari E Development and validation of a self-determination theory-based measure of motivation to Exercise and Diet in children [Article] Frontiers In Psychology 2020 11 1299 10.3389/fpsyg.2020.01299 32714231 Pietrabissa G Rossi A Simpson S Tagliagambe A Bertuzzi V Volpi C Fava G Manzoni GM Gravina G Castelnuovo G Evaluation of the reliability and validity of the italian version of the schema mode inventory for eating disorders: short form for adults with dysfunctional eating behaviors [Article] Eating and Weight Disorders 2020 25 3 553 565 10.1007/s40519-019-00644-5 30734224 Pyszczynski T Greenberg J Solomon S Arndt J Schimel J Why do people need Self-Esteem? A theoretical and empirical review Psychological Bulletin 2004 130 3 435 468 10.1037/0033-2909.130.3.435 15122930 Raykov T Scale validity evaluation with congeneric measures in hierarchical designs British Journal of Mathematical and Statistical Psychology 2011 64 3 427 438 10.1348/000711009X484286 21973095 Raykov, T. (2012). Scale construction and development using structural equation modeling. In R. H. Hoyle (Ed.), Handbook of structural equation modeling (pp. 472–492). The Guilford Press. Raykov, T., & Marcoulides, G. A. (2011). Introduction to psychometric theory. Routledge. Revelle, W. (2018). Psych: procedures for personality and Psychological Research. Northwestern University. Rosenberg, M. (1965). Society and the adolescent self-image. Princeton University Press. Rossi, A., Panzeri, A., Pietrabissa, G., Manzoni, G. M., Castelnuovo, G., & Mannarini, S. (2020). The Anxiety-Buffer Hypothesis in the Time of COVID-19: When Self-Esteem Protects From the Impact of Loneliness and Fear on Anxiety and Depression [Article]. Front Psychol, 11, Article 2177. 10.3389/fpsyg.2020.02177 Rossi, A. A., Marconi, M., Taccini, F., Verusio, C., & Mannarini, S. (2021). From fear to hopelessness: the buffering effect of patient-centered communication in a sample of oncological patients during covid-19 [Article]. Behavioral Sciences, 11(6), 10.3390/bs11060087. Rossi, A. A., Marconi, M., Taccini, F., Verusio, C., & Mannarini, S. (2022a). Screening for Distress in Oncological Patients: The Revised Version of the Psychological Distress Inventory (PDI-R) [Original Research]. Front Psychol, 13. 10.3389/fpsyg.2022.859478 Rossi, A. A., Panzeri, A., & Mannarini, S. (2022). The Italian Version of the Difficulties in Emotion Regulation Scale - Short Form (IT-DERS-SF): A Two-Step Validation Study. Journal of Psychopathology and Behavioral Assessment. https://doi.org/10.1007/s10862-022-10006-8 Rossi Ferrario S Panzeri A Exploring illness denial of LVAD patients in cardiac rehabilitation and their caregivers: a preliminary study Artificial Organs 2020 10.1111/aor.13630 Salzman, M. B., & Halloran, M. J. (2004). Cultural trauma and recovery: Cultural meaning, self-esteem and the reconstruction of the cultural anxiety buffer. Santini ZI Jose PE Cornwell Y Koyanagi E Nielsen A Hinrichsen L Meilstrup C Madsen C Koushede V Social disconnectedness, perceived isolation, and symptoms of depression and anxiety among older Americans (NSHAP): a longitudinal mediation analysis The Lancet Public Health 2020 5 1 e62 e70 10.1016/S2468-2667(19)30230-0 31910981 Sareen J Cox BJ Stein MB Afifi TO Fleet C Asmundson GJG Physical and Mental Comorbidity, disability, and suicidal Behavior Associated with Posttraumatic stress disorder in a large community sample Psychosomatic Medicine 2007 69 3 242 248 10.1097/PSY.0b013e31803146d8 17401056 Schipolowski S Schroeders U Wilhelm O Pitfalls and challenges in constructing short forms of cognitive ability measures Journal of Individual Differences 2014 35 190 200 10.1027/1614-0001/a000134 Shevlin, M., Butter, S., McBride, O., Murphy, J., Gibson-Miller, J., Hartman, T. K., Levita, L., Mason, L., Martinez, A. P., McKay, R., Stocks, T. V. A., Bennett, K., Hyland, P., & Bentall, R. P. (2021). Refuting the myth of a ‘tsunami’ of mental ill-health in populations affected by COVID-19: evidence that response to the pandemic is heterogeneous, not homogeneous. Psychological Medicine, 1–9. 10.1017/s0033291721001665. Silva Junior, F. J. G., d., Sales, J. C. S., Monteiro, C. F., d., S., Costa, A. P. C., Campos, L. R. B., Miranda, P. I. G., Monteiro, T. A. d., Lima, S., R. A. G., & Lopes-Junior, L. C. (2020). Impact of COVID-19 pandemic on mental health of young people and adults: a systematic review protocol of observational studies. BMJ Open, 10(7), e039426. 10.1136/bmjopen-2020-039426 Silver RC Surviving the trauma of COVID-19 Science 2020 369 6499 11 11 10.1126/science.abd5396 32631871 Solomon, S., Greenberg, J., & Pyszczynski, T. (2004). The Cultural Animal: twenty years of Terror Management Theory and Research. Handbook of experimental existential psychology (pp. 13–34). Guilford Press. Sowislo JF Orth U Does low self-esteem predict depression and anxiety? A meta-analysis of longitudinal studies Psychological Bulletin 2013 139 1 213 240 10.1037/a0028931 22730921 Starr LR Davila J Responding to anxiety with rumination and hopelessness: mechanism of anxiety-depression Symptom Co-Occurrence? Cognitive Therapy and Research 2012 36 4 321 337 10.1007/s10608-011-9363-1 22865943 Tabachnick, B. G., & Fidell, L. S. (2014). Using Multivariate Statistics. Pearson. Tessitore F Parola A Margherita G Mental Health risk and protective factors of nigerian male asylum seekers hosted in Southern Italy: a culturally sensitive quantitative investigation J Racial Ethn Health Disparities 2022 10.1007/s40615-022-01260-3 Thakur V Jain A COVID 2019-suicides: a global psychological pandemic Brain Behavior and Immunity 2020 88 952 953 10.1016/j.bbi.2020.04.062 32335196 Thompson EA Mazza JJ Herting JR Randell BP Eggert LL The mediating roles of anxiety, Depression, and hopelessness on adolescent suicidal behaviors Suicide and Life-Threatening Behavior 2005 35 1 14 34 10.1521/suli.35.1.14.59266 15843321 Torales J O’Higgins M Castaldelli-Maia JM Ventriglio A The outbreak of COVID-19 coronavirus and its impact on global mental health International Journal of Social Psychiatry 2020 66 4 317 320 10.1177/0020764020915212 32233719 van de Schoot R Lugtig P Hox J A checklist for testing measurement invariance European Journal of Developmental Psychology 2012 9 4 486 492 10.1080/17405629.2012.686740 van den Berg YHM Burk WJ Cillessen AHN Roelofs K Emerging adults’ Mental Health during the COVID-19 pandemic: a prospective longitudinal study on the importance of Social Support Emerging Adulthood 2021 9 5 618 630 10.1177/21676968211039979 34925969 VanderWeele T Vansteelandt S Mediation analysis with multiple mediators Epidemiologic Methods 2014 2 1 95 115 10.1515/em-2012-0010 25580377 Wang, C., Pan, R., Wan, X., Tan, Y., Xu, L., Ho, C. S., & Ho, R. C. (2020). Immediate Psychological Responses and Associated Factors during the Initial Stage of the 2019 Coronavirus Disease (COVID-19) Epidemic among the General Population in China. Int J Environ Res Public Health, 17(5), 1729. https://www.mdpi.com/1660-4601/17/5/1729 Waseem S Nayar SK Hull P Carrothers A Rawal J Chou D Khanduja V The global burden of trauma during the COVID-19 pandemic: a scoping review Journal of Clinical Orthopaedics and Trauma 2021 12 1 200 207 10.1016/j.jcot.2020.11.005 33223749 Weiss, D. S. (2007). The impact of event scale: revised. In J. P. Wilson, & C. S. Tang (Eds.), Cross-cultural Assessment of Psychological Trauma and PTSD (pp. 219–238). Springer US. 10.1007/978-0-387-70990-1_10. Weiss, D. S., & Marmar, C. R. (1996). The impact of event scale - revised. In J. Wilson, & T. M. Keane (Eds.), Assessing psychological trauma and PTSD (pp. 399–411). Guilford Press. Yaribeygi H Panahi Y Sahraei H Johnston TP Sahebkar A The impact of stress on body function: a review EXCLI Journal 2017 16 1057 1072 10.17179/excli2017-480 28900385 Zhang L Pan R Cai Y Pan J The prevalence of post-traumatic stress disorder in the General Population during the COVID-19 pandemic: a systematic review and single-arm Meta-analysis Psychiatry Investig 2021 18 5 426 433 10.30773/pi.2020.0458	0	PMC9734448	NO-CC CODE	2022-12-14 23:28:28	no	J Child Adolesc Trauma. 2022 Dec 7;:1-19	utf-8	J Child Adolesc Trauma	2,022	10.1007/s40653-022-00503-z	oa_other
==== Front Updates Surg Updates Surg Updates in Surgery 2038-131X 2038-3312 Springer International Publishing Cham 36479676 1411 10.1007/s13304-022-01411-5 Original Article Routine placement of abdominal drainage in pouch surgery does not impact on surgical outcomes http://orcid.org/0000-0002-1202-9447 Luberto Antonio 2 http://orcid.org/0000-0003-4090-0989 Crippa Jacopo 2 http://orcid.org/0000-0001-9920-0368 Foppa Caterina 12 http://orcid.org/0000-0002-7576-9629 Maroli Annalisa 2 http://orcid.org/0000-0001-8313-4909 Sacchi Matteo 2 http://orcid.org/0000-0003-2233-0871 De Lucia Francesca 2 http://orcid.org/0000-0002-8650-2321 Carvello Michele 12 http://orcid.org/0000-0002-1493-1768 Spinelli Antonino [email protected] 12 1 grid.452490.e Department of Biomedical Sciences, Humanitas University, Via Rita Levi Montalcini 4, 20090 Pieve Emanuele, Milan Italy 2 grid.417728.f 0000 0004 1756 8807 IRCCS Humanitas Research Hospital, via Manzoni 56, 20089 Rozzano, Milan Italy 7 12 2022 18 2 6 2022 22 10 2022 © Italian Society of Surgery (SIC) 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 evidence does not support the routine use of abdominal drainage (AD) in colorectal surgery. However, there is no data on the usefulness of AD, specifically, after ileal pouch-anal anastomosis (IPAA). The aim of this study is to assess post-operative outcomes of patients undergoing IPAA with or without AD at a high volume referral center. A retrospective analysis of prospectively collected data of consecutive patients undergoing IPAA with AD (AD group) or without AD (NAD group) was performed. Baseline characteristics, operative, and postoperative data were analyzed and compared between the two groups. A total of 97 patients were included in the analysis, 46 were in AD group and 51 in NAD group. AD group had a higher BMI (23.9 ± 3.9 kg/m2 vs 21.9 ± 3.0 kg/m2; p = 0.007) and more commonly underwent two-stage proctocolectomy with IPAA compared to the NAD group (50.0% vs 3.9%; p < 0.001). There was no difference in anastomotic leak rate (6.5% AD vs 5.9% NAD group; p = 1.000), major post-operative complication (8.6% vs 7.9%; p = 0.893); median length of stay [IQR] (5 [5–7] days vs 5 [4–7] days; p = 0.305) and readmission < 90 days (8.7% vs 3.9%; p = 0.418). The use of AD does not impact on surgical outcome after IPAA and question the actual benefit of its routine placement. Keywords Colorectal surgery IBD Laparoscopic surgery Drain Pouch surgery ==== Body pmcIntroduction Restorative proctocolectomy (RP) with ileal pouch–anal anastomosis (IPAA) is the gold standard for refractory ulcerative colitis (UC) or dysplasia and cancer in medically refractory UC, familial adenomatous polyposis (FAP), multiple synchronous colorectal cancer, and selected Crohn's disease (CD) patients [1–3]. RP and IPAA provide reasonable long-term functional results, as reflected in health-related quality of life studies [4–6]. However, postoperative pelvic sepsis, which is primarily responsible for significant long-term functional impairment and, eventually, for pouch failure, is the most feared complication [4]. Abdominal drainage (AD) in colorectal has been traditionally used to early detect intra-abdominal complications such as anastomotic leakage (AL), bleeding and collections [7, 8]. However, the evidence does not support its routine use in elective colorectal surgery [9]. Indeed, potential AD pitfalls such as increased post-operative pain at the drain site, surgical site infection, and abdominal obstruction have been reported [9, 10]. According to the most recent evidence, enhanced recovery program guidelines do not advise AD placement after colorectal surgery [11]. Their recommendation was corroborated by the results of a Cochrane systematic review [12] and a systematic review and meta-analysis [13] both concluding that pelvic and peritoneal AD did not decrease AL, mortality, wound infection and reoperation rates. There is no specific data on the usefulness and role of AD in pouch surgery. The aim of this study is to investigate post-operative outcomes of patients undergoing IPAA with or without AD placement. Methods Patients A prospectively maintained database was screened to collect all consecutive patients undergoing IPAA from January 2016 to June 2021 at a tertiary referral center. While before 2016, patients were routinely given AD after IPAA, over the study time-window, this practice gradually changed to a no drain policy. Data were collected according to the following variables: age at surgery, previous abdominal surgery, pathological diagnosis, pre-operative lab test (hemoglobin, creatinine and albumin), body mass index (BMI), American society of anesthesiologists (ASA) score, details on surgical procedure, length of stay (LOS), post-operative complication rate and Clavien–Dindo classification, post-operative unplanned readmission. Post-operative complications included AL, sepsis, ileus and complications other than leak. AL was registered according to the Italian multi-society consensus on the definition and management of anastomotic leak [2]. A defect of the intestinal wall at the anastomotic site leading to a communication between the intra- and extraluminal compartments diagnosed by surgical procedure, endoscopy, contrast enema was considered as AL. A pelvic abscess close to the anastomosis diagnosed by CT scan, even without any evident communication with the colonic lumen, was also considered as AL. Sepsis was defined as systemic inflammatory response syndrome (SIRS) signs with at least two of the following: fever > 38 °C, leukocytosis, heart rate > 90 bpm, respiratory rate > 20/min and confirmed or suspected infection. Pouch patients were included in the study if they met the following criteria: diagnosis of UC, multiple synchronous colorectal cancer or FAP. Patients were divided in two groups: patients who were placed an AD (AD group), and patients not receiving AD (NAD group). During the progressive transition to a no drain policy, AD was placed according to patients’ characteristics and assessment of intra-operative hazards. One experienced colorectal surgeons performed the surgical operations during the study period. The primary end-point was the rate post-operative complications (overall rate and rated according to the Clavien–Dindo score) [14]. The secondary end-points were AL, LOS, readmission, and stoma closure rate. Patients were followed-up until ileostomy closure or at least 90 days after index procedure. The study was conducted according to the ethical standards and the principles of the Declaration of Helsinki. This study complies with the ‘strengthening the reporting of observational studies in epidemiology’ (STROBE) statement for observational studies [15]. Surgical technique and post-operative care Patients underwent either a 2-, modified 2- or 3-stage RP. Minimally invasive approach (including trans-stomal single port access after subtotal colectomy with end-ileostomy) was the preferred approach whenever feasible. A 15 cm ileal J-pouch was created using two fires of 100-mm linear stapler. Proctectomy was carried out by up-to-down approach for conventional double-stapled IPAA or transanal transection and single stapling anastomosis technique and by bottom-up approach for transanal IPAA as previously described [6, 16]. A stapled IPAA was fashioned using a 31 mm circular stapler either through double-stapled or double purse-string single-stapled technique. A Foley catheter was left transanally into the pouch for decompression. When an AD was used, a capillary drain was placed posteriorly to the pouch into the pelvic cavity. In patients with AD, the drained fluid was evaluate daily for quantity and quality. Statistical analysis All the statistical calculations were performed using JMP Pro 15.0 (SAS Institute Inc., Cary, NC, USA). Continuous data were described as the means ± standard deviation (SD) and analyzed through a t-test when they were normally distributed. Continuous data that were not normally distributed were expressed as medians and interquartile range (IQR) and analyzed using Mann–Whitney U tests. Both the categorical data and ordinal data were presented as the number of cases and percentages. Categorical data were analyzed using a χ2 test or Fisher’s exact tests, while the ordinal data were subsequently analyzed using Mann–Whitney U tests. All analyses were two-sided, and p < 0.05 was considered as statistically significant. Results A total of 97 consecutive patients underwent IPAA surgery during the study time-window, 51 belong to NAD group and 46 to AD group. Clinical and pathological characteristics of the study population are described in Table 1. Patients in the AD group had a higher BMI (23.9 ± 3.9 kg/m2 vs 21.9 ± 3.0 kg/m2; p = 0.007). When compared to the NAD group, the AD group had a higher rate of 2-stage RP (50% vs 3.9%; p < 0.001) likely justified by the higher rate of FAP (19.6% vs 2.0%; p < 0.001), colorectal cancer on UC (23.9% vs 2.0%; p < 0.001). Table 2 describes intra e post-operative characteristics. Patients in the AD group had a longer mean operative time (344.1 ± 78.1 min vs 252.8 ± 51.2 min; p < 0.001) as a consequence of the higher rate of 2-stage procedures. AL rate was comparable between AD and NAD groups (6.5% [n = 3] vs 5.9% [n = 3]; p = 1.000, respectively). Using the Clavien–Dindo classification system, score’s grades were homogeneous (p = 0.835) between the two groups. A total of eight patients developed a severe complication (Clavien–Dindo ≥ III). Of 4 (8.6%) patients in the AD group, 1 had hemoperitoneum occurred at the first post-operative day, 1 patient underwent laparoscopic exploration for bowel obstruction, in 1 patient was placed a percutaneous drainage for a pelvic collection and 1 patient underwent surgery to treat a pouch-vaginal fistula. In the NAD group, 4 (7.9%) patients developed severe complications, of which 2 underwent peritoneal lavage and IPAA revision for AL, 1 had a stoma reversal for intestinal obstruction at ileostomy site and 1 patient had an endoscopic hemostasis to treat acute pouch bleeding after stoma closure. The median LOS [IQR] was similar between groups (5 [5–7] days in AD group vs 5 [4–7] days in NAD group; p = 0.305). Readmission within 90 days was 8.7% for the AD group and 3.9% for the NAD group and (p = 0.418). A total of 6 patients were readmitted. Reasons for readmission were post-operative fever due to pouch-vaginal fistula in 1 case, splenic vein thrombosis in 1 case and post-operative ileus in 2 patients in AD group; 1 case of anastomotic bleeding and 1 case of IPAA stricture in NAD group. A total of 45 (97.8%) patients in AD group and 48 (94.1%) patients in NAD group received an ileostomy after IPAA procedure. Stoma reversal rate was similar between groups (88.9% vs 93.8%; p = 0.752). The majority of patients who did not undergo stoma reversal declined surgery as a personal decision despite the absence of contraindications. Table 3 shows intra- e post-operative characteristics of 72 patients undergoing 3-stage procedures. Among these, 23 (31.9%) were in AD group and 49 (68.1%) were in NAD group. The 2 groups were different in the conversion rate with a higher rate in AD group (21.7% vs 2.0%; p = 0.011) and mean operative time [IQR] with longer time procedure in the same group (299 [269–338] min vs 236 [215–274] min; p < 0.001). Figure 1 shows the proportion of patients receiving AD according to study year. AD was utilized in 90.9% of patients treated in 2016. Thereafter, the highest proportion of no drain policy was in 2019 when only 30% of patients belonged to the AD group.Table 1 Pre-operative characteristics Drain n = 46 No drain n = 51 P Age 45.1 (12.8) 45.2 (14.2) 0.951 Female 23 (50%) 25 (49%) 0.923 BMI kg/m2 23.9 (3.9) 21.9 (3.0) 0.007* ASA 1 16 (34.8%) 23 (45.1%) 0.641 2 28 (60.9%) 26 (51%) 3 2 (4.3%) 2 (3.9%) S-Albumin g/dl 42.1 (3.1) 42.4 (2.5) 0.201 Hemoglobin g/dl 13.1 (2.0) 13.2 (1.7) 0.911 Creatinine mg/dl 0.8 (0.3) 0.8 (0.2) 0.486 Procedure 3-stage 23 (50.0%) 49 (96.1%) < 0.001* 2-stage 23 (50.0%) 2 (3.9%) Diagnosis UC 26 (56.5%) 49 (96.1%) < 0.001* UC with cancer 11 (23.9%) 1 (2.0%) FAP 9 (19.6%) 1 (2.0%) Values are expressed in mean ± SD or n° of patients and percentage ASA American society of anesthesiologist score, BMI body mass index. SD standard deviation Statistically significant Table 2 Intra and post-operative characteristics Drain n = 46 No drain n = 51 P Post-operative complications (overall) 15 (32.6%) 18 (35.3%) 0.780 Anastomotic leak^ 3 (6.5%) 3 (5.9%) 1.000 Conversion to open^ 5 (10.9%) 1 (2.0%) 0.098 Operative time (min) 344.1 (78.1) 252.8 (51.2) < 0.001 Temporary diversion 45 (97.8%) 48 (94.1%) 0.129 Clavien-Dindo^ 0 31 (67.4%) 33 (64.7%) 0.835 I 5 (10.9%) 4 (7.8%) II 6 (13.0%) 10 (19.6%) IIIa 2 (4.3%) 1 (2.0%) IIIb 2 (4.3%) 3 (5.9%) IV 0 (0.0%) 0 (0.0%) Sepsis^ 6 (13.0%) 5 (9.8%) 0.752 Complications other than leak 12 (26.1%) 17 (33.3%) 0.435 Ileus 6 (13.0%) 6 (11.8%) 1.000 Length of stay (days)° (IQR) 5 (5–7) 5 (4–7) 0.305 Readmission < 90 days^ 4 (8.7%) 2 (3.9%) 0.418 n = 45 n = 48 P Stoma closure 40 (88.9%) 45 (93.8%) 0.752 Values are expressed in mean ± SD or n° of patients and percentage, median (IQR) for length of stay SD standard deviation, IQR interquartile range Statistically significant °Wilcoxon’s signed rank test ^Fisher’s exact test Table 3 Intra and post-operative characteristics for 3-stage procedures Drain n = 23 No drain n = 49 P Post-operative complications (overall)^ 4 (17.4%) 18 (36.7%) 0.110 Anastomotic leak^ 0 (0.0%) 3 (6.1%) 0.546 Conversion to open^ 5 (21.7%) 1 (2.0%) 0.011 Operative time (min)° (IQR) 299 (269–338) 236 (215–274) < 0.001* Temporary diversion^ 22 (95.7%) 46 (93.9%) 1.000 Clavien–Dindo^ 0 19 (82.6%) 31 (63.3%) 0.499 I 2 (8.7%) 4 (8.2%) II 2 (8.7%) 10 (20.4%) IIIa 0 (0.0%) 1 (2.0%) IIIb 0 (0.0%) 3 (6.1%) IV 0 (0.0%) 0 (0.0%) Sepsis^ 1 (4.3%) 5 (10.2%) 0.657 Complications other than leak^ 3 (13.0%) 17 (34.7%) 0.089 Ileus^ 2 (8.7%) 6 (12.2%) 1.000 Length of stay (days)° (IQR) 5 (5–6) 5 (4–7) 0.578 Readmission < 90 days^ 0 (0.0%) 2 (4.1%) 0.199 n = 22 n = 46 P Stoma closure 21 (95.5%) 44 (95.7%) 0.752 Values are expressed in median (IQR) or n° of patients and percentage SD standard deviation, IQR interquartile range *Statistically significant °Wilcoxon’s signed rank test ^Fisher’s exact test Fig. 1 Cases/drain positioning according to year Discussion In our study, the use of prophylactic AD for pouch surgery did not affect surgical outcome. Furthermore, AL rate and grade was similar between those who received an AD and those who did not. The evidence on the value of drainage placement in rectal surgery for cancer does not support its routine use [10, 13, 17, 18]. In fact, a randomized trial [10] assessing the role of pelvic drain in rectal surgery for cancer showed no benefits in terms of reoperation rate, LOS, pelvic sepsis incidence and overall surgical morbidity. Furthermore, a recent multicenter prospective study evaluated the efficacy of the routinary use of intra-abdominal drainage in elective colorectal surgery, showing that its use is not associated with the early detection of collections, while it prolongs hospital stay and surgical site infections [9]. However, specific data on the role of AD after IPAA are not available. Our colorectal surgery division, has progressively transitioned to a no drain policy in the last years as testified by the steady decreasing in the use of AD during the study time-window (Fig. 1). Even though gradually abandoned, the choice to use the AD was undertaken according to patients’ characteristics and assessment of intra-operative hazards. The main reasons for AD placement were represented by the expected amount of post-operative serum secretion due to greater presence of adipose tissue, the necessity of performing an extensive adhesiolysis, and intra-operative complications that may pose the IPAA at higher risk, such as postoperative bleeding or anastomotic leak. This is reflected by the unequal BMI and operative time in the two groups (higher for the AD group), even though this difference is likely explained also by the different rate of 3- and 2-stage procedures. The choice of placing a drain in overweighted or obese patients is supported by previous studies reporting an increased risk for patients with a high BMI to develop postoperative complications after colorectal cancer surgery [19, 20]. Akiyoshi et al. analyzed 1194 consecutive patients who underwent laparoscopic resection in a single center and they classified patients according to BMI [19]. The study showed that BMI > 35 kg/m2 is an independent predictive factor of developing AL. The BMI may also have an impact on IPAA procedure [21]. In fact, an inadequate mesenteric length, which is often the case in obese patients, could determine high tension on the anastomosis and a consequent higher risk of developing an AL [22]. Our study showed that operative time was significantly longer in the AD group than the NAD group. The difference, despite the clear relation with higher number of two stage procedure for AD group, may be also related to surgeon’s decision to place an AD after a more demanding surgery. Indeed, a case–control study demonstrated a stronger association between prolonged operative time and the development of post-operative AL after colorectal surgery [23]. However, a large multicenter evaluation did not consider a long operative time as a risk of AL after IPAA surgery and only an high BMI, ASA score > 2 and a long disease course were independent risk factor for AL [24]. The rate of AL after IPAA reported in the literature ranges between 8 and 15% [6, 25]. In our series of 90-day short-term outcomes, we reported an overall leak rate of 6.2%. For instance, we mainly performed a 3-stage procedure, while in the above-mentioned studies more modified 2 stage procedures (with no protective stoma at the point of IPAA construction) were performed. The almost constant presence of ileostomy in our series may have influenced the AL rate towards a greater number of undetected leaks compared to other series. However, in our series, the use of AD did not increase the number of leak detection nor decreased its severity. This finding is supported by a systematic review and meta-analysis of Urbach et al. which questioned the use of drainage to detect AL in colon and rectal anastomosis [26]. They described only 1 out of 20 diagnosis of AL observing pus or enteric content in drainage fluid while the drain was in place. Hence, they concluded that a drain is rarely useful in expelling enteric material once a leak occurs. It is therefore unlikely that a drain may be useful even for the purpose of controlling a leak if one occurs. Interestingly, LOS was similar in both groups in our study, with a median value of 5 days. This may be explained by the early drain removal policy in our department. In fact, drains were generally withdrawn between the third and the fifth post-operative day if the postoperative course was uneventful. Despite no difference in LOS, Moloo et al. advised against the use of abdominal drains underlaying potential pitfalls associated with drain placement [27]. The presence of an abdominal drain may allow for bacterial infections, increase risk of incisional hernia or could cause mechanical bowel obstruction. An interesting animal study by Nora et al. shows that 34% of abdominal drains revealed bacterial growths from cultures of their interior portion [28]. This finding suggest that bacteria could migrate into the abdomen via the drain, and may represent a reason to avoid the routine use of an AD. The rate of post-operative ileus (POI) was 11.8% in the NAD group vs 13.0% in the AD group (p = 1.000). Our results are similar to those reported beforehand [29, 30]. Indeed, previous articles did not report the presence of drainage as a cause of POI. They describe open surgery, male gander, conversion to open, ileostomy diversion and AL as possible risk factor for developing POI [29, 30]. Instead, abdominal drainage was reported as a cause of bowel mechanical obstruction only in case report [31]. Our series include all consecutive laparoscopic assisted procedures. Indeed, our unit offers a minimal invasive option as a gold standard to all our patients undergoing IPAA surgery. Minimal invasive surgery provides multiple advantages in IPAA surgery. Indar et al. [32] evaluated the amount of post-operative adhesions during loop ileostomy closure in patients who underwent IPAA and they reported fewer pelvic adhesions to the abdominal wall and gynecologic organs in the laparoscopic group than in the open group. This results could explain the lower rate of female infertility after IPAA when a minimally-invasive approach is utilized [33]. Additional benefits of minimal invasive surgery compared with open surgery comprise shorter LOS, reduced post-operative pain with improvement of short-term outcomes along with better cosmetic results. Moreover, the quality of life is similar regardless of the surgical technique [34–36]. In the last few years, Enhanced Recovery Protocols (ERP) were shown to be effective in reducing post-operative complications in patients undergoing colorectal surgery through a standardization of perioperative procedures, including a routine no drain policy [37]. Most of the literature on this topic is based on colorectal cancer patients, whereas knowledge about ERP implementation in inflammatory bowel disease (IBD) has been barely investigated [38]. Vigorita et al. [39] showed in a recent systematic review that studies on the subject tend to include only patients with Crohn's disease or to include patients with Crohn's disease and ulcerative colitis despite the different disease settings. Based on our experience, we have included the no drain policy as part of our IBD-ERP. In our sub-analysis of patients who underwent a 3-stage procedure, a higher conversion rate in the AD group was observed (5/23 [21.7%] vs 1/49 [2.0%]; p = 0.011). The reason for conversion to open was in all patients the presence of diffuse abdominal adhesions. This is also reflected by the longer median operative time [IQR] in the AD group compared to the NAD group (299 [269–338] min vs 236 [215–274] min; p < 0.001). Even though a decreasing trend in the use of drain was registered in our study, in 2020 a higher rate of AD was registered in IPAA surgery (Fig. 1). One of the main reasons may be the impact of the COVID-19 pandemic [40], which severely affected our surgical practice, switching mainly to oncological or emergency procedures. As a result, a higher number of two-stage procedures, which is the gold standard in case of total proctocolectomy for polyposis or multiple concomitant colorectal cancer, was performed, while surgery for benign disease such as medically refractory UC, a three-stage procedure in our practice, was mostly deferred. Given the higher complexity of a 2-stage procedure, these patients were more likely to receive an intra-operative drain, according to the reasons expressed above. On the other hand, when a subanalysis of patients undergoing a 2- or 3-stage procedure, no differences were registered in post-operative complications, as reported by others [41]. Our study is limited by its retrospective single center design. Further, the lack of homogeneous pre-agreed policy for drain placement might be considered a selection bias. Of note, the present study reports the real life data of a high volume dedicated colorectal surgery division and highlights the possible lack of benefits in terms of postoperative outcome behind the use of surgical drain during elective IPAA surgery posing a point of discussion, worth of further investigations. Learning curve and surgical experience may definitively have played a role in affecting post-operative outcomes, considering that in our study the accrual date started prior to the introduction of a ‘no drain’ policy. Nevertheless, we present a cohort of consecutive patients treated in the same setting in a span of only 5 years. Further, even in the latest years the rate of patients who received an intraoperative drain was approximately 35%. These factors may mitigate experience as a confounder. In conclusion, our experience has shown similar short-term outcomes in patients undergoing IPAA surgery with or without AD placement and questions the usefulness of AD in pouch surgery. Funding No funding was needed for the writing of this manuscript. Data availability The authors declare that no patient data appear in this article. Declarations Conflict of interest AS acted as speaker for Johnson and Johnson and Takeda; MC acted as speaker for Takeda and Pfizer, the other authors have no conflicts of interest to declare. Protection of human and animal subjects The authors declare that no experiments were performed on humans or animals for this study. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Panis Y Poupard B Nemeth J Lavergne A Hautefeuille P Valleur P Ileal pouch/anal anastomosis for Crohn's disease Lancet 1996 347 9005 854 857 10.1016/s0140-6736(96)91344-6 8622390 2. Spinelli A Anania G Arezzo A Berti S Bianco F Bianchi PP Italian multi-society modified Delphi consensus on the definition and management of anastomotic leakage in colorectal surgery Updates Surg 2020 72 3 781 792 10.1007/s13304-020-00837-z 32613380 3. Spinelli A Allaix ME Restorative proctocolectomy in IBD: does it fit elderly patients too? J Crohns Colitis 2017 11 6 647 648 10.1093/ecco-jcc/jjx001 28130328 4. Fazio VW Ziv Y Church JM Oakley JR Lavery IC Milsom JW Schroeder TK Ileal pouch-anal anastomoses complications and function in 1005 patients Ann Surg 1995 222 2 120 127 10.1097/00000658-199508000-00003 7639579 5. McGuire BB Brannigan AE O'Connell PR Ileal pouch-anal anastomosis Br J Surg 2007 94 7 812 823 10.1002/bjs.5866 17571291 6. Chandrasinghe P Carvello M Wasmann K Foppa C Tanis P Perry-Woodford Z Transanal ileal pouch-anal anastomosis for ulcerative colitis has comparable long-term functional outcomes to transabdominal approach: a multicentre comparative study J Crohns Colitis 2020 14 6 726 733 10.1093/ecco-jcc/jjz174 31637417 7. Sehapayak S McNatt M Carter HG Bailey W Baldwin A Jr Continuous sump-suction drainage of the pelvis after low anterior resection: a reappraisal Dis Colon Rectum 1973 16 6 485 489 10.1007/BF02588875 4769224 8. Hilsabeck JR The presacral space as a collector of fluid accumulations following rectal anastomosis: tolerance of rectal anastomosis to closed suction pelvic drainage Dis Colon Rectum 1982 25 7 680 684 10.1007/BF02629540 7128370 9. EuroSurg Collaborative Intraperitoneal drain placement and outcomes after elective colorectal surgery: international matched, prospective, cohort study Br J Surg 2022 109 6 520 529 10.1093/bjs/znac069 35576382 10. Denost Q Rouanet P Faucheron JL Panis Y Meunier B Cotte E French Research Group of Rectal Cancer Surgery (GRECCAR) To drain or not to drain infraperitoneal anastomosis after rectal excision for cancer: the GRECCAR 5 randomized trial Ann Surg 2017 265 3 474 480 10.1097/SLA.0000000000001991 27631776 11. Gustafsson UO Scott MJ Hubner M Nygren J Demartines N Francis N Rockall TA Guidelines for perioperative care in elective colorectal surgery: enhanced recovery after surgery (ERAS®) society recommendations: 2018 World J Surg 2019 43 3 659 695 10.1007/s00268-018-4844-y 30426190 12. Jesus EC Karliczek A Matos D Castro AA Atallah AN Prophylactic anastomotic drainage for colorectal surgery Cochrane Database Syst Rev 2004 2004 4 CD002100 10.1002/14651858.CD002100.pub2 15495028 13. Podda M Di Saverio S Davies RJ Atzeni J Balestra F Virdis F Prophylactic intra-abdominal drainage following colorectal anastomoses. A systematic review and meta-analysis of randomized controlled trials Am J Surg 2020 219 1 164 174 10.1016/j.amjsurg.2019.05.006 31138400 14. Dindo D Demartines N Clavien PA Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey Ann Surg 2004 240 2 205 213 10.1097/01.sla.0000133083.54934.ae 15273542 15. von Elm E Altman DG Egger M Pocock SJ Gøtzsche PC Vandenbroucke JP STROBE Initiative The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies J Clin Epidemiol 2008 61 4 344 349 10.1016/j.jclinepi.2007.11.008 18313558 16. Spinelli A Carvello M D'Hoore A Foppa C Integration of transanal techniques for precise rectal transection and single-stapled anastomosis: a proof of concept study Colorectal Dis 2019 21 7 841 846 10.1111/codi.14631 30943327 17. Sagar PM Hartley MN Macfie J Mancey-Jones B Sedman P May J Randomized trial of pelvic drainage after rectal resection Dis Colon Rectum 1995 38 3 254 258 10.1007/BF02055597 7882787 18. Brown SR Seow-Choen F Eu KW Heah SM Tang CL A prospective randomised study of drains in infra-peritoneal rectal anastomoses Tech Coloproctol 2001 5 2 89 92 10.1007/s101510170005 11862564 19. Akiyoshi T Ueno M Fukunaga Y Nagayama S Fujimoto Y Konishi T Effect of body mass index on short-term outcomes of patients undergoing laparoscopic resection for colorectal cancer: a single institution experience in Japan Surg Laparosc Endosc Percutan Tech 2011 21 6 409 414 10.1097/SLE.0b013e31822e5fdc 22146162 20. Pikarsky AJ Saida Y Yamaguchi T Martinez S Chen W Weiss EG Is obesity a high-risk factor for laparoscopic colorectal surgery? Surg Endosc 2002 16 5 855 858 10.1007/s004640080069 11997837 21. Browning SM Nivatvongs S Intraoperative abandonment of ileal pouch to anal anastomosis–the Mayo clinic experience J Am Coll Surg 1998 186 4 441 445 10.1016/s1072-7515(98)00056-8 9544959 22. Khasawneh MA McKenna NP Abdelsattar ZM Johnson A Dozois EJ Pemberton JH Mathis KL Impact of BMI on ability to successfully create an IPAA Dis Colon Rectum 2016 59 11 1034 1038 10.1097/DCR.0000000000000686 27749478 23. Telem DA Chin EH Nguyen SQ Divino CM Risk factors for anastomotic leak following colorectal surgery: a case-control study Arch Surg 2010 145 4 371 376 10.1001/archsurg.2010.40 20404288 24. Sahami S Bartels SA D'Hoore A Fadok TY Tanis PJ Lindeboom R A multicentre evaluation of risk factors for anastomotic leakage after restorative proctocolectomy with ileal pouch-anal anastomosis for inflammatory bowel disease J Crohns Colitis 2016 10 7 773 778 10.1093/ecco-jcc/jjv170 26417046 25. Pellino G Keller DS Sampietro GM Carvello M Celentano V Coco C Italian Society of Colorectal Surgery SICCR Inflammatory bowel disease position statement of the Italian society of colorectal surgery (SICCR): ulcerative colitis Tech Coloproctol 2020 24 8 907 10.1007/s10151-020-02259-w 32564234 26. Urbach DR Kennedy ED Cohen MM Colon and rectal anastomoses do not require routine drainage: a systematic review and meta-analysis Ann Surg 1999 229 2 174 180 10.1097/00000658-199902000-00003 10024097 27. Moloo H Etzioni DA Intraoperative adjuncts in colorectal surgery Surg Clin North Am 2013 93 1 33 43 10.1016/j.suc.2012.09.007 23177064 28. Nora PF Vanecko RM Bransfield JJ Prophylactic abdominal drains Arch Surg 1972 105 2 173 176 10.1001/archsurg.1972.04180080027005 5044542 29. Millan M Biondo S Fraccalvieri D Frago R Golda T Kreisler E Risk factors for prolonged postoperative ileus after colorectal cancer surgery World J Surg 2012 36 1 179 185 10.1007/s00268-011-1339-5 22083434 30. Wolthuis AM Bislenghi G Lambrecht M Fieuws S van Overstraeten ADB Boeckxstaens G D’Hoore A Preoperative risk factors for prolonged postoperative ileus after colorectal resection Int J Colorectal Dis 2017 32 6 883 890 10.1007/s00384-017-2824-6 28444506 31. Poon CM Leong HT Abdominal drain causing early small bowel obstruction after laparoscopic colectomy JSLS 2009 13 4 625 627 20202408 32. Indar AA Efron JE Young-Fadok TM Laparoscopic ileal pouch-anal anastomosis reduces abdominal and pelvic adhesions Surg Endosc 2009 23 1 174 177 10.1007/s00464-008-0139-y 18855064 33. Beyer-Berjot L Maggiori L Birnbaum D Lefevre JH Berdah S Panis Y A total laparoscopic approach reduces the infertility rate after ileal pouch-anal anastomosis: a 2-center study Ann Surg 2013 258 2 275 282 10.1097/SLA.0b013e3182813741 23360923 34. Larson DW Davies MM Dozois EJ Cima RR Piotrowicz K Anderson K Sexual function, body image, and quality of life after laparoscopic and open ileal pouch-anal anastomosis Dis Colon Rectum 2008 51 4 392 396 10.1007/s10350-007-9180-5 18213489 35. Larson DW Cima RR Dozois EJ Davies M Piotrowicz K Barnes SA Safety, feasibility, and short-term outcomes of laparoscopic ileal-pouch-anal anastomosis: a single institutional case-matched experience Ann Surg 2006 243 5 667 670 10.1097/01.sla.0000216762.83407.d2 16633002 36. Larson DW Dozois EJ Piotrowicz K Cima RR Wolff BG Young-Fadok TM Laparoscopic-assisted vs. open ileal pouch-anal anastomosis: functional outcome in a case-matched series Dis Colon Rectum 2005 48 10 1845 1850 10.1007/s10350-005-0143-4 16175324 37. Ripollés-Melchor J Abad-Motos A Cecconi M Pearse R Jaber S Slim K Francis N Spinelli A Association between use of enhanced recovery after surgery protocols and postoperative complications in colorectal surgery in Europe: the EuroPOWER international observational study J Clin Anesth 2022 80 110752 10.1016/j.jclinane.2022.110752 35405517 38. Spinelli A Bazzi P Sacchi M Danese S Fiorino G Malesci A Gentilini L Poggioli G Montorsi M Short-term outcomes of laparoscopy combined with enhanced recovery pathway after ileocecal resection for Crohn's disease: a case-matched analysis J Gastrointest Surg 2013 17 1 126 132 10.1007/s11605-012-2012-5 22948838 39. Vigorita V Cano-Valderrama O Celentano V Vinci D Millán M Spinelli A Pellino G Inflammatory bowel diseases benefit from enhanced recovery after surgery [ERAS] protocol: a systematic review with practical implications J Crohns Colitis 2022 16 5 845 851 10.1093/ecco-jcc/jjab209 34935916 40. Wexner SD Cortés-Guiral D Gilshtein H Kent I Reymond MA COVID-19: impact on colorectal surgery Colorectal Dis 2020 22 6 635 640 10.1111/codi.15112 32359223 41. Lee GC Deery SE Kunitake H Hicks CW Olariu AG Savitt LR Comparable perioperative outcomes, long-term outcomes, and quality of life in a retrospective analysis of ulcerative colitis patients following 2-stage versus 3-stage proctocolectomy with ileal pouch-anal anastomosis Int J Colorectal Dis 2019 34 3 491 499 10.1007/s00384-018-03221-x 30610435	36479676	PMC9734453	NO-CC CODE	2022-12-14 23:28:28	no	Updates Surg. 2022 Dec 7;:1-8	utf-8	Updates Surg	2,022	10.1007/s13304-022-01411-5	oa_other
==== Front Arch Dermatol Res Arch Dermatol Res Archives of Dermatological Research 0340-3696 1432-069X Springer Berlin Heidelberg Berlin/Heidelberg 36471086 2497 10.1007/s00403-022-02497-y Review Lichen planus after COVID-19 infection and vaccination http://orcid.org/0000-0002-0041-3839 Zou Henry [email protected] 1 http://orcid.org/0000-0002-6951-2894 Daveluy Steven 2 1 grid.17088.36 0000 0001 2150 1785 Michigan State University College of Human Medicine, 15 Michigan St NE, Grand Rapids, MI 49503 USA 2 grid.254444.7 0000 0001 1456 7807 Department of Dermatology, Wayne State University School of Medicine, Detroit, MI 48201 USA 5 12 2022 18 22 10 2022 19 11 2022 28 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. Lichen planus (LP) is an inflammatory disorder believed to result from CD8 + cytotoxic T-cell (CTL)-mediated autoimmune reactions against basal keratinocytes. We present a review of LP following COVID-19 infection and vaccination. Literature searches were conducted on PubMed and Google Scholar from 2019 to 7/2022. 36 articles were selected based on subject relevance, and references within articles were also screened. 39 cases of post-vaccination LP and 6 cases of post-infection LP were found among case reports and case series. 152 cases of post-vaccination LP and 12 cases of post-infection LP were found in retrospective and prospective studies. LP is a rare complication of COVID-19 infection and vaccination that may be mediated by overstimulation of T-cell responses and proinflammatory cytokine production. However, it does not represent a limitation against COVID-19 vaccination, and the benefits of vaccination considerably outweigh the risks. Keywords Lichen planus (LP) Oral lichen planus (OLP) CD8 + cytotoxic T-cell (CTL) CD4 + helper T-cell (Th1) Interleukin (IL) Tumor necrosis factor (TNF) Interferon (INF) Oral lichenoid lesions (OLL) ==== Body pmcIntroduction Lichen planus (LP) is a chronic inflammatory disorder of unknown origin that frequently involves the skin and mucosa. Skin lesions classically present as flat-topped, purple papules that can be pruritic [1]. Oral lichen planus (OLP) is a subset of LP that can present as white reticular or erythematous lesions, papules, plaques, or painful erosions [1, 2]. LP pathogenesis is believed to result from an autoimmune reaction involving CD8+ cytotoxic T-cell (CTL) attack against basal keratinocytes in the epidermis and other unknown antigens [1]. LP has been associated with hepatitis C viral infection and autoimmune disorders including alopecia areata and ulcerative colitis [1]. However, there has been limited inquiry into the potential association between LP and COVID-19 infection and vaccination. We present a review of LP following COVID-19 infection and vaccination and its implications for adverse event monitoring. Methods Literature searches were conducted on PubMed and Google Scholar ranging from 2019 to 7/2022. Thirty-six articles were selected based on subject relevance; novel onset and flares of LP after COVID-19 infection and vaccination were included. References within selected articles were also screened. Selected articles included one review of LP, one prospective observational study, one retrospective registry-based study, one retrospective cohort study, one prospective cross-sectional study, one commentary, four case series, two letters responding to previously published studies, and twenty-four case reports. Results To date (7/2022), there have been 39 cases of LP after COVID-19 vaccination (Mage = 55.97 years, Rage = 28-86 years, Male:Female = 17:22) and 6 cases of LP after COVID-19 infection (Mage = 53.17 years, Rage = 41-63 years, Male:Female = 2:4) among published case reports and case series (Table 1, Appendix). Nine of the post-vaccination cases were flares. Tozinameran (Pfizer-BioNTech) was linked to 16 cases, Spikevax (Moderna) to four cases, Vaxzevria (Oxford-AstraZeneca) to eight cases, Sinopharm to eight cases, CoronaVac to two cases, and Jcovden (Johnson and Johnson) to one case; the administered vaccine was unspecified in two post-vaccination LP cases. Retrospective and prospective studies yielded 152 cases of LP after COVID-19 vaccination and 12 cases of LP after COVID-19 infection (Table 2, Appendix). Discussion Multiple authors have hypothesized that exposure to the COVID-19 spike protein antigen via infection or vaccination may trigger immune dysregulation including altered T-cell activity and elevated cytokines that mediate LP pathogenesis [2, 5, 7–10]. SARS-CoV-2 antigens in COVID-19 vaccines induce B-cell activation and a strong CD8+ cytotoxic T-cell (CTL) response that can escalate into an autoimmune reaction against basal keratinocytes in the epidermis, triggering keratinocyte apoptosis and subsequent LP development [2, 5]. Furthermore, the vaccines also activate CD4+ helper T-cells (Th1), which release proinflammatory cytokines including interleukin-2 (IL-2), tumor necrosis factor-ɑ (TNF-ɑ), and interferon-γ (IFN-γ) that maintain the CTL response, further upregulate Th1 activity, and induce tissue damage [2, 7–10]. TNF-ɑ, and IFN-γ result in basal keratinocyte apoptosis, the hallmark of LP. Their upregulation may thus help explain LP pathogenesis after COVID-19 infection and vaccination [4]. COVID-19 infection has also been associated with dysregulation of the mammalian target of rapamycin (mTOR) signaling pathway, which has been implicated in dysfunctional T-cell proliferation and OLP pathogenesis [2]. Moreover, it has been hypothesized that SARS-CoV-2 triggers the overexpression of TRIM21 (tripartite motif containing-21), which stimulates antiviral CTLs, increases cytokine production, and has been identified in OLP lesions using immunohistochemistry [2]. Another hypothesis is that molecular mimicry is responsible for triggering the autoimmune CTL and Th1 responses that mediate LP in both infection and vaccination [2, 4, 22, 32]. The SARS-CoV-2 antigen has demonstrated cross-reactivity with multiple endogenous human antigens, including those found on the basal keratinocytes of the epidermis [2, 4, 22, 32]. Some attribute this antigen cross-reactivity to genetic similarities or shared epitopes [4, 22]. Specifically, SARS-CoV-2 proteins demonstrated similarities to human mitochondrial M2 proteins, F-actin, and TPO proteins on selective epitope mapping [2]. However, others suggest that the propensity for SARS-CoV-2 to target the ACE2 receptor for host cell entry may be implicated, as ACE2 receptors are found in abundance among cells in the skin and oral mucosa [2, 13, 33]. Binding of the SARS-CoV-2 spike protein to ACE2 receptors on epidermal cells may trigger Th1 recruitment and the subsequent autoimmune cascade responsible for LP pathogenesis [13, 33]. Some also suggest that COVID-19 infection and vaccination can induce a hyperinflammatory reaction mediated by the reticuloendothelial system, leading to the development of LP or LP-like lesions [18, 30]. Meanwhile, specific ingredients in the formulations of COVID-19 vaccines might trigger type IV hypersensitivity reactions that can manifest as oral lichenoid lesions (OLL) [28]. Finally, there are concerns that immunocompromising comorbidities including hypertension, diabetes, vitamin D deficiency, and vitiligo are risk factors that may increase susceptibility to LP after COVID-19 infection or vaccination [2, 25]. Diabetes and hypertension have been identified as risk factors for OLP development and COVID-19 mortality, and vitamin D has been found to modulate Th1 cells and regulate T-cell-mediated immune activity [2]. The association between COVID-19 infection and LP remains under debate. A prospective observational study of 74 COVID-19 positive patients found that 16.2% of them had oral lesions attributed to LP [33]. However, the authors did not specify whether the diagnosis was confirmed by histopathological analysis or only based on clinical findings [35]. The potential relationship between COVID-19 vaccination and OLP was investigated through a retrospective cohort study that matched 217,863 vaccinated patients to 217,863 unvaccinated patients using the TriNetX database [28]. Incidence of OLP/OLL was significantly higher among vaccinated patients relative to unvaccinated patients (risk difference = 0.04%; p < 0.001; 95% confidence interval = 0.00027; 0.00053) [28]. The authors acknowledged that they were unable to clinically differentiate between OLL and OLP or entirely eliminate distribution differences in the frequency of NSAID use between the two cohorts [28]. Such adverse reactions are rare, often experience spontaneous remission, and should not be considered a contraindication to COVID-19 vaccination at a population level [28]. Both the retrospective cohort study (N = 435,726) and another retrospective registry-based study (N=58) found that mRNA-based vaccines were most commonly implicated in post-vaccination LP onset [28, 34]. Similarly, mRNA-based vaccines (Tozinameran and Spikevax) accounted for 20/39 cases of post-vaccination LP identified by this review. We hypothesize that the stronger immune responses induced by mRNA-based vaccines relative to other vaccines correlate with a higher risk of autoimmune T-cell-mediated reactions that can manifest as LP. Conclusions LP is a rare complication following COVID-19 infection and vaccination, and patients with immunocompromising comorbidities may be particularly vulnerable. OLP and OLL are considered premalignant, and healthcare providers should carefully monitor for LP-like adverse effects among vaccinated and unvaccinated patients as well as those with a history of COVID-19. Nonetheless, there is no definitive causal link between COVID-19 vaccination and LP. Moreover, there is scientific consensus that LP-related adverse effects do not constitute a contraindication against vaccination and that the benefits of COVID-19 vaccination continue to outweigh the risks significantly. Appendix Table 1 Case reports and series of Lichen planus after COVID-19 infection and vaccination Patient age and sex Infection or vaccination? COVID-19 Vaccine type Latency (days) Distribution Treatment Outcome 29-y.o. F [3] (Bularca et al.) Vaccination Tozinameran, 2nd dose (Pfizer-BioNTech) 7 days Dorsum of the hands, wrists, eyelids, sub-mammary region, lower extremities, and oral mucosa Methotrexate Unspecified 63-y.o. F [4] (Paolino and Rongioletti) Vaccination Tozinameran, 2nd dose 3 days Palms, wrists, and soles Acitretin 25 mg/day, topical calcipotriene/betamethasone dipropionate foam Total resolution of lesions in 1 month, but with residual palmar hyperpigmentation 49-y.o. M [5] (Troelzsch et al.) Vaccination Jcovden, single dose (Johnson & Johnson) 6 days Oral mucosa Topical clobetasol mouth irrigation solution 0.5 mg/mL Significant improvement after 4 weeks 28-y.o. F [6] (Kaomongkolgit et aland Sawangarun) Vaccination Tozinameran, 2nd dose 7 days Oral mucosa Fluocinolone acetonide 0.1% in orabase paste Significant improvement after 2 weeks 82-y.o. F [7] (Hlaca et al.) Vaccination Tozinameran, 2nd dose 14 days Axillae, flexural wrists and forearms, ankles, buttocks, lower back, and abdomen Prednisolone 20 mg/day Gradual improvement after 6 week taper 68-y.o. F [7] (Hlaca et al.) Vaccination Spikevax, 2nd dose (Moderna) 14 days Trunk, buttocks, extremities, ankles, forearms, flexural wrists, axillae, and palms Prednisolone 30 mg/day Resolution after 6 week taper 56-y.o. F [8] (Merhy et al.) Vaccination Tozinameran, 2nd dose 7 days Trunk Unspecified Unspecified 54-y.o. M [9] (Zagaria et al.) Vaccination Tozinameran, 1st dose 10 days Trunk, upper and lower limbs Oral prednisolone 25 mg daily for 7 days, then tapered for up to 4 weeks Rapid resolution without side effects or recurrence 72-y.o. M [10] (Alabdulaaly et al.) Vaccination Spikevax, 1st and 2nd doses ~ 30–60 days Gingiva and upper lip High potency topical steroids Improved within 3 months 61-y.o. M [10] (Alabdulaaly et al.) Vaccination Spikevax, 2nd dose ~ 30–45 days (flare) Gingiva and tongue Continued topical pimecrolimus 1% cream Recovered to baseline 12 weeks after 2nd dose 65-y.o. F [10] (Alabdulaaly et al.) Vaccination Tozinameran, 2nd dose 7 days (flare) Buccal mucosa and tongue Topical vitamin A 0.025% gel and clobetasol 0.05% gel four times daily Resolution after 4 weeks 65-y.o. F [10] (Alabdulaaly et al.) Vaccination Tozinameran, 2nd dose 1 day (flare) Gingiva and vestibular mucosa Topical clobetasol and bethanechol Significant reduction in erythema after 1 month 51-y.o. M [10] (Alabdulaaly et al Vaccination Tozinameran, 2nd dose 14 days (flare) Posterior buccal mucosa Topical pimecrolimus cream and turmeric supplementation Returned to baseline after 2 months 40-y.o. M [11] (Caggiano et al.) Vaccination Tozinameran, 2nd dose 30–31 days Buccal mucosa Replacement of amalgam fillings No clinical improvement after 6 months 82-y.o. F [12] (Baba et al.) Vaccination Tozinameran, 2nd dose 7 days Trunk, upper and lower limbs Unspecified Unspecified 49-y.o. M [13] (Zengarini et al.) Vaccination Vaxzevria (Oxford-AstraZeneca), 2nd dose 11 days Trunk, upper and lower limbs Topical steroids and systemic antihistamines Near complete resolution with mild residual erythema after 1 month 59-y.o. F [14] (Herzum et al.) Vaccination Tozinameran, 2nd dose 14 days (flare) Medial ankles and feet Topical high-potency corticosteroids Resolution after 3 weeks 46-y.o. M [15] (Alrawashdah et al.) Vaccination Vaxzevria, 1st dose 5 days Forehead, abdomen, back, and legs Topical clobetasol propionate 0.1% cream twice daily, hydroxyzine hydrochloride 25 mg three times daily. After 4 weeks, added hydroxychloroquine 200 mg twice daily Mild improvement after 4 weeks of topical steroids/oral antihistamines. Significant reduction in pruritus after 2 months of adding hydroxychloroquine, but minimal improvement in skin lesions 56-y.o. F [16] (Hiltun et al.) Vaccination Tozinameran, 2nd dose 2 days (flare) Ankles, flexural wrists and forearms, periumbilical area, breasts, and axillary folds High-potency topical corticosteroids Unspecified 86-y.o. M [17] (Gamonal et al.) Vaccination Vaxzevria, 1st and 2nd doses 7 days (1st dose), exacerbated after 2nd dose Upper and lower limbs, trunk, buttocks Halobetasol propionate 0.05% cream Unspecified 60-y.o. F [18] (Diab et al.) Vaccination Vaxzevria, 2nd dose 14 days (flare) Cheeks, forehead, and scalp Intralesional corticosteroids and Tofacitinib General improvement in follow-up visits 55-y.o. F [18] (Diab et al.) Vaccination Sinopharm, 1st dose 3 days Lower limbs and buttocks Metronidazole 500 mg twice daily Improvement on follow-up 45-y.o. F [19] (Shakoei et al.) Vaccination Sinopharm, 1st dose 14 days Arm, forearms, and ankle Topical corticosteroid Significant improvement 40-y.o. M [19] (Shakoei et al.) Vaccination Sinopharm, 1st and 2nd doses 10 days Wrist and forearms Topical corticosteroid Significant improvement 38-y.o. M [19] (Shakoei et al.) Vaccination Sinopharm, 1st and 2nd doses 21 days (flare) Arm and forearms Topical corticosteroid, calcineurin inhibitor Significant improvement 45-y.o. M [19] (Shakoei et al.) Vaccination Sinopharm, 1st and 2nd doses 7 days Forearms and chest Topical corticosteroid Significant improvement 45-y.o. M [19] (Shakoei et al.) Vaccination Vaxzevria, 1st dose 7 days Acral Systemic prednisolone Significant improvement 49-y.o. F [19] (Shakoei et al.) Vaccination Sinopharm, 1st dose 10 days Acral N/A Significant improvement 32-y.o. M [19] (Shakoei et al.) Vaccination Sinopharm, 2nd dose 10 days Extremities Oral prednisolone Significant improvement 65-y.o. F [20] (Kulkarni and Sollecito) Vaccination Unspecified “Immediately following the administration” (flare) Left buccal mucosa Unspecified Regressed to baseline after 3 weeks 44-y.o. M [21] (Awada et al.) Vaccination Vaxzevria, 2nd dose 14 days Axillae Betamethasone cream once daily Resolution after 4 weeks 65-y.o. F [22] (Masseran et al.) Vaccination Vaxzevria, 1st and 2nd doses 10 days (1st dose), 7 days (2nd dose) Arms, legs, buttocks, and abdomen Clobetasol propionate 0.05% cream Near-complete remission in 4 weeks with residual pruritus and pigmentation 35-y.o. F [23] (Sharda et al.) Vaccination Unspecified 14 days Buccal and gingival mucosa Short term steroids course Responded well 52-y.o. F [24] (Babazadeh et al.) Vaccination Sinopharm, 1st and 2nd doses 10 days (1st dose), 7 days (2nd dose) Extremities, inguinal and axillary folds, lips, and buccal mucosa Oral antihistamines, topical calcipotriol and triamcinolone Favorable response to treatment 64-y.o. F [25] (Piccolo et al.) Vaccination Tozinameran, 1st dose 5 days Lateral aspects of dorsal hands, in areas previously affected by vitiligo Topical and systemic corticosteroids Unspecified 64-y.o. F [26] (Sun et al.) Vaccination Vaxzevria, 1st dose 14 days Inframammary folds, axillae, lower back, and groin Topical betamethasone 0.05% ointment Minor clinical improvement after 2 months 81-y.o. M [27] (Picone et al.) Vaccination Spikevax, 1st dose 7 days Flexural wrists, lumbosacral region, posterior thighs, dorsal feet Topical clobetasol propionate and cetirizine 10 mg daily × 10 days Clinical remission after 15 days, no recurrence at 1-month follow-up 50-y.o. M [28] (Hertel et al.) Vaccination Tozinameran, 2nd dose 9 days Buccal mucosa Unspecified Unspecified 57-y.o. F [28] (Hertel et al.) Vaccination Tozinameran, 2nd dose 14 days Upper and lower vestibules Unspecified Unspecified 63-y.o. M [29] (Saleh et al.) Infection N/A 30–31 days Oral mucosa Topical corticosteroids 3 times daily for 10 days followed by a symptom-dependent taper Marked improvement after 4 weeks (decreased pain and size of lesions) 41-y.o. M [10] (Alabdulaaly et al.) Infection N/A 14 days Bilateral buccal mucosa and gingival margins Fluocinonide 0.05% gel Resolution in 1 month 56-y.o. F [10] (Alabdulaaly et al.) Infection N/A 30–31 days Buccal mucosa Fluocinonide 0.05% gel Unspecified 51-y.o. F [30] (Gimeno Castillo et al.) Infection N/A 21 days Lumbar area, feet, and hands Tapered oral prednisone, then clobetasol cream Responded to prednisone, but relapsed and only partially responded to clobetasol 56-y.o. F [31] (Burgos-Blasco et al.) Infection N/A 49 days Buccal mucosa Unspecified Unspecified 52-y.o. F [32] (Diaz-Guimaraens et al.) Infection N/A 5 days Right shin, buccal mucosa Clobetasol propionate 0.05% cream twice daily Resolution of pruritus after 10 days with a residual brown patch on shin Table 2 Retrospective and prospective studies of Lichen planus after COVID-19 infection and vaccination Authors Fidan et al. [33] Hertel et al. [28] McMahon et al. [34] Cebeci Kahrman et al. [36] Infection or vaccination? Infection Vaccination Vaccination Vaccination Study type Prospective observational Retrospective cohort Retrospective registry-based Prospective cross-sectional study All patients in sample Number of patients 74 435,726 (Cohort I of 217,863 vaccinated matched to Cohort II of 217,863 unvaccinated) 58 (patients with post-vaccination cutaneous reactions who had available biopsy samples) 2290 Percentage of males 66.2% Cohort I: 43.88% Cohort II: 44.20% Unspecified 43.6% Percentage of females 33.8% Cohort I: 56.12% Cohort II: 55.80% 62% 56.4% Mean age ± SD (years) 45.6 ± 12.8 Cohort I: 53.10 ± 21.81 Cohort II: 53.00 ± 22.54 Mean age unspecified Median age = 61 50.4 ± 17.9 Age range (years) 19–78 Cohort I: 12–90 Cohort II: 12–90 Range unspecified Interquartile range = 44–77 20–96 Lichen planus patients in sample Number of patients 12 Cohort I: 146 Cohort II: 59 4 2 Percentage of total sample 16.2% Cohort I: 0.067% Cohort II: 0.027% 6.90% 0.1% Age range (years) Unspecified Unspecified 31–72 “60 s” Location of lesions Tongue (n = 3), buccal mucosa (n = 4), gingiva (n = 4), palate (n = 1) Unspecified Trunk and extremities Bilateral forearms Vaccines N/A Cohort I only: 88 received mRNA-based, 58 received adenovirus vector-based Tozinameran (n = 3), Spikevax (n = 1) CoronaVac (n = 2) Acknowledgements None. Author contributions HZ (lead) and Dr. SD (supporting) were responsible for conceptualization, data curation, formal analysis, investigation, methodology, project administration, and original draft preparation. Funding acquisition, resources, and software are not applicable for this study. Dr. SD (lead) and HZ (supporting) were responsible for supervision, validation, and visualization. HZ (equal) and Dr. SD (equal) wrote, reviewed, and edited the manuscript. Funding The authors declare no source of funding. Data availability All data generated or analyzed during this study are included in this published article. Declarations Conflict of interest The authors have no conflicts of interest to declare. IRB approval status Exempt. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Le Cleach L Chosidow O Lichen planus N Engl J Med 2012 366 723 732 10.1056/NEJMcp1103641 22356325 2. Sood A Raghavan S Batra P Sharma K Talwar A Rise and exacerbation of oral lichen planus in the background of SARS-CoV-2 infection Med Hypotheses 2021 156 110681 10.1016/j.mehy.2021.110681 34601231 3. Bularca E Monte-Serrano J Villagrasa-Boli P Lapeña-Casado A de-le Fuente S Reply to “COVID vaccine–induced lichen planus on areas previously affected by vitiligo” J Eur Acad Dermatol Venereol 2022 10.1111/jdv.18001 4. Paolino G Rongioletti F Palmoplantar lichenoid drug eruption following the administration of Pfizer-BioNTech COVID-19 vaccine JAAD Case Rep 2022 21 182 184 10.1016/j.jdcr.2022.01.014 35097176 5. Troeltzsch M Gogl M Berndt R Troeltzsch M Oral lichen planus following the administration of vector-based COVID-19 vaccine (Ad26.COV2.S) Oral Dis 2021 10.1111/odi.14025 6. Kaomongkolgit R Sawangarun W Oral lichen planus following mRNA COVID-19 vaccination Oral Dis 2022 10.1111/odi.14182 7. Hlaca N Zagar T Kastelan M Peternel S Brajac I Prpic-Massari L New-onset lichen planus and lichen planus flare in elderly women after COVID-19 vaccination J Cosmet Dermatol 2022 10.1111/jocd.15185 8. Merhy R Sarkis A-S Kaikati J El Khoury L Ghosn S Stephan F New-onset cutaneous lichen planus triggered by COVID-19 vaccination J Eur Acad Dermatol Venereol 2021 10.1111/jdv.17504 9. Zagaria O Villani A Ruggiero A Potestio L Fabbrocini G Gallo L New-onset lichen planus arising after COVID-19 vaccination Dermatol Ther 2022 10.1111/dth.15374 10. Alabdulaaly L Sroussi H Epstein JB New onset and exacerbation of oral lichenoid mucositis following SARS-CoV-2 infection or vaccination Oral Dis 2022 10.1111/odi.14257 11. Caggiano M Amato M Di Spirito F Galdi M Sisalli L mRNA COVID-19 vaccine and oral Lichen planus: a case report Oral Dis 2022 10.1111/odi.14184 12. Baba A Kanekura T Tsubouchi H Lichenoid drug eruption after coronavirus disease 2019 vaccination J Dermatol 2022 10.1111/1346-8138.16294 13. Zengarini C Piraccini BM La Placa M Lichen Ruber Planus occurring after SARS-CoV-2 vaccination Dermatol Ther 2022 10.1111/dth.15389 14. Herzum A Burlando M Molle MF Micalizzi C Cozzani E Parodi A Lichen planus flare following COVID-19 vaccination: a case report Clin Case Rep 2021 10.1002/ccr3.5092 15. Alrawashdah HM, Al-Habahbeh O, Naser AY, Abu Serhan H, Hamdan O, Sweiss K, et al (2022) Lichen planus eruption following oxford-astrazeneca COVID-19 vaccine administration: a case report and review of literature. Cureus https://www.cureus.com/articles/85859-lichen-planus-eruption-following-oxford-astrazeneca-covid-19-vaccine-administration-a-case-report-and-review-of-literature. Accessed 13 Jul 2020 16. Hiltun I Sarriugarte J Martínez-de-Espronceda I Garcés A Llanos C Vives R Lichen planus arising after COVID-19 vaccination J Eur Acad Dermatol Venereol 2021 10.1111/jdv.17221 17. Gamonal SBL Gamonal ACC Marques NCV Adário CL Lichen planus and vitiligo occurring after ChAdOx1nCoV-19 vaccination against SARS-CoV-2 Dermatol Ther 2022 10.1111/dth.15422 18. Diab R Araghi F Gheisari M Kani ZA Moravvej H Lichen planus and lichen planopilaris flare after COVID-19 vaccination Dermatol Ther 2022 10.1111/dth.15283 19. Shakoei S Kalantari Y Nasimi M Tootoonchi N Ansari MS Razavi Z Cutaneous manifestations following COVID-19 vaccination: a report of 25 cases Dermatol Ther 2022 10.1111/dth.15651 20. Kulkarni R Sollecito TP COVID-19 vaccination: possible short-term exacerbations of oral mucosal diseases Int J Dermatol 2022 10.1111/ijd.15779 21. Awada B Abdullah L Kurban M Abbas O Inverse lichen planus post Oxford-AstraZeneca COVID-19 vaccine J Cosmet Dermatol 2022 21 883 885 10.1111/jocd.14738 34997985 22. Masseran C Calugareanu A Caux F Bohelay G Extensive cutaneous lichen planus triggered by viral vector COVID-19 vaccination (ChAdOx1 nCoV-19) J Eur Acad Dermatol Venereol 2022 10.1111/jdv.17899 23. Sharda P Mohta A Ghiya BC Mehta RD Development of oral lichen planus after COVID-19 vaccination – a rare case report J Eur Acad Dermatol Venereol 2022 10.1111/jdv.17718 24. Babazadeh A Miladi R Barary M Shirvani M Ebrahimpour S Aryanian Z COVID-19 vaccine-related new-onset lichen planus Clin Case Rep 2022 10.1002/ccr3.5323 25. Piccolo V Mazzatenta C Bassi A Argenziano G Cutrone M Grimalt R COVID vaccine-induced lichen planus on areas previously affected by vitiligo J Eur Acad Dermatol Venereol 2022 10.1111/jdv.17687 26. Sun L Duarte S Soares-de-Almeida L Case of lichen planus pigmentosus–inversus after Oxford–AstraZeneca COVID-19 vaccine: cause or coincidence? J Eur Acad Dermatol Venereol 2022 10.1111/jdv.18058 27. Picone V Fabbrocini G Martora L Martora F A case of new-onset Lichen planus after COVID-19 vaccination Dermatol Ther 2022 12 801 805 10.1007/s13555-022-00689-y 28. Hertel M Schmidt-Westhausen A-M Wendy S Heiland M Nahles S Preissner R Onset of oral lichenoid lesions and oral lichen planus following COVID-19 vaccination: a retrospective analysis of about 300,000 vaccinated patients Vaccines 2022 10 480 10.3390/vaccines10030480 35335112 29. Saleh W SHawky E Halim GA Ata F Oral lichen planus after COVID-19, a case report Ann Med Surg. 2021 72 103051 10.1016/j.amsu.2021.103051 30. Gimeno Castillo J de la Torre Gomar FJ Saenz AA Lichen planus after SARS-CoV-2 infection treated with lopinavir/ritonavir Med Clínica 2021 156 468 469 10.1016/j.medcli.2020.12.014 31. Burgos-Blasco P Fernandez-Nieto D Selda-Enriquez G Melian-Olivera A De Perosanz-Lobo D Dominguez-Santas M COVID-19: a possible trigger for oral lichen planus? Int J Dermatol 2021 60 882 883 10.1111/ijd.15529 33686658 32. Diaz-Guimaraens B Dominguez-Santas M Suarez-Valle A Fernandez-Nieto D Jimenez-Cauhe J Ballester A Annular lichen planus associated with coronavirus SARS-CoV-2 disease (COVID-19) Int J Dermatol 2021 60 246 247 10.1111/ijd.15338 33320353 33. Fidan V Koyuncu H Akin O Oral lesions in Covid 19 positive patients Am J Otolaryngol 2021 42 102905 10.1016/j.amjoto.2021.102905 33497914 34. McMahon DE Kovarik CL Damsky W Rosenbach M Lipoff JB Tyagi A Clinical and pathologic correlation of cutaneous COVID-19 vaccine reactions including V-REPP: a registry-based study J Am Acad Dermatol 2022 86 113 121 10.1016/j.jaad.2021.09.002 34517079 35. Bin DA Letter to the editor: oral lesions in Covid 19 positive patients Am J Otolaryngol 2021 42 103176 10.1016/j.amjoto.2021.103176 34446327 36. Cebeci Kahraman F Savaş Erdoğan S Aktaş ND Albayrak H Türkmen D Borlu M Cutaneous reactions after COVID -19 vaccination in Turkey: a multicenter study J Cosmet Dermatol 2022 21 3692 3703 10.1111/jocd.15209 35780311	36471086	PMC9734460	NO-CC CODE	2022-12-14 23:28:28	no	Arch Dermatol Res. 2022 Dec 5;:1-8	utf-8	Arch Dermatol Res	2,022	10.1007/s00403-022-02497-y	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 36478269 1470 10.1007/s40615-022-01470-9 Article Early Pandemic Access to COVID-19 Testing in the Somali Community in King County, Washington, USA: a Mixed-Methods Evaluation Abdi Najma 1 Ebengho Sabrina 12 Mohamed Nasra 3 Scallon Andrea 4 Mohamed Ayan 1 Ahmed Asiya 12 Abdi Abdifatah 2 Ahmed Ruweida 5 Mohamed Farah 2 Ibrahim Anisa 26 Ali Ahmed 2 West Kathleen McGlone 7 http://orcid.org/0000-0001-5625-4884 Ronen Keshet [email protected] 4 1 grid.34477.33 0000000122986657 School of Public Health, University of Washington, Seattle, WA USA 2 Somali Health Board, Tukwila, WA USA 3 grid.34477.33 0000000122986657 University of Washington, Bothell, WA USA 4 grid.34477.33 0000000122986657 Department of Global Health, University of Washington, Seattle, WA USA 5 grid.34477.33 0000000122986657 School of Medicine, University of Washington, Seattle, WA USA 6 grid.34477.33 0000000122986657 Department of Pediatrics, University of Washington, Seattle, WA USA 7 grid.34477.33 0000000122986657 Department of Health Systems and Population Health, University of Washington, Seattle, WA USA 7 12 2022 114 24 7 2022 11 11 2022 21 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. Background Racial and ethnic disparities in COVID-19 infection and outcomes have been documented, but few studies have examined disparities in access to testing. Methods We conducted a mixed methods study of access to COVID-19 testing in the Somali immigrant community in King County, Washington, USA, early during the COVID-19 pandemic. In September 2020–February 2021, we conducted quantitative surveys in a convenience sample (n = 528) of individuals who had accessed PCR testing, recruited at King County testing sites near Somali population centers and through social media outreach in the Somali community. We compared self-identified Somali and non-Somali responses using Chi-square and Wilcoxon rank sum tests. We also conducted three Somali-language focus groups (n = 26) by video conference to explore Somali experiences with COVID-19 testing, and in-depth interviews with King County-based policymakers and healthcare workers (n = 13) recruited through the research team’s professional network to represent key demographics and roles. Data were analyzed using qualitative rapid analysis to explore the county’s COVID-19 testing landscape. Results Among 420 survey respondents who had received COVID-19 testing in the prior 90 days, 29% of 140 Somali vs. 11% of 280 non-Somali respondents tested because of symptoms (p = 0.001), with a trend for longer time from symptom onset to testing (a measure of testing access) among Somali respondents (median 3.0 vs. 2.0 days, p = 0.06). Focus groups revealed barriers to testing, including distrust, misinformation, stigma, language, lack of awareness, and transportation. Stakeholders responding from all sectors highlighted the importance of community partnership to improve access. Conclusion Somali communities experience barriers to COVID-19 testing, as evidenced by the longer time from symptom onset to testing and corroborated by our qualitative findings. These barriers, both structural and community-derived, may be overcome through partnerships between government and community to support community-led, multilingual service delivery and racial representation among medical staff. Supplementary Information The online version contains supplementary material available at 10.1007/s40615-022-01470-9. Keywords COVID-19 Testing Somali Immigrant Population Health Initiative, University of Washington ==== Body pmcIntroduction In the USA, the Coronavirus Disease 2019 (COVID-19) pandemic has exacerbated long-standing health disparities. Black Americans have an increased risk of COVID-19 infection, hospitalization, and mortality [1, 2], with a COVID-19 death rate 2–3 times higher than White individuals [3]. Compared to those born in the USA, immigrants have elevated COVID-19 infection risk, attributable to poverty, multigenerational and overcrowded housing, and occupations that hinder physical distancing [4–6]. Well-documented disparities in a variety of healthcare and prevention services based on race/ethnicity, income, transportation, English proficiency, and access to technology apply to COVID-19 [7]. Timely COVID-19 testing is critical to epidemic control as it enables outbreak containment and treatment. While disparities in COVID-19 acquisition, hospitalization, and mortality by race, ethnicity, English proficiency, and immigration status were well-documented early in the pandemic, fewer studies have examined disparities in testing [8–13], and routine public health data have not reported disaggregated testing rates by demographic characteristics [14]. King County, Washington, is home to approximately 30,000 Somali immigrants, comprising about 20% of the county’s Black population [15]. Many Somali immigrants have limited English proficiency, are low-income, and experience attitude and structural barriers to prevention and care services for a variety of conditions [16, 17]. Data from King County and Washington State indicate that several immigrant groups with limited English proficiency are at higher risk of infection and hospitalization than those who speak English fluently [18, 19]. A study of COVID-19 testing in a single healthcare system in King County early in the epidemic documented disparities among various language speakers, with lower testing rates and higher test positivity proportions among non-English speakers, including Somali speakers, vs. English speakers [8]. Anecdotal observations of barriers to accessing COVID-19 testing led our team of university and community partners to evaluate the King County Somali community’s access to testing at community sites. We conducted a mixed-methods study to quantitatively determine access disparities using the established measure of time from symptom onset to testing and to qualitatively identify barriers and recommendations to augment local testing services. Methods We employed a convergent mixed methods design, collecting Somali community member data from a quantitative survey and focus group discussions concurrently [20]. Preliminary analysis of data from focus groups and the survey were used to inform the development of in-depth interview guides. Toward the end of the interviews, we presented a preliminary summary of these data to policy maker and healthcare worker interviewees for their reflection on what we found in the community. Quantitative data were analyzed separately from qualitative data and then compared to formulate a contextualized view of COVID-19 testing disparities in the Somali community. Below, we describe the three data collection methods. Quantitative Data Collection We conducted an anonymous survey of Somali and non-Somali participants seeking COVID-19 testing in King County, WA, between September 2020 and February 2021. Eligible participants were aged ≥ 18 and had undergone polymerase chain reaction (PCR) COVID-19 testing, using the deep nasal swab method. Respondents were recruited using two approaches. (1) Bilingual study staff recruited a convenience sample of individuals presenting for testing at King County-hosted community-based PCR testing sites near Somali population centers. Staff at Somali Health Board branded tables made the survey available to all individuals presenting during data collection hours, and administered the survey in Somali or English, based on respondent preference, entering data using REDCap software [21]. (2) Somali community members who had been tested at any time were purposively recruited by disseminating paper and online flyers through the study team’s social networks; respondents self-administered the survey online using REDCap software. We aimed to reach a target sample of 200, recruiting as long as resources allowed. The survey ascertained participants’ demographic characteristics, time from symptom onset to testing, perceived barriers to testing access, and factors influencing decision to test from a predetermined set of options. Survey questions were developed by the study team and are available as a supplemental file for this manuscript. Participants provided informed consent verbally if the survey was staff-administered, or electronically on the consent form page of the online survey if it was self-administered. Qualitative Data Collection Focus Groups We conducted 3 focus group discussions (FGDs) with Somali community members who both had and had not been tested for COVID-19. Discussions were conducted in Somali during November 2020, with an additional group held in February 2021, by a bilingual, bicultural investigator who is also a known Somali community member. A semi-structured discussion guide covered experiences with, views of, and barriers and facilitators to COVID-19 testing. FGDs were chosen over in-depth interviews (IDIs) per Somali team members’ advice, as FGDs are familiar and comfortable to community members. Due to resource constraints, we aimed to have one group of men and one of women, to align with common practices of gender separation. Later, we conducted an additional group targeting older adults (> 60 years) to balance the skewing of our survey sample toward younger ages. Interviews After completing FGDs with community members, IDIs were conducted with healthcare workers serving the Somali community and local government stakeholders working at the state, county, and city levels in areas with high Somali population density. Interviews were conducted in English during January-March 2021. A semi-structured interview guide addressed interviewees’ roles in COVID-19 testing implementation or policymaking, testing successes and challenges for local immigrant communities, community testing models and government testing goals, and lessons learned about testing equity. Interviewees were given and asked to comment on the preliminary results of the community FGD data in their professional capacity. All FGD and IDI participants were recruited purposively [22] through investigators’ community and professional networks via email and phone. FGD participants were selected to represent a range of characteristics. IDI participants were selected to reach stakeholders at multiple levels of government (city, county, state), and healthcare workers in multiple healthcare professions and representing both Somali and non-Somali ethnicities. Sessions were conducted and recorded via video conference, lasting 30–60 min. Interviews were conducted individually, except one group interview that involved four government stakeholders who assumed varying roles within the same office. Participants provided verbal informed consent and, when ethically permitted, received a $50 gift for their time. The study team transcribed all recordings and translated Somali discussions into English. Quantitative Data Analysis Somali identity was the exposure of interest. Outcomes of interest included time from symptom onset to testing and motivations, perceived accessibility, and barriers to testing. We compared participant characteristics and outcomes of interest between respondents who self-identified as Somali vs. non-Somali (this could include first generation Somali immigrants as well as second generation Somali Americans). Proportions were compared by Chi-square test; if any categories contained < 5 individuals, Fisher’s exact test was used. Days from symptom onset to testing were compared between Somali and non-Somali participants using the Wilcoxon rank sum test. The threshold for statistical significance was a p-value 0.05. Univariate and multivariate regressions were not conducted due to sample size constraints. Statistical analyses were conducted using RStudio version 1.3.1093 (2020–10-10). Qualitative Data Analysis We deductively developed a structured coding template based on each IDI/FGD guide using the rapid analysis approach to qualitative analysis [23]. In addition, our FGD coding template contained codes for the 5 dimensions of healthcare access theorized by Penchansky and Thomas (affordability, availability, accessibility, accommodation, acceptability) [24]. After training by the team’s senior qualitative researcher (KW), all analysts used Microsoft Excel to apply the template to one transcript for calibration and template revision, consistent with the rapid analysis approach. Teams of two analysts independently double coded each IDI and FGD transcript and resolved coding discrepancies through discussion. Analyst pairs’ reconciled coding templates for a single stakeholder group were compiled. The same process was applied to each stakeholder group. The combined templates were used to draft stakeholder-specific narrative summaries [23]. The full analytic team read summaries for commonalities across stakeholder groups and stakeholder-specific views. Ethics Approval This study was determined to be exempt by the University of Washington Institutional Review Board. Results Survey Findings Participant Characteristics In total, 528 individuals completed the survey. To reduce recall bias, we limited analyses to those who had tested within the previous 90 days (n = 420). Of these, 140 participants (33%) were Somali and 280 (67%) were non-Somali (Table 1). All Somali participants identified as Black/African American/African-born. Non-Somali participants were 48% White, 19% Asian/Asian American, 18% Black/African American/African-born, 10% Hispanic/Latinx, and 4% Native Hawaiian/Pacific Islander. The study population was young, with 187 (45%) participants aged 18–29 and 112 (27%) aged 30–39. The majority spoke English fluently or proficiently (83%) and were employed either full- or part-time (76%). Around half (53%) had private health insurance, 131 (32%) had public insurance, and 62 (15%) were uninsured.Table 1 Survey participant characteristics Overall (N = 420) Non-Somali (N = 280) Somali (N = 140) p-value Somali vs non-Somali Characteristic N n or median % or IQR N n or median % or IQR N n or median % or IQR Race/Ethnicity* 420 280 140 African American/ Black/ African-Born 191 45.5% 51 18.2% 140 100.0% American Indian/Alaska Native 2 0.0% 2 0.7% 0 0.0% Asian/Asian American 53 12.6% 53 18.9% 0 0.0% Hispanic/Latinx 27 6.4% 27 9.6% 0 0.0% Native Hawaiian/Pacific Islander 12 2.9% 12 4.3% 0 0.0% White 135 32.1% 135 48.2% 0 0.0% Other 8 1.9% 8 2.9% 0 0.0% Age Group 418 278 140 0.10 18–29 187 44.7% 115 41.4% 72 51.4% 30–39 112 26.8% 76 27.3% 36 25.7% 40–49 54 12.9% 35 12.6% 19 13.6% 50–59 38 9.1% 31 11.2% 7 5.0% 60 + 27 6.5% 21 7.6% 6 4.3% Comfort in English 416 276 140 < 0.001 ≤ Intermediate 73 17.5% 31 11.2% 42 30.0% ≥ Proficient 343 82.5% 245 88.8% 98 70.0% Employment* 397 263 134 Full-time 233 58.7% 174 66.2% 59 44.0% < 0.001 Part-time 67 16.9% 34 12.9% 33 24.6% 0.004 Unemployed 80 20.2% 48 18.3% 32 23.9% 0.18 Student 32 8.1% 10 3.8% 22 16.4% < 0.001 Health Insurance 412 273 139 < 0.001 Uninsured 62 15.0% 46 16.8% 16 11.5% Public Insurance 131 31.8% 45 16.5% 86 61.9% Private Insurance 219 53.2% 182 66.7% 37 26.6% Reason for testing 418 278 140 < 0.001 Symptoms 74 17.7% 31 11.2% 43 30.7% In contact with positive individual 107 25.6% 73 26.3% 34 24.3% In preparation for travel/Visit Vulnerable Person 57 13.6% 42 15.1% 15 10.7% Other 180 43.1% 132 47.5% 48 34.3% Days since symptom onset 2.0 1.0–3.0 3.0 1.5–4.0 0.06 Accessibility of Testing 348 223 125 0.13 Hard to access 55 15.8% 30 13.5% 25 20.0% Easy to access 293 84.2% 193 86.5% 100 80.0% Motivators for testing site choice* 414 277 137 Saw it advertised 34 8.2% 15 5.4% 19 13.9% 0.005 Close to home 209 50.5% 136 49.1% 73 53.3% 0.54 Free 99 23.9% 45 16.2% 54 39.4% < 0.001 Interpretation 2 0.5% 2 0.7% 0 0.0% 0.55 Other 130 31.4% 110 39.7% 20 14.6% < 0.001 Hesitations to test* 420 280 140 None 294 70.0% 203 72.5% 91 65.0% 0.14 Wait time 35 8.3% 20 7.1% 15 10.7% 0.29 Cost of testing 14 3.3% 5 1.8% 9 6.4% 0.02 Transport to site 15 3.6% 12 4.3% 3 2.1% 0.40 Perceived test utility** 29 6.9% 7 2.5% 22 15.7% < 0.001 Knowledge of free COVID-19 testing 417 278 139 < 0.001 Aware 334 80.1% 238 85.6% 96 69.1% Not aware 42 10.1% 24 8.6% 18 12.9% Unsure 41 9.8% 16 5.8% 25 18.0% Sum is greater than 100% as participants could select more than one response Perceived test utility includes not trusting the accuracy of the result, feeling the result would not influence their behavior due to long turnaround or inability to quarantine, and feeling confident they did not have COVID-19 Compared with non-Somali participants, fewer Somali participants were proficient in English (70% vs. 89%, p < 0.001), fewer were employed full-time (44% vs. 66%, p < 0.001), and more were employed part-time (25% vs. 13%, p = 0.004). Somali participants were also significantly more likely to be on public insurance options of Medicaid, Medicare, or Veterans Affairs (62% vs. 17%, p < 0.001). Factors Influencing Testing Overall, 30% of participants reported some hesitation about COVID-19 testing (Table 1). The proportion of Somali participants who expressed any hesitation was higher than non-Somalis, though this did not reach statistical significance (35% vs. 28%, p = 0.14). Hesitation due to cost and perceived utility were both higher in Somali participants (6%, vs. 2%, p = 0.02, and 16% vs. 3%, p < 0.001, respectively). A smaller proportion of Somali participants reported awareness that COVID-19 testing was available for free in King County (69% vs. 86%, p < 0.001). When asked about their choice of test location, Somali participants were significantly more likely to report no-cost testing (39% vs. 16%, p < 0.001) and seeing the location advertised (14% vs. 5%, p = 0.005) as deciding factors. About a quarter of participants (107, 26%) tested due to COVID-19 exposure and 57 (14%) tested ahead of contact with others. Only 74 (18%) tested due to COVID-19 symptoms. Somali participants were significantly more likely to test due to symptoms than non-Somali participants (31% vs. 11%, p < 0.001). Accessibility of COVID-19 Testing As an indicator of timely testing access, we compared the median number of days from symptom onset to testing between Somali and non-Somali participants who tested due to COVID-19 symptoms (n = 74). We found a non-significant trend for longer time among Somali participants: median 3 days (interquartile range, IQR: 1.5–4 days) among Somalis vs. 2 days (IQR: 1–3 days), p = 0.06. Participant assessments of COVID-19 testing accessibility in King County revealed that a minority of participants (55, 16%) perceived testing as hard or very hard to access. No significant differences were found in perceived accessibility of testing between Somali and non-Somali participants (p = 0.13). Interview and Focus Group Findings Participant Characteristics FGDs were conducted with 26 Somali participants. Most identified as male (60%) (Table 2). Ten participants were at least 60 years old (36%) and the remainder were under 50 years old. Twelve participants (43%) reported having intermediate or higher English proficiency. Twenty participants (71%) had resided in the USA at least 10 years. Thirteen participants (46%) were not employed, while 8 were employed full-time (29%). Most participants (16, 57%) had below college education and most (17, 61%) had been tested for COVID-19 once.Table 2 Focus group discussion participant characteristics Characteristics (n = 28) n % Age 18–29 2 7.1 30–39 6 21.4 30–40 7 25.0 40–49 3 10.7 ≥ 60 10 35.7 Gender Female 11 39.3 Male 17 60.7 English proficiency None 2 7.1 Elementary 5 17.9 Limited working 9 32.1 Intermediate 2 10.7 Fluent 9 32.1 Duration in the USA < 10 years 8 28.6 ≥ 10 years 20 71.4 Employment status Part-time 3 10.7 Full-time 8 28.6 Self-employed 1 3.6 Unemployed or stay at home parent 13 46.4 Social assistance 2 7.1 Retired 1 3.6 Education completed No formal schooling 2 7.1 Less than high school 1 3.6 High school 13 46.4 Some college 5 17.9 Bachelor’s degree 3 10.7 Graduate degree 4 14.3 Previously tested for COVID-19 Never 17 60.7 Once or more 11 39.3 In-depth interviews (IDIs) were conducted with 5 healthcare workers and 8 government stakeholders. Most interviewees identified as male (7, 54%), 8 identified as Black/African American/African-born, 4 as White, and 1 as Hispanic/Latinx (Table 3). The majority held Master’s degrees (7, 54%). Healthcare workers had worked in their field for a median of 10 years. Government stakeholders had worked in their field for a median of 5.5 years and held positions in the local, county, and state governments. Four interviewees (31%) identified as part of the Somali community, while 6 (46%) provided services for the Somali community and 4 (30%) had a professional or volunteer relationship with the Somali community.Table 3 Key informant interview participant characteristics Characteristics Healthcare Workers (n = 5) Policy makers (n = 8) Total (n = 13) n or median % or IQR n or median % or IQR n or median % or IQR Gender Female 2 40.0 4 50 6 46.2 Male 3 60.0 4 50 7 53.8 Race/ethnicity Black/African American/African-born 3 60.0 5 62.5 8 61.5 White 2 40.0 2 25.0 4 30.8 Hispanic/Latinx - 1 12.5 1 7.7 Education completed Bachelor’s degree or RN 1 20.0 2 25.0 3 23.0 Master’s degree (MPH, M.Ed., MPP) 2 40.0 5 62.5 7 53.8 Terminal degree (MD, PhD) 2 40.0 1 12.5 3 23.1 How long in their professional field (in years) 10 9–12 5.5* 3.8–5.5 9*** 7–12 Relationship* to Somali community Service provider for Somali community 5 100.0 1 12.5 6 46.2 Somali community member 2 40.0 2 25.0 4 30.8 Volunteer or involved with SHB and other community programs 2 40.0 - 2 15.4 Professional work with Somali community members and organizations - 2 25.0 2 15.4 Personal relationships with community members - 1 12.5 1 7.7 None - 2 25.0 2 15.4 Sum is greater than 100% due to multiple relationships per individual. All responses were open-ended, so categories were not pre-determined, and some individuals might have chosen an additional option if pre-determined choices were offered n = 4 had data on time in profession **n = 9 had data on time in profession Interview and Focus Group Themes We report our findings across all three groups (Somali community members, healthcare workers, and government stakeholders) in two domains: (1) testing barriers, and (2) overcoming barriers and future pandemic preparedness. Participant quotes illustrating each theme are displayed in Table 4. We identified in our data all five dimensions of the Penchansky and Thomas concept of access [24], and note these dimensions within the barriers, which are reported by salience across stakeholders.Table 4 Illustrative quotes from key informant interviews and focus group discussions Domain 1: Barriers to testing Logistical barriers Site locations and availability “So, we saw that in South King County, there were testing deserts up until recently… we had to advocate for…more communication around accessibility and interpretation being available at these sites…These are all sort of the social determinants of health that the county may not have considered, and we were not as preventative as we could have been.” (County-level government stakeholder, Black woman) “There are a lot of logistical challenges and not enough…supplies. And those kinds of things were challenging… The budget issues have also been challenging, in the sense that these have been supported with COVID funds at this point. It was CARES [temporary Federal aid] funds. And there have been several rounds of these. So, it is hard to plan.” (City/county-level government stakeholder, Latinx man) “Transportation is, indeed, another challenge for our community considering… majority of them don't even have a driver's license…Even the people that they could have called to get a ride from, now they're not allowed to see that person. So, you have people…not getting tested simply because they can't access the site.” (Nurse, Somali man) “I think there have been more testing sites…However, I think it's still more around certain hours…I think we can still do better at making sure that [testing is available] every day…because people are working different jobs and just don't have the time to go during those specific hours.” (State-level government stakeholder, Black man) Technology “It’s gonna be a nightmare for you to find out your results because they give us a paper with a barcode and you have to have…a smartphone…or a computer so…you can find out what your result is…Imagine if my mother or my grandmother went into that test center, she will not figure out her results, ever…It is tough even to find out what is the barcode and how many numbers you have to put in the website to access your health record.” (Internal medicine clinician, Somali man) Cost “If you go to testing sites set up near the road, [there are] signs that advertise that it is free…but if you go to hospital…they will charge your insurance. They will tell you that testing here is not free and to go [elsewhere] if you want free testing. But if you are in a hurry, you would just do it there and not go to the free testing site…It depends on the location that you go to.” (Male community member, 30–39 years old, college degree) Language barriers Lack of Somali language at sites “At the testing site there are no Somali people…I took my mother and if I was not with her, there would be nobody there to translate for her. So we really need translation services, they have it at other places but not with Somali people.” (Female community member, 30–39 years old, middle school education) “From what I’ve seen there has been an emphasis on Spanish and making sure that there are Spanish interpreters and less so for other languages. I know there hasn’t been as much for people that are of African descent, or even Pacific Islander descent, which is two highly populated populations in [the region].” (State-level government stakeholder, Black man) Distrust and fear Government and health systems “I initially definitely was trusting and messaging around [testing]… the messaging was you know only certain demographic age 65 and above seniors those with you know who were had health risks such as diabetes and whatnot could get tested and when that message was no longer true, and testing was widely available to all Community members.” (County-level government stakeholder, Black woman) When they don’t have…accurate information immediately, people hear rumors and stories from their friends and colleagues and community members and they didn’t have the…most current information…Building that trust and that information lag was definitely part of the reason why that happened.” (County-level government stakeholder, Black woman) Lack of Somali representation “[Somalis] are entitled to get those tests, but…they are dealing with ‘can I do this?’, ‘what will I see?’, and ‘will it be welcoming if I go?’… He has a list of questions to answer for himself before he drives to that site. So, someone who has those kinds of questions, and nobody is answering them. That is the biggest barrier for me.” (Public Health Professional, Somali man) Fearing positive result: Economic loss & stigma “Dealing with a pandemic that caused a lot of people to lose the small businesses, like our people lost their jobs, and the number of hours that they have been working reduced due to the pandemic…the main thing they’re thinking about now is, ‘how do I feed my kids, how do I pay my utilities, how do I pay my rent?’” ( Nurse, Somali man) Risks of the test “I too, am uneasy about the nasal swab because it goes too deep. Someone I know went before me and told me that their experience was good, which prompted me to go get tested and it was very easy and simple. That is why I got tested.” (Male community member, > 60 years old, graduate degree) Domain 2: Overcoming barriers to testing and preparing for future pandemics Centering communities to overcome language, distrust and fear barriers Leadership from the most impacted communities “It was these partnerships, which are led by community and…trusted leaders. I think the partnership that happened with the Somali Health Board, the testing event…it was very well attended…I think that's at least one of the lessons learned as we move into vaccination…and we know whether it's with groups like Somali Health Board or with faith-based groups – those are critical for us to be able to support those kinds of efforts.” (City/county-level government stakeholder, Latinx man) “I think the first support that's needed from the youth, public health or any other health field students, and nurses are to be available for the community. It takes our parents time to grasp and take youth seriously, that they are educated enough, so we need more of your time…It needs a full well-rounded community, including youth and other junior or senior staff from the community working hand-in-hand and supporting each other…That is really what makes communities to improve their health.” (Public Health Professional, Somali man) Representation and culturally responsive, proactive outreach and education “When you're in the public health world you know sometimes you can get a little disconnected and so…collaborating with different community members and working through our advisory group kind of…gives us that eye level of what's going on in the community. And then we are giving our partners tools to educate community members about testing sites, vaccinations – different ways to communicate different policies and procedures and safety precautions…Our Community Navigator, I think that is one strategy, especially for trust. Even when you're going into communities and you’re a representative of that community, there still can be a level of mistrust.” (City/county-level government stakeholder, Black woman) “We need people who are Somali and when they’re calling with positive results, they can understand both the cultural and religious part of the situation and can calmly deliver and reassure that a positive result isn’t as bad as other people can portray it…We need someone who understands the culture when delivering such news.” (Male community member, > 60 years old, high school education) Prioritize services in communities experiencing health disparities “Not having those testing sites at those locations was another example of the areas with the greatest needs not receiving the services that they needed right away…I think it was an oversight, what they were trying to do is have a regional sort of response…all throughout the region and try to do it in an equal sort of way, but really what needed to be centered was equity, making sure that we…really look at the areas with the greatest needs and start there.” (County-level government stakeholder, Black woman) Logistical barriers Site-related “Now what, to me, is going really well is the fact that [testing] is available to people, regardless of their immigration status, insurance status, there are also options…One of the things that we've done was advocate really hard for more testing sites, more testing sites to be placed in South King County and I think now what we're seeing is the availability of testing across the region.” (County-level government stakeholder, Black woman) Sustainability: Maintain and leverage infrastructure established during COVID-19 “[Previously] we didn't have this large community-facing effort which is community mitigation and recovery… So, I think we've been able to do some things and bring people together to both help us and be an extension of the response, but more importantly also inform us in a real way whether it's through that advisory group and saying we need to be shifting in a different direction or through community navigators who are telling us that certain things are not working for community.” (City/county-level government stakeholder, Latinx man) Funding “There's a disconnect between the state level and the local level, and we need to find people that can make those connections and so that's part of the problem. I do know that there's funding available that was earmarking but who's going to get that funding, we want to make sure it's the right organizations like you said, the communities of color-led organizations that are facilitating that funding and it's in the right hands.” (State-level government stakeholder, Black man) Domain 1: Barriers to Testing Our quantitative survey results identified that Somali respondents reported more delayed access to testing, less knowledge of free testing, and greater hesitation to test due to test cost and utility. In our qualitative data, we elucidated greater depth in and potential reasons for these patterns. Barriers noted across stakeholder groups in our qualitative study included logistics, language-related barriers, and distrust and fear. Logistical Barriers Accessibility of Site Locations All government stakeholders noted early disparities in testing locations, with sites located far from the most affected communities. Healthcare workers reported that drive-up testing sites did not accommodate clients who used public transit. Healthcare workers and community members expressed that testing site location and schedule information was inconsistent and confusing. These responses differ somewhat with our quantitative finding that seeing a site advertised or known to be without cost was a key factor in test site choice. Importantly, the survey respondents had all been tested for COVID-19, while the FGD participants represented a mix of testers and never testers, so the qualitative data offer additional insights beyond the survey regarding test site location as a barrier to testing. Availability of Resources Government stakeholders highlighted early difficulty obtaining test supplies and unreliability of government funding for wrap-around services, including testing and grocery delivery for those who must self-isolate. Accommodation Several community members reported that site hours did not accommodate work schedules. Multiple healthcare workers reported supporting non-native English speakers and those with limited technology access or technological proficiency to schedule testing through online portals and access test results through QR codes. As many test sites only offered online scheduling and results, technology presented a barrier for many with limited technological proficiency. Affordability Although free testing was available throughout the county, healthcare workers noted commonly held concerns about cost and limited awareness of free sites preventing some community members from seeking testing. One male community member succinctly combined the financial insecurity caused by the pandemic with concern for test cost: “Do I pay rent or for a COVID-19 test?”. This concern is consistent with the quantitative survey finding that Somali respondents were more likely to state cost of testing was a perceived barrier to testing, more likely to state free testing was a factor in selecting a testing site, and less likely to be aware that free testing was available. Language Barriers Accommodation All stakeholder groups reported that translated materials available online and at testing sites were lacking. Participants described poor dialect and cultural appropriateness of available translations, lack of live interpreters at testing sites, and difficulties with receiving test results. Government stakeholders noted that phone interpretation was available but often not actively offered, and community members did not note awareness of such services. Distrust and Fear Acceptability Several barriers arose within the theme of distrust and fear, which falls under the Penchansky and Thomas concept of acceptability [24]. Government and Health Systems Government stakeholder acknowledged communities’ existing distrust of the healthcare system and that this was exacerbated by delays in information dissemination to communities and services developed without community input. Community members expressed distrust of how privacy and confidentiality were maintained, especially among undocumented individuals. Some feared how the government would handle a positive case. “A lot of people in the community were afraid about the contact tracing and feared the county would lock them in rooms for isolation.” (Female community member). Lack of Somali Representation Community members and healthcare workers identified the lack of Somali and culturally attuned workers administering and supporting testing as a barrier to trust. Fearing Economic Loss and Stigma Community members feared lost wages after a positive test result, compounding the overall pandemic economic hardship. Worries about stigma prevented some community members from disclosing a positive result to those they may have exposed. “People within the community spread the word of those families' positive results and don’t do it for others to stay cautious but do it to alienate that family from the community. It has gotten to the point where families have started to hide their positive result in fear of backlash.” (Male community member). Fearing the Test Some community members feared contact with testing staff who are potentially in contact with positive cases. One community member worried about test safety, observing his sister becoming sick after testing and worrying that testing would trigger his asthma. Healthcare workers and community members noted fearing pain from the nasal swab. The theme of distrust and fear is consistent with quantitative survey findings that Somali respondents were more likely to identify lack of utility of the test as a cause of hesitation to test. Our qualitative results go beyond lack of utility to emphasize fear and harm from the test, an important difference in degree not captured in the survey. Domain 2: Overcoming Barriers to Testing and Preparing for Future Pandemics When asked about effective efforts and lessons learned, participants across all groups discussed effectiveness of community-led interventions, logistical improvements, and the importance of sustainable actions (Table 4). Centering Communities to Overcome Barriers Prioritize Services in Communities Experiencing Health Disparities Government stakeholders recognized that default social structures maximizing overall reach led to racial and ethnic inequities in test access. They recommended that future responses prioritize communities of color early on using data already available to health departments to ensure access to free high-throughput testing sites and accurate information in appropriate languages. Suggested alternative testing approaches—some of which were subsequently adopted by the County—included supporting testing events organized by community organizations in places of worship or gathering, providing more mobile and at-home testing, and increasing the number of sites with walk-up capacity. Leadership and Representation from the Most Impacted Communities While IDI participants were not surprised by the results of our community FGDs when presented with the summary, participants from all stakeholder groups emphasized the primacy of partnering with and relying on community leadership to guide or carry out efforts. For nearly all barriers mentioned, effective solutions stemmed from community-driven actions. “There is no one-size-fits-all even from the same community…depending on the people's [length of] stay in the country, educational level…You have to live through that experience and someone from the same culture only can relay and then get it back to the assurance they need… That's what we have to do, finding the audience-specific, community-centered approach that will address all those kinds of barriers that each community is facing.” (County-level government stakeholder). Healthcare workers believed that informing communities with high levels of distrust of government and health systems about the virus, precautions, and timely testing may foster trust. Community members suggested that Somali youth could bridge gaps between older generations and the healthcare system by providing elders with COVID-19-related education and outreach services framed with religious and cultural relevance. Testing offered by community organizations was widely supported as a strategy to dispel misinformation and fears, explaining cost, billing, and implications of positive tests. The presence of Somali staff leading testing efforts centering cultural context was viewed as ideal, but if infeasible, the presence of Black staff or staff of color could provide some comfort. To illustrate the value of cultural concordance in testing, a public health professional shared, “If there is a gap, I think that would be providing the language support [and] cultural support, which is not only the language interpretation, but also someone they can connect to.” (County-level government stakeholder). Sustainability Maintain Infrastructure Established During COVID-19 Government stakeholders applauded the invaluable community partnerships that were bolstered during the pandemic. They advocated to prioritize maintaining these relationships and leverage partnered approaches enabled by the pandemic to improve access and reduce health disparities more broadly. They acknowledged that this depends on sustainable public funding and shared that short-term emergency funds cannot be relied upon for ongoing community services at it breeds confusion and difficult future planning. They suggested that public–private partnerships could strengthen mobile testing services and pop-up clinics, and that funds should be dispersed directly to community organizations without government agency mediators. Discussion In this mixed-methods, cross-sectional, community-based study, we analyzed access to timely COVID-19 testing among the King County Somali community and explored stakeholder views on meeting community testing needs. Compared with non-Somali participants, Somali participants were significantly more likely to test due to symptoms, to choose a testing site based on cost, to hesitate to test due to distrust and cost, and were less aware that free testing was available in King County. We found a trend for longer time to testing among Somali participants, suggesting elevated access barriers. Qualitative interviews and focus groups highlighted several barriers faced by the Somali community, including inaccessible testing locations, confusion about cost (consistent with quantitative findings about the influence of cost), lack of interpretation services, and distrust resulting from misinformation, discrimination, and economic precarity (consistent with quantitative findings about distrust as a deterrent to testing). Participants made several recommendations to overcome these barriers in ongoing and future pandemic responses, primarily by amplifying community-led work and centering the needs of the most marginalized communities at the outset. Our quantitative findings of reduced access to testing and more testing due to COVID-19 symptoms in the Somali community are consistent with previous studies reporting racial and economic disparities in access to COVID-19 testing. Studies conducted throughout 2020 using surveillance data from Washington, Missouri, West Virginia, Illinois, North Carolina, and New York reported lower testing rates in areas with a larger Black, Latinx and low-income population [9–13]. Regions with the most testing access were not the regions with the highest test positivity rates, indicating inequitable distribution of testing sites [13, 25]. Similarly, a study among patients at a medical system in King County, Washington, reported that non-English-speaking patients, including Somali speakers, were less likely to receive testing but more likely to have a positive result than English speakers [8]. Participants in our qualitative data collection identified barriers across all 5 dimensions of healthcare access in Penchansky and Thomas’s model of access to health services: accessibility, affordability, availability, acceptability, and accommodation [24]. Accessibility of testing locations was frequently mentioned, with community members reporting inaccessibility of sites far from home, and that drive-through-only sites were inaccessible to those who do not drive. Limited availability with regard to testing site hours of operation was identified as a barrier to testing for community members who could not take time off work to test during business hours. Despite free testing, affordability was a frequent community member concern, in both quantitative and qualitative data, possibly resulting from delayed communication about free testing to the Somali community, or opportunity costs associated with receiving a positive result. Accommodation barriers included lack of translation, interpretation, and culturally attuned staff at testing facilities, as well as use of digital tools for community members with limited technology fluency. Barriers to acceptability included distrust rooted in discrimination, misinformation, lack of community engagement by public health authorities, and a mismatch between guidelines and community members’ own realities. Similar barriers to COVID-19 testing have been documented in other immigrant communities and communities of color [26]. A study conducted in Washington State found that accessibility, affordability, acceptability and accommodation of COVID-19 testing and treatment in Latinx communities was impaired by anti-immigrant policies, lack of health insurance, lack of interpretation services, and misinformation [19, 27, 28]. To our knowledge, our study is the first to gather multiple stakeholder perspectives on these issues, including community members, service providers, and government stakeholders; this approach supports the development of strategies that simultaneously address the needs of the distinct groups affected. Our study is also the first to report on barriers to testing specifically in a Somali immigrant community. Analysis of barriers and strategies in specific communities is critical to developing strategies that are nuanced and responsive to community needs. While several published studies have documented testing barriers, few have elicited stakeholders’ proposed solutions to improve access to COVID-19 testing among immigrants, low-income communities, and communities of color. Our participants’ primary recommendation was to elevate and partner with community leaders to deliver representative and culturally coherent COVID-19 testing services. Community and government stakeholders recommended prioritizing site locations by proximity to underserved communities and access via public transport. These recommendations are consistent with development of community-based testing in neighborhoods with large populations of immigrants and people of color in Chicago, Baltimore, New Orleans, Cleveland, Minnesota, and San Francisco [29–36]. Other studies have highlighted the importance of community leaders in epidemic control and prevention. For example, religious leaders or community advocates can be trusted messengers in proving accurate COVID-19 information or addressing misinformation [37]. These suggestions have also been raised specifically in the Somali and East African immigrant communities for the prevention of HPV and HIV [38–40]. Our participants highlighted medical distrust and suggested this can be addressed with more Somali or Black representation at testing sites to appropriately provide information, language services and comfort for community members. They also noted that in King County’s response, although areas with large communities of immigrants and people of color ultimately gained priority for testing and community partners were ultimately engaged, these approaches should have been implemented from the start of the public health response. Our findings must be interpreted in the context of the COVID-19 pandemic’s ever-changing landscape. At the time of our data collection (late 2020 and early 2021), timely testing and behavior change were the only prevention tools available; at-home testing was not available. At the time of publication, at-home antigen tests are widely available, and a multitude of prevention tools exists. This means disparities in access to facility-based PCR testing per se may not contribute as substantially to COVID-19 epidemic control now as they did then. However, our findings remain applicable to current disparities in epidemic control. The disparities and access barriers we and others have reported to testing mirror current disparities and barriers to COVID-19 vaccination in immigrant communities and communities of color [41–43]. Community member, healthcare worker, and government stakeholder participants in our qualitative data collection highlighted that the issues of access and distrust in testing also applied to vaccine roll-out. Furthermore, the themes we identified in our data were consistent with themes identified in barriers to other preventative health services in the Somali community [38, 39], highlighting the persistent shortcomings of service delivery and the continued need for tailoring to marginalized and underserved communities. Our participants’ recommendations for promoting testing equity may be relevant to promoting equity in access to COVID-19 vaccination and treatment. Our study’s strengths include its mixed-methods design, exploration of multiple stakeholder perspectives, data disaggregation to describe a specific community of Black immigrants, and leveraging a partnership between a Somali community-based organization (Somali Health Board) and an academic institution (University of Washington). The Somali Health Board is a widely known and trusted grassroots organization in King County, and their established community relationships and cultural fluency supported collection of detailed views from a highly impacted community that were not already widely documented in published literature. There are limitations to this study. While sample size for our quantitative survey was considerable, the subgroup who reported testing due to symptoms was smaller, limiting statistical power to compare time to testing among Somali and non-Somali respondents, possibly attenuating the resulting association. To minimize the risk of recall bias, we limited responses to individuals who had tested in the past 90 days. Still, as with any self-reported data, ours may be prone to recall bias and misclassification. Additionally, our two-pronged sampling approach (at testing sites and through outreach in the Somali community) was intended to increase inclusion of Somali respondents, but did not produce a sample representative of King County. Recruiting survey respondents at COVID-19 testing sites may have introduced selection bias by excluding people who were unable to attend due to the most severe access barriers. Additionally, our survey sample was skewed towards younger participants, a result we determined after survey data collection ended. To address these potential threats to internal validity, we oversampled older adults and included never-testers in our qualitative data collection to gain insights into barriers that may not have been experienced by survey respondents who were able to test. Our qualitative aim used purposive sampling of our professional and social networks to reach a broad representation of individuals within three stakeholder groups. As the Somali Health Board is well-known and respected in the community, Somali survey respondents likely knew the Board, potentially influencing their participation. All Somali FGD and IDI participants had some relationship with the Somali investigators, which might have influenced their responses either to be more positive, or to be more open and truthful with known community members. Given the language needs, having known Somali speakers was the best option to communicate effectively and provide a comforting space for Somali participants. We strove to include government stakeholders at multiple levels of government, and healthcare workers in varying healthcare roles; however, resource limitations limited the breadth of voices we could include. Many intersectionalities are not represented, including both Somali and non-Somali representatives at each level of government or in all healthcare roles. The focus groups resulted in differing levels of depth by gender. All groups were conducted by a male Somali member of the team, which may have resulted in different gender-based dynamics across the groups. Gender separation is common in many activities in the Somali community, so having a gender concordant interviewer in the men’s group and a gender discordant interviewer in the women’s group may have explained the different richness of responses across groups. Additionally, the older adults were enrolled at a later point in time (February 2021) compared with the younger men’s and women’s groups, both held in November 2020, which may have resulted in recall bias or responses based on newer information. However, we believe that including their views was necessary to gain a broader representation of the community and the range of access barriers that may exist. Conclusions While our study focused on COVID-19 testing within the first year of the pandemic and much has since changed, timely testing remains an essential tool in monitoring and controlling pandemic surges. The barriers and strategies identified in our data may be generalizable to ongoing efforts to ensure equitable access to COVID-19 testing, vaccination, and treatment. This study also highlights a continued need for community leadership in health responses, as well as disaggregation of county-level data to identify and address disparities found in many immigrant and refugee communities across the USA. Supplementary information Below is the link to the electronic supplementary material.Supplementary file1 (26.9 KB) Author Contribution Ahmed Ali, Kathleen McGlone West, and Keshet Ronen contributed to the study conception and design. Data collection was performed by Najma Abdi, Nasra Mohamed, Ayan Mohamed, Asiya Ahmed, Abdifatah Abdi, Ruweida Ahmed, Farah Mohamed, Kathleen McGlone West. Data analysis was performed by Najma Abdi, Sabrina Ebengho, Nasra Mohamed, Andrea Scallon, Ayan Mohamed, Asiya Ahmed, Abdifatah Abdi, Ruweida Ahmed, Farah Mohamed, Kathleen McGlone West, Keshet Ronen. The first draft of the manuscript was written by Najma Abdi, Sabrina Ebengho, Kathleen McGlone West, and Keshet Ronen. All authors reviewed and approved the final manuscript. Funding This work was supported by the University of Washington Population Health Initiative. Data Availability Data are available by request from the corresponding author. Declarations Ethics Approval This study was determined to be exempt by the University of Washington Institutional Review Board. Consent to Participate Informed consent was obtained from all individual participants included in the study. 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. Najma Abdi and Sabrina Ebengho contributed equally to this work. Kathleen McGlone West and Keshet Ronen contributed equally to this work. ==== Refs References 1. CDC. COVID Data Tracker [Internet]. Cent. Dis. Control Prev. 2020 [cited 2021 Sep 19]. Available from: https://covid.cdc.gov/covid-data-tracker 2. Price-Haywood EG, Burton J, Fort D, Seoane L. Hospitalization and mortality among Black Patients and White Patients with Covid-19. N Engl J Med. 2020;NEJMsa2011686. 3. Stokes EK Zambrano LD Anderson KN Marder EP Raz KM El Burai FS Coronavirus Disease 2019 Case Surveillance — United States, January 22–May 30, 2020 Morb Mortal Wkly Rep 2020 69 759 765 10.15585/mmwr.mm6924e2 4. Bogart LM Ojikutu BO Tyagi K Klein DJ Mutchler MG Dong L COVID-19 Related Medical Mistrust, Health Impacts, and Potential vaccine hesitancy among Black Americans living with HIV JAIDS J Acquir Immune Defic Syndr 2021 86 200 207 10.1097/QAI.0000000000002570 33196555 5. OECD Policy Responses to Coronavirus (COVID-19) What is the impact of the COVID-19 pandemic on immigrants and their children? OECD [Internet]. 2020 Oct 19 [cited 2021 Jun 20]; Available from: https://www.oecd.org/coronavirus/policy-responses/what-is-the-impact-of-the-covid-19-pandemic-on-immigrants-and-their-children-e7cbb7de/ 6. Clark E Fredricks K Woc-Colburn L Bottazzi ME Weatherhead J Disproportionate impact of the COVID-19 pandemic on immigrant communities in the United States Brookes VJ editor PLoS Negl Trop Dis. 2020 14 e0008484 10.1371/journal.pntd.0008484 32658925 7. Yaya S Yeboah H Charles CH Otu A Labonte R Ethnic and racial disparities in COVID-19-related deaths: counting the trees, hiding the forest BMJ Glob Health 2020 5 e002913 10.1136/bmjgh-2020-002913 8. Kim HN Lan KF Nkyekyer E Neme S Pierre-Louis M Chew L Assessment of disparities in COVID-19 Testing and infection across language groups in Seattle Washington JAMA Netw Open 2020 3 e2021213 10.1001/jamanetworkopen.2020.21213 32970156 9. Hendricks B Paul R Smith C Wen S Kimble W Amjad A Coronavirus testing disparities associated with community level deprivation, racial inequalities, and food insecurity in West Virginia Ann Epidemiol 2021 59 44 49 10.1016/j.annepidem.2021.03.009 33812965 10. Holden TM. Geographic and demographic heterogeneity of SARS-CoV-2 diagnostic testing in Illinois, USA, March to December 2020. 2021;13. 11. Lieberman-Cribbin W Tuminello S Flores RM Taioli E Disparities in COVID-19 testing and positivity in New York City Am J Prev Med 2020 59 326 332 10.1016/j.amepre.2020.06.005 32703702 12. Mody A Pfeifauf K Bradley C Fox B Hlatshwayo MG Ross W Understanding Drivers of Coronavirus Disease 2019 (COVID-19) Racial disparities: a population-level analysis of COVID-19 testing among Black and White populations Clin Infect Dis 2021 73 e2921 e2931 10.1093/cid/ciaa1848 33315066 13. Seto E, Min E, Ingram C, Cummings BJ, Farquhar SA. Community-level factors associated with COVID-19 cases and testing equity in King County, Washington. Int J Environ Res Public Health. Multidisciplinary Digital Publishing Institute; 2020;17:9516. 14. Public Health - Seattle & King County. COVID-19 race and ethnicity data dashboard [Internet]. Available from: https://kingcounty.gov/depts/health/covid-19/data/race-ethnicity.aspx. Accessed 1 July 2022. 15. Washington State Office of Financial Management. Washington State Office of Financial Management Forecasting and Research Division. Small Area Demographic Estimates (SADE) by Age, Sex, Race and Hispanic Origin [Internet]. Available from: https://ofm.wa.gov/washington-data-research/population-demographics/population-estimates/estimates-april-1-population-age-sex-race-and-hispanic-origin. Accessed 1 July 2022. 16. Berge JM Fertig A Tate A Trofholz A Neumark-Sztainer D Who Is Meeting the Healthy People 2020 Objectives?: Comparisons Between Racially/Ethnically Diverse and Immigrant Children and Adults Fam Syst Health J Collab Fam Healthc 2018 36 451 470 10.1037/fsh0000376 17 Hassan SA Mohamed F Sheikh N Basualdo G Daniel NA Schwartz R “They Wait until the Disease Has Taking over You and the Doctors Cannot Do Anything about It”: Qualitative Insights from Harambee! 2.0 Int J Environ Res Public Health. 2021 18 12706 10.3390/ijerph182312706 34886432 18. Washington State Department of Health. COVID-19 morbidity and mortality by race, ethnicity and spoken language in Washington state [Internet]. Washignton State Department of Health; 2021 Sep. Available from: https://www.doh.wa.gov/Portals/1/Documents/1600/coronavirus/data-tables/COVID-19MorbidityMortalityRaceEthnicityLanguageWAState.pdf. Accessed 1 July 2022. 19. Baquero B, Gonzalez C, Ramirez M, Chavez Santos E, Ornelas IJ. Understanding and addressing Latinx COVID-19 disparities in Washington State. Health Educ Behav SAGE Publications Inc; 2020;47:845–9. 20. Creswell JW, Clark VLP. Designing and conducting mixed methods research. SAGE Publications; 2011. 21. Harris PA Taylor R Minor BL Elliott V Fernandez M O’Neal L The REDCap consortium: building an international community of software platform partners J Biomed Inform 2019 95 103208 10.1016/j.jbi.2019.103208 31078660 22. Patton MQ. Qualitative research & evaluation methods. Los Angeles: SAGE; 2015. 23. Gale RC Wu J Erhardt T Bounthavong M Reardon CM Damschroder LJ Comparison of rapid vs in-depth qualitative analytic methods from a process evaluation of academic detailing in the Veterans Health Administration Implement Sci BioMed Central 2019 14 1 12 24. Penchansky R Thomas JW The concept of access: definition and relationship to consumer satisfaction Med Care 1981 19 127 140 10.1097/00005650-198102000-00001 7206846 25. Brandt K Goel V Keeler C Bell GJ Aiello AE Corbie-Smith G SARS-CoV-2 testing in North Carolina: racial, ethnic, and geographic disparities Health Place 2021 69 102576 10.1016/j.healthplace.2021.102576 33915376 26 Egelko A Arnaout L Garoon J Streed C Berger Z “Do I Have to Be Tested?”: understanding reluctance to be screened for COVID-19 American Public Health Association 2020 110 1769 71 10.2105/AJPH.2020.305964 27. Page KR, Venkataramani M, Beyrer C, Polk S. Undocumented US immigrants and Covid-19. N Engl J Med Massachusetts Medical Society; 2020;382:e62. 28. Page KR Flores-Miller A Lessons We’ve Learned — Covid-19 and the undocumented Latinx community N Engl J Med Massachusetts Medical Society 2021 384 5 7 10.1056/NEJMp2024897 29. Bharmal N, Bailey J, Johnson V, Alejandro-Rodriguez M, Holmes JC, Li-Ng M, et al. Addressing COVID-19 health disparities through a regional community health response. Cleve Clin J Med. 2021; 30. Bigelow BF Saxton RE Flores-Miller A Mostafa HH Loss MJ Phillips KH Community Testing and SARS-CoV-2 Rates for Latinxs in Baltimore Am J Prev Med 2021 60 e281 e286 10.1016/j.amepre.2021.01.005 33775510 31 Brewer LC Woods C Patel A Weis J Jones C Abbenyi A Establishing a SARS-CoV-2 (COVID-19) drive-through collection site: a community-based participatory research partnership with a federally qualified health center Am J Public Health 2021 111 658 62 10.2105/AJPH.2020.306097 33600248 32. English K, Lei U, Shipman-Amuwo F, Burkey M, MSP1, González JG, et al. Community-based testing for SARS-CoV-2 - Chicago, Illinois, May-November 2020. Morb Mortal Wkly Rep 2021;70:707–11. 33. Hernandez JH Karletsos D Avegno J Reed CH Is Covid-19 community level testing effective in reaching at-risk populations? Evidence from spatial analysis of New Orleans patient data at walk-up sites BMC Public Health 2021 21 632 10.1186/s12889-021-10717-9 33789647 34. Kerkhoff AD Sachdev D Mizany S Rojas S Gandhi M Peng J Evaluation of a novel community-based COVID-19 ‘Test-to-Care’ model for low-income populations PLOS ONE 2020 15 e0239400 10.1371/journal.pone.0239400 33035216 35. Tande AJ Challener DW Shah AS O’Horo JC Beam E Community-based drive-through and walk-through testing centers Mayo Clin Proc 2020 95 S20 S22 10.1016/j.mayocp.2020.06.041 32829902 36. Kim SJ, Watson K, Khare N, Shastri S, Pintpo CDG, Nazir NT. Addressing racial/ethnic equity in access to COVID-19 testing through drive-thru and walk-in testing sites in Chicago. Med Res Arch [Internet]. 2021 [cited 2021 Dec 7];9. Available from: https://esmed.org/MRA/mra/article/view/2430 37. Gilmore B Ndejjo R Tchetchia A de Claro V Mago E Diallo AA Community engagement for COVID-19 prevention and control: a rapid evidence synthesis BMJ Glob Health 2020 5 e003188 10.1136/bmjgh-2020-003188 38. Ko LK Taylor VM Mohamed FB Do HH Gebeyaw FA Ibrahim A “We brought our culture here with us”: A qualitative study of perceptions of HPV vaccine and vaccine uptake among East African immigrant mothers Papillomavirus Res 2019 7 21 25 10.1016/j.pvr.2018.12.003 30594650 39. Roberts DA Abera S Basualdo G Kerani RP Mohamed F Schwartz R Barriers to accessing preventive health care among African-born individuals in King County, Washington: A qualitative study involving key informants PLOS ONE 2021 16 e0250800 10.1371/journal.pone.0250800 33970923 40. Daniel NA, Hassan SA, Mohamed F, Sheikh N, Basualdo G, Schwartz R, et al. Harambee! 20: The Impact of HIV-Related and Intersectional Stigmas on HIV Testing Behaviors Among African Immigrant Communities in Seattle Washington. AIDS Behav 2022;26:149–64. 41. Williams N Tutrow H Pina P Belli HM Ogedegbe G Schoenthaler A Assessment of racial and ethnic disparities in access to COVID-19 vaccination sites in Brooklyn New York JAMA Netw Open 2021 4 e2113937 10.1001/jamanetworkopen.2021.13937 34143195 42. Routledge I Takahashi S Epstein A Hakim J Janson O Turcios K Using sero-epidemiology to monitor disparities in vaccination and infection with SARS-CoV-2 Nat Commun 2022 13 2451 10.1038/s41467-022-30051-x 35508478 43. Public Health - Seattle & King County. Summary of COVID-19 vaccination among King County residents [Internet]. Available from: https://kingcounty.gov/depts/health/covid-19/data/vaccination.aspx. Accessed 1 July 2022.	36478269	PMC9734463	NO-CC CODE	2022-12-14 23:28:28	no	J Racial Ethn Health Disparities. 2022 Dec 7;:1-14	utf-8	J Racial Ethn Health Disparities	2,022	10.1007/s40615-022-01470-9	oa_other
==== Front Cell Mol Life Sci Cell Mol Life Sci Cellular and Molecular Life Sciences 1420-682X 1420-9071 Springer International Publishing Cham 36460750 4643 10.1007/s00018-022-04643-1 Original Article Neutrophils play a major role in the destruction of the olfactory epithelium during SARS-CoV-2 infection in hamsters Bourgon Clara 1 Albin Audrey St 1 Ando-Grard Ophélie 1 Da Costa Bruno 1 Domain Roxane 2 Korkmaz Brice 2 Klonjkowski Bernard 3 Le Poder Sophie 3 http://orcid.org/0000-0002-2769-647X Meunier Nicolas [email protected] 1 1 grid.507621.7 Unité de Virologie et Immunologie Moléculaires (UR892), INRAE, Université Paris-Saclay, Jouy-en-Josas, France 2 grid.12366.30 0000 0001 2182 6141 INSERM UMR-1100, “Research Center for Respiratory Diseases” and University of Tours, 37032 Tours, France 3 grid.428547.8 0000 0001 2169 3027 UMR 1161 Virologie, INRAE-ENVA-ANSES, École Nationale Vétérinaire d’Alfort, Maisons-Alfort, 94704 Paris, France 3 12 2022 2022 79 12 6168 6 2022 2 11 2022 22 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 loss of smell (anosmia) related to SARS-CoV-2 infection is one of the most common symptoms of COVID-19. Olfaction starts in the olfactory epithelium mainly composed of olfactory sensory neurons surrounded by supporting cells called sustentacular cells. It is now clear that the loss of smell is related to the massive infection by SARS-CoV-2 of the sustentacular cells in the olfactory epithelium leading to its desquamation. However, the molecular mechanism behind the destabilization of the olfactory epithelium is less clear. Using golden Syrian hamsters infected with an early circulating SARS-CoV-2 strain harboring the D614G mutation in the spike protein; we show here that rather than being related to a first wave of apoptosis as proposed in previous studies, the innate immune cells play a major role in the destruction of the olfactory epithelium. We observed that while apoptosis remains at a low level in the damaged area of the infected epithelium, the latter is invaded by Iba1+ cells, neutrophils and macrophages. By depleting the neutrophil population or blocking the activity of neutrophil elastase-like proteinases, we could reduce the damage induced by the SARS-CoV-2 infection. Surprisingly, the impairment of neutrophil activity led to a decrease in SARS-CoV-2 infection levels in the olfactory epithelium. Our results indicate a counterproductive role of neutrophils leading to the release of infected cells in the lumen of the nasal cavity and thereby enhanced spreading of the virus in the early phase of the SARS-CoV-2 infection. Supplementary Information The online version contains supplementary material available at 10.1007/s00018-022-04643-1. Keywords Post-viral olfactory disorder (PVOD) Pathophysiology Innate immunity ANRCORAR Meunier Nicolas DIM One HealthRegion Centre Val de LoirePirana Korkmaz Brice issue-copyright-statement© Springer Nature Switzerland AG 2022 ==== Body pmcIntroduction Loss of smell (anosmia) is a major symptom of COVID-19 pandemic. With omicron’s increased transmission, hundreds of thousands of people per day still get infected worldwide. Despite omicron’s reduced anosmia prevalence [1], loss of smell will likely affect millions more [2, 3] and 10% of anosmic patients might not recover their sense of smell 6 months after the disease onset [4, 5]. The full olfactory recovery could even take up to 1 year and some patients may never recover their sense of smell [6]. A recent study estimates that in the USA about 720,000 people actually suffer from chronic olfactory disorder related to COVID-19 [7]. The loss of smell negatively impacts life quality by disrupting feeding behavior potentially leading to malnutrition; and by exposing to food poisoning and to inhalation of dangerous chemicals [8]. In severe and persistent cases, anosmic patients could possibly suffer from chronic depression [9]. It is thus crucial to understand the cellular basis of anosmia. Olfaction starts in the olfactory epithelium (OE) which contains olfactory sensory neurons (OSNs) surrounded by supporting cells called sustentacular cells. Both cell types are regenerated regularly due to multipotent basal cells [10]. Among these cells, only sustentacular cells express significantly angiotensin-converting enzyme 2 (ACE2) and transmembrane serine protease 2 (TMPRSS2) required for SARS-CoV-2 cellular entrance [11–13]. We and others observed in the golden Syrian hamster model that SARS-CoV-2 infects massively the sustentacular cells in the OE leading to its desquamation as well as olfactory neurons deciliation and death [14–16]. Although very rarely OSNs may be infected by SARS-CoV-2 [17], a recent study in humans confirms that anosmia arises primarily from infection of sustentacular cells of the OE followed by the disruption of OE integrity without OSN infection [18]. In this study, we focused on the early events following SARS-CoV-2 infection of the nasal cavity to explore the mechanism of the unusually extensive OE damage following SARS-CoV-2 infection. Several studies reported that most cells of the infected OE including OSNs undergo apoptosis [19–22]. A similar phenomenon has been reported during influenza infection [23] and has been considered for SARS-CoV-2 as a defense mechanism to limit a potential invasion of the central nervous system by pathogens using the olfactory route [24]. Alternatively, innate immune cells could trigger directly the desquamation of the OE through inflammation as observed in the lung [25]. Indeed, innate immune cells invade massively the SARS-CoV-2-infected OE [14]. Iba1 (ionized calcium-binding adapter molecule 1) is a marker of microglia/macrophages [26] which are the most studied innate immune cells in the nasal cavity [23]. In the central nervous system, Iba1+ microglial cells ensure viral clearance by phagocytizing viral particles and infected cells [27, 28] and can induce cell death as observed in the hippocampus using the Theiler’s virus model of encephalitis [29]. As the OE is not protected by the blood–brain barrier, neutrophils and monocytes/macrophages classically recruited during the early event of inflammation could also be involved in the OE damage following SARS-CoV-2 infection [30, 31]. Neutrophils are well known for their ability to induce tissue damage, notably through the release of elastase-like proteinases [32, 33] as well as the production of reactive oxygen species (ROS) by the myeloperoxidase (MPO) and formation of toxic neutrophil extracellular traps [34, 35]. Macrophages are known for their ability to phagocyte pathogens, produce cytokines and activate other immune cells [36]. Although they are involved in the regeneration of the OE [37, 38], they can also lead to tissue damage during viral infections notably through NLRP3 (NOD-, LRR- and pyrin domain-containing protein 3) inflammasome activation and metalloproteinases activity [31]. In the current work, we observed that apoptosis remains at a low level in the infected areas of the OE while innate immune cells were systematically present in the damaged area of the OE. By manipulating neutrophil activity by two different complementary approaches, we show that they play a major role in the early events of OE destabilization following SARS-CoV-2 infection. Material and methods Study design The study was performed to understand the cellular mechanisms leading to the SARS-CoV-2-induced damage in the OE using hamsters as an animal model. Hamsters experiments were planned in accordance with the principles of the 3Rs (replacement, reduction and refinement). Body weight and animal behavior were monitored before and during the experiments. Different parameters in the nasal cavity were measured by quantitative polymerase chain reaction (qPCR) and by immunohistochemistry. SARS-CoV-2 replication was measured in vitro to evaluate a potential inhibition by the drugs used to modulate neutrophils activity. Sample size for each experiment is indicated in figure legends. During analysis, all data points were included except because of technical failure to process the sample. Animals were randomized to the experimental groups. All analyses were performed blindly of the treatment. SARS-CoV-2 isolates In vivo experiments were carried out with SARS-CoV-2 strain BetaCoV/France/IDF/200107/2020, which was isolated by Dr. Paccoud from the La Pitié-Salpétrière Hospital in France. This strain was kindly provided by the Urgent Response to Biological Threats (CIBU) hosted by Institut Pasteur (Paris, France), headed by Dr. Jean-Claude Manuguerra. Cell culture experiments were performed with the SARS-CoV-2 strain France/IDF0372/2020 kindly provided by Sylvie van der Werf. Both strains have been isolated in the beginning of the pandemic in Europe in March 2020. Animals Fifty-six 8-week-old male hamsters were purchased from Janvier’s breeding Center (Le Genest, St Isle, France). Animal experiments were carried out in the animal biosafety level 3 facility of the UMR Virologie (ENVA, Maisons-Alfort); approved by the ANSES/EnvA/UPEC Ethics Committee (CE2A16) and authorized by the French ministry of Research under the number APAFIS#25384-2020041515287655. Infection was achieved by nasal instillation (40 µL in each nostril with 5.103 TCID50 of SARS-CoV2 strain BetaCoV/France/IDF/200107/2020) on anesthetized animals under isoflurane. Such viral infection dose gives robust infection level in the nasal cavity [14, 39]. Seven mock-infected animals received only Dulbecco’s minimal essential medium. For neutrophil depletion experiments, hamsters were injected intraperitoneally with either PBS or 150 mg/kg and 100 mg/kg of cyclophosphamide (CAS: 6055-19-2; PHR1404; Sigma-Aldrich), respectively, at 3 and 1 days before SARS-CoV-2 infection. Animals were sacrificed at 1 dpi (days post-infection) and at 2 dpi (n = 4 in each group). To inhibit neutrophil elastase-like proteases, we used a specific synthetic cathepsin C inhibitor (IcatCXPZ-01; [40] diluted in 10% (2-Hydroxypropyl)-β-cyclodextrin (CAS: 128446-35-5; C0926; Sigma-Aldrich) suspended in citrate buffer 50 mM at pH 5 (vehicle) as described previously [41]. Hamsters were injected intraperitoneally twice a day with either vehicle (n = 4) or IcatCXPZ-01 at 4.5 mg/kg for 10 days before infection by SARS-CoV-2 (n = 4). Animals were sacrificed at 1 dpi. For all experiments except IcatCXPZ-01 treatment, the head was divided sagittally into two halves, of which one was used for immunohistochemistry experiments. Nasal turbinates were extracted from the other half for qPCR analysis. Only histological analysis was performed on tissues from IcatCXPZ-01 treatment experiments. Histology, immunohistochemistry and quantifications The immunohistochemistry analysis of the olfactory mucosa tissue sections was performed as described previously in mice [43]. Briefly, the animal hemi-heads were fixed for 3 days at room temperature in 4% paraformaldehyde (PFA) and decalcified in Osteosoft (101728; Merck Millipore; Saint-Quentin Fallavier; France) for 3 weeks. Blocks were cryo-protected in 30% sucrose. Cryo-sectioning (12 µm) was performed in coronal sections of the nasal cavity, perpendicular to the hard palate in order to examine the vomeronasal organ (VNO), olfactory epithelium (OE), Steno’s gland and olfactory bulb. Sections were stored at − 80 °C until use. Non-specific staining was blocked by incubation with 2% bovine serum albumin (BSA) and 0.05% Tween. The sections were then incubated overnight with primary antibodies directed against SARS nucleocapsid protein (1/500; mouse monoclonal; clone 1C7C7; Sigma-Aldrich), ionized calcium-binding adapter molecule 1 (Iba1) (1/500; rabbit monoclonal; clone EPR16588; Abcam), myeloperoxidase protein (MPO) (1/500; rabbit monoclonal; clone EPR20257; Abcam), CD68 (1/200; rabbit polyclonal; PA1518; Boster), cleaved caspase 3 (C3C) (1/200; rabbit polyclonal; #9661; Cell signaling), Golf (1/300; rabbit polyclonal; C-18; Santa Cruz), Olfactory Marker Protein (OMP) (1/500; goat polyclonal; 544-10001; Wako) and CK18 (1:50; mouse polyclonal; MAB3234—RGE53, Sigma-Aldrich). Fluorescence staining was performed using goat anti-mouse-A555; goat anti-rabbit-A488 and donkey anti-goat-A546 (1/800; Molecular Probes A21422; A11056; A11008, respectively; Invitrogen). Images were taken using a 1X71 Olympus microscope equipped with an Orca ER Hamamatsu cooled CCD camera (Hamamatsu Photonics France; Massy; France) or with a Zeiss LSM 700 confocal 187 microscope for cleaved caspase 3 co-staining with either OMP or CK18 experiment and MPO co-staining with Hoechst stained multi-lobal nuclei (MIMA2 Platform, INRAe). Whole section images were reconstructed from 3 images taken at × 50 magnification using a Leica MZ10F Fluorescent binocular microscope. We used these images to display the level of infection revealed by TRITC fluorescence. As this binocular did not possess any UV filter to display Hoechst fluorescence, we used the FITC channel green autofluorescence to display the turbinates structures. To assess olfactory epithelium damage, we scored the integrity of the OE from 1 to 9 based on Hoechst staining according to missing nuclei area, irregularity of the OE structure and increased distance between nuclei indicating loosening of the epithelium leading to desquamation (Supp. Fig. 1). To evaluate the correlation of apoptosis and innate immune cell presence with damage, OE areas were divided in two groups: undamaged areas (damage score equal to 1 or 2) and damaged areas (damage score between 5 and 9). Apoptosis level, infiltration of immune cells in the OE and its underlying lamina propria were quantified as the percentage of the area positive for C3C (cleaved caspase 3); Iba1 (macrophages/microglia), CD68 (activated bone-marrow-derived macrophages) and MPO (neutrophils). For each animal, the percentage of stained OE was averaged over 4 distinct areas in the beginning of olfactory turbinates at 1 dpi and in the medial part of the nasal cavity containing Steno’s gland and NALT at 2 dpi. In cyclophosphamide and in IcatCXPZ-01 experiments, we performed both IHC and classical hematoxylin and eosin (HE) staining as described previously [42]. We examined two independent sections of nasal turbinates (separated by 500 µm) in the middle of the nasal cavity containing NALT and Steno’s gland. For each section, we measured the total infected area of the OE, the area of desquamated cells in the lumen (based on Hoechst nuclear staining) and the percentage of infected desquamated cells in the lumen (based on N protein immunostaining). These experiments were designed to measure the global damage of the OE following SARS-CoV-2 infection when neutrophil action was impaired. At 2 dpi, the level of neutrophil infiltration was so high that complete quantification could not be achieved. Instead, we set a global score from 1 to 9 of 1/OE damage based on the integrity of the OE and 2/neutrophil infiltration based on the overall presence of MPO signal in nasal mucosa and in nasal cavity lumen. We verified that our scoring system and total quantification gave similar results for MPO+ cells presence at 1 dpi where infection is restricted to the most ventro rostral part of the nasal turbinates. All quantifications were made with ImageJ (Rasband, W.S., ImageJ, US National Institutes of Health, Bethesda, Maryland, USA, http://imagej.nih.gov/ij/, 1997–2012) to threshold specific staining and performed blindly of the treatment. RNA extraction and RT-qPCR analysis Total RNA was extracted from frozen nasal turbinates using the Trizol-chloroforme method as described previously [43]. Oligo-dT first strand cDNA synthesis was performed from 5 µg total RNA with iScript Advance cDNA Synthesis Kit for RT-qPCR (BioRad; #1725038) following manufacturer’s recommendations. qPCR was carried out using 125 ng of cDNA added to a 15 µL reaction mix. This reaction mix contained 10 µL iTaq Universal Sybr Green SuperMix (BioRad; #1725124), and primers at 500 nM (sequences in Supp Table 1). The reaction was performed with a thermocycler (Mastercycler ep Realplex, Eppendorf). Fluorescence during qPCR reaction was monitored and measured by Realplex Eppendorf software. A dissociation curve was plotted at the end of the forty amplification cycles of the qPCR to check the ability of theses primers to amplify a single and specific PCR product. Quantification of initial specific RNA concentration was done using the ΔΔCt method. Standard controls of specificity and efficiency of the qPCR were performed. The mRNA expression of each gene was normalized with the expression level of G3PDH. A correction factor was applied to each primer pair according to their efficiency [44]. Measure of antiviral activity of cyclophosphamide and cathepsin C inhibitor against SARS-CoV-2 in cell culture Vero E6 cells (CRL-1586, ATCC maintained at 37 °C; 5% CO2) were seeded at 2.104 cells per well in a 96-well plate in Dulbecco’s Modified Eagle’s Medium, 5% fetal bovine serum (FBS-12A, Capricorn Scientific, Clinisciences). For cyclophosphamide antiviral activity evaluation, cells were treated with 0.15 mg/mL (corresponding to the maximum dose potentially present in hamsters) or 0.45 mg/mL cyclophosphamide diluted in sterile PBS. For IcatCXPZ-01 antiviral activity evaluation, cells were treated with 4.5 µg/mL (corresponding to the maximum dose potentially present in hamsters) or 13.5 µg/mL IcatCXPZ-01 diluted in 10% dextrin, citrate buffer 50 mM, pH = 5. Cells were treated with PBS as control (n = 6 for each condition) and molecule cytotoxicity was tested as well without infection at the highest used concentration. All treatments were started one hour prior to infection with SARS-CoV-2 strain France/IDF0372/2020 at 5 × 103 pfu per well diluted in DMEM, 10% fetal bovine serum. Loss of cell viability reflecting the efficiency of viral infection was measured 3 days after infection by adding 100 µL Cell Titer-Glo reagent to each well (CellTiterGlo Luminescent Cell Viability Assay, Promega #G7571), according to the manufacturer’s protocol. Cell luminescence of each well was then quantified using an Infinite M200Pro TECAN and normalized to the control condition. Statistical analysis All comparisons were made using Prism 5.0 (GraphPad). Statistical significance between groups was assessed using nonparametric Mann–Whitney tests. For correlation analyses, we used Spearman nonparametric test. Error bars indicate the SEM. Detailed information on statistical test used, sample size and P value are provided in the figure legends. Results Apoptosis occurs after cell desquamation following SARS-CoV-2 infection of the olfactory epithelium We previously observed that as soon as two days following nasal instillation of SARS-CoV-2 in Syrian gold hamsters, the sustentacular cells of the OE were massively infected along with strong cellular loss and cellular debris filling the lumen of the nasal cavity [14]. In order to understand the events leading to this desquamation, we chose to focus on the early stages of infection at 1 and 2 dpi. To evaluate the importance of apoptosis in the damage of the OE following SARS-CoV-2 infection, we measured the level of cleaved caspase 3 signal in uninfected animals, and in infected zones of the OE that were either intact or damaged (Fig. 1). Basal level of apoptosis occurring in the OE was not increased in either zone at 1 or 2 dpi (Fig. 1D). However, we observed a strong cleaved caspase 3 signal co-localizing partly with desquamated cell in the lumen of the nasal cavity. The cleaved caspase 3 signal in the lumen of the nasal cavity was increased 5- and 14-fold compared to the OE at 1 and 2 dpi, respectively, which was statistically significant at 2 dpi (Mann–Whitney, p = 0.0286) and nearly significant at 1 dpi (p = 0.0525). In order to examine the potential OE origin of the C3C positive cells, we performed double staining with cleaved caspase 3 and CK18 or OMP (specific markers of sustentacular cells and OSN, respectively). We observed that most apoptotic cells co-stained with OMP but not with CK18, indicating that OSN but not sustentacular cells undergo Caspase 3 apoptosis once released in the lumen of the nasal cavity (Supp. Fig. 2).Fig. 1 Apoptosis occurs in desquamated cells in the lumen of the nasal cavity following SARS-CoV-2 infection but not in the olfactory epithelium. Representative images of an infected intact (A), infected damaged (B) area of the olfactory epithelium at 2 days post-infection (dpi) and in a control animal (C). Apoptotic cells in the olfactory epithelium are indicated by a white arrow (OE; olfactory epithelium/LP lamina propria). The lumen of the nasal cavity is indicated by a white asterisk and is filled with cells, some of which colocalize in their nucleus cleaved caspase 3 signal (orange arrow). (D) Cleaved caspase 3 signal in the olfactory epithelium normalized to control (log 10, Mean ± SEM, n = 4, p < 0.01 (Mann–Whitney test)) Damage of the infected olfactory epithelium is correlated with infiltration of innate immune cells Since apoptosis does not significantly occur in the OE during the initial phase of infection, the desquamation of the infected OE may be related to immune cell infiltration [14, 15]. So far, the immune cells in the nasal cavity have been poorly characterized. Neutrophils and macrophages are known for their importance in clearing infected tissue [45], but only Iba1+ cells presence is well characterized in the OE [23]. Iba1+ cells are described as microglia/macrophages, but CD68 is more classically used as a marker of monocytes and macrophages [46]. Concerning neutrophils, the presence of neutrophil cytosol factor 2 (Ncf2; [47]) and myeloperoxidase (MPO) have been used successfully to characterize these cells in hamsters [48]. As MPO is only expressed during neutrophil maturation in the bone marrow [49], we used Ncf2 as a marker of neutrophil presence by qPCR and MPO by immunohistochemistry. We first evaluated at 1 dpi and 2 dpi by qPCR the expression of Iba1, CD68 and Ncf2 along with classical inflammatory markers (TNFα and IL6) and the presence of the virus (Supp. Fig. 3). At 1 dpi, SARS Nucleocapsid protein (SARS N) was already abundantly expressed in the OE at a similar level as at 2 dpi, and TNFα and IL6 transcripts increased gradually (Mann–Whitney, p < 0.05). Iba1 and CD68 expression related to macrophage presence in the OE did not rise significantly at 1 dpi compared to control (Mann–Whitney, p = 0.164 and 0.128, respectively) but did at 2 dpi (Mann–Whitney, p < 0.05). Concerning neutrophils, Ncf2 expression was strongly enhanced at 1 dpi and was still increasing at 2 dpi (Mann–Whitney, p < 0.05). These results suggest that neutrophils are already recruited at 1 dpi and that their recruitment continues at 2 dpi along with the arrival of Iba1+ and CD68+ cells. We next focused on immunostaining to characterize the presence of Iba1+, CD68+ and MPO+ cells. In the OE of an uninfected hamster, Iba1+ cells were already present and mainly localized in the lamina propria while CD68 signal was absent (Fig. 2A), indicating that Iba1+ cells do not express the classical CD68 marker of macrophages. This was confirmed in the infected areas of the OE where we observed a very different presence of Iba1+ and CD68+ cells. Iba1+ cells were massively present as soon as 1 dpi in the damaged parts of the infected OE as well as in the desquamated cells in the lumen of the nasal cavity. CD68+ cells were less abundant in the damaged part of the OE and mainly present in the desquamated cells filling the lumen of the nasal cavity (Fig. 2B). A double staining against Iba1 and CD68 of the desquamated cells in the lumen of the nasal cavity did not reveal any overlap of the two markers (Supp. Fig. 4), showing that Iba1+ cells do not express CD68 once they are located among the desquamated cells. Similar to CD68, MPO signal was absent in uninfected OE and appears during infection partly in the damaged OE and mainly in the lumen of the nasal cavity along with desquamated cells. The MPO+ cells possess typical multi-lobal nuclei characteristic of neutrophils (Supp. Fig. 5). Overall, these results show that Iba1+ cells are much more abundant in the infected OE than CD68+ macrophages and MPO+ neutrophils cells, both being mainly present in the desquamated cells filling the lumen of the SARS-CoV-2-infected nasal cavity.Fig. 2 Iba1+ (microglia/monocyte lineage), CD68+ (macrophages) and MPO+ (neutrophils) cells presence in the olfactory epithelium before and during SARS-CoV-2 infection. Immunostaining on successive slides of the olfactory epithelium from a non-infected (A) or 1 dpi hamster (B). Only Iba1+ cells are present in the uninfected olfactory epithelium (OE) and in the lamina propria (LP). In the infected epithelium, Iba1+ cells are massively present in the OE while CD68+ and MPO cells are mostly present in the desquamated cells (red asterisk) in the lumen of the nasal cavity (white asterisk) If these innate immunity cells are involved in the desquamation of the OE, we should always observe their presence in the damaged infected area of the OE. To investigate their infiltration in the OE and its correlation with damage, we focused on three zones similarly as for apoptosis quantification: 1/uninfected, infected 2/without or 3/with damage at 1 and 2 dpi. The infiltration level of Iba1+ cells in the OE was increased in the damaged infected zone but not in the undamaged one (Fig. 3). This difference was statistically significant at 2 dpi (Mann–Whitney, p = 0.0286) and nearly significant at 1 dpi (p = 0.0525). The infiltration of these cells was similarly increased in the lamina propria underneath the previous OE zones with a significant difference at 2 dpi (p = 0.0286). We observed a significant correlation between the damage of the OE and their presence in both the OE and the underlying lamina propria (Spearman test, p = 0.0098 and 0.0006, respectively).Fig. 3 Iba1+ cell infiltration increases with the damage in the OE. Representative images of the olfactory epithelium from an uninfected animal (A), infected but undamaged (B) and infected and damaged (C) area of the olfactory epithelium (OE) at 2 days post-infection (dpi). The lumen of the nasal cavity is indicated by a white asterisk. (D) Iba1+ signal in the olfactory epithelium (OE, left) and lamina propria (LP, right) in control animals (CTL) or at 1 or 2 dpi (Mean normalized to control ± SEM, n = 4, p < 0.01 (Mann–Whitney test)). (E) Correlation between score damage of the olfactory epithelium and the percentage of Iba1+ signal in the olfactory epithelium (left panel) and the lamina propria (right panel). Spearman test p value We similarly examined whether the presence of CD68+ macrophages and MPO+ neutrophils in the OE was associated with the damage of the OE after SARS-CoV-2 infection. Both CD68 and MPO signals were increased in the damaged infected zone but not in the undamaged one (Fig. 4). This difference was statistically significant in the infected damaged zones at 1 and 2 dpi for both markers compared to control and infected undamaged zones of the OE and lamina propria (Mann–Whitney, p < 0.05). We observed a significant correlation between the damage of the OE and the presence of both CD68+ and MPO+ cells in the OE and the lamina propria (Spearman test, p < 0.001).Fig. 4 CD68+ macrophage and MPO+ neutrophil cells are associated with damage of the olfactory epithelium during SARS-CoV-2 infection. CD68+ (A1) and MPO+ (B2) signal in the olfactory epithelium (OE, left) and lamina propria (LP, right) in either control animals (CTL) or at 1 or 2 days post-infection (dpi) (Mean normalized to control ± SEM, n = 4, p < 0.05 (Mann–Whitney test)). Correlation between score damage and percentage of CD68+ (A1) and MPO+ (B2) signal in the olfactory epithelium (left panel) and the lamina propria (right panel). Spearman test p value Neutropenia reduces damage of the OE related to SARS-CoV-2 infection as well as OE infected area Neutrophils are the main actors of damage to the olfactory epithelium during Poly(I:C)-induced inflammation [50]. We therefore evaluated whether a neutropenic treatment based on cyclophosphamide would reduce the damage induced by SARS-CoV-2 infection in the OE. Such treatment causes apoptosis of bone-marrow-derived cells and has previously been used successfully on hamsters to induce neutropenia [48]. We first monitored in control animals how the treatment impacts circulating immune cells. Neutrophils population was decreased by ~ tenfold, lymphocytes were also decreased by ~ threefold, and monocytes by ~ fivefold (Supp. Fig. 6A). Since cyclophosphamide can impact basal cell proliferation and thus OE structure, we examined its effect in uninfected animals and did not observe any evident damage on the OE structure (Supp Fig. 6B). We next examined the impact of this treatment on the expression of genes related to the innate immune system in the nasal turbinates during SARS-CoV-2 infection. At 1 dpi, the expression of Iba1 and CD68 was not statistically different between vehicle and cyclophosphamide treated animals but a decrease of Ncf2 expression reflecting a reduced presence of neutrophils almost reached significance (Mann–Whitney; p = 0.0571; Fig. 5A). We observed a tendency of TNFα and IL6 expression reduction which did not reach significance either (p = 0.1143). Despite the overall tendency of a decrease of innate immune system response at 1 dpi, the level of SARS-CoV-2 infection reflected by N protein expression was decreased and the difference almost reached significance (p = 0.0571). At 2 dpi, the expression of all genes related to innate immune cell presence as well as TNFα was decreased (p < 0.05). We examined histologically the neutrophil presence, damage, level of infection in the OE (Fig. 5B, C and Supp. Figs. 6C, 7). MPO signal was significantly decreased at 1 dpi in the OE of cyclophosphamide treated animals compared to control (p < 0.05) (Fig. 5D1). The damage in the OE was significantly decreased at 1 and 2 dpi (p < 0.01 and p < 0.05, respectively) (Fig. 5D2 and Supp. Figs. 8, 9). The reduction tendency of the virus presence measured by the N protein expression was confirmed by immunostaining in the OE at 1 dpi only (p < 0.05; Fig. 5D3) and seems specific to the OE as the infection of the Steno’s gland epithelium lining the maxillary sinus was similar in both conditions (Supp. Fig. 8). We hypothesize that this reduction could be linked to less infected desquamated cells released into the lumen of the nasal cavity following the OE damage induced by the neutrophils. We thus quantified the area of desquamated cells in the lumen which was significantly decreased at 1 dpi and almost reached significance at 2 dpi (Fig. 5D4; p < 0.001 and p = 0.0905, respectively). In the desquamated cells of the lumen, the percentage of infected cells was significantly diminished at 1 dpi (Fig. 5D5; p < 0.001) but not at 2 dpi when infected desquamated cells were in the lumen of the nasal cavity in some treated animals (Supp Fig. 6C). Since the reduction of OE infected area of immunocompromised animal was unexpected, we verified that a dose three times higher than the maximum dose of cyclophosphamide potentially present in the hamsters during infection did not limit the virus replication in vitro (Supp. Fig. 10A, C).Fig. 5 Immunosuppression induced by cyclophosphamide reduces damage of the olfactory epithelium as well as OE infection area. (A) Expression of innate immune genes in the nasal turbinates with or without cyclophosphamide treatment at 1 and 2 days post-infection (dpi). Iba1, CD68 and Ncf2 are related to the presence of microglia/macrophages, monocytes/macrophages and neutrophils, respectively; TNFα and IL6 are two cytokines expressed during inflammation; SARS-CoV-2 N expression is related to the SARS-CoV-2 infection. Results represent the Mean ± SEM relative to vehicle-treated hamsters (n = 4, p < 0.05; Mann–Whitney test). Representative images of the infected olfactory epithelium immunostained for MPO (neutrophil marker) and SARS-CoV-2 N protein in (B) vehicle and (C) cyclophosphamide treated animal (olfactory epithelium (OE), lamina propria (LP)). In the vehicle condition, the lumen (white asterisk) is filled with desquamated cells (red asterisk) containing MPO signal. In the cyclophosphamide condition, MPO signal is absent and the lumen is mostly free of cellular debris. Quantification in the OE of (D1) MPO+ neutrophil presence (D2) damage score (D3) SARS-CoV-2-infected area and in the lumen of the nasal cavity of (D4) desquamated cells area and (D5) percentage of SARS-CoV-2-infected area in the desquamated cells (Mean ± SEM, n = 8 areas of the nasal cavity from 4 different animals, p < 0.05, p < 0.01, *p < 0.001 (Mann–Whitney)) Fig. 6 Inhibition of neutrophil proteinases reduces damage of the olfactory epithelium as well as infected area. Representative images of the infected olfactory epithelium immunostained for MPO (neutrophil marker) and SARS-CoV-2 N protein in (A) vehicle and (B) cathepsin C inhibitor (IcatCXPZ-01) treated animal (olfactory epithelium (OE), lamina propria (LP)). In the vehicle condition, the lumen (white asterisk) is filled with desquamated cells (red asterisk) containing MPO signal. Under cathepsin C inhibition, MPO signal is less abundant and the lumen is mostly free of cellular debris. Quantification (C1) in the OE of MPO+ neutrophil presence; damage score; SARS-CoV-2-infected area and (C2) in the lumen of the nasal cavity of desquamated cell area and percentage of SARS-CoV-2-infected area in the desquamated cells (Mean ± SEM, n = 8 areas of the nasal cavity from 4 different animals, p < 0.05, p < 0.01, *p < 0.001 (Mann–Whitney test)) Inhibition of neutrophil proteinases reduces damage of the OE related to SARS-CoV-2 infection as well as infected OE area In order to confirm our results on cyclophosphamide treatment which affects immune cells other than neutrophils, we treated animals with an inhibitor of cathepsin C (IcatCXPZ-01) which is essential for the maturation of elastase-like proteinases of neutrophils. This inhibitor has been used successfully to almost completely eliminate the elastase-like activity of neutrophils in vivo [41]. We chose to focus on the histological impact of IcatCXPZ-01 treatment at 1 dpi as it gave the most significant results during cyclophosphamide treatment. The inhibition of elastase-like proteinases of neutrophils gave similar results as cyclophosphamide with the exception of some limited neutrophil infiltration in the infected OE (Fig. 6A, B). Globally, the MPO+ neutrophil presence in the nasal turbinates and the damage in the infected area of the OE were significantly reduced compared to vehicle treated animals (Fig. 6C1 and Supp. Figs. 11 and 12; Mann–Whitney; p < 0.01). We also observed significantly less desquamated cells in the lumen of the nasal cavity which were also less infected (Fig. 6C2; Mann–Whitney; p < 0.01). Since we observed again that the inhibition of neutrophil action limited SARS-CoV-2 presence in the OE, we verified that a dose three times higher than the maximum dose of IcatCXPZ-01 potentially present in the hamsters during infection did not impair the virus replication in vitro (Supp. Fig. 10B, C). We also observed that the reduction of SARS-CoV-2 infection by IcatCXPZ-01 was specific to the OE as the infection of Steno’s gland epithelium was not affected in presence of cathepsin C inhibitor (Supp. Fig. 11). Discussion The anosmia induced by SARS-CoV-2 infection is now clearly linked to the infection of the olfactory epithelium with a main tropism for sustentacular cells [18, 51]. We and others have observed that following this infection, the OE undergoes massive damage leading to cell desquamation and cellular debris filling the lumen of the nasal cavity [14, 21, 22], but the mechanism of this destruction is less clear. Several studies reported an increase in apoptosis especially in olfactory sensory neurons [20–22] and assumed that it led to the destruction of the OE. Here, we first examined the apoptosis level based on cleaved caspase 3 level in the OE of uninfected animals and infected areas of the OE, either intact or damaged. If apoptosis initiates the desquamation process then it should increase in the damaged areas of the OE. However, we observed a similar level of apoptosis level in all these areas (Fig. 1D), which was consistent with the basal level of apoptosis that we previously measured in adult mice and rats OE [42, 52]. While the apoptosis in the infected damaged OE was low, we observed an increased level of apoptosis in cells present in the lumen of the nasal cavity. The discrepancy with previous studies may be due to different models as some were performed using transgenic mice expressing hACE2, but it should be noted that these studies did not perform any quantification and observed as well apoptosis staining in the lumen of the nasal cavity [22]. We observed that most apoptotic cells co-stained with OMP specifically expressed by mature OSNs but not CK18 specific of sustentacular cells (Supp Fig. 2B), indicating that the OSNs undergo C3C-linked apoptosis once released in the lumen of the nasal cavity. The induction of apoptosis by loss of cell contact is well described [53], a phenomenon known as anoikis [54] and observed in human airway epithelial cells during SARS-CoV-2 infection [55]. Desquamated cells in the OE may thus be sufficiently preserved to be able to enter apoptosis after the desquamation process is initiated following OE SARS-CoV-2 infection. The C3C-linked apoptosis may also be related to inflammation as well described for SARS-CoV-2 infection [56]. The remaining cells ongoing apoptosis in the lumen may be immune cells as described during SARS-CoV-2 infection [57] and further experiments are required to identify them. Since we previously observed that the infected area of the OE is infiltrated by immune cells [14], we next explored whether innate immune cells are involved in this process, especially macrophages and neutrophils known to be involved in damage of epithelial cells during acute inflammation [30, 31, 58]. If so, we should systematically observe their presence in the damaged area of the OE. We first characterized the presence of these cells in the infected OE (Fig. 2). We observed that CD68, a classical marker of macrophages [46], was expressed in a different population than Iba1+ cells previously described as a microglia/macrophages cellular population [26]. We observed a continuum of Iba1+ cells staining in the OE and in the olfactory bulb similarly as others [59], indicating that they may be more related to microglia than activated macrophages but they require further phenotyping. In the following, we will thus distinguish Iba1+ cells and macrophages as CD68+ cells to avoid confusion between these two cells types. Immunostaining revealed that at 1 dpi, some parts of the OE in the most rostral part of the nasal cavity were already significantly damaged. We observed that in the infected and damaged area of the OE, Iba1+ cells were mainly recruited while macrophages and neutrophils appeared in the zones desquamating toward the lumen of the nasal cavity. While our qPCR results indicate that neutrophils are recruited more abundantly at 1 dpi than Iba1+ cells and macrophages (Supp. Fig. 3), we observed that contrary to macrophages and neutrophils, Iba1+ cells are already present in the lamina propria of uninfected zones and are rapidly gathering in the infected part of the OE (Fig. 3). The basal presence of Iba1+ cells in the lamina propria could explain the apparent discrepancy between qPCR and immunohistochemistry results. Indeed, at the beginning of infection, the increase of Iba1+ cells could simply arise from infiltration of adjacent cells in the nasal turbinates, while macrophages and neutrophils may migrate from the blood following chemotaxis. We can thus hypothesize that Iba1+ cells are first infiltrating the SARS-CoV-2-infected OE. Similarly, Iba1+ microglia are recruited around infected cells of the central nervous system within hours [27], which is consistent with our observation in the OE. Further studies are required to decipher the origin and specific role of these three cells population during the early events of SARS-CoV-2 infection in the infected nasal turbinates. Neutrophils are known to induce epithelial damage and an elegant study demonstrated their major role during Poly(I:C) (an artificial double-stranded RNA agonist of TLR3 receptor) nasal instillation leading to damage of the OE [50]. In order to evaluate the importance of the neutrophils in the damage induced by SARS-CoV-2, we first induced neutropenia based on cyclophosphamide treatment successfully used in hamsters [48]. We confirmed that such treatment mainly affects neutrophils but also reduced to a lesser degree other leucocyte populations (Fig. 5A). As expected such treatment reduced neutrophil infiltration in infected areas of the OE and we observed that damage of the infected OE was significantly reduced as well (Fig. 5D1, D2). In order to confirm the role of neutrophils in the damage of the OE following SARS-CoV-2 infection, we treated hamsters with an inhibitor of cathepsin C (IcatCXPZ-01) specifically reducing the neutrophil elastase-like proteinases activity [60]. We observed that similar to cyclophosphamide treatment, the damage of the OE was greatly reduced in this context (Fig. 6C1). Surprisingly, the global infiltration of neutrophils was reduced as well, even though we observed that some neutrophils were still present in the most infected area of the OE. Since neutrophils are mainly present among the desquamated cells present in the lumen of the nasal cavity, the reduction in neutrophil infiltration during IcatCXPZ-01 treatment may be linked to a decrease in the OE damage as elastase-like proteinase action increases inflammation [61]. It suggests that the damage of the OE initiated by neutrophils may participate in the increase in infiltration of neutrophils leading in fine to massive damage of the infected OE areas. Additionally, part of the damage of the OE could simply reflect the destruction of sustentacular cells by SARS-CoV-2 as the treatments on neutrophils that we used were not sufficient to completely eliminate neutrophils or their proteases activity. Overall, these results show that neutrophils have a major causative role in the destruction of the OE following SARS-CoV-2 infection by releasing elastase-like proteinases. Iba1+ cells and macrophages seem clearly to be involved as well as they are recruited during the event leading to the damage of the infected OE. Our results with histochemistry show that Iba1+ cells arrive earlier than neutrophils and macrophages. Iba1+ cells could be involved in the initial events leading to the damage of the OE as well as phagocytosis. However, their importance in the OE damage should be much lower than neutrophils. Indeed, in the CP treatment which did not impact Iba1+ cell presence in the OE at 1 dpi, we still observed a ~ threefold OE damage reduction with only partial inhibition of neutrophils. In the central nervous system, Iba1+ cells are essential to trigger innate immunity response [28]. We can thus hypothesize that Iba1+ cells may play a chemo-attractive role for macrophages and neutrophils in the infected area of the OE. Further experiments are required to decipher their precise importance in the event leading to OE damage following SARS-CoV-2 infection. We observed that at 1 dpi, the level of SARS-CoV-2 infection was reduced in cyclophosphamide and IcatCXPZ-01-treated hamsters (Figs. 5D3 and 6C1). Such a result was unexpected as neutrophils should effectively destroy infected cells and thus their action should reduce infection progression. We observed that these treatments did not impair SARS-CoV-2 replication in the Steno’s gland epithelium and in vitro at a dose three times higher than the maximum dose potentially circulating in hamsters during infection (Supp. Fig. 10). As both treatments act on neutrophils, we thus hypothesize that neutrophils damage of the infected OE may have a counterproductive effect by releasing infected cells into the lumen of the nasal cavity. These infected cells could allow the virus to spread more easily in the OE as it was recently observed in human airway epithelial cells during SARS-CoV-2 infection [55]. This process may also be enhanced by the recently demonstrated impairment of mucociliary clearance during SARS-CoV-2 infection [62]. Such a hypothesis is consistent with our results, showing that cyclophosphamide and IcatCXPZ-01 treatment significantly reduced the amount of infected desquamated cells filling the lumen of the nasal cavity (Figs. 5D4, D5 and 6C2). The shedding of SARS-CoV-2 by release of damaged infected cells has been observed recently in vitro in respiratory epithelial cell culture where infection is enhanced in the presence of neutrophils [63]. In this preliminary study, SARS-CoV-2 alone did not significantly increase cytokine production but the neutrophil presence did, showing a major role of the neutrophils in the epithelial response to SARS-CoV-2 infection. In our study, at 2 dpi, the cyclophosphamide treatment inducing neutropenia was less effective than at 1 dpi to prevent OE damage related to SARS-CoV-2 infection (Fig. 5D2). As the neutropenia was only partial, the remaining neutrophils may be more effectively recruited when infection progresses (Fig. 5D1) showing the strong ability of SARS-CoV-2 infection in the OE to recruit innate immune cells. Overall, our results show that the SARS-CoV-2 infection does not directly induce the massive damage of the OE but that neutrophils play a major role by releasing elastase-like proteinase in the infected OE. This probably leads to the destabilization of the OE structures and shedding of infected cells. In the early phase of the infection, the shedding of infected cells could enhance the virus spread in the OE (Fig. 7). We observed damaged areas of the OE as soon as 1 dpi, indicating that the innate immune system is extremely efficient in detecting SARS-CoV-2-infected cells to destroy them. The signal triggering this very fast action remains to be explored. The host’s immune defense system that may be present to prevent pathogen invasion from the nose to the brain seems beneficial for SARS-CoV-2 to achieve a much more extensive infection of the OE than any previous virus, resulting in unprecedented olfactory dysfunction in the COVID-19 pandemic.Fig. 7 Model of innate immune cell signaling leading to olfactory epithelium desquamation. The olfactory epithelium is mainly composed of olfactory sensory neurons (OSN) surrounded by supporting cells (sustentacular cells) and basal cells able to regenerate all cell types of the epithelium. During the infection of sustentacular cells (turning red), Iba1+ cells become activated and infiltrate the olfactory epithelium followed by neutrophils and macrophages. Neutrophils release elastase-like proteinase leading to destabilization of the epithelium structures and the expulsion of cells including non-infected neurons into the lumen of the nasal cavity. The release of infected cells may contribute to an increased spreading of the virus in the OE Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 18314 KB) Abbreviations OE Olfactory epithelium OSN Olfactory sensory neuron MPO Myeloperoxidase Acknowledgements We would like to thank all VIM members for their helpful discussion, Christopher von Bartheld and Birte Nielsen for improvement of the manuscript, Dr Pierre Deshuillers from the BioPole plateform of the National Veterinary School of Alfort for hamsters’ blood count, and all the people from the PRBM platform of ENVA who helped us in the BSL3 animal facility. We would also like to thank Bertrand Bryche, Georges Saade, Mustapha Si-Tahar, Laëtitia Merle and Déborah Diakite for their helpful discussion and technical help. Author contribution NM and CB were involved in conceptualization; CB, ASA, OAG, BDC, RD, BK, BK, SLP, and NM helped in investigation; CB and NM contributed to formal analysis; NM and CB helped in writing with input from all authors. Funding NM is supported by INRAe SA department and ANR (Grant CORAR). CB is supported by the “DIM One Health.” BK is supported by the “Région Centre Val de Loire (Project Pirana). Availability of data and material Data will be made available on reasonable request, not applicable for material. Declarations Conflict of interest All authors do not have conflict of interest. Ethics approval and 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. Butowt R Bilińska K von Bartheld C Why does the omicron variant largely spare olfactory function? Implications for the pathogenesis of anosmia in COVID-19 J Infect Dis 2022 10.1093/infdis/jiac113 2. von Bartheld CS Hagen MM Butowt R Prevalence of chemosensory dysfunction in COVID-19 patients: a systematic review and meta-analysis reveals significant ethnic differences ACS Chem Neurosci 2020 10.1021/acschemneuro.0c00460 3. Vaira LA De Vito A Lechien JR New onset of smell and taste loss are common findings also in patients with symptomatic COVID-19 after complete vaccination Laryngoscope 2021 10.1002/lary.29964 4. Boscolo-Rizzo P Menegaldo A Fabbris C Six-month psychophysical evaluation of olfactory dysfunction in patients with COVID-19 Chem Senses 2021 10.1093/chemse/bjab006 5. Lechien JR Chiesa-Estomba CM Beckers E Prevalence and 6-month recovery of olfactory dysfunction: a multicentre study of 1363 COVID-19 patients J Intern Med 2021 10.1111/joim.13209 6. Renaud M Thibault C Le Normand F Clinical outcomes for patients with anosmia 1 year after COVID-19 diagnosis JAMA Netw Open 2021 4 e2115352 10.1001/jamanetworkopen.2021.15352 34165581 7. Khan AM Kallogjeri D Piccirillo JF Growing public health concern of COVID-19 chronic olfactory dysfunction JAMA Otolaryngol Head Neck Surg 2021 10.1001/jamaoto.2021.3379 8. Meunier N Briand L Jacquin-Piques A COVID 19-induced smell and taste impairments: putative impact on physiology Front Physiol 2021 10.3389/fphys.2020.625110 9. Rebholz H Braun RJ Ladage D Loss of olfactory function-early indicator for Covid-19, other viral infections and neurodegenerative disorders Front Neurol 2020 11 569333 10.3389/fneur.2020.569333 33193009 10. Schwob JE Jang W Holbrook EH Stem and progenitor cells of the mammalian olfactory epithelium: taking poietic license J Comp Neurol 2017 525 1034 1054 10.1002/cne.24105 27560601 11. Bilinska K Jakubowska P Bartheld CS Butowt R Expression of the SARS-CoV-2 entry proteins, ACE2 and TMPRSS2, in cells of the olfactory epithelium: identification of cell types and trends with age ACS Chem Neurosci 2020 10.1021/acschemneuro.0c00210 12. Brann DH Tsukahara T Weinreb C Non-neuronal expression of SARS-CoV-2 entry genes in the olfactory system suggests mechanisms underlying COVID-19-associated anosmia Sci Adv 2020 10.1126/sciadv.abc5801 13. Fodoulian L Tuberosa J Rossier D SARS-CoV-2 receptors and entry genes are expressed in the human olfactory neuroepithelium and brain iScience 2020 10.1016/j.isci.2020.101839 14. Bryche B St Albin A Murri S Massive transient damage of the olfactory epithelium associated with infection of sustentacular cells by SARS-CoV-2 in golden Syrian hamsters Brain Behav Immun 2020 89 579 586 10.1016/j.bbi.2020.06.032 32629042 15. Urata S Kishimoto-Urata M Kagoya R Prolonged and extended impacts of SARS-CoV-2 on the olfactory neurocircuit bioRxiv 2021 10.1101/2021.11.04.467274 16. Storm N Crossland NA McKay LGA Griffiths A Comparative pathogenicity of SARS-CoV-2 Omicron and Delta variants in Syrian hamsters mirrors the attenuated clinical outlook of Omicron in COVID-19 irrespective of age Microbiology 2022 24 103530 17. Butowt R Meunier N Bryche B von Bartheld CS The olfactory nerve is not a likely route to brain infection in COVID-19: a critical review of data from humans and animal models Acta Neuropathol 2021 141 809 822 10.1007/s00401-021-02314-2 33903954 18. Khan M Yoo S-J Clijsters M Visualizing in deceased COVID-19 patients how SARS-CoV-2 attacks the respiratory and olfactory mucosae but spares the olfactory bulb Cell 2021 10.1016/j.cell.2021.10.027 19. Seo JS Yoon SW Hwang SH The microvillar and solitary chemosensory cells as the novel targets of infection of SARS-CoV-2 in Syrian golden hamsters Viruses 2021 10.3390/v13081653 20. Ye Q Zhou J He Q SARS-CoV-2 infection in the mouse olfactory system Cell Discovery 2021 7 49 10.1038/s41421-021-00290-1 34230457 21. Yu P Deng W Bao L Comparative pathology of the nasal epithelium in K18-hACE2 Tg mice, hACE2 Tg mice, and hamsters infected with SARS-CoV-2 Vet Pathol 2022 10.1177/03009858211071016 22. Zhang AJ Lee AC-Y Chu H Severe acute respiratory syndrome coronavirus 2 infects and damages the mature and immature olfactory sensory neurons of hamsters Clin Infect Dis 2021 73 e503 e512 10.1093/cid/ciaa995 32667973 23. Mori I Goshima F Imai Y Olfactory receptor neurons prevent dissemination of neurovirulent influenza A virus into the brain by undergoing virus-induced apoptosis J Gen Virol 2002 83 2109 2116 10.1099/0022-1317-83-9-2109 12185263 24. Le Bon SD Horoi M Is anosmia the price to pay in an immune-induced scorched-earth policy against COVID-19? Med Hypotheses 2020 143 109881 10.1016/j.mehy.2020.109881 32474381 25. Burgoyne RA Fisher AJ Borthwick LA The role of epithelial damage in the pulmonary immune response Cells 2021 10 2763 10.3390/cells10102763 34685744 26. Imai Y Ibata I Ito D A novel geneiba1in the major histocompatibility complex class III region encoding an EF hand protein expressed in a monocytic lineage Biochem Biophys Res Commun 1996 224 855 862 10.1006/bbrc.1996.1112 8713135 27. Fekete R Cserép C Lénárt N Microglia control the spread of neurotropic virus infection via P2Y12 signalling and recruit monocytes through P2Y12-independent mechanisms Acta Neuropathol 2018 136 461 482 10.1007/s00401-018-1885-0 30027450 28. Wheeler DL Sariol A Meyerholz DK Perlman S Microglia are required for protection against lethal coronavirus encephalitis in mice J Clin Invest 2018 128 931 943 10.1172/JCI97229 29376888 29. Käufer C Chhatbar C Bröer S Chemokine receptors CCR2 and CX3CR1 regulate viral encephalitis-induced hippocampal damage but not seizures Proc Natl Acad Sci USA 2018 115 E8929 E8938 10.1073/pnas.1806754115 30181265 30. Kolaczkowska E Kubes P Neutrophil recruitment and function in health and inflammation Nat Rev Immunol 2013 13 159 175 10.1038/nri3399 23435331 31. Meidaninikjeh S Sabouni N Marzouni HZ Monocytes and macrophages in COVID-19: friends and foes Life Sci 2021 269 119010 10.1016/j.lfs.2020.119010 33454368 32. Janoff A Human granulocyte elastase. Further delineation of its role in connective tissue damage Am J Pathol 1972 68 579 592 4342329 33. Kawabata K Hagio T Matsuoka S The role of neutrophil elastase in acute lung injury Eur J Pharmacol 2002 451 1 10 10.1016/S0014-2999(02)02182-9 12223222 34. Brinkmann V Reichard U Goosmann C Neutrophil extracellular traps kill bacteria Science 2004 303 1532 1535 10.1126/science.1092385 15001782 35. Saffarzadeh M Juenemann C Queisser MA Neutrophil extracellular traps directly induce epithelial and endothelial cell death: a predominant role of histones PLoS ONE 2012 7 e32366 10.1371/journal.pone.0032366 22389696 36. Arango Duque G Descoteaux A Macrophage cytokines: involvement in immunity and infectious diseases Front Immunol 2014 10.3389/fimmu.2014.00491 37. Borders AS Getchell ML Etscheidt JT Macrophage depletion in the murine olfactory epithelium leads to increased neuronal death and decreased neurogenesis J Comp Neurol 2007 501 206 218 10.1002/cne.21252 17226772 38. Borders AS Hersh MA Getchell ML Macrophage-mediated neuroprotection and neurogenesis in the olfactory epithelium Physiol Genomics 2007 31 531 543 10.1152/physiolgenomics.00008.2007 17848607 39. Hoagland DA Moller R Uhl SA Leveraging the antiviral type I interferon system as a first line of defense against SARS-CoV-2 pathogenicity Immunity 2021 10.1016/j.immuni.2021.01.017 40. Korkmaz B Lesner A Wysocka M Structure-based design and in vivo anti-arthritic activity evaluation of a potent dipeptidyl cyclopropyl nitrile inhibitor of cathepsin C Biochem Pharmacol 2019 164 349 367 10.1016/j.bcp.2019.04.006 30978322 41. Guarino C Hamon Y Croix C Prolonged pharmacological inhibition of cathepsin C results in elimination of neutrophil serine proteases Biochem Pharmacol 2017 131 52 67 10.1016/j.bcp.2017.02.009 28193451 42. Bryche B Dewaele A Saint-Albin A IL-17c is involved in olfactory mucosa responses to Poly(I:C) mimicking virus presence Brain Behav Immun 2019 10.1016/j.bbi.2019.02.012 43. Bryche B Fretaud M Saint-Albin Deliot A Respiratory syncytial virus tropism for olfactory sensory neurons in mice J Neurochem 2019 10.1111/jnc.14936 44. Muller PY Janovjak H Miserez AR Dobbie Z Processing of gene expression data generated by quantitative real-time RT-PCR Biotechniques 2002 32 1372 1374 12074169 45. Bouchery T Harris N Neutrophil–macrophage cooperation and its impact on tissue repair Immunol Cell Biol 2019 97 289 298 10.1111/imcb.12241 30710448 46. Holness C Simmons D Molecular cloning of CD68, a human macrophage marker related to lysosomal glycoproteins Blood 1993 81 1607 1613 10.1182/blood.V81.6.1607.1607 7680921 47. Leto TL Lomax KJ Volpp BD Cloning of a 67-kD neutrophil oxidase factor with similarity to a noncatalytic region of p60 c-src Science 1990 248 727 730 10.1126/science.1692159 1692159 48. Peniche AG Bonilla DL Palma GI A secondary wave of neutrophil infiltration causes necrosis and ulceration in lesions of experimental American cutaneous leishmaniasis PLoS ONE 2017 12 e0179084 10.1371/journal.pone.0179084 28591228 49. Borregaard N Theilgaard-Mönch K Sørensen OE Cowland JB Regulation of human neutrophil granule protein expression Curr Opin Hematol 2001 8 23 27 10.1097/00062752-200101000-00005 11138622 50. Kanaya K Kondo K Suzukawa K Innate immune responses and neuroepithelial degeneration and regeneration in the mouse olfactory mucosa induced by intranasal administration of Poly(I:C) Cell Tissue Res 2014 357 279 299 10.1007/s00441-014-1848-2 24744264 51. Fumagalli V Ravà M Marotta D Administration of aerosolized SARS-CoV-2 to K18-hACE2 mice uncouples respiratory infection from fatal neuroinvasion Sci Immunol 2022 7 eabl9929 10.1126/sciimmunol.abl9929 34812647 52. Raynaud A Meunier N Acquistapace A Bombail V Chronic variable stress exposure in male Wistar rats affects the first step of olfactory detection Behav Brain Res 2015 291 36 45 10.1016/j.bbr.2015.05.013 26003942 53. Bates RC Buret A van Helden DF Apoptosis induced by inhibition of intercellular contact J Cell Biol 1994 125 403 415 10.1083/jcb.125.2.403 8163556 54. Frisch S Francis H Disruption of epithelial cell-matrix interactions induces apoptosis J Cell Biol 1994 124 619 626 10.1083/jcb.124.4.619 8106557 55. Morrison CB Edwards CE Shaffer KM SARS-CoV-2 infection of airway cells causes intense viral and cell shedding, two spreading mechanisms affected by IL-13 Proc Natl Acad Sci USA 2022 119 e2119680119 10.1073/pnas.2119680119 35353667 56. Li S Zhang Y Guan Z SARS-CoV-2 triggers inflammatory responses and cell death through caspase-8 activation Sig Transduct Target Ther 2020 5 235 10.1038/s41392-020-00334-0 57. Junqueira C Crespo Â Ranjbar S FcγR-mediated SARS-CoV-2 infection of monocytes activates inflammation Nature 2022 10.1038/s41586-022-04702-4 58. Hasegawa-Ishii S Shimada A Imamura F Lipopolysaccharide-initiated persistent rhinitis causes gliosis and synaptic loss in the olfactory bulb Sci Rep 2017 7 11605 10.1038/s41598-017-10229-w 28912588 59. Ueha R Ito T Furukawa R Oral SARS-CoV-2 inoculation causes nasal viral infection leading to olfactory bulb infection: an experimental study Front Cell Infect Microbiol 2022 12 924725 10.3389/fcimb.2022.924725 35770069 60. Guarino C Legowska M Epinette C New selective peptidyl di(chlorophenyl) phosphonate esters for visualizing and blocking neutrophil proteinase 3 in human diseases J Biol Chem 2014 289 31777 31791 10.1074/jbc.M114.591339 25288799 61. Döring G The role of neutrophil elastase in chronic inflammation Am J Respir Crit Care Med 1994 150 S114 S117 10.1164/ajrccm/150.6_Pt_2.S114 7952645 62. Li Q Vijaykumar K Philips SE Mucociliary transport deficiency and disease progression in Syrian hamsters with SARS-CoV-2 infection Microbiology 2022 5 562 63. Calvert B, Quiroz E, Lorenzana Z et al (2021) Neutrophilic inflammation promotes SARS-CoV-2 infectivity and augments the inflammatory responses in airway epithelial cells. 10.1101/2021.08.09.455472	36460750	PMC9734468	NO-CC CODE	2022-12-15 23:15:21	no	Cell Mol Life Sci. 2022 Dec 3; 79(12):616	utf-8	Cell Mol Life Sci	2,022	10.1007/s00018-022-04643-1	oa_other
==== Front J Neuroimmune Pharmacol J Neuroimmune Pharmacol Journal of Neuroimmune Pharmacology 1557-1890 1557-1904 Springer US New York 36464726 10054 10.1007/s11481-022-10054-7 Original Article Angiotensin Type 2 Receptor Pharmacological Agonist Relieves Neurocognitive Deficits via Reducing Neuroinflammation and Microglial Engulfment of Dendritic Spines Shen Liang 1 Chen Dan-yang 2 Lou Qian-qian 2 Cao Peng 2 Hu Rui 2 Jin Yan 2 Wang Di 13 http://orcid.org/0000-0002-0058-4750 Hu Shan-shan [email protected] 4 1 Anhui Provincial Hospita, Anhui Provincial Hospital Affiliated to Anhui Medical University, Anhui Medical University, Hefei, 230036 China 2 grid.59053.3a 0000000121679639 Department of Neurobiology, Division of Life Sciences and Medicine, University of Science and Technology of China, Hefei, 230001 China 3 grid.59053.3a 0000000121679639 Department of Anesthesiology, First Affiliated Hospital of USTC (Anhui Provincial Hospita), Division of Life Sciences and Medicine, University of Science and Technology of China (USTC), Hefei, 230001 China 4 grid.59053.3a 0000000121679639 Department of Clinical Laboratory, First Affiliated Hospital of USTC (Anhui Provincial Hospita), Division of Life Sciences and Medicine, University of Science and Technology of China (USTC), Hefei, 230001 China 5 12 2022 117 14 7 2022 8 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. Mechanically ventilated patients suffering critical illness are at high risk of developing neurocognitive impairments. Angiotensin type 2 receptor (AGTR2) has been demonstrated to be anti-inflammatory and neuroprotective. The present study thus aimed to investigate whether AGTR2 can alleviate cerebral dysfunction in mice subjected to cochallenge with lipopolysaccharide (LPS) and mechanical ventilation (MV), and to reveal the underlying mechanism. We utilized a mice model that received a single injection of LPS (1 mg/kg, intraperitoneally) followed 2 h later by MV (10 ml/kg, lasting for 2 h). Pretreatment with the AGTR2 pharmacological agonist C21 (0.03, 0.3, and 3 mg/kg, intraperitoneally, once daily, lasting for 10 days). Locomotor activity and behavioral deficits were evaluated 24 h post-MV by open-field and fear-condition tests. Brain hippocampus and prefrontal cortex tissues were collected for immunofluorescence staining and western blotting to evaluate the resulting impacts on microglia, including morphological traits, functional markers, synaptic engulfment, superoxide production, and signaling molecules. Compared with vehicle-control, pre-administrated C21 reduced the branch endpoints and length of microglia processes in a dose-dependent manner in mice subjected to LPS/MV. The neuroprotective effect of AGTR2 was behaviorally confirmed by the improvement of memory decline in LPS/MV-treated mice following C21 pretreatment. In addition to morphological alterations, C21 reduced microglial functional markers and reduced microglial-dendrite contact and microglial engulfment of synaptic protein markers. In terms of the underlying molecular mechanism, AGTR2 stimulation by C21 leads to activation of protein phosphatase 2A, which subsequently mitigates microglial PKCδ and NF-κB activation, and inhibites NOX2-derived ROS production. The AGTR2 agonist C21 alleviates behavioral deficits in those mice subjected to LPS/MV, via mechanisms that involve reactive microglia and abnormal synaptic plasticity in NOX2-derived ROS and the PKCδ-NFκB pathway. Graphical Abstract Supplementary Information The online version contains supplementary material available at 10.1007/s11481-022-10054-7. Keywords Neurocognitive disturbances Mechanical ventilation Systemic inflammation Microglia Synaptic engulfment Oxidative stress AGTR2 http://dx.doi.org/10.13039/501100012165 Key Technologies Research and Development Program 2021ZD0203105 Wang Di http://dx.doi.org/10.13039/501100001809 National Natural Science Foundation of China 82171218 81703524 Jin Yan Hu Shan-shan ==== Body pmcIntroduction Cerebral dysfunction ranging from acute delirium to long-term neurocognitive impairment may occur among up to 80% of critically ill patients after intensive care unit (ICU) discharge (Pandharipande et al. 2013; Sasannejad et al. 2019). During the COVID-19 pandemic, this problem has been particularly emphasized in severe coronavirus-infected patients who were admitted to ICUs and received life-saving support (Helms et al. 2020; Pun et al. 2021). Considerable progress has been made in delineating the pathophysiological mechanism that underlies the concomitant neurocognitive disorder. Some risk factors have been identified, such as systemic inflammation and mechanical ventilation (MV), which have both independently been shown to be associated with cerebral dysfunction in patients surviving critical illness (Wilcox et al. 2013). Sepsis, a lethal syndrome characterized by a systemic inflammation that develops in response to the infection, is the most common reason for ICU stay and is associated with a high prevalence of cerebral dysfunction termed sepsis‐associated encephalopathy (Gofton and Young 2012). In addition to sepsis-associated encephalopathy, MV may remotely affect cerebral function due to brain-lung interactions and is associated with neurocognitive complications in a considerable portion of critically ill, mechanically ventilated patients (Slutsky and Tremblay 1998; Ely et al. 2004). While the pathophysiology of neurocognitive deficits in critically ill patients is thought to be complex and multifactorial, the immune activation and systemic inflammation elicited by either the septic condition or the use of ventilatory support have been proposed to play a critical role in neuropsychological alterations (Chen et al. 2015; Turon et al. 2018; Sparrow et al. 2021). Systemic inflammation has a substantial impact on the central nervous system (CNS), as evidenced by the neuroinflammation pathologically occurring in the brain via an impaired blood–brain barrier. Numerous efforts have been made to clarify the mechanism underlies how neuroinflammation develops in response to systemic infection or mechanical ventilation and further contributes to resulting cerebral consequences (Hoogland et al. 2015; Giordano et al. 2021). For example, accumulating experimental evidence in rodents has suggested that systemic challenge caused by intraperitoneal injection of lipopolysaccharide (LPS) or MV is significantly associated with reactive microglia, the resident macrophage-like cells within the brain for monitoring and sustaining the balance of the microenvironment within the CNS (Terrando et al. 2010; Chen et al. 2012, 2016). Microglia are crucial for the maintenance of synaptic integrity by synaptic remodelling and have also been implicated in pathological synapse loss and dysfunction following injury, inflammation, or nervous system degeneration (Werneburg et al. 2020; Cao et al. 2021). For example, microglial engulfment of neuronal synapses contributes to cognitive dysfunction, and synapse numbers are decreased in the dorsolateral prefrontal cortex of patients suffering mental illness, such as suppression (Kang et al. 2012). NADPH oxidase 2 (NOX2) enzymes critically contribute to reactive microglia through the generation of superoxide anions, such as reactive oxygen species (ROS), and subsequently lead to neuronal dysregulation such as synaptic plasticity impairment and neuronal death, which affects neurocognitive function (Qiu et al. 2016; Huang et al. 2020). Many molecular targets have been proposed to address the mechanism of reactive microglia and the resulting neuronal dysfunction or damage, such as angiotensin II type 2 receptor (AGTR2). In addition to the cardiovascular and renal systems, abundant levels of AGTR2 within the CNS locally expressed on a variety of cell types, including microglia, astrocytes, and neurons of the prefrontal cortex, hippocampus, and basal ganglia, have been extensively studied in the context of neuroprotection and cognition (Lenkei et al. 1996; de Kloet et al. 2016; Jackson et al. 2018). Recently, using AGTR2-deficient mice or an orally effective AGTR2 agonist, compound 21 (C21), it has been reported that direct AGTR2 stimulation is neuroprotective and improves cognitive impairment in animal models with cerebral ischemia stroke (Alhusban et al. 2015; Schwengel et al. 2016; Eldahshan et al. 2019)and Alzheimer’s disease (Jing et al. 2012; Royea et al. 2020). This neuroprotective action of AGTR2 has been thought to be related to a microglia-dependent mechanism. It has been demonstrated that AGTR2 stimulation using C21 lessens NADPH oxidase (NOX)-mediated reactive microglia and participates in the modulation of aberrant microglial polarization (Bhat et al. 2019; Jackson et al. 2020; Jackson-Cowan et al. 2021). It is largely unclear whether and how AGTR2 is involved in the development of cognitive deficits caused by a critical illness. Based on these findings, we hypothesized that AGTR2 activation might be a suitable strategy for mitigating cognitive deficits under critically ill conditions, and this beneficial effect was related to the reversal of reactive microglia and synaptic abnormalities. To test this hypothesis, we utilized a clinically relevant mice model subjected to a single intraperitoneal injection of LPS followed 2 h later by MV (10 ml/kg, lasting for 2 h) to mimic, at least to some extent, the complex settings of ventilator-treated patients with preexisting sepsis in the present study (Ding et al. 2013, 2018; Zhang et al. 2021). Both the hippocampus and prefrontal cortex were selected throughout the following study because areas commonly thought to be highly linked to neurocognitive processes are often reported brain regions of interest in the majority of studies when it has been mentioned that reactive microglia respond to systemic inflammation (Hoogland et al. 2015). By using a combination of pharmacological tools, behavioral testing paradigm, microglial structural reconstruction, and confocal imaging, we demonstrated here that the AGTR2 agonist C21 alleviates behavioral deficits in mice subjected to LPS/MV, via mechanisms that involve reactive microglia and abnormal synaptic plasticity in NOX2-derived ROS and the PKC-NFκB pathway. Our findings provide experimental evidence to support the molecular and cellular basis for AGTR2 as a target candidate against cognitive deficits after exposure to critically ill conditions. Methods Animals All animal protocols were approved by the Animal Use and Ethics Committee, Anhui Medical University. Male C57BL/6 J mice (purchased from Jackson Laboratory) and Cx3cr1-GFP mice, Thy1-YFP mice (kindly offered by Professor Zhi Zhang), aged 8–12 weeks, weighing approximately 25 g, were used in the experiments. Mice were housed in a specific pathogen-free environment with controlled ambient temperature (22 ± 25℃) and humidity (50% ± 10%) under a 12 h light–dark cycle (lights on from 07:00 to 19:00) with water and food available ad libitum. The animals were acclimated to the housing conditions for 1 week before the experiments. Animal Model A two-hit model with LPS challenge followed by MV was used, as described previously (Ding et al. 2018; Zhang et al. 2021). Briefly, 2 h before receiving MV, mice were anesthetized using isoflurane and pretreated with an intraperitoneal injection of LPS (Sigma, L-2880, Escherichia coli, serotype O26: 1 mg/kg) to induce systemic inflammation. Afterward, mice were anesthetized with an intraperitoneal injection of pentobarbital (50 mg/kg) before endotracheal intubation. Mice were ventilated for 120 min with a tidal volume of 10 mL/kg, a respiratory rate of 70 breaths/min, and a 1:1 inspiratory: expiratory ratio with room air (FiO2 of 0.21) and without positive end-expiratory pressure, using a volume-driven small animal ventilator (RWD407, RWD Life Science: Shenzhen, China). The body temperature of the animals was maintained at 36 °C throughout the procedure using a heating pad. In this study, the control treatment was defined as those mice receiving the same volume of saline injection 2 h before anesthesia and being allowed to breathe spontaneously. Open Field Test All behavioral procedures were carried out in a sound-isolated room. Tests were recorded by the same investigator blinded to the grouping of mice. The apparatus was made up of a white Plexiglas chamber (50 cm × 50 cm × 25 cm) made of gray polyvinyl chloride. The mice were gently placed in the center. A camera recorded the routes of the mice as they moved for a period of 6 min. Total distance travelled were calculated by EthoVision XT 14 software (Noldus, Wageningen, the Netherlands). The chamber was cleaned with 75% ethanol and clean water after each test to remove olfactory cues. Fear Conditioning Test On the training day, the animals were placed into an enclosed training chamber and allowed to explore for 180 s. Then, the mice were exposed to a tone (30 s, 70 dB, 3 kHz), followed by a 2 s foot shock (0.75 mA). Afterward, the mice were left in the chamber for an additional 30 s before being returned to their home cage. On the test day, the mice were re-exposed in the same chamber for 5 min for the contextual fear conditioning test. The animal was placed 2 h later into another test chamber for the cued fear conditioning test, which had a different context and smells from the first test chamber in relative lightroom. After a 3 min exploratory period, the tone was then turned on for 3 cycles, each cycle for 30 s followed by a 1-min intercycle interval (4.5 min in total). The chamber was thoroughly cleaned with 75% ethanol after each trial. Animal behavior in these two chambers was video recorded. The length of freezing behavior in 5 min in the first chamber (context-related) and 4.5 min in the second chamber (tone-related) was recorded by an observer who was blinded to the group assignment. Freezing was defined as a completely immobile posture except for respiration. Treatment Grouping To determine the optimal dose of C21 against the reactive microglia, C21 at 0.03, 0.3, or 3 mg/kg dissolved in saline solution was injected intraperitoneally once daily for seven consecutive days before the two-hit model. Animals in the control groups received the same volume of vehicle (normal saline) at the corresponding time points. Immunofluorescent Staining Mice were deeply anesthetized with an intraperitoneal injection of pentobarbital (50 mg/kg) and perfused intracardially with saline and 4% (w/v) paraformaldehyde. The brain was removed and postfixed in 4% (w/v) paraformaldehyde overnight, before being placed in 30% (w/v) sucrose overnight for dehydration. Coronal frozen Sects. (40 μm thick) were cut by cryostat (Leica CM1860, Germany) for further immunoreactivity staining. The sections were blocked in 5% (w/v) bovine serum and 0.5% (v/v) Triton X-100 for 1 h at room temperature. Following blocking, the sections were incubated with primary antibodies diluted with blocking solution, including anti-ionized calcium-binding adapter molecule 1 (Iba1) (1:500, rabbit, Wako and 1:500, goat, Abcam), anti-MHCII (1:500, mouse, Abcam), anti-CD68 (1:500, mouse, Abcam) and anti-PSD95 (1:500, Goat, Abcam) and anti-synaptophysin (1:100, rabbit, Cell signaling), anti-GFAP (1:200, mouse, Cell signaling), anti-NeuN (1:500, mouse, Abcam), and anti-NOX2 (1:500, mouse, Santa Cruz), at 4 °C overnight. Next, the sections were incubated with corresponding fluorophore-conjugated secondary antibodies diluted with blocking solution for 1.5 h at room temperature, followed by staining with 4,6-diamidino-2-phenylindole (DAPI; 1:2000, Sigma) for 3 min. Finally, coverslips were mounted after adding antifade mounting media (VECTASHIELD Vibrance), and fluorescence signals were visualized using a Leica DM2500 camera and a Zeiss LSM880 microscope. Further analyses, such as analysis of fluorescence intensity and colocalization, were conducted using ImageJ software (Fiji edition, NIH). All the experimental details are described elsewhere (Pan et al. 2022). Morphological Analysis For morphometric analysis, as mentioned previously (Pan et al. 2022), confocal images of Iba1-positive cells were visualized and acquired by a confocal laser-scanning microscope (Carl Zeiss LSM880, Germany) with a 40 × objective at 30-μm intervals along the z-axis. We opened the confocal Z-stacks image file and selected “Add New Filaments \| Create \| Calculate Diameter of Filaments from Image”. The diameter should be between 0.25 μm and 15 μm. We modulated the “Starting Point” and “Seed Point Thresholds” of dendrites according to the actual size and selected “Remove Seed Points Around Starting Points” and set the diameter of “Sphere Regions” as 30 μm. The number of balls was the number of myeloid cells in the image counted per mm3 according to the volume of the image. We adjusted the threshold of the dendrites. In the spine “Points Diameter” step, we deselected the “Detect Spines” option. Finally, we clicked “statistics \| Detailed \| Specific Values \| dendrite length” and saved the data as a.xls file. The data were used as a measure of myeloid cell morphology, such as the number of endpoints and process length. Two images were randomly picked from each mouse, and the mean result was used for morphological analysis. Three-dimensional reconstruction of myeloid cells was performed using IMARIS software. Microglial engulfment was analyzed using IMARIS software to create a 3D surface rendering of the microglia, with a threshold established to ensure accurate reconstruction of microglial processes, which was then used for subsequent reconstructions. PSD95+ and synaptophysin+ puncta were reconstructed using the IMARIS ‘‘Spots’’ function. We used the IMARIS MATLAB-based (MathWorks) plugin ‘‘Split into Surface Objects’’ to assess the number of PSD95+ and synaptophysin+ puncta entirely within the microglial surface. Two images randomly picked from each mouse were reconstructed, each having at least ten cells, for six mice in each group. The mean result was used for morphological analysis. All the experimental details were described previously (Cao et al. 2021). To assess dendritic spine elimination, images were acquired from Thy1-YFP mice and dendritic spines were randomly selected by an observer blinded to the experimental purpose. The selected spines were then assessed to ensure that they were resolved on the images. In detail, approximately 30 dendrites from three mice in each group were analyzed using the plugin NeuronJ 1.4.3 in ImageJ software (Fiji edition, NIH). Thereafter, the plugin Cell Counter was used to calculate the number of spines on these dendrites. All the experimental details were described previously (Cao et al. 2021). For analysis of microglial processes and neuronal dendrite contacts, imaging was performed on a Zeiss LSM880 microscope using a 100 × /1.4 NA oil objective, and imaging parameters (laser power, gain and offset) were consistent across all experiments. Both Iba1+ microglia and YFP+ dendrites were accurately reconstructed using the ‘‘Surface’’ function in IMARIS, and initially-established thresholds were then used for subsequent reconstructions. The MATLAB-based IMARIS plugin ‘‘Surface-Surface Contact Area,’’ was used to measure the size of contact areas between microglial processes and neuronal dendrites. All the experimental details were described previously (Cao et al. 2021). Fluorescence-activated Cell Sorting Mice were anesthetized with an intraperitoneal injection of pentobarbital (50 mg/kg). Subsequently, mice were perfused intracardially with 20 ml of 0.1 M cold Hanks’ balanced salt solution (HBSS), followed by a rapid collection of hippocampus and prefrontal cortex tissues (10 pooled animals per N), which were washed with cold HBSS and chopped into small pieces on ice. Small tissue was mechanically homogenized using a 23G needle to produce single-cell suspensions, which were filtered through a 70 μm cell strainer. After 70%-30% Percoll gradient (Sigma, US) separation, the single-cell suspension was isolated from the interface and filtered with a 200 μm nylon mesh prior to staining with antibody staining. Microglia were labelled with CD11b-PC5.5 (1:200, Biolegend), and sorted by BD FACSAria III (BD, USA) for subsequent qPCR experiments. Real-time PCR Total RNA was extracted from isolated microglia using TRIzol reagent (Vazyme, Nanjing, China) and quantified using NanoVue plus (GE, US). Approximately 500 ng of total RNA was reverse transcribed using StarScript II First-strand cDNA Synthesis Mix with gDNA Remover (GenStar, Beijing, China) in 20 μl reactions following the manufacturer’s protocol. Quantitative real-time PCR (qPCR) for IL-1β, TNF-α, BDNF, and NOX2 was performed on an Applied Biosystems StepOne™ Real-Time PCR System using the 2 × TSINGKE Master qPCR Mix (SYBR Green I) (Tsingke Biotechnology, Beijing, China). Gene-specific primers were purchased from Tsingke Biotechnology. The primer sequences were as follows:IL-1β (forward 5’-TTCAGGCAGGCAGTATCACTC-3’, reverse 5’-GAAGGTCCACGGGAAAGACAC-3’) TNF-α (forward 5’-CCTGTAGCCCACGTCGTAG-3’, reverse 5’-GGGAGTAGACAAGGTACAACCC-3’) BDNF (forward 5’-GCCTTTGGAGCCTCCTCCTCTAC-3’, reverse 5’-GCGGCATCCAGGTAATTTT-3’) NOX2 (forward 5’-GGGAACTGGGCTGTGAATGA-3’, reverse 5’-CAGTGCTGACCCAAGGAGTT-3’) In situ ROS Assay Detection of ROS production in vivo in mice was conducted by using dihydroethidium (DHE, Beyotime, China) as previously described (Zhang et al. 2019; Liu et al. 2020). Briefly, mice were injected with DHE (0.01 mg/g body weight, intraperitoneal injection) 4 h before sacrifice. Oxidized DHE (ox-DHE) emits bright red fluorescence, which can be detected at an emission wavelength of 570 nm. The fluorescent signal was captured with a Zeiss LSM880 microscope, and the threshold-based area of DHE-positive staining was quantified using the ImageJ system. Western Blotting Mice were deeply anesthetized and subsequently intracardially perfused with 20 ml of 0.1 M cold PBS. The hippocampus and prefrontal cortex were collected and washed twice with ice-cold PBS and lysed in ice-cold RIPA lysis buffer containing 1 mM PMSF. The supernatant was obtained by centrifuging at 12,000 g at 4 °C for 15 min. The protein concentration was detected with PierceTM BCA protein assay kit (Thermo, USA). Twenty micrograms of protein per lane was loaded on a 10% SDS–polyacrylamide gel and subsequently transferred to a nitrocellulose membrane, rinsed in TRIS buffered saline (TBS), and blocked in 5% nonfat milk in TBS with 0.1% Tween (TBS-T) for 1 h at room temperature. After rinsing in TBS-T, membranes were incubated 4 °C overnight in primary antibodies against β-Actin (1:500, Absin Bioscience, China), anti-IL-1β (1:500, Absin Bioscience, China), anti-TNF-α (1:500, Absin Bioscience, China), anti-BDNF (1:500, AbCam, UK), anti-phospho-NCF1 (1:500, Bioss, China), anti-phospho-PP2A (1:500, Bioss, China), anti-phospho-PKCδ (1:500, Bioss, China), anti-p65 (1:1000, CST, US), anti-phospho-p65 (1:1000, CST, US), anti-I-κB (1:500, Affinity, US), and anti-phospho-I-κB (1:500, Affinity, US). Blots were washed in TBS-T and incubated for 1 h in peroxidase-labelled secondary antibody (1:5000, Thermo Scientific) at room temperature. Protein bands were visualized by chemiluminescence reagent (Biosharp, China) and quantified using ImageJ software. Statistical Analysis GraphPad Prism 8 (GraphPad Software, Inc., US) was used for statistical analysis and graphing. The QQ-plots were used to verify the assumptions before the t-test, ANOVA, and descriptive statistics. We conducted statistical comparisons between two groups using Student’s t-tests. ANOVA and Bonferroni post hoc analysis were used in analyses with multiple experimental groups. All data are expressed as the mean ± SEM and significance levels are indicated as * p < 0.05, p < 0.01 and * p < 0.001. P values are not provided as exact values when they are less than 0.0001. Results Microglial Activation Develops in Response to Cotreatment with LPS/MV Given that the hippocampus and prefrontal cortex are the two most vulnerable areas in neurocognitive dysfunction, we selected those two areas throughout the following study. To determine whether microglia were activated in response to cotreatment with LPS/MV, we performed morphological reconstruction of microglia at consecutive time points by using Iba1-staining and further assessed the temporal changes in processes in branch endpoints and length, which is known to correlate well with reactive microglial status (Cao et al. 2021; Pan et al. 2022). Experimental studies previously have shown that microglia were moderately active within hours after a single challenge with LPS, reaching their profound activation state after 8 h to 2 days and subsequently returning to their normal surveilling state after 7 days (Hoogland et al. 2015). In line with the published results, our findings revealed that a significant status of reactive microglia, which was manifested by the shorter processes and decreased branch points, substantially occurred 24 h after a cotreatment of LPS/MV (Fig. 1A-F).Fig. 1 Microglial activation develops in response to LPS/MV cotreatment. A Representations from Paxinos & Franklin mouse atlas of regions of interest analysed. IL region of mPFC. B Representative images from LPS/MV-treated mice at different time points. Iba1 immunostaining and three-dimensional (3D) reconstruction of microglia in the mPFC was shown. Scale bars, 40 μm (overview) and 10 μm (Zoom and Rendering). C Quantification of Iba1+ number and intensity, soma size, Imaris-based semi-automatic quantification of Iba1+ microglia morphometry in the mPFC. D Representations from Paxinos & Franklin mouse atlas of regions of interest analysed. CA1 region of the dorsal hippocampus. E Representative images from LPS/MV-treated mice at different time points. Iba1 immunostaining and three-dimensional (3D) reconstruction of microglia in the hippocampus was shown. Scale bars, 40 μm (overview) and 10 μm (Zoom and Rendering). F Quantification of Iba1+ number and intensity, soma size, Imaris-based semi-automatic quantification of Iba1+ microglia morphometry in the hippocampus. G Schematic timeline of behavioral tests. H The analysis results of FCT in control and LPS/MV-treated mice. I The analysis results of OFT in control and LPS/MV-treated mice. All data were presented as mean ± SEM. * p < 0.05; p < 0.01; * p < 0.001; ns, not significant. For detailed statistics information, see Table S1. Abbreviations: mPFC, medial prefrontal cortex; LPS, lipopolysaccharide; MV, mechanical ventilation; OFT, Open Field Test; FCT, Fear Conditioning Test Locomotor activity and behavioural phenotypes were evaluated 24 h after the co-treatment of LPS/MV (Fig. 1G). In the FCT, the cotreatment of LPS/MV significantly decreased the percentage of freezing time compared with the control group (Fig. 1H). In the OFT, there was no significant difference in the total distance or time/distance spent in the center area between the control-treated group and the LPS/MV-treated group (Fig. 1I). These data have shown that the mice subjected to LPS/MV display memory impairments, without any changes in locomotor activity. AGTR2 Agonist Rescues Microglia Activation and Memory Decline Given that the neuroprotective role of AGTR2 is thought to be related to its suppressive action on microglia-mediated neuroinflammation, we utilized the AGTR2 pharmacological agonist C21 at different dosages (0.03, 0.3, and 3 mg/kg) to observe the resulting impacts on the reactive microglia (Fig. 2A). Compared with the vehicle-control, pretreatment of C21 with 7 continuous days at 0.3 and 3 mg/kg, but without 0.03 mg/kg, significantly altered the microglial structural traits both in the hippocampus and prefrontal cortex in mice subjected to LPS/MV (Fig. 2BF). Our findings showed that C21 restrained the microglia in a dose-dependent manner, and a dose of 0.3 mg/kg was chosen for the following experiments. For further validation, we subsequently sought to determine the impact of the AGTR2 agonist C21 on behavioural deficits induced by cotreatment with LPS/MV. Compared with the vehicle-control, the FCT results indicated that AGTR2 stimulation rescued memory impairments in C21-treated mice following cotreatment with LPS/MV (Fig. 2G), but without any alterations in locomotor activity according to OFT results (Fig. 2H).Fig. 2 AGTR2 agonist rescues microglia activation and memory decline. A Schematic timeline of experiment. Control and LPS/MV-stimulated mice were pre-treated with vehicle, and LPS/MV-stimulated mice pre-treated with different concentrations of C21 were designated as LPS/MV + C21 (0.03, 0.3, and 3 mg/kg) groups. B The results of Iba1 immunostaining and 3D reconstruction of microglia in the mPFC was shown. Scale bars, 40 μm (overview) and 10 μm (Zoom and Rendering). C Quantification of Iba1+ number, intensity, soma size, and Iba1+ microglia morphometry was measured by Imaris-based semi-automatic quantification in the mPFC. D The results of Iba1 immunostaining and 3D reconstruction of microglia in the hippocampus was shown. Scale bars, 40 μm (overview) and 10 μm (Zoom and Rendering). E Quantification of Iba1+ number and intensity in the hippocampus. F Quantification of Iba1+ soma size, and Imaris-based semi-automatic quantification of Iba1+ microglia morphometry in the hippocampus. G The results of FCT was analysed in Veh and C21-treated mice. H The results of OFT was analysed in Veh and C21-treated mice. All data were presented as mean ± SEM. * p < 0.05; p < 0.01; * p < 0.001; ns, not significant. For detailed statistics information, see Table S1. Abbreviations: AGTR2, angiotensin type 2 receptor; Veh, vehicle; C21, compound 21 In addition to the alterations of morphological traits, activated microglia are highly characterized by producing high levels of proinflammatory cytokines (such as IL-1β and TNF-α) and brain-derived neurotrophic factor (BDNF) (Fig. 3A). We thus isolated the CD11b-labelled microglia using FACS from the hippocampus and prefrontal cortex tissues, and further quantified the mRNA levels of IL-1β, TNF-α, and BDNF using qPCR (Fig. 3B, C). As shown in Fig. 3D, an increase in IL-1β, TNF-α, and BDNF levels in response to LPS/MV was observed in isolated microglia, which was prevented by pretreatment with C21. Western blot analysis of these protein levels also showed the same results (Fig. 3E, F). Furthermore, CD68 and MHCII markers, which are associated with reactive microglia status, were also found to be increased in mice subjected to cotreatment with LPS/MV. In line with the findings that C21-elicited suppression of microglial morphological phenotypes, pretreatment with C21 significantly reduced the enhancement of CD68 and MHCII levels in those LPS/MV-treated mice (Fig. 3G-J).Fig. 3 AGTR2 agonist reduces microglial functional markers and inflammatory factors. A Schematic timeline, and mice were randomly divided into three groups as control + Veh, LPS/MV + Veh, and LPS/MV + C21. B-C Workflow diagram and the scheme of flow cytometry and cell sorting. D The gene mRNA expression of IL-1 β, TNF α, and BDNF in CD11b-labelled microglia from the mPFC (left panel) and hippocampus (right panel). E Western blot analysis of IL-1 β, TNF α, and BDNF protein levels in the mPFC (left panel) and hippocampus (right panel). F Semi-quantification for different proteins expression were normalized in the mPFC (left panel) and hippocampus (right panel). G Representative images (left panel) and analysis results (right panel) of CD68 in Iba1+ cell from the areas of mPFC. Scale bar, 20 μm. H Representative images (left panel) and analysis results (right panel) of CD68 in Iba1 + cell from the areas of hippocampus. Scale bar, 20 μm. I Representative images (left panel) and quantitative analysis (right panel) of MHCII in the areas of mPFC. Scale bar, 20 μm. J Representative images (left panel) and quantitative analysis (right panel) of MHCII in the areas of hippocampus. Scale bar, 20 μm. All data were presented as mean ± SEM. * p < 0.05; p < 0.01; * p < 0.001; ns, not significant. For detailed statistics information, see Table S1 AGTR2 Agonist Reduces Microglial Engulfment of Neuronal Spines Microglia are important phagocytic cells that participate in synapse remodelling by conducting synaptic pruning, and disrupted synaptic pruning is associated with concomitant reactive microglia in neurological disease (Tonnies and Trushina 2017; Cobley et al. 2018). We observed synaptic dysfunction in response to a cochallenge with LPS/MV, which was morphologically manifested by a reduction in the number of synaptic spines in LPS/MV-treated Thy1-YFP mice (Fig. 4A, B). We next examined the interactions of microglial processes and neuronal dendritic spines in the hippocampus and prefrontal cortex. First, we investigated the microglia-dendrite contact using Iba1-staining to reconstruct the morphology of microglial processes in Thy1-YFP mice, which can be visualized with dendritic spines of neurons expressing YFP. The contact of microglial processes (red) and neuronal dendrites (green) is shown in Fig. 4C, D and the size of microglia-dendrite contact areas was abundantly elevated in LPS/MV-treated mice, which was significantly reduced by pretreatment with C21 (Fig. 4C, D). In addition to microglia-dendrite contact, three-dimensional reconstruction further revealed that abundant immunoreactive puncta (red) of synaptic protein markers, such as postsynaptic marker PSD95 or presynaptic marker synaptophysin, and Cx3cr1-labelled microglial processes (green) colocalized in Cx3cr1-GFP mice following a co-treatment of LPS/MV, but not control-treated mice (Fig. 4E-H). Compared with the vehicle control, these synaptic alterations were significantly reduced by pretreatment with C21 (Fig. 4E-H). These results collectively showed that LPS/MV led to an increase in synaptic loss due to microglial phagocytosis and that these phenotypes were reversed upon intraperitoneal injection of preadministered C21.Fig. 4 AGTR2 agonist reduces microglial engulfment of neuronal spines and phagocytosis of synapses. A Representative images of neuronal dendrites (left panel) and summarized data for spine numbers per 10 μm (right panel) were displayed in the mPFC from Thy1-YFP mice. Scale bar, 10 μm. B Representative images of neuronal dendrites (left panel) and summarized data for spine numbers per 10 μm (right panel) were displayed in the hippocampus from Thy1-YFP mice. Scale bar, 10 μm. C Reconstructed images of Iba1+ microglia (red) and YFP+ neuronal dendrites (Thy1-YFP mice) (left panel), and summarized data for the size of microglia-dendrite contacts (right panel) in the mPFC. Scale bars, 5 μm (overview) and 2 μm (inset). D Reconstructed images of Iba1+ microglia (red) and YFP+ neuronal dendrites (Thy1-YFP mice) (left panel), and summarized data for the size of microglia-dendrite contacts (right panel) in the hippocampus. Scale bars, 5 μm (overview) and 2 μm (inset). E Representative images and 3D surface rendering of Iba1+ microglia containing synaptophysin puncta (left panel), and quantification of synaptophysin puncta in microglia (right panel) in the mPFC as indicated. Scale bars, 10 μm (overview) and 2 μm (inset and rendering). F Representative images and 3D surface rendering of Iba1+ microglia containing synaptophysin puncta (left panel), and quantification of synaptophysin puncta in microglia (right panel) in the hippocampus as indicated. Scale bars, 10 μm (overview) and 2 μm (inset and rendering). G Representative images and 3D surface rendering of Iba1+ microglia containing PSD95+ puncta (left panel), and quantification of PSD95+ puncta in microglia (right panel) in the mPFC as indicated. Scale bars, 10 μm (overview) and 2 μm (inset and rendering). H Representative images and 3D surface rendering of Iba1+ microglia containing PSD95+ puncta (left panel), and quantification of PSD95 + puncta in microglia (right panel) in the hippocampus as indicated. Scale bars, 10 μm (overview) and 2 μm (inset and rendering). All data were presented as mean ± SEM. * p < 0.05, p < 0.01, and * p < 0.001; ns, not significant. For detailed statistics information, see Table S1 AGTR2 Agonist Declines NOX2-derived ROS and PKCδ-NFκB in Microglia To elucidate how the AT2R agonist C21 rescues the aberrant reactive microglia and dysregulated synaptic pruning, we further determined whether C21 produced any impacts on the high level of superoxides, such as ROS, which are thought to initiate or facilitate the downstream reactive microglia and neuronal dysfunction (Qiu et al. 2016; Huang et al. 2020). Given that ROS production can be mediated through upregulation of NOX2, our data supported the idea that NOX2 is highly expressed in microglia from the hippocampus and prefrontal cortex, which may underpin the increased ROS generation observed following LPS/MV treatment (Fig. 5A, B). For further validation, we determined the change in mRNA levels of NOX2 from isolated microglia in control-treated and LPV/MV-treated mice. We observed that NOX2 mRNA was significantly enriched in response to LPS/MV in isolated microglia (Fig. 5C), and the same results were found at the protein level (Fig. 5D). Our findings indicated that NOX2, which is mainly detectable in Iba1-labelled microglia, in LPS/MV-treated mice was significantly higher than that in control-treated mice, and was subsequently lessened by pretreatment with C21 (Fig. 5E, F). We further identified ROS production by tracking the signals of oxidized dihydroethidine (ox-DHE), a marker for intracellular superoxide, and we found that ox-DHE staining was increased in response to LPS/MV, which was consistent with the increase in microglial NOX2 (Fig. 5G, H). Compared with the vehicle-control, pretreatment with C21 lessened ox-DHE staining in C21-treated mice following exposure to LPS/MV.Fig. 5 AGTR2 agonist inhibits NOX2-derived ROS production in microglia. A Representative staining in the mPFC from LPS/MV-treated mice was shown. GFAP (violet), NeuN (violet), Iba1 (green), NOX2 (red), and DAPI (blue), and the co-localization of NOX2/Iba1, NOX2/NeuN, or NOX2/GFAP was analyzed by The Pearson’s Correlation Coefficient test in the mPFC. Scale bar, 20 μm. B Representative staining in the hippocampus from LPS/MV-treated mice was shown. GFAP (violet), NeuN (violet), Iba1 (green), NOX2 (red), and DAPI (blue), and the co-localization of NOX2/Iba1, NOX2/NeuN, or NOX2/GFAP was analyzed by The Pearson’s Correlation Coefficient test in the hippocampus. Scale bar, 20 μm. C Gene expression of NOX2 from CD11b-labelled microglia in the areas of mPFC (left) and hippocampus (right). D Western blot analysis of NOX2 protein levels (left), and Semi-quantification for proteins expression were normalized (right) in the mPFC and hippocampus. E Representative images (left) and the analysis (right) of NOX2 in the Iba1+ cell of mPFC. Scale bar, 20 μm. F Representative images (left) and the analysis (right) of NOX2 in the Iba1+ cell of hippocampus. Scale bar, 20 μm. G Representative images (left) and quantitative analysis (right) of immunostaining for DHE in the mPFC. Scale bar, 100 μm. H Representative images and quantitative analysis of immunostaining for DHE in the hippocampus. Scale bar, 100 μm. All data were presented as mean ± SEM. * p < 0.05, p < 0.01, and * p < 0.001; ns, not significant. For detailed statistics information, see Table S1. Abbreviations: NOX2, NADPH oxidase 2; DHE, dihydroethidine On the other hand, given that relevant signaling molecules, such as PKC, MAPKs, and NF-κB, have been tightly linked to LPS-elicited reactive microglia and accompanying NOX2-induced ROS generation (Wen et al. 2011), we included NF-κB, to explore whether it was involved in AT2R-elicited beneficial actions and to identify whether AT2R activation prevented p65 NF-κB phosphorylation. In the present study, a substantial decrease in the ratio of total-p65/phosphorylated-p65 is attributed to higher levels of the phosphorylated p65 subunit, which indicates the activation of the NF-κB signaling pathway after a C21 pretreatment (Fig. 6A-D). As an upstream signaling molecule, PKCδ is likely to be the major player that orchestrates microglial function in response to LPS insults by initiating NF-κB-mediated transcriptional activation. It has been reported that the functional knockdown of PKC-δ attenuated ERK1/2 and p38 MAPKs phosphorylation, which subsequently led to the blockade of NF-κB activation and iNOS overproduction in microglial cells. Findings from other works have demonstrated that AT2R induces the activation of PP2A, a protein phosphatase that leads to the dephosphorylation and inactivation of PKC in microglia (Bhat et al. 2019, 2021). In the present study, our results indicated that C21 phosphorylated PP2A and reduced phosphorylated PKC-δ (Fig. 6A-D). Our data supported the idea that stimulation of AT2R signaling by C21 prevented microglial activation through NOX-2-induced ROS generation and PP2A-mediated inhibition of PKC and NF-κB.Fig. 6 AGTR2 agonist inhibits PKCδ-NFκB in microglia. A The expression of PP2A, PKC-δ, NF-κB p65, p-NF-κB p65, IκBα and p-IκBα were detected in the mPFC using western blot. B The expression of PP2A, PKC-δ, NF-κB p65, p-NF-κB p65, IκBα and p-IκBα were detected in the hippocampus using western blot. C Semi-quantification for different proteins expression were normalized in the mPFC. D Semi-quantification for different proteins expression were normalized in the hippocampus. All data were presented as mean ± SEM. * p < 0.05, p < 0.01, and * p < 0.001; ns, not significant. For detailed statistics information, see Table S1 Discussion The objectives of the current study were to determine whether pretreatment with C21 lessened cognitive deficits and weakened reactive microglia in mice subjected to LPS/MV, and to elucidate the possible underlying mechanism. Although AGTR2 exerts a beneficial role in improving cognitive function in multiple models of neurological diseases, this is the first study to demonstrate the potential role of the AGTR2 agonist C21 on concomitant neurocognitive disorder in a mice model with high relevance to the critical care field. Given that a compelling relationship is implicated between systemic inflammation and reactive microglia in mediating acute brain dysfunction, we evaluated the impacts of C21 on reactive microglia and functional alterations in response to LPS/MV. Our findings indicated that pretreatment with C21 reduced reactive microglia, accompanied by suppression of microglia-mediated engulfment of neuronal synapses, in mice subjected to LPS/MV. In terms of the underlying mechanism, the C21-elicited ameliorative effect on reactive microglia and functions may be related to the NOX2-derived ROS PKCδ-NFκB pathway. The pathophysiological mechanisms of acute brain dysfunction in critically ill patients are poorly understood, but one of the main triggers that have been proposed is inflammation, which is commonly elicited by systemic infection, traumatic injury, surgical operation, and mechanical ventilatory support (Mei et al. 2021; Saito et al. 2021; Manabe and Heneka 2022). Inflammatory signals can be transmitted to brain by direct or indirect pathways and remotely affect the brain function as a consequence. Microglial cells, which are characterized by a very low threshold of activation, rapidly respond to noxious signals and exhibit highly functional plasticity in response to environmental alterations. In the present study, following the “two-hit” challenge of LPS plus MV, shorter processes and decreased branch points were found in the microglia of the mice subjected to LPS/MV. In addition to morphological changes, dramatic increases in MHCII, and CD68 markers, and the loss of synaptic protein markers were detected in microglia from PLS/MV-treated mice. Once surveilling microglia transform into reactive states in response to stress challenges, reactive microglia are implicated in the resulting neurocognitive outcome through a process known as microglia-dependent synaptic remodelling. This causal relationship has been evidenced by the fact that systemic inflammation-induced reactive microglia are accompanied by deficits in synaptic refinement, neuronal connectivity, and the proper extent of synaptic pruning (Wang et al. 2020; Cao et al. 2021). For example, Cao et al. found that early-life inflammation encodes long-lasting maladaptation of neurons to stress through excessive microglial engulfment of neuronal spines, resulting in the development of depression-like symptoms during adolescence (Cao et al. 2021). Microglia have been found to be highly active in constantly responding to any type of brain homeostatic disturbance and are rapidly transformed from a highly ramified process to an amoeboid morphology under pathological conditions (Furube et al. 2018). In addition to morphological alterations, it has been convincingly shown that the proliferation of microglia is a concomitant event that is associated with disease pathological conditions, coinciding with the activation of microglia (Furube et al. 2018). In the mouse and human brain, microglial density remains remarkably stable, because microglial proliferation in adult rodent brains is slow under physiologically healthy conditions. In contrast, microglia increase their population by both proliferation of resident microglia and recruitment of blood-derived immune cells under pathological brain conditions, for example, LPS induces systemic inflammation and disruption of the blood–brain barrier occurs. A large amount of evidence indicates that there is a causative relationship between microglial proliferation and neurocognitive outcomes. This is further confirmed by the fact that the inhibition of microglial proliferation by pharmacological inhibitors against colony stimulating factor receptor 1 (CSF1R) produced an improvement in neurocognitive deficits (Olmos-Alonso et al. 2016). Our findings suggest that C21 treatment efficiently reduces the proliferation of microglia as well as reactive microglia. Numerous efforts have recently been made to explain the molecular mechanism by which AGTR2 mitigates reactive microglia and their subsequent function. Traditionally, AGTR2 is believed to be highly expressed in the fetus but rapidly declines to an extremely low level after birth (Lenkei et al. 1996; de Kloet et al. 2016). Recently, it has been increasingly acknowledged that inducible AGTR2 expression is abundantly elicited on activated microglia in response to challenges, and AGTR2 stimulation might exert anti-inflammatory action via the inhibition of microglia, which leads to an improved outcome in neurological disease (Jackson et al. 2018; Rivas-Santisteban et al. 2021). Our findings strengthen the beneficial role of AGTR2 on neurocognitive outcome, which is linked to a reduction in reactive microglia, and microglial engulfment of the synapse. NADPH oxidase activation and subsequent ROS overproduction are important upstream events that can activate microglia and amplify microglial dysfunction (Dohi et al. 2010; Kumar et al. 2016). There are multiple sources of the production of ROS, among which the most studied include NADPH oxidase and mitochondrial ROS (Circu and Aw 2010). Despite a disagreement on the extent to which mitochondria produce ROS in vivo, NADPH oxidase is regarded as the predominant source of ROS in phagocytes such as microglia, which is upregulated in SAE and POCD and essential for microglia-mediated amyloid neurotoxicity (Bhat et al. 2019). We demonstrated here that pretreatment with C21 lessened microglial NOX2 expression and resulted in ROS production in the prefrontal cortex and hippocampus in mice subjected to LPS/MV. In addition to NOX2-derived ROS, it cannot be neglected that microglia produce a wide spectrum of neural active substances to participate in the neuroinflammatory cascade and synaptic dysfunction, such as IL-1 and TNF as well as neuroprotective factors, including BDNF which is critically involved in synaptic maturation and plasticity (Jiang et al. 2022). Although others and our data revealed that the substantial increase in IL-1, TNF, and BDNF levels occurs concomitantly with reactive microglia, there is a limitation that we did not deeply investigate whether and how the alteration of these substances drives the subsequent neurological dysfunction. Furthermore, the findings from our and others’ works have demonstrated that stimulation of AT2R induces the activation of PP2A, which is a protein phosphatase that leads to the dephosphorylation of PKCδ that typically results in the subsequent NF-κB-mediated transcriptional activation in microglia. Conclusions In summary, our findings revealed that the AGTR2 agonist C21 alleviated behavioral deficits in mice subjected to LPS/MV cochallenge in a microglia-dependent manner via inhibition of aberrant reactive microglia and microglial engulfment of neuronal synapses. The findings of this study provide experimental evidence supporting the pharmacological stimulation of AGTR2 by C21 and may be valuable for the development of prevention or mitigation strategies for neurological morbidity in these critically ill patients. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 52 KB) Acknowledgements The authors very much appreciate Professor Zhi Zhang for kindly providing the Cx3cr1-GFP and Thy1-YFP mice. Authors' Contributions D. WANG, L. SHEN, and D. CHEN designed the studies, conducted most of the experiments and data analysis and wrote the draft manuscript. Q. LOU, P. CAO, and R.HU conducted the behavioral experiments and data analysis and wrote the text of the final manuscript. Y. JIN, S. HU, and D. WANG were involved in the overall design of the study and approved the final manuscript. Funding This work was supported by the National Key Research and Development Program of China (2021ZD0203105), National Natural Science Foundation of China (81703524, 82171218), Natural Science Foundation of Anhui Province (1608085QH210), Fundamental Research Funds for the Central Universities (WK9100000030), USTC Research Funds of the Double First-Class Initiative (YD9100002018), and Youth Innovation Promotion Association CAS. Data Availability The data that support the findings of this study are available from the corresponding author (Dr. Shan-shan Hu) upon request. Declarations Ethics Approval and Consent to Participate This study was approved by experimental Animal Ethics Committee of Anhui Medical University (No. LLSC20160121), and all experiment protocols were approved by the Institutional Animal Care and Use Committee of Anhui Medical University. Consent for Publication This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Competing Interests The authors declare that they have no competing interests. Liang Shen, Dan-yang Chen, and Qian-qian Lou these authors have contributed equally to this work and share first authorship. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Alhusban A Fouda AY Bindu P Ishrat T Soliman S Fagan SC Compound 21 is pro-angiogenic in the brain and results in sustained recovery after ischemic stroke J Hypertens 2015 33 170 180 10.1097/HJH.0000000000000364 25304472 Bhat SA Sood A Shukla R Hanif K AT2R activation prevents microglia pro-inflammatory activation in a NOX-Dependent manner: inhibition of PKC activation and p47(phox) phosphorylation by PP2A Mol Neurobiol 2019 56 3005 3023 10.1007/s12035-018-1272-9 30076526 Bhat SA Fatima Z Sood A Shukla R Hanif K The protective effects of AT2R agonist, CGP42112A, against angiotensin ii-induced oxidative stress and inflammatory response in astrocytes: role of AT2R/PP2A/NFkappaB/ROS signaling Neurotox Res 2021 39 1991 2006 10.1007/s12640-021-00403-4 34529240 Cao P Early-life inflammation promotes depressive symptoms in adolescence via microglial engulfment of dendritic spines Neuron 2021 109 2573–2589 e2579 Chen C Zhang Z Chen T Peng M Xu X Wang Y Prolonged mechanical ventilation-induced neuroinflammation affects postoperative memory dysfunction in surgical mice Crit Care 2015 19 159 10.1186/s13054-015-0882-0 25887955 Chen T Chen C Zhang Z Zou Y Peng M Wang Y Toll-like receptor 4 knockout ameliorates neuroinflammation due to lung-brain interaction in mechanically ventilated mice Brain Behav Immun 2016 56 42 55 10.1016/j.bbi.2016.04.004 27067748 Chen Z Jalabi W Shpargel KB Farabaugh KT Dutta R Yin X Kidd GJ Bergmann CC Stohlman SA Trapp BD Lipopolysaccharide-induced microglial activation and neuroprotection against experimental brain injury is independent of hematogenous TLR4 J Neurosci 2012 32 11706 11715 10.1523/JNEUROSCI.0730-12.2012 22915113 Circu ML Aw TY Reactive oxygen species, cellular redox systems, and apoptosis Free Radic Biol Med 2010 48 749 762 10.1016/j.freeradbiomed.2009.12.022 20045723 Cobley JN Fiorello ML Bailey DM 13 reasons why the brain is susceptible to oxidative stress Redox Biol 2018 15 490 503 10.1016/j.redox.2018.01.008 29413961 de Kloet AD Wang L Ludin JA Smith JA Pioquinto DJ Hiller H Steckelings UM Scheuer DA Sumners C Krause EG Reporter mouse strain provides a novel look at angiotensin type-2 receptor distribution in the central nervous system Brain Struct Funct 2016 221 891 912 10.1007/s00429-014-0943-1 25427952 Ding N Wang F Xiao H Xu L She S Mechanical ventilation enhances HMGB1 expression in an LPS-induced lung injury model PLoS ONE 2013 8 e74633 10.1371/journal.pone.0074633 24058610 Ding X Tong Y Jin S Chen Z Li T Billiar TR Pitt BR Li Q Zhang LM Mechanical ventilation enhances extrapulmonary sepsis-induced lung injury: role of WISP1-alphavbeta5 integrin pathway in TLR4-mediated inflammation and injury Crit Care 2018 22 302 10.1186/s13054-018-2237-0 30445996 Dohi K Ohtaki H Nakamachi T Yofu S Satoh K Miyamoto K Song D Tsunawaki S Shioda S Aruga T Gp91phox (NOX2) in classically activated microglia exacerbates traumatic brain injury J Neuroinflammation 2010 7 41 10.1186/1742-2094-7-41 20659322 Eldahshan W Ishrat T Pillai B Sayed MA Alwhaibi A Fouda AY Ergul A Fagan SC Angiotensin II type 2 receptor stimulation with compound 21 improves neurological function after stroke in female rats: a pilot study Am J Physiol Heart Circ Physiol 2019 316 H1192 H1201 10.1152/ajpheart.00446.2018 30822121 Ely EW Shintani A Truman B Speroff T Gordon SM Harrell FE Jr Inouye SK Bernard GR Dittus RS Delirium as a predictor of mortality in mechanically ventilated patients in the intensive care unit JAMA 2004 291 1753 1762 10.1001/jama.291.14.1753 15082703 Furube E Kawai S Inagaki H Takagi S Miyata S Brain Region-dependent Heterogeneity and Dose-dependent Difference in Transient Microglia Population Increase during Lipopolysaccharide-induced Inflammation Sci Rep 2018 8 2203 10.1038/s41598-018-20643-3 29396567 Giordano G Pugliese F Bilotta F Neuroinflammation, neuronal damage or cognitive impairment associated with mechanical ventilation: a systematic review of evidence from animal studies J Crit Care 2021 62 246 255 10.1016/j.jcrc.2020.12.017 33454552 Gofton TE Young GB Sepsis-associated encephalopathy Nat Rev Neurol 2012 8 557 566 10.1038/nrneurol.2012.183 22986430 Helms J Kremer S Merdji H Clere-Jehl R Schenck M Kummerlen C Collange O Boulay C Fafi-Kremer S Ohana M Anheim M Meziani F Neurologic features in severe SARS-CoV-2 infection N Engl J Med 2020 382 2268 2270 10.1056/NEJMc2008597 32294339 Hoogland IC Houbolt C van Westerloo DJ van Gool WA van de Beek D Systemic inflammation and microglial activation: systematic review of animal experiments J Neuroinflammation 2015 12 114 10.1186/s12974-015-0332-6 26048578 Huang WY Liu KH Lin S Chen TY Tseng CY Chen HY Wu HM Hsu KS NADPH oxidase 2 as a potential therapeutic target for protection against cognitive deficits following systemic inflammation in mice Brain Behav Immun 2020 84 242 252 10.1016/j.bbi.2019.12.006 31841660 Jackson-Cowan L, Eldahshan W, Dumanli S, Dong G, Jamil S, Abdul Y, Althomali W, Baban B, Fagan SC, Ergul A (2021) Delayed Administration of Angiotensin Receptor (AT2R) agonist C21 improves survival and preserves sensorimotor outcomes in female diabetic rats post-stroke through modulation of microglial activation. Int J Mol Sci 22:1356 Jackson L, Eldahshan W, Fagan SC, Ergul A (2018) Within the Brain: The Renin Angiotensin System. Int J Mol Sci 19:876 Jackson L Dong G Althomali W Sayed MA Eldahshan W Baban B Johnson MH Filosa J Fagan SC Ergul A Delayed administration of angiotensin II type 2 receptor (AT2R) agonist compound 21 prevents the development of post-stroke cognitive impairment in diabetes through the modulation of microglia polarization Transl Stroke Res 2020 11 762 775 10.1007/s12975-019-00752-5 31792796 Jiang J Tang B Wang L Huo Q Tan S Misrani A Han Y Li H Hu H Wang J Cheng T Tabassum S Chen M Xie W Long C Yang L Systemic LPS-induced microglial activation results in increased GABAergic tone: a mechanism of protection against neuroinflammation in the medial prefrontal cortex in mice Brain Behav Immun 2022 99 53 69 10.1016/j.bbi.2021.09.017 34582995 Jing F Mogi M Sakata A Iwanami J Tsukuda K Ohshima K Min LJ Steckelings UM Unger T Dahlof B Horiuchi M Direct stimulation of angiotensin II type 2 receptor enhances spatial memory J Cereb Blood Flow Metab 2012 32 248 255 10.1038/jcbfm.2011.133 21971355 Kang HJ Voleti B Hajszan T Rajkowska G Stockmeier CA Licznerski P Lepack A Majik MS Jeong LS Banasr M Son H Duman RS Decreased expression of synapse-related genes and loss of synapses in major depressive disorder Nat Med 2012 18 1413 1417 10.1038/nm.2886 22885997 Kumar A Barrett JP Alvarez-Croda DM Stoica BA Faden AI Loane DJ NOX2 drives M1-like microglial/macrophage activation and neurodegeneration following experimental traumatic brain injury Brain Behav Immun 2016 58 291 309 10.1016/j.bbi.2016.07.158 27477920 Lenkei Z Palkovits M Corvol P Llorens-Cortes C Distribution of angiotensin II type-2 receptor (AT2) mRNA expression in the adult rat brain J Comp Neurol 1996 373 322 339 10.1002/(SICI)1096-9861(19960923)373:3<322::AID-CNE2>3.0.CO;2-4 8889931 Liu Z Yao X Jiang W Li W Zhu S Liao C Zou L Ding R Chen J Advanced oxidation protein products induce microglia-mediated neuroinflammation via MAPKs-NF-kappaB signaling pathway and pyroptosis after secondary spinal cord injury J Neuroinflammation 2020 17 90 10.1186/s12974-020-01751-2 32192500 Manabe T Heneka MT Cerebral dysfunctions caused by sepsis during ageing Nat Rev Immunol 2022 22 444 458 10.1038/s41577-021-00643-7 34764472 Mei B Li J Zuo Z Dexmedetomidine attenuates sepsis-associated inflammation and encephalopathy via central alpha2A adrenoceptor Brain Behav Immun 2021 91 296 314 10.1016/j.bbi.2020.10.008 33039659 Olmos-Alonso A Schetters ST Sri S Askew K Mancuso R Vargas-Caballero M Holscher C Perry VH Gomez-Nicola D Pharmacological targeting of CSF1R inhibits microglial proliferation and prevents the progression of Alzheimer's-like pathology Brain 2016 139 891 907 10.1093/brain/awv379 26747862 Pan TT Gao W Song ZH Long DD Cao P Hu R Chen DY Zhou WJ Jin Y Hu SS Wei W Chai XQ Zhang Z Wang D Glutamatergic neurons and myeloid cells in the anterior cingulate cortex mediate secondary hyperalgesia in chronic joint inflammatory pain Brain Behav Immun 2022 101 62 77 10.1016/j.bbi.2021.12.021 34973395 Pandharipande PP Girard TD Jackson JC Morandi A Thompson JL Pun BT Brummel NE Hughes CG Vasilevskis EE Shintani AK Moons KG Geevarghese SK Canonico A Hopkins RO Bernard GR Dittus RS Ely EW Investigators B-IS Long-term cognitive impairment after critical illness N Engl J Med 2013 369 1306 1316 10.1056/NEJMoa1301372 24088092 Pun BT Prevalence and risk factors for delirium in critically ill patients with COVID-19 (COVID-D): a multicentre cohort study Lancet Respir Med 2021 9 239 250 10.1016/S2213-2600(20)30552-X 33428871 Qiu LL Ji MH Zhang H Yang JJ Sun XR Tang H Wang J Liu WX Yang JJ NADPH oxidase 2-derived reactive oxygen species in the hippocampus might contribute to microglial activation in postoperative cognitive dysfunction in aged mice Brain Behav Immun 2016 51 109 118 10.1016/j.bbi.2015.08.002 26254234 Rivas-Santisteban R Lillo J Munoz A Rodriguez-Perez AI Labandeira-Garcia JL Navarro G Franco R Novel interactions involving the mas receptor show potential of the renin-angiotensin system in the regulation of microglia activation: altered expression in parkinsonism and dyskinesia Neurotherapeutics 2021 18 998 1016 10.1007/s13311-020-00986-4 33474655 Royea J Lacalle-Aurioles M Trigiani LJ Fermigier A Hamel E AT2R's (Angiotensin II Type 2 Receptor's) role in cognitive and cerebrovascular deficits in a mouse model of Alzheimer disease Hypertension 2020 75 1464 1474 10.1161/HYPERTENSIONAHA.119.14431 32362228 Saito M Fujinami Y Ono Y Ohyama S Fujioka K Yamashita K Inoue S Kotani J Infiltrated regulatory T cells and Th2 cells in the brain contribute to attenuation of sepsis-associated encephalopathy and alleviation of mental impairments in mice with polymicrobial sepsis Brain Behav Immun 2021 92 25 38 10.1016/j.bbi.2020.11.010 33181271 Sasannejad C Ely EW Lahiri S Long-term cognitive impairment after acute respiratory distress syndrome: a review of clinical impact and pathophysiological mechanisms Crit Care 2019 23 352 10.1186/s13054-019-2626-z 31718695 Schwengel K Namsolleck P Lucht K Clausen BH Lambertsen KL Valero-Esquitino V Thone-Reineke C Muller S Widdop RE Denton KM Horiuchi M Iwai M Boato F Dahlof B Hallberg A Unger T Steckelings UM Angiotensin AT2-receptor stimulation improves survival and neurological outcome after experimental stroke in mice J Mol Med (berl) 2016 94 957 966 10.1007/s00109-016-1406-3 26983606 Slutsky AS Tremblay LN Multiple system organ failure. Is mechanical ventilation a contributing factor? Am J Respir Crit Care Med 1998 157 1721 1725 10.1164/ajrccm.157.6.9709092 9620897 Sparrow NA Anwar F Covarrubias AE Rajput PS Rashid MH Nisson PL Gezalian MM Toossi S Ayodele MO Karumanchi SA Ely EW Lahiri S IL-6 inhibition reduces neuronal injury in a murine model of ventilator-induced lung injury Am J Respir Cell Mol Biol 2021 65 403 412 10.1165/rcmb.2021-0072OC 34014798 Terrando N Rei Fidalgo A Vizcaychipi M Cibelli M Ma D Monaco C Feldmann M Maze M The impact of IL-1 modulation on the development of lipopolysaccharide-induced cognitive dysfunction Crit Care 2010 14 R88 10.1186/cc9019 20470406 Tonnies E Trushina E Oxidative stress, synaptic dysfunction, and Alzheimer's disease J Alzheimers Dis 2017 57 1105 1121 10.3233/JAD-161088 28059794 Turon M Fernandez-Gonzalo S de Haro C Magrans R Lopez-Aguilar J Blanch L Mechanisms involved in brain dysfunction in mechanically ventilated critically ill patients: implications and therapeutics Ann Transl Med 2018 6 30 10.21037/atm.2017.12.10 29430447 Wang C Yue H Hu Z Shen Y Ma J Li J Wang XD Wang L Sun B Shi P Wang L Gu Y Microglia mediate forgetting via complement-dependent synaptic elimination Science 2020 367 688 694 10.1126/science.aaz2288 32029629 Wen J Ribeiro R Zhang Y Specific PKC isoforms regulate LPS-stimulated iNOS induction in murine microglial cells J Neuroinflammation 2011 8 38 10.1186/1742-2094-8-38 21510893 Werneburg S Jung J Kunjamma RB Ha SK Luciano NJ Willis CM Gao G Biscola NP Havton LA Crocker SJ Popko B Reich DS Schafer DP Targeted complement inhibition at synapses prevents microglial synaptic engulfment and synapse loss in demyelinating disease Immunity 2020 52 167–182 e167 10.1016/j.immuni.2019.12.004 Wilcox ME Brummel NE Archer K Ely EW Jackson JC Hopkins RO Cognitive dysfunction in ICU patients: risk factors, predictors, and rehabilitation interventions Crit Care Med 2013 41 S81 98 10.1097/CCM.0b013e3182a16946 23989098 Zhang B Bailey WM McVicar AL Stewart AN Veldhorst AK Gensel JC Reducing age-dependent monocyte-derived macrophage activation contributes to the therapeutic efficacy of NADPH oxidase inhibition in spinal cord injury Brain Behav Immun 2019 76 139 150 10.1016/j.bbi.2018.11.013 30453022 Zhang C Hu S Zosky GR Wei X Shu S Wang D Chai X Paracoxib alleviates ventilator-induced lung injury through functional modulation of lung-recruited CD11bloLy6Chi monocytes Shock 2021 55 236 243 10.1097/SHK.0000000000001591 32590697	36464726	PMC9734469	NO-CC CODE	2022-12-14 23:28:28	no	J Neuroimmune Pharmacol. 2022 Dec 5;:1-17	utf-8	J Neuroimmune Pharmacol	2,022	10.1007/s11481-022-10054-7	oa_other
==== Front Ann Oper Res Ann Oper Res Annals of Operations Research 0254-5330 1572-9338 Springer US New York 5112 10.1007/s10479-022-05112-5 Original Research Non-parametric generalised newsvendor model Ghosh Soham [email protected] 1 http://orcid.org/0000-0002-8089-4248 Mukhoti Sujay [email protected] 2 1 grid.450280.b 0000 0004 1769 7721 Humanities and Social Sciences, Indian Institute of Technology Indore, Indore, Madhya Pradesh 453552 India 2 grid.466775.1 0000 0001 1535 7334 Operations Management and Quantitative Techniques Area, Indian Institute of Management Indore, Indore, Madhya Pradesh 453556 India 3 12 2022 126 28 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 present paper we generalise the classical newsvendor problem for critical perishable commodities having more severe costs than its linear alternative. Piece wise polynomial cost functions are introduced to accommodate the excess severity. Stochastic demand is assumed to follow a completely unknown probability distribution. Non parametric estimator of the optimal order quantity has been developed from an estimating equation using a random sample. Strong consistency of the estimator is proved for unique optimal order quantity and the result is extended for multiple solutions. Simulation results indicate that non parametric estimator is efficient in terms of mean square error. Real life application of the proposed non-parametric estimator has been demonstrated with Avocado demand in the United States of America and Covid-19 test kit demand during second wave of SARS-COV2 pandemic across 86 countries. Keywords Stochastic programming Non-parametric estimation Strong consistency Monte-Carlo simulation Newsvendor problem Non-linear optimisation http://dx.doi.org/10.13039/501100001691 Indian Institute of Management Indore RS/09/2019-20 Ghosh Soham Department of Science and Technology, Government of IndiaINSPIRE/IF-190728 Mukhoti Sujay ==== Body pmcIntroduction Newsvendor problem deals with determination of optimal order quantity of a perishable commodity by offsetting piece-wise linear shortage and excess costs. We assume a newsvendor model with the following assumptions:single perishable item over a single period positive random demand with a continuous but unknown probability distribution no backlog or pre-booking is allowed absence of any influencing factors like marketing efforts,promotions,discounts etc. instantaneous replenishment of order quantities The decision for a single period problem is taken at the beginning, i.e. before the random demand is realised (see Chernonog & Goldberg, 2018, and the references therein). However, perishable critical resources would often warrant shortage and excess costs to be more severe than linear. In addition, a popular practice is to assume a parametric demand distribution with known parameters. In reality the distribution remains unknown most often. In this paper, we propose non-parametric estimation method of the optimal order quantity in the generalised newsvendor problem, which accommodates the severity of loss through a piece-wise non-linear function (Ghosh et al., 2021). In many real life situations, shortage and excess costs become more excruciating than a piece-wise linear newsvendor model, which accommodates only the quantity lost. For example, chemotherapy drugs are administered to patients as per a schedule. Shortage of the drug on the scheduled day would result in breaking of the treatment cycle. Here the loss is more severe than merely the quantity lost. Similarly, excess inventory of critical drugs or chemical resources might cause vast environmental and microbial hazards during disposal of the excess material. Piece-wise non-linearity is thus an appropriate choice for shortage and excess cost models. Non-linear newsvendor problem has been studied only recently in the literature. Parlar and Rempala (1992) considered the periodic review inventory problem and derived the solution of a newsvendor problem with a quadratic cost function. Gerchak and Wang (1997) described optimal order quantity determination from a newsvendor problem with linear excess but quadratic shortage cost. Pal et al. (2015) used exponential weight function of order quantity to the holding cost and linear excess cost in a newsvendor set-up. Kyparisis and Koulamas (2018) addressed the newsvendor problem for quadratic utility function. Khouja (1995), Chandra and Mukherjee (2005), among others, considered optimisation of reliability function of the stochastic cost. In this paper, we consider generalisation of the classical newsvendor model of Ghosh et al. (2021). This version of the generalised newsvendor model considers severity of losses by modelling the severity of shortage and excess costs using measurable and continuous non-linear weights and the conditions for existence of the optimal order quantity has been established for exponential and uniform demands. A critical issue with the optimal order quantity determination in classical newsvendor problem is the lack of knowledge on random demand. Majority of the works assume a completely specified demand distribution, whereas in reality, it is seldom so. In case of unknown demand distribution, parametric and distribution-free estimation of the optimal order quantity has been considered more recently. Parametric estimation of the optimal order quantity has been studied by Nahmias (1994) and more recently, Kevork (2010) for Normal demand. Agrawal and Smith (1996) estimated the order quantity for negative binomial demand. Rossi et al. (2014) has given bounds on the optimal order quantity using confidence interval for parametric demand distributions. Ghosh et al. (2021) estimated optimal order quantity for uniform and exponential demands in non-linear newsvendor problem. Distribution free estimation of optimal order quantity, on the other hand, has been studied in two parallel ways in the context of classical newsvendor problem. In the first case, the investigator has access to population summary measures like mean, variance etc, but the demand distribution remains unknown (Bai et al., 2020). Scarf (1958) and later Moon and Gallego (1994) studied the min-max optimal order quantity in such cases. The second approach considers the estimation problem based on an uncensored random sample from the unknown demand distribution. Pal (1996), Bookbinder and Lordahl (1989) discussed construction of bootstrap based point and interval estimator of the optimal order quantity using demand data. The sampling average approximation (SAA) method (see Kleywegt et al., 2001; Linderoth et al., 2006), replaces the expected cost by the sample average of the corresponding objective function and then optimises it. Levi et al. (2015) provides bounds of the relative bias of estimated optimal cost using SAA based on uncensored demand data. More recently data driven non-parametric approaches has become quite popular in studying different variations of classical newsvendor problems. He et al. (2012) studied the impact of availability of data for newsvendor model using nurse staffing data from a hospital. Ban and Rudin (2019) developed single step machine-learning algorithm for a classical newsvendor problem with historical data on demand and several related features. Punia et al. (2020) proposed machine-learning based solution for multi-item newsvendor model in presence of capacity constraint. Keskin et al. (2021) proposed a data driven estimation method for optimum order quantity when demand is a non-stationary time-series in a classical newsvendor set-up. Lin et al. (2022) discusses the data driven decision making of a risk-averse newsboy by maximising expected profit under the presence of value-at-risk constraint. However, not much work has been done on non-parametric estimation in non-linear newsvendor problems to the best of our knowledge. In this paper we explain the existence of optimal order quantity and devise a non-parametric technique to estimate it in the generalised newsvendor model. Our study makes two unique contributions to the literature. First, we develop a non-parametric estimator of the optimal order quantity in a generalised newsvendor set-up with non-linear cost function of higher degree. The non-parametric estimator is developed from an estimating equation using an uncensored random sample on stochastic demand. The feasibility of obtaining real positive solutions to the estimating equation has been derived in almost-sure sense. We have studied the strong consistency property of the estimated optimal order quantity when the true solution is unique and its extension to the cases, when true optimal order quantity is not unique or both the true and estimated optimal order quantities are not unique. Our second contribution is a detailed simulation study of the performance of the non-parametric estimator of optimal order quantity in a generalised newsvendor model. It may be remarked here that analytical measurement of performance of the non-parametric estimator seems to be very difficult for severe cases. Here we present the empirical distributions of the non-parametric estimators of optimal order quantities obtained from the simulation for different severity levels and per-unit costs. We also present comparison between the non-parametric estimator and its parametric counterparts having Uniform and Exponential demands using simulation. For this purpose we have computed the bias and mean square error (MSE) and optimal order quantities (Ghosh et al., 2021) using simulated data from the respective distributions. Asymptotic behaviour is presented through bias and MSE plots. Our simulation study is based on 3.15 million numerical experiments on the mentioned parametric demand distributions. We present the real-life application of the non-parametric method developed in this paper for estimating optimal order quantity in a generalised newsvendor set-up. First real-life scenario describes the amount of avocado imported in United States of America over consecutive weeks starting from January,2020 to July,2022. Second one describes the total number of daily tests performed during the second wave of COVID-19 pandemic. In both the cases we have estimated the optimal order quantity using non-parametric and parametric models with uniform and exponential demands. We compared the performance of the non-parametric estimator using percentage savings in estimated optimal cost (see Keskin et al., 2021). The paper concludes with a discussion on the findings. Symmetric generalised newsvendor problem We begin this section with the note that a table of major notations used in the following sections has been given in the appendix (Table 1) for ready reference of the readers. Our work in this paper considers a case where the severity of the excess and shortage are more than the quantity lost (i.e. the gap between inventory and demand). Let the stochastic demand be represented by a random variable X with a compact support X⊆R+ defined over the complete probability space (Ω,F,P), where F is the σ-algebra over Ω. Since we do not consider pre-booking, we have 0∈X. Further, let Ce(∈R+) and Cs(∈R+) be the excess and shortage costs per unit respectively. Then the cost function in classical newsvendor set-up at an inventory level Q is given by1 C(Q,X)=Ce(Q-X),ifX≤QCs(X-Q),ifX>Q Related stochastic programming problem under the assumption of existence of EG[X], is given by2 argminQ∈XEG[C(Q,X)] where G(·) is the induced probability distribution of X defined over the measurable space (R+,B+), B+ being the corresponding Borel-algebra. We consider generalisation of quadratic cost function by introducing polynomial weights (in Q and X) of degree m (say, P1,m(Q,X) and P2,m(Q,X)) to shortage and excess respectively. Degree of the polynomials (m) represents the (equal) severity of shortage and excess. The severity polynomials should satisfy the following properties: for a given X, Pi,m(Q,X) is continuously differentiable with respect to Q(∈X) for i=1,2, up to order m The mth derivative of Pi,m(Q,X) is finite, i=1,2. If for any convergent sequence {Xn} in X, Xn→a.s.Q, then Pi,m(Q,Xn)→a.s.0 for i=1,2. Based on the above properties, a natural choice for the severity polynomials are as follows:3 P1,m(Q,X)=∑j=0m-1(-1)m-1-jm-1jQjXm-1-j=(Q-X)m-1 4 P2,m(Q,X)=∑j=0m-1(-1)m-1-jm-1jQm-1-jXj=(X-Q)m-1 The constant m is integer valued and m-1 could be interpreted as the severity constant. As m increases, more severe is the loss. For m=1, no extra severity is implicated and the problem reduces to the classical newsvendor problem. Thus the new cost function for generalised newsvendor is given by5 Cm(Q,X)=Ce(Q-X)m,ifX≤QCs(X-Q)m,ifX>Q The new cost functions could also be interpreted as a generalisation of constant costs per unit (Ce,Cs) model to demand and inventory dependent cost models, viz. Ce(Q-X)m-1 and Cs(X-Q)m-1 respectively. In view of the above weight function structure, we now make the following assumptions about the probability distribution of demand (X): X is independent of Q G is continuous and strictly increasing over the support X Xp is G-integrable ∀p≥0 The assumption A1 is required to avoid the trivial solution of zero order quantity, which may arise for certain choices of demand distribution, the degree of severity (m) and the costs (Ce,Cs). For example, if the demand is Uniform(0, 2Q) then for Ce=Cs, the optimum order quantity would become zero. Hence, we make further assumption of Ce≠Cs. The expected cost function in this case can be written as,6 EG[Cm(Q,X)]=∫SQCe(Q-x)P1,m(Q,x)dG+∫SQ′Cs(x-Q)P2,m(Q,x)dG where SQ={ω∈Ω∣X(ω)∈(0,Q)}, SQ′=X\SQ and EG denotes expectation with respect to G. Differentiating Eq. 6 with respect to Q using Leibnitz rule, we get the first order condition for the minimisation problem stated above as follows7 ∂EG[Cm(Q,X)]∂Q=0⇒∫SQCe(Q-X)m-1dG=∫SQ′Cs(X-Q)m-1dG⇒Ce∫SQ(Q-X)m-1dG=Cs∫X(X-Q)m-1dG-∫SQ(X-Q)m-1dG⇒∫SQ(Q-X)m-1dG=CsCe+Cs(-1)m-1∫X(X-Q)m-1dG⇒EG(Q-X)m-1I(SQ)EG[(X-Q)m-1]=km where, I(SQ) is an indicator function over the set SQ and km=CsCe+(-1)m-1Cs. Denoting ∫SQ(Q-X)idG=θ1,i and E(X-Q)i=θ2,i, ∀i=1,2,…, Eq. 7 can be written as8 Let us define the jth partial raw moment of X as δj=∫SQXjdG and the jth raw moment of X by μj′=∫XXjdG ∀j=1,2,…. Further let, the optimal expected cost be denoted by φm∗ and the corresponding set of optimal order quantities by U∗, which are obtained by solving the population stochastic minimisation problem given later in this section (Eq. 9). Next we show that U∗ is non-empty, i.e. at least one feasible solution to Eq. 8 exists. Theorem 2.1 Consider the stochastic minimisation problem in a SyGen-NV set-up as follows,9 argminQ∈XEG[Cm(Q,X)] where X is the positive demand defined over the probability space (Ω,F,P) and Q is order quantity. Under the assumptions A1-A3, at least one positive solution to the stochastic minimisation problem exists. Proof From the first order condition in Eq. 8, we notice that10 ∫SQ(Q-X)m-1dG=km(-1)m-1∫X(Q-X)m-1dG,(Q∈X)⇒∫SQ∑j=0m-1m-1jQm-1-j(-1)jXjdG=km(-1)m-1∫X(Q-X)m-1dG⇒∑j=0m-1m-1jQm-1-j(-1)j∫SQXjdG-km(-1)m-1∫XXjdG=0.⇒∑j=0m-1m-1jQm-1-j(-1)j[δj-(-1)m-1kmμj′]=0⇒∑j=0m-1(-1)jβjQm-1-j=0,whereβj=m-1j[δj-(-1)m-1kmμj′] Odd values of m (m=2d+1) ensures that km lies in the interval (0, 1) and βj=2dj[δj-k2d+1μj′]. Letting Q→0, it is observed that, δ2d→0, resulting in limQ→0β2d=-k2d+1μ2d′<0 so that limQ→0∑j=02d(-1)jβjQ2d-j=β2d<0 . On the other hand, it is possible to choose a large Q, say Q0, so that δj≈μj′,∀j=0,1,…2d, whenever Q≥Q0. In that case, βj→τj, where, τj=2djμj′(1-k2d+1)>0,∀j=0,1,…2d. Choosing Q0=maxτ2j+1τ2j:j=0,1,…(d-1), we, therefore, obtain∑j=02d(-1)jτjQ2d-j=τ0Q2d-τ1Q2d-1+⋯+τ2d-2Q2-τ2d-1Q+τ2d=Q2d-1(τ0Q-τ1)+Q2d-3(τ2Q-τ3)+⋯+Q(τ2d-2Q-τ2d-1)+τ2d>0,forQ>Q0 Thus, the function in Eq. 10 is negative when Q→0 and is positive for large Q (i.e. Q>Q0). Hence, presence of a positive solution of Eq. 10 follows from the well known Bolzano’s theorem on zero of continuous functions. Even values of m (m=2d) ensures that either km>0 or km<-1. The first case is given by 0<k2d and βj=2d-1j[δj+k2dμj′]. Letting Q→0, it is observed that, δ2d-1→0, resulting in limQ→0β2d-1=k2dμ2d-1′>0 so that limQ→0∑j=02d-1(-1)jβjQ2d-1-j=(-1)2d-1β2d-1=-β2d-1<0 . Choosing a large Q, say Q1, implies δj≈μj′,∀j=0,1,…2d-1, whenever Q≥Q1. Here, βj→τj=2d-1jμj′(1+k2d)>0,∀j=0,1,…2d-1. Q1 is selected as Q1=maxτ2j+1τ2j:j=0,1,…(d-1). Therefore the function Eq. 10 is obtained as,∑j=02d-1(-1)jτjQ2d-1-j=τ0Q2d-1-τ1Q2d-2+⋯+τ2d-2Q-τ2d-1=Q2d-2(τ0Q-τ1)+Q2d-4(τ2Q-τ3)+⋯+τ2d-2Q-τ2d-1>0,forQ>Q1 similar argument as the previous case guarantees that a positive solution of the stochastic minimisation problem exists. The second case is given by k2d<-1 and βj=2d-1j[δj+k2dμj′]. Letting Q→0, it is observed that, δ2d-1→0, resulting in limQ→0β2d-1=k2dμ2d-1′<0 so that limQ→0∑j=02d-1(-1)jβjQ2d-1-j=(-1)2d-1β2d-1=β2d-1>0 . A large value of Q, say Q2, indicates that δj≈μj′,∀j=0,1,…2d-1, whenever Q≥Q2. In that case, βj→τj, where, τj=2d-1jμj′(1-\|k2d\|)=-2d-1jμj′(\|k2d\|-1)=-κj<0,∀j=0,1,…2d-1. κj is obtained as κj=2d-1jμj′(\|k2d\|-1)>0,∀j=0,1,…2d-1. The function is written as∑j=02d-1(-1)j+1κjQ2d-1-j=-κ0Q2d-1+κ1Q2d-2+⋯-κ2d-2Q+κ2d-1=Q2d-2(κ1-κ0Q)+Q2d-4(κ3-κ2Q)+⋯+κ2d-1-κ2d-2Q<0,forQ>Q2 The choice of Q2 is described as Q2=maxκ2j+1κ2j:j=0,1,…(d-1). Similar argument as previous one establishes the existence of the positive solution. Since there could be many positive roots, we select the one with maximum magnitude. □ Non-parametric optimal order quantity estimation in SyGen-NV In this section, we present non-parametric estimation of the optimal order quantity, when the demand distribution is completely unknown, but historical uncensored demand data are available. Let us denote an uncensored random sample of size n by drawn from G. We define two statistics as T1n=1n∑i=1n(Q-Xi)m-1I(Xi≤Q) and T2n=1n∑i=1n(Xi-Q)m-1. Then the sample version of the first order condition in Eq. 9 can be constructed by replacing θi,m-1 with corresponding Tin, i=1,2. The estimating equation can be written as11 Further, we define sample partial and complete raw moments of order j as dj=1n∑i=1nXijI(Xi≤Q) and mj′=1n∑i=1nXij. It can be easily observed that the sample raw moments dj and mj′ are unbiased estimators of δj and μj′. Hence, β^j=m-1j[dj-(-1)m-1kmmj′] is the unbiased estimator of βj. We then construct the sample version of the first order condition provided in Eq. 10 as12 ∑j=0m-1(-1)jβ^jQm-1-j=0 where β^j is as defined above. We would refer to as estimating function and the function in the alternative form of the first order condition in Eq. 12 as the random estimating function or simply random function. Properties of Tn∼ Some important properties of Tin,i=1,2 are as follows. Ti,n is unbiased for θi,m-1, i=1,2. Ti,n→a.s.θi,m-1 as n→∞ n(Tin-θi,m-1)→LN(0,σi,n2), where nσi,n2=θi,2m-2-θi,m-12,i=1,2 and the symbol →L stands for convergence in distribution. Proof of P1 is immediate by taking expectation of Ti,n. P2 follows from Kolmogorov’s strong law of large number (see pp-115 Rao, 1973) and the fact that each of Ti,n,i=1,2 is an average of independently and identically distributed (iid) random variables satisfying existence of variance by assumption A3 stated above. P3 is also straight forward from Lindeberg-Levy central limit theorem for iid samples Rao (1973). Properties of h(Tn∼;Q) We begin with the statement of the following properties of . P4 is a measurable function over (R+n,Bn) for every Q∈X. P5 is continuously differentiable with respect to Q within the compact set X a.e Bn. Property P4 of is straight forward from the fact that it is a ratio of two measurable functions for every Q∈X. The next property follows from the facts that T1n and T2n are positive a.eR+n for every Q∈X and ratio of non-zero functions are differentiable. In what follows, we provide the asymptotic distribution of the random function for every Q∈X. First we state an important result, called the delta method for asymptotic normality of a one time differentiable function. Theorem 3.1 (Delta Method DasGupta (2008)) Suppose is a sequence of k-dimensional random vectors such that . Let g:Rk→R be once differentiable at θ with the gradient vector g(1)(θ). Then13 We now prove the asymptotic normality of in the following theorem. Theorem 3.2 Consider the estimating function in Eq. 11. Then for large n14 where Σ is the dispersion matrix of , is the 1st vector derivative of with respect to evaluated at and Proof The co-variance between T1n and T2n is15 σ12;n=Cov(T1n,T2n)=Cov1n∑i=1n(Q-Xi)m-1I(Xi≤Q),1n∑i=1n(Xi-Q)m-1=1n2∑i=1nCov((Q-Xi)m-1I(Xi≤Q),(Xi-Q)m-1)=1n(-1)m-1θ1,2m-2-θ1,m-1θ2,m-1 From the property P3 and Eq. 15, it could be easily seen that is asymptotically multivariate normal with dispersion matrix Σ=((σij;n)),i,j=1,2 and σii;n=σi,n2. Also, note that Tin>0a.e.R+n,i=1,2 and is once differentiable for every Q∈X. We denote the 1st derivative of by , where for i=1,2. Thus, using routine algebra it can be easily shown that The proof of the theorem is then immediate from the delta method (Th. 3.1). □ Solution of the estimating equation In this section we present the statistical properties of the estimated optimal order quantity and the optimal value function. We denote by φ^m∗ the estimated optimal cost function and the corresponding set of estimated optimal order quantities are denoted by U^∗. In the following theorem we prove that U^∗ is non-empty with probability (wp) 1, i.e. there exists at least one positive solution to Eq. 11wp 1. Theorem 3.3 Under the regularity assumptions A1-A3, the random function ∑j=0m-1(-1)jβ^jQm-1-j will have positive zeroes almost surely. where β^j=dj-(-1)m-1kmmj′,∀j=1,2…m-1. Proof Notice that, dj→a.sδj and mj′→a.sμj′, which implies in turn that β^j→a.sβj. Thus the proof of this theorem is same as that of Th. 2.1 in almost sure sense. We omit the details to avoid repetition. □ Next we show that any solution to the estimating equation converges to the true optimal order quantity in SyGen-NV problem. Let the solution of the estimating equation Eq. 11 (or Eq. 12) be denoted by Q^n∗. We show that the solution is strongly consistent for the solution to the stochastic optimisation problem argminQ∈XEGCm(Q,X) under mild regularity conditions. First we state the following theorem without proof on existence of optima of a continuous function on a compact set. Theorem 3.4 (Extreme value theorem (see Stein & Shakarchi, 2010)) A continuous function on a compact set X is bounded and attains a maximum and minimum on X. We state the next lemma on the compactness of the complement of an open subset of a compact set. Lemma 3.5 Let X be a compact set and O be an open subset of X. Then O′=X\O, denoting the complement of O in X, is also a compact set. The proof is a routine exercise in real analysis and hence is omitted. Theorem 3.6 Let Q^n∗∈X be the unique solution to the estimating equation and Q∗ uniquely solves the stochastic programming problemargminQ∈XEGCm(Q,X) Then16 Q^n∗→a.s.Q∗ Proof Let O⊆X denote an arbitrary open neighbourhood of Q∗. From lemma 3.5, the complement of O, O′=X\O is also a compact set. Notice that the expected cost EG[Cm(X,Q)](=φm(Q),say), is a continuous function of Q. Hence, from Theorem 3.4, the stochastic optimisation problem argminQφm(Q) will have a solution in O′ with unique minimum value of φm(Q). Let us denote, r=minQ∈O′φm(Q)-φm(Q∗)>0. Also, from property P2 of Tin,(i=1,2) and the continuous mapping theorem, it can be easily seen that . Since Q^n∗∈X, there would exist n0(ϵ) for every ϵ>0, such that , ∀n≥n0(ϵ), wp 1. Therefore ∃n>n0(ϵ) for every 0<ϵ<r2, so that17 \|h(θ,Q^n∗)-h(θ,Q∗)\|<ϵ,∀n>n0(ϵ),wp1 This implies Q^n∗∉O′. O being arbitrary, Q^n∗→a.s.Q∗. □ The roots of the FOC (Eq. 10) may not be unique. Let the set of corresponding distinct roots be denoted by Q∗={Q1∗,Q2∗…Qk∗},k=1,2…m-1. Similarly, there could be p(≥1) roots of the random function (Eq. 12), say Q^∗={Q^1∗,Q^2∗…Q^p∗}. In the next two corollaries, we extend Theorem 3.6 for multiple roots. Corollary 3.6.1 Let Q^∗ be the set of distinct roots of the random function (Eq. 12) and Q∗ be unique solution to the stochastic minimisation problem (9). Then Q^max∗→a.sQ∗, where Q^max∗=max{Q^∗}. Proof Notice, the maximum of Q∗^ is unique. Hence, from Th. 3.6, the proof is immediate. □ Corollary 3.6.2 Let Q^n∗ be the unique solution to the random function Eq. 12 and Q∗ be the set of distinct solutions to the stochastic minimisation problem (9). Then Q^∗→a.sQi∗; for exactly one i; i=i=1,2,…,k. Proof Let Oi denote an arbitrary open neighbourhood around Qi∗ selected in such a way that Oi’s are disjoint. Then, O=∪i=1kOi is also an open set. Implementing the same argument as Theorem 3.6 we ensure that Q^n∗∈O. Disjoint property of Oi indicates Q^n∗∈Oi for exactly one i. □ Corollary 3.6.3 Let Q^∗ be the set of distinct solutions to the random function Eq. 12 and Q∗ is the set of distinct solutions of the FOC Eq. 10, then Q^max∗→a.sQi∗; for exactly one i; i=i=1,2,…,k. Proof Proof immediately follows from previous two corollaries. □ From the above theorem, it can be easily seen that the estimated optimal cost φ^n∗=φm(Q^∗) almost surely converges to the true optimal cost φm∗, using the continuity of the cost function φm(Q). Numerical experiments In this section we present the results of numerical experiments on the non-parametric estimator of the optimal order quantity in SyGen-NV set-up. We use Monte-Carlo simulation as well as real data to gauge the performance of the non-parametric estimator. For comparison purpose, we consider two parametric counter parts, viz. Exponential and Uniform demand distributions for estimating the optimal order quantity in this problem. Monte-Carlo simulation We consider here two known probability distributions for the demand, viz. Uniform(0, 1) and Exponential(1). The severity index m is assumed to be known (∈{2,3,4,5,10}). Further, we take the excess-to-shortage cost ratio, η(=CeCs)∈{0.25,0.45,0.65,0.85,1.05,1.25,1.45,1.65,1.85}. For each of the (m,η) pairs, we compute numerically the optimal order quantities for both Uniform and Exponential true demands. Further, we conduct 3.15 million Monte-Carlo simulation experiments for each of the demand distributions to understand the small and large sample properties of the non-parametric estimator. In particular, we draw random samples of size n (=20,50,100,500,1000,5000, 10000) for each combination of (η,m) and estimate the optimal order quantities Q^n∗ therefrom. We repeat this process for M times (M=5000). We study the sampling properties of Q^n∗ from these M estimates. Uniform demand distribution The optimal order quantity in the SyGen-NV problem with Uniform(0, 1) demand is given by (Ghosh et al., 2021)Qunif∗=11+η1m The non-parametric estimator, Q^n∗, can be obtained from the estimating equation (Eq. 12). The probability distribution of the estimated order quantity is presented in the form of box-plots in Fig. 4. For η<1, the probability distributions of Q^n∗ are stochastically larger with increasing severity levels, the distribution for m=2 being centred at the highest value among all others. For η>1, the distributions of estimated order quantity for even m are different than those of the odd m. Odd severity seems to result in stochastically smaller distribution of Q^n∗. The variation, on the other hand, seems to decrease with severity for all η. Next we present the performance study of Q^n∗ using the mean square error (MSE) computed from the M estimates as MSE=1M∑i=1M(Q^in∗-Qn∗)2. Figure 5a–i in the appendix presents the MSE’s plotted against sample sizes. It could be seen that for η<1, the MSEs converge to 0 with increasing n for all m, with worst performance of Q^n∗ observed at m=2. For η>1, however, the convergence is slow in case of even m. Exponential demand distribution The optimal order quantity in the SyGen-NV problem with Exponential(1) demand can be obtained from the random function (Eq. 10) by replacing the partial and full raw moments by those for the Exponential(1) distribution. The modified equation is given as (Ghosh et al., 2021)∑j=0m-1(-1)jQm-j-11(m-j-1)!=e-QCsCe-(-1)m As described in the uniform case, Q^n∗ can be obtained from the estimating equation (Eq. 12). Unlike the uniform demand case, probability distribution of the estimated optimal order quantity increases stochastically with severity for all η (see Fig. 6). Not only the location, the scale (or variance) of the distribution also increases with m. In terms of MSE, Q^n∗ performs well asymptotically as the MSE (vs. n) curve (see Fig. 7a–h) decreases to zero with increasing sample size (for all m and η), the worst performance being observed for m=10. The best estimator, in the MSE sense, is obtained for m=2 when η<1. However, for η>1 performance of Q^n∗ for m=2 worsens in small samples. Relative performance of non-parametric estimator We compare the performance of the non-parametric estimator with its parametric counter parts in terms of the estimated optimum cost. Following Ghosh et al. (2021), the parametric demand distributions are assumed to be Exponential(λ) and Uniform(0, b), so that the mean remains same (200). Optimum expected cost can be estimated by replacing Q with Q^∗ and the parameters by their maximum likelihood estimators. For uniform demand, closed form expression of the optimum cost is18 φm,Unif∗=Cs×η(Q∗)m+1+(b-Q∗)m+1b(m+1) For exponential demand the optimal cost function could be obtained as19 φm,Exp∗=Cs∑j=0mmj(-1)m-jηΓ(m-j+1)λm-j(Q∗)jγQ∗(m-j+1,λ)+Γ(j+1)λj(Q∗)m-jΓQ∗(j+1,λ) where γQ∗(m-j+1,λ) and ΓQ∗(j+1,λ) are the cumulative distribution function and survival function of Gamma distribution evaluated at Q∗. To estimate the above optimal costs, we replace Q∗ by Q^∗ and the parameters by their maximum likelihood estimators (MLE) (see Ghosh et al., 2021) in the above expressions. In the non-parametric case, optimum cost is estimated by replacing Q with respective Q^∗ in Eq. (6). We use the percentage savings in estimated optimal costs (see Keskin et al., 2021), given by Δ^=1-φ^m,D∗φ^m,NP∗×100 to compare the performance of the non-parametric estimator with the parametric alternatives (D∈{Unif,Exp}). Two parametric demand distributions that we consider here are thin tailed (Uniform(0, b)) and heavy tailed (Exponential(λ)). Thin tailed distribution don’t admit extreme observations whereas heavy tailed distributions are more likely to do so. It is also known that parametric inference is more efficient (in the MSE sense) if the data is generated from a correctly specified demand distribution. On the other hand non-parametric methods are more robust with respect to extreme values (or tail fatness) and incorrect specification of the parametric form of the distribution. Noticing these considerations, we have generated the data from N(200,652) distribution for fair comparison between parametric and non-parametric estimates. It is worth mentioning here that Normal is moderately tailed and not too close to either of the parametric models under comparison, viz. Uniform and Exponential. Since Q^n∗ is asymptotically unbiased, we have generated a large random sample of size 5000. The estimated percentage savings (Δ^) has been presented for all m and η in the Tables (2a-2b) in appendix A.1. All values in both the tables are negative indicating that non-parametric estimates work better than the parametric ones. Among the parametric models, Uniform is better than the Exponential, which we believe is due to heavier tail of the latter, i.e. due to higher likelihood of observing extreme demand. Variation in Δ^ with respect to the cost ratio (η) does not show any pattern. However, percentage savings in cost increases with degree of severity. Real data applications In this section we study the impact of different degrees of severity on optimal order quantity and related optimal cost estimation in the SyGen-NV set-up with real data. We use Avocado import data from Haas Avocado Board (2022) and Covid-19 daily new infection rate per day for 86 countries during second corona infection surge. We assume that the data are iid samples from the corresponding demand distributions. We consider the log-transformed data as the observations are quite large. Assuming a SyGen-NV setup, we model the demand as Uniform(0, b) or Exponential(λ) for parametric estimation of the optimal order quantity. Similarly, we also use the non-parametric method described above to estimate the optimal order quantity using the same data. In both the cases, we study the estimators and compare them for different degrees of severity (m) and cost-ratio (η) detailed in the following two subsections. Avocado import data First we consider the import data of Avocado from Haas Avocado Board (2022). The data contains weekly arrival volume of Avocado in the United States of America market over a time-period of two and half years (January,2020-July,2022). The volumes (in log-scale) are considered here as iid samples on avocado demand. For parametric inference, we assume that the demand distribution is either Uniform(0, b) or Exponential(λ) and estimate the corresponding optimal order quantities. The MLEs of b and λ are 7.89 and 7.71 respectively. Figure 2a–c shows the optimal order quantity plot against the cost ratio for different degrees of severity. Both uniform and exponential demands depict a decreasing trend in estimated optimal order quantity with cost ratio. The estimated optimal order quantity plots in Fig. 2a reveal a tendency of Q^∗ towards 4 (approximately) with increasing severity level (curve corresponding to m=10). Exponential demand exhibits a risk taking choice for the newsvendor as Q^∗ increases with m. The non-parametric estimates of Q∗, on the other hand exhibits an overall decreasing trend within a small interval compared to the parametric counterparts. The patterns are not very smooth as in the parametric cases due to restricted sample size. Comparing the percentage gain (Δ^) in non-parametric estimator of optimal cost given in tables (3a-3b), we observe that the non-parametric estimators provide lower cost and hence is more useful. Also, among the two parametric demand distributions, Uniform performs better in percentage cost savings. Covid-19: test kit demand In this section we analyse data set containing number of tests carried out to detect Corona Virus infection during the second wave of pandemic (01/03/2021–30/07/2021). The data were obtained from Ritchie et al. (2020) and we use the log-transformed number of tests for 86 countries during the said period as iid samples on demand of test kits. We aim to determine the optimal number of test kits that would be required if such a time appears again based on the iid sample. For Uniform(0, b) demand, MLE of b is b^=8.54. Here, the estimated optimal order quantities show a decreasing pattern in cost ratio with a tendency to b^/2=4.25 (approximately) for increasing severity (curve corresponding to m=10). In case of Exponential(λ) demand, the estimated optimal order quantity plots show a decreasing pattern with cost ratio, but for higher severity (m) the estimates increase sharply compared to the other two cases. In case of non-parametric estimators of the optimal order quantity the parameter estimates show an unsmooth decreasing pattern with asymptotic order quantity estimate seemingly close to 6.25 except for m=3. Non-parametric estimators of the optimal cost outperforms both the parametric counterparts. The tables (4a -4b) show that percentage gain are all highly negative for every η and m. In fact for a given η, the percentage gains improve in favour of non-parametric estimator with m. In other words, non-parametric estimator performs better than its parametric counterparts irrespective of the magnitude of cost ratio and severity level. Discussion In this paper we have discussed non-parametric estimation of the optimal order quantity in case of a general newsvendor problem, where the severity of the losses are much more than merely the quantity lost. Major contributions of this paper are two-fold. First we have constructed a non-parametric estimation method for the optimal order quantity in the SyGen-NV problem with power type shortage and excess. Secondly, we have studied the properties and performances of the estimators of the optimal order quantities. Our contribution in the non-parametric estimation of the optimal order quantity starts with formulation of an estimating equation from the first order condition using uncensored demand data. We have presented strong consistency of the estimating function and its asymptotic distribution has been derived. Further, we have established feasibility of solution to the estimating equation by establishing existence of the zeroes of the random function in almost sure sense. We have also proven the strong consistency of the estimated optimal order quantity. The theoretical results in this paper has been supported by an exhaustive set of simulation experiments and real data analysis. In particular, we have considered known uniform and exponential as true demand distributions. The distribution of the estimated optimal order quantities suggests that odd and even order of severity influences the estimates differently for uniform demand, whereas for exponential demand, the estimate increases uniformly with severity. Comparing the mean square errors for different sample sizes, severity and cost-ratio, it has been found that the estimators perform well in the MSE sense when severity is high in case of uniform demand and the opposite for exponential distribution. To show how well non-parametric estimators work, we have considered one synthetic data set of 5000 observations simulated from N(200,652) and two real data sets, viz. avocado import to USA and number of Covid-19 tests carried out in 86 countries during second wave of the pandemic (both in log-scale). Analysis of simulated data show that non-parametric method outperforms the parametric alternatives across all cost-ratio and severity values. The results derived from the analysis of real data sets show that estimated optimal order quantities increase with severity in exponential demand whereas the same tends to stabilise to a constant (b/2) with increase in m for uniform demand. In other words, exponential demand leads to an aggressive or risk taking newsvendor whereas uniform demand reflects a risk neutral newsvendor. We argue that this observed differences between uniform and exponential cases (order quantity or cost) happens due to the prospect of selling more in case of exponential model as it admits extreme demand. Uniform demand resembles, on the other hand, a risk neutral choice for the newsvendor since it is equally informative (or non-informative) about higher or lower demand and does not allow extreme observations. Q∗ decreases with m in this case and the newsvendor attempts to sell at an average level (b/2) when severity is very high, neglecting the role of η. In case of non-parametric analysis, the optimum order quantity estimates for both avocado and Covid-19 data shows not very smooth patterns. This lack of regularity in Figs. 2c and 3c could be argued as an aftermath of poor density estimation caused by the restricted sample size (see Fig. 1a and b). We conclude the paper with comments on future scope of research. A natural extension of the SyGen-NV problem would be to consider asymmetric weight functions for shortage and excess. Complexity arises due to different dimensions of the two costs as a result of asymmetric weighing. Baraiya and Mukhoti (2019) discussed, in an unpublished manuscript, selection of weights so that the shortage and excess costs remain comparable. However, estimation of optimal order quantity in such asymmetric generalised newsvendor problem remains open. Appendix A.1 Tables See Tables 1, 2, 3 and 4.Table 1 Table of notations X Random demand X Compact support of Demand distribution G Probability distribution of Demand EG expectation with respect to G B+ Borel Algebra over R+ F σ-algebra defined over Ω P Probability measure Ce Excess cost per unit Cs Shortage cost per unit Q∗ Optimal order quantity C(Q, X) Cost function for classical newsvendor Pm(Q,X) Polynomial in Q and X of degree m →a.s. Almost sure convergence →L Convergence in Distribution a.e Almost everywhere SQ {ω∈Ω∣X(ω)∈(0,Q)} SQ′ Complement of SQ (X\SQ) I(SQ) an indicator function over the set SQ δj jth order partial raw moment (∫SQXjdG) μj′ jth raw moment of X (∫XXjdG) Historical demand data (X1,X2,…,Xn) φm∗ Optimal cost function U∗ Set of optimal order quantities φ^m∗ Estimated optimal cost function U^∗ Set of estimated optimal order quantities dj jth order sample partial raw moment (1n∑i=1nXijI(Xi≤Q)) mj′ jth order sample raw moment (1n∑i=1nXij) Q^n∗ Non parametric estimator of optimal order quantity Q^∗ Set of distinct solutions to the estimating equation Q∗ Set of distinct solutions of First order condition Q^max∗ Largest member in the set {Q^∗} η Excess-to-shortage cost ratio (CeCs) γQ∗(m-j+1,λ) Cumulative Distribution Function (CDF) of Gamma distribution evaluated at Q∗ ΓQ∗(j+1,λ) Survival function of Gamma distribution evaluated at Q∗ Table 2 Percentage gain in non-parametric estimate of optimal cost over Uniform and Exponential alternatives with Normal(200,652) data η m=2 m=3 m=4 m=5 m=10 (a) Δ^ for Uniform demand 0.25 -54.98 -22.62 -328.48 -220.94 -778.49 0.45 -40.13 -77.04 -249.82 -167.27 -11.51 0.65 -52.44 -71.98 -163.67 -397.73 -422.99 0.85 -90.43 -68.11 -206.91 -178.43 -465.26 1.05 -34.38 -166.53 -248.44 -129.57 -461.06 1.25 -65.42 -223.31 -213.25 -324.12 -297.64 1.45 -80.11 -159.23 -245.71 -264 -482.44 1.65 -28.14 -65.71 -174.67 -322.1 -657.93 1.85 -54.03 -95.61 -236.74 -134.2 -439.38 (b) Δ^ for Exponential demand 0.25 -818.66 -207.53 -1.4E+04 -2.7E+04 -6.9E+08 0.45 -658.28 -470.89 -1.4E+04 -3.7E+04 -4.4E+08 0.65 -510.55 -670.18 -1.3E+04 -3.6E+04 -1.3E+09 0.85 -438.03 -581.97 -1.2E+04 -3.8E+04 -1.4E+09 1.05 -320.22 -834.53 -1.1E+04 -3.5E+04 -1.1E+09 1.25 -307.7 -778.99 -1.1E+04 -2.8E+04 -8.6E+08 1.45 -311.49 -661.01 -9.5E+03 -3.2E+04 -1.0E+09 1.65 -206.11 -545.98 -6.4E+03 -3.0E+04 -1.2E+09 1.85 -210.83 -607.93 -6.2E+03 -2.9E+04 -4.2E+08 Table 3 Percentage gain in non-parametric estimate of optimal cost over Uniform and Exponential alternatives for AVOCADO data η m=2 m=3 m=4 m=5 m=10 (a) Δ^ for Avocado data with Uniform demand 0.25 -3.8E+04 -2.6E+06 -8.3E+06 -3.8E+07 -8.2E+12 0.45 -3.7E+04 -2.3E+06 -7.5E+06 -5.1E+07 -1.2E+13 0.65 -3.3E+04 -2.2E+06 -7.1E+06 -4.9E+07 -2.7E+13 0.85 -3.1E+04 -2.1E+06 -6.6E+06 -5.1E+07 -4.8E+13 1.05 -2.8E+04 -2.0E+06 -6.2E+06 -5.2E+07 -5.4E+13 1.25 -2.7E+04 -1.7E+06 -5.3E+06 -5.3E+07 -5.8E+13 1.45 -2.5E+04 -1.4E+06 -4.5E+06 -4.8E+07 -5.6E+13 1.65 -2.3E+04 -1.3E+06 -4.1E+06 -4.9E+07 -5.9E+13 1.85 -1.7E+04 -1.2E+06 -4.0E+06 -4.9E+07 -6.0E+13 (b) Δ^ for Avocado data with Exponential demand 0.25 -9.7E+05 -5.2E+07 -7.5E+09 -1.2E+11 -1.8E+22 0.45 -6.7E+05 -6.7E+07 -4.9E+09 -1.6E+11 -2.2E+22 0.65 -4.8E+05 -6.8E+07 -3.9E+09 -1.5E+11 -4.5E+22 0.85 -3.8E+05 -6.4E+07 -3.2E+09 -1.5E+11 -7.2E+22 1.05 -3.0E+05 -5.9E+07 -2.7E+09 -1.5E+11 -7.7E+22 1.25 -2.6E+05 -5.0E+07 -2.2E+09 -1.4E+11 -7.8E+22 1.45 -2.2E+05 -4.2E+07 -1.7E+09 -1.3E+11 -7.3E+22 1.65 -1.9E+05 -3.8E+07 -1.5E+09 -1.3E+11 -7.3E+22 1.85 -1.3E+05 -3.5E+07 -1.4E+09 -1.2E+11 -7.2E+22 Table 4 Percentage gain in non-parametric estimate of optimal cost over Uniform and Exponential alternatives for Covid-19 test data η 2 3 4 5 10 (a) Δ^ for Covid-19 data with Uniform demand 0.25 -3.0E+02 -6.0E+02 -2.1E+03 -3.0E+03 -4.5E+04 0.45 -3.5E+02 -7.2E+02 -2.2E+03 -3.5E+03 -9.9E+04 0.65 -3.7E+02 -7.5E+02 -2.2E+03 -3.7E+03 -1.3E+05 0.85 -3.6E+02 -7.8E+02 -2.2E+03 -3.8E+03 -1.4E+05 1.05 -3.6E+02 -6.2E+02 -2.2E+03 -3.8E+03 -1.4E+05 1.25 -3.5E+02 -5.4E+02 -2.2E+03 -3.8E+03 -1.4E+05 1.45 -3.5E+02 -5.5E+02 -2.1E+03 -3.7E+03 -1.4E+05 1.65 -3.4E+02 -5.6E+02 -2.1E+03 -3.8E+03 -1.4E+05 1.85 -3.3E+02 -4.8E+02 -2.1E+03 -3.7E+03 -1.4E+05 (b) Δ^ for Covid-19 data with Exponential demand 0.25 -6.3E+03 -7.2E+03 -7.5E+05 -3.1E+06 -9.3E+12 0.45 -4.9E+03 -1.2E+04 -5.7E+05 -3.5E+06 -1.7E+13 0.65 -4.0E+03 -1.3E+04 -4.9E+05 -3.6E+06 -2.0E+13 0.85 -3.4E+03 -1.3E+04 -4.4E+05 -3.5E+06 -1.9E+13 1.05 -2.9E+03 -1.1E+04 -3.9E+05 -3.4E+06 -1.8E+13 1.25 -2.6E+03 -9.4E+03 -3.6E+05 -3.2E+06 -1.8E+13 1.45 -2.3E+03 -9.4E+03 -3.3E+05 -3.1E+06 -1.7E+13 1.65 -2.1E+03 -9.3E+03 -3.1E+05 -3.0E+06 -1.6E+13 1.85 -1.9E+03 -8.0E+03 -2.9E+05 -2.9E+06 -1.6E+13 A.2 Figures See Figs. 1, 2, 3, 4, 5, 6 and 7.Fig. 1 Density plots of Avocado and Covid test data (in log scale) Fig. 2 Estimated optimal order quantities (Q^∗) for Avocado between 2020-22 Fig. 3 Estimated optimal order quantities (Q^∗) for Covid-19 test data Fig. 4 Boxplot of estimated order quantity for different degrees of severity (m) for Uniform demand Fig. 5 MSE of estimated order quantity for different degrees of severity (m) for Uniform demand Fig. 6 Boxplot of estimated order quantity for different degrees of severity (m) for Exponential demand Fig. 7 MSE of estimated order quantity for different degrees of severity (m) for Exponential demand Acknowledgements The authors would like to thank the anonymous referees for their valuable comments, which has been very helpful in improving the manuscript. Work of first author was supported by INSPIRE Fellowship Grant, Department of Science and Technology, Govt. of India (Grant No. 190728) and work of second author was supported by the Indian Institute of Management Indore SEED grant (Grant No. SM/09/2019-20). The authors would also like to thank Dr. Abhirup Banerjee, Institute of Biomedical Engineering, University of Oxford for helpful suggestions on the simulation experiments. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Agrawal N Smith SA Estimating negative binomial demand for retail inventory management with unobservable lost sales Naval Research Logistics (NRL) 1996 43 6 839 861 10.1002/(SICI)1520-6750(199609)43:6<839::AID-NAV4>3.0.CO;2-5 Bai Q Xu J Zhang Y The distributionally robust optimization model for a remanufacturing system under cap-and-trade policy: A newsvendor approach Annals of Operations Research 2020 2020 1 30 Ban G-Y Rudin C The big data newsvendor: Practical insights from machine learning Operations Research 2019 67 1 90 108 10.1287/opre.2018.1757 Baraiya, R. & Mukhoti, S. (2019). Generalization of the newsvendor problem with gamma demand distribution by asymmetric losses. Technical Report WP/03/2019-20/OM &QT, Indian Institute of Management Indore, India, 2019. Bookbinder J Lordahl A Estimation of inventory reorder level using the bootstrap statistical procedure IIE Transactions 1989 21 302 312 10.1080/07408178908966236 Chandra A Mukherjee SP Some alternative methods of finding the optimal order quantity in inventory models Calcutta Statistical Association Bulletin 2005 57 1–2 121 10.1177/0008068320050110 Chernonog T Goldberg N On the multi-product newsvendor with bounded demand distributions International Journal of Production Economics 2018 203 38 47 10.1016/j.ijpe.2018.05.027 DasGupta A Asymptotic theory of statistics and probability 2008 Berlin Springer Gerchak Y Wang S Liquid asset allocation using “newsvendor” models with convex shortage costs Insurance: Mathematics and Economics 1997 20 1 17 21 Ghosh S Sahare M Mukhoti S Sinha BK Bagchi SB A new generalized newsvendor model with random demand and cost mis-specification Strategic management, decision theory and decision sciences: contributions to policy issues. chapter 14 2021 Berlin Springer Nature Haas Avocado Board. Volume data and category data, 2022. https://hassavocadoboard.com/. He B Dexter F Macario A Zenios S The timing of staffing decisions in hospital operating rooms: incorporating workload heterogeneity into the newsvendor problem Manufacturing & Service Operations Management 2012 14 1 99 114 10.1287/msom.1110.0350 Keskin, N. B., Min, X., & Song, J.-S. J. (2021). The nonstationary newsvendor: Data-driven nonparametric learning. SSRN 3866171. 10.2139/ssrn.3866171. Kevork IS Estimating the optimal order quantity and the maximum expected profit for single-period inventory decisions Omega 2010 38 3–4 218 227 10.1016/j.omega.2009.09.005 Khouja M The newsboy problem under progressive multiple discounts European Journal of Operational Research 1995 84 2 458 466 10.1016/0377-2217(94)00053-F Kleywegt A Shapiro A Homem-De-Mello T The sample average approximation method for stochastic discrete optimization SIAM Journal on Optimization 2001 12 479 502 10.1137/S1052623499363220 Kyparisis GJ Koulamas C The price-setting newsvendor problem with nonnegative linear additive demand European Journal of Operational Research 2018 269 2 695 698 10.1016/j.ejor.2018.02.019 Levi R Perakis G Uichanco J The data-driven newsvendor problem: New bounds and insights Operations Research 2015 63 6 1294 1306 10.1287/opre.2015.1422 Lin S Chen Y Li Y Shen Z-JM Data-driven newsvendor problems regularized by a profit risk constraint Production and Operations Management 2022 31 4 1630 1644 10.1111/poms.13635 Linderoth J Shapiro A Wright S The empirical behavior of sampling methods for stochastic programming Annals of Operations Research 2006 142 1 215 241 10.1007/s10479-006-6169-8 Moon I Gallego G Distribution free procedures for some inventory models Journal of the Operational research Society 1994 45 6 651 658 10.1057/jors.1994.103 Nahmias S Demand estimation in lost sales inventory systems Naval Research Logistics (NRL) 1994 41 6 739 757 10.1002/1520-6750(199410)41:6<739::AID-NAV3220410605>3.0.CO;2-A Pal B Sana SS Chaudhuri K A distribution-free newsvendor problem with nonlinear holding cost International Journal of Systems Science 2015 46 7 1269 1277 10.1080/00207721.2013.815828 Pal M Asymptotic confidence intervals for the optimal cost in newsboy problem Calcutta Statistical Association Bulletin 1996 46 3–4 245 252 10.1177/0008068319960308 Parlar M Rempala R A stochastic inventory problem with piecewise quadratic costs International Journal of Production Economics 1992 26 1–3 327 332 10.1016/0925-5273(92)90083-J Punia S Singh SP Madaan JK From predictive to prescriptive analytics: A data-driven multi-item newsvendor model Decision Support Systems 2020 136 113340 10.1016/j.dss.2020.113340 Rao CR Linear statistical inference and its applications 1973 New York Wiley Ritchie, H., Mathieu, E., Rodés-Guirao, L., Appel, C., Giattino, C., Ortiz-Ospina, E., Hasell, J., Macdonald, B., Beltekian, D. & Roser, M. (2020). Coronavirus pandemic (Covid-19). Our world in data. Rossi R Prestwich S Tarim SA Hnich B Confidence-based optimisation for the newsvendor problem under binomial, Poisson and exponential demand European Journal of Operational Research 2014 239 3 674 684 10.1016/j.ejor.2014.06.007 Scarf H Arrow K Karlin S Scarf H A min-max solution of an inventory problem Studies in the mathematical theory of inventory and production 1958 Stanford Stanford University Press 201 209 Stein EM Shakarchi R Complex analysis 2010 Princeton Princeton University Press	0	PMC9734470	NO-CC CODE	2022-12-14 23:28:28	no	Ann Oper Res. 2022 Dec 3;:1-26	utf-8	Ann Oper Res	2,022	10.1007/s10479-022-05112-5	oa_other
==== Front Sozial Extra Sozial Extra 0931-279X 1863-8953 Springer Fachmedien Wiesbaden Wiesbaden 540 10.1007/s12054-022-00540-w Durchblick: Junge Menschen, Politikberatung und Beteiligung Weg vom Katzentisch Zum Stellenwert von Kindern und Jugendlichen in der Politikberatung Fanroth Yola 12004. Seit 2019 Mitglied im Jugendbeirat des Deutschen Kinderhilfswerks und Schulsprecherin an ihrer Schule. Hofmann Holger [email protected] 11966. Dipl. Sozialarbeiter, Mediator, seit 2012 Bundesgeschäftsführer des Deutschen Kinderhilfswerkes und Mitglied des jugendpolitischen Beirats des Bundesfamilienministeriums. Sipeer Vincent [email protected] 22001. Studium der Staats- und Rechtswissenschaften, Mitglied der Initiative „Starke Kinder- und Jugendparlamente“ und im Jugendpolitischen Beirat des Bundesfamilienministeriums. 1 Berlin, Deutschland 2 Erfurt, Deutschland 6 12 2022 15 13 9 2022 15 9 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 eine anderer Vertragspartnerin) hält die ausschließlichen Nutzungsrechte an diesem Artikel kraft eines Verlagsvertrags mit dem/den Autorin(nen) oder anderen Rechteinhaberin(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. Es braucht verbindliche Strukturen, innerhalb derer Kinder und Jugendliche ihre Interessen zu allen für sie relevanten gesellschaftlichen Themen einbringen können. Dafür müssen Beteiligungsstrukturen ausgebaut, Qualitäten entwickelt und Beteiligungsrechte gesetzlich verankert werden. Junge Menschen müssen dabei viel öfter selbst mit am Verhandlungstisch sitzen, weil sie es können. Diese Aspekte werden von Yola Fanroth, Holger Hofmann und Vincent Sipeer diskutiert. Schlüsselwörter Beteiligung Mitwirkung Kinderhilfswerk Generationengerechtigkeit ==== Body pmcUnsere Demokratie ist abhängig davon, dass es gelingt, sowohl den Interessen aktueller als auch künftiger Generationen gleichermaßen gerecht zu werden. Aus Sicht des Deutschen Kinderhilfswerks werden die Interessen junger Menschen in politischen Prozessen nur nachrangig berücksichtigt. Sollte sich dieser Trend nachhaltig bestätigen, steht unsere Gesellschaft vor einer Zerreißprobe. Ein Gespräch mit Yola Fanroth (Schülerin), Vincent Sipeer (Student) und Holger Hofmann (Deutsches Kinderhilfswerk). Holger Hofmann: Während der Hochphase der Corona-Pandemie haben wir erlebt, dass Kinder nicht als Grundrechtsträger wahrgenommen wurden, die nach der UN-Kinderrechtskonvention ein Recht auf Beteiligung haben, sondern, zugespitzt ausgedrückt, als Virenschleudern und Krankheitsüberträger_innen. Ihre eigenen Interessen und Bedürfnisse wurden erst sehr spät, zu spät in den Blick genommen. Gleichzeitig wurden sie auf ihre Rolle als Schüler_innen reduziert, ihre Mitwirkung an Lösungen wurde von der Politik nicht angestrebt. Eine Ausnahme bildete die Schulleitline S 3, die unter Federführung des Universitätsklinikums Hamburg-Eppendorf entwickelt wurde. Yola, du warst eine der Jugendlichen, die an dieser Leitlinie mitwirken konnten. Welche Erfahrungen hast du gemacht? Yola Fanroth: Für mich war das eine sehr positive Erfahrung. Ich hatte eigentlich keine große Erwartungshaltung, sondern mich vor allem darüber gefreut, dass ich mitwirken darf und nach meiner Meinung gefragt wurde. Dabei habe ich sehr viel Neues gelernt und war positiv überrascht von der gemeinschaftlichen Arbeit mit den Erwachsenen, die ganz viele verschiedene Blickwinkel mitbringen. Am Ende war ich stolz, dass unsere Leitlinie von vielen Schulen genutzt wurde. Es macht für mich viel Sinn, dass Kinder und Jugendliche an Themen mitwirken, die sie selbst betreffen und erleben, wie ihre Mitwirkung eine Wirkung entfaltet. Holger Hofmann: Wie wurde der Blinkwinkel der Kinder und Jugendlichen berücksichtigt? Yola Fanroth: Die ersten Vorschläge kamen von den Erwachsenen. Die Jugendlichen konnten jedoch sagen: „Das finden wir nicht gut oder wir haben etwas zu ergänzen“. Jedenfalls wurden wir als Expertinnen und Experten in eigener Sache angesehen und alle hatten dieselben Rederechte. Holger Hofmann: Ich stelle mir vor, dass die medizinischen Aspekte für Kinder und Jugendliche mitunter schwierig nachzuvollziehen waren. Yola Fanroth: Natürlich, manchmal war es schwer für mich, spezielle Inhalte zu verstehen, aber dann hat mir geholfen, dass ich eine Ansprechperson hatte, bei der ich nachfragen konnte und die mir es dann in Ruhe erklärt hat. Es war gut für mich, dass ich immer wieder gefragt wurde, ob ich Fragen habe. Alle wurden mitgenommen, egal welche Vorkenntnisse sie hatten und am Ende haben viele Jugendliche und Erwachsene hervorgehoben, dass ein gemeinsames Ergebnis geschaffen wurde. Holger Hofmann: Das war aus meiner Sicht an dem Ergebnis sichtbar und es liegt nahe, dass dies dann auch die Akzeptanz unter denjenigen erhöht hat, die diese Leitlinie umsetzen müssen, also die Schülerinnen und Schüler selbst. Yola Fanroth: Das sehe ich auch so und es wäre wichtig, wenn wir nicht nur in besonderen Situationen wie in so einer Krise Jugendliche beteiligen, sondern kontinuierlich. Vincent, du bist Vorsitzender des Dachverbandes der Kinder- und Jugendgremien in Thüringen. Du begleitest damit schon seit langem eine Form der kontinuierlichen Politikberatung. Wie erlebst du den Einfluss von Kindern und Jugendlichen auf die Politik der Erwachsenen? Vincent Sipeer: Es gibt in Deutschland ungefähr 500 Kinder- und Jugendparlamente und nochmal 300 Kinder- und Jugendforen. Das zeigt, dass die Kinder und Jugendbeteiligung von unten nach oben wächst und darüber in den Kommunen und Landkreisen recht gut in der Fläche etabliert ist. Aber das sagt noch nichts über die Qualität aus. Du hast ja gefragt, ob die Beteiligung dort schon besonders gut klappt. Das würde ich verneinen. Oftmals sitzen die Kinder und Jugendlichen nur am Katzentisch, diskutieren für sich und werden nicht unmittelbar angehört, wenn Entscheidungen getroffen werden. Dabei wird das besondere Potenzial nicht erkannt, wenn Kinder und Jugendliche gleichberechtigt mit am Tisch sitzen. Die Ergebnisse von Politik, die Entscheidungen und Maßnahmen, werden bedarfsgerechter, die Akzeptanz ist größer und die Entscheidungen sind nachhaltiger, wenn man Kinder und Jugendliche mit an den Tisch bittet. Denn Kinder und Jugendliche haben ein besonderes Gespür dafür, was die Themen der Zeit sind und dafür was eine Kommune anpacken muss, um zukunftsfähig zu werden. Yola Fanroth: Du warst selbst auch schon zu einem Hearing im Landtag eingeladen. Wie war das für dich? Vincent Sipeer: Für mich persönlich, wie für den Dachverband der Kinder- und Jugendgremien Thüringen, war das etwas ganz Besonderes, zur parlamentarischen Anhörung eines Gesetzentwurfes eingeladen zu werden. In Thüringen gibt es 25 Kinder- und Jugendparlamente. Gemeinsam mit den Jugendverbänden haben wir als Türöffner darauf hingewirkt, dass das Ausführungsgesetz zum Kinder- und Jugendhilfegesetz in Thüringen vorsieht, junge Menschen in die Jugendhilfeausschüsse reinzubringen. Und seitdem sitzen im Landesjugendhilfeausschuss tatsächlich zehn junge Menschen. Die saßen da vorher nicht und wir waren sehr stolz darauf, dass heute Kinder und Jugendliche in diesem Gremium vertreten sind, in dem elementare Fachpolitik für junge Menschen gemacht wird. Persönlich betrachtet, war es für mich ein sehr bewegender Moment, weil ich dort als 17-jähriger mit sehr vielen Erwachsenen in einem Raum war, die alle viel erfahrener waren als ich. Dank der guten Vorbereitung unseres Statements für diese Anhörung habe ich mich so bestärkt gefühlt, dass ich dort auch überzeugend vortragen konnte. Ich denke rückblickend, das war ein Meilenstein, weil wir eben nicht am Katzentisch saßen, sondern direkten Einfluss hatten. Das sollte man Kindern und Jugendlichen bei allen sie betreffenden Themen gewähren. Yola Fanroth: Ich frage mich, was man tun kann, dass solche Erfahrungen ganz viele andere Kinder und Jugendliche machen. Vincent Sipeer: So wie ich das gerade in Thüringen erlebt habe, sind wir eine der ersten Generationen, die Beteiligungsräume in der Landespolitik, in den Kommunen, in den Landkreisen erkämpfen. Deshalb habe ich auch den Begriff „Türöffner“ gebraucht, weil die jugendfreundlichen Regelungen in den Sitzungen erst geschaffen werden müssen. Es geht jedoch nicht nur darum, Kinder und Jugendliche anzuhören, sondern man muss auch darüber nachdenken, wann fangen die Sitzungen an und wie lange dauern sie? Wo finden sie statt? Was sind eigentlich die Zugänge und können sich alle gut vorbereiten? Ich erinnere mich an ganz viele Vorbereitungen, etwa mit der Jugendamtsleiterin, in der wir uns gemeinsam durch einen Wust von Unterlagen gearbeitet haben. Das war für alle Beteiligten sicher anstrengend, aber nur so wird am Ende ein Schuh draus. Holger Hofmann: Noch eine persönliche Frage, Vincent. Glaubst du, dass genau diese Verantwortung, die du da bekommen hast, dich angesteckt hat, weiterzumachen? Vincent Sipeer: In den Beteiligungsnetzwerken in Thüringen haben wir innerhalb der letzten Jahre eine ganze Menge junge Leute mitten reingebracht ins Spiel, die haben Reden gehalten vor hundert Leuten. Das hätten sie sich selbst nie geträumt. Mich hat es bestärkt, mehr Verantwortung zu übernehmen. Gleichzeitig wachsen bei solchen Gelegenheiten eine ganze Menge persönliche Fähigkeiten und Kenntnisse und man traut sich mehr zu. Holger, nach den Zahlen eures letzten Kinderreports sind es vor allem Haupt- und Gesamtschüler, die ihre Interessen nicht berücksichtigt sehen. Wie erklärst du dir, dass diese Gruppe sich außenvor fühlt und ist das nicht ein alarmierendes Ergebnis? Holger Hofmann: Das ist ähnlich wie bei den Erwachsenen. Auch bei den Kindern und Jugendlichen haben viele das Gefühl, dass nur denjenigen zugehört wird, die sich gut ausdrücken können oder die über ihre Eltern Einflussmöglichkeiten haben. Und da sehen sich viele Gymnasiasten ganz vorne. Und noch etwas möchte ich hier anführen: Während früher die Straßenkindheit eine Unterschichtskindheit war, ist es heute die der Familien mit mehr Geld im Portemonnaie. Straßenkindheit ist hier in einem breiten Sinne zu verstehen. Arme Kinder und Jugendliche sind nicht nur weniger beim Spielen draußen anzutreffen, sondern auch weniger in der Musik- oder Kunstschule, in Vereinen aktiv. Und haben deshalb das Gefühl, überhaupt nicht mehr gesehen zu werden. Daher gilt es, Straßensozialarbeit, Angebote in belasteten Stadtteilen, niedrigschwellige Kunst und Kultur und attraktive Aufenthaltsorte für junge Menschen im öffentlichen Raum zu fördern. Yola Fanroth: Und was hat das mit unserem Thema Politikberatung zu tun? Holger Hofmann: Eine ganze Menge, weil der Weg zur politischen Mitbestimmung über die persönliche Selbstbestimmung führt. Wie sehr Kinder und Jugendliche in der Lage sind, Verantwortung zu übernehmen, ihre Meinung in der Öffentlichkeit zu vertreten, Fähigkeiten wie Toleranz und Empathie zu entwickeln, hängt von ihrer sozialen Entwicklung ab. Wir haben vor einigen Jahren eine Umfrage unter amtierenden Bürgermeisterinnen und Bürgermeistern gemacht und gefragt, was in frühen Jahren prägend dafür war, dass sie heute dieses Amt ausüben können. Es gab natürlich eine Reihe unterschiedlicher Antworten, aber fast alle haben geantwortet, dass ihnen früh zugetraut wurde bzw. sie die Möglichkeit dazu hatten, Verantwortung zu übernehmen. Beispielweise in der Schule oder im Verein. Vincent Sipeer: Gleichzeitig sind in unseren Kinder- und Jugendparlamenten nach eurer eigenen Studie Kinder und Jugendliche aus allen sozialen Schichten zu finden. Wie erklärst du dir das? Holger Hofmann: Kinder- und Jugendparlamente sind im günstigsten Fall in ein Netzwerk von unterschiedlichen Beteiligungsformaten eingebunden und in Kommunen zu finden, die beispielsweise auch in Kita und Schule auf Beteiligungsangebote achten. Weiterhin werden Kinder- und Jugendparlamente in der Regel von einer Fachkraft betreut. Alles Faktoren, die dazu führen sollten, dass keine soziale Selektion stattfindet. Würdest du zustimmen, Vincent, du bist da näher dran? Vincent Sipeer: Aus meiner Sicht ist noch zu ergänzen, dass es auch auf ein eigenes Budget für Kinder- und Jugendparlamente ankommt, damit es nicht reine Debattierclubs sind und, dass in den gemeinsamen Gremien mit den Erwachsenen den jungen Menschen auf Augenhöhe begegnet wird. Yola Fanroth: Was heißt das für dich? Vincent Sipeer: Zunächst würde ich die Rolle der Erwachsenen in Gremien unterscheiden, in denen ausschließlich Kinder und Jugendliche sitzen, das ist bei Kinder- und Jugendparlamenten der Fall, und anderen Gremien, in denen sowohl junge Menschen als auch Erwachsene im besten Fall kollegial zusammenarbeiten. In Jugendgremien sollten Erwachsene beratend unterstützen und als Ermöglicher auftreten. Es sind dann im besten Fall Kinder und Jugendliche selbst, die die Tagesordnung schreiben, die durch die Sitzung führen und die Abstimmungen durchführen. In gemischten Gremien ist es komplexer. Dort sollten die Erwachsenen die Position von jungen Menschen empathisch nachvollziehen. Wenn Jugendliche mitwirken, sollte der Umgangston insgesamt weniger konfrontativ sein. Holger Hofmann: Müssen die Erwachsenen in Kauf nehmen, dass die Beratungen mehr Zeit in Anspruch nehmen? Vincent Sipeer: Nur auf den ersten Blick, denn maßgeblich ist das Ergebnis. Wenn der Auftrag klar ist, die Strukturen und die flankierenden Maßnahmen stimmen und bei allen Beteiligten die jugend- und partizipationsfreundliche Haltung stimmt, dann werden die Prozesse und die Ergebnisse eines Gremiums mit Jugendbeteiligung besser sein als jene eines vergleichbaren Gremiums ohne Anhörung und Beteiligung der Jugend. Vielleicht sollten wir auch darauf eingehen, welche Themen mit Kindern und Jugendlichen besprochen werden können. Yola, du bist jünger als ich, wie sieht du das? Yola Fanroth: Für mich lässt sich das nicht eingrenzen. Natürlich sollten es Themen sein, die Kinder und Jugendliche unmittelbar betreffen, aber in der Regel betrifft Kinder und Jugendliche schon heute fast alles, also Themen wie Schule, Freizeitgestaltung, Corona, Krieg oder es betrifft sie in der Zukunft. Dazu zählt der Klimawandel genauso wie das Thema Generationengerechtigkeit. Holger Hofmann: Generationengerechtigkeit ist ein recht komplexes Thema. Meinst du, Kinder und Jugendliche haben Lust auf Diskussionen wie Rentenpolitik? Yola Fanroth: Naja, das Thema ist für uns weit weg, aber ich frage mich, warum wir nicht in der Schule mehr darüber diskutieren. Es ist doch wichtig zu wissen, wie das im Alter funktioniert, was man dafür tun kann, dass man als Rentnerin nicht jeden Euro zweimal umdrehen muss. Vincent Sipeer: Ich stimme dir zu. Es sollte keine Politikfelder geben, die allein Erwachsenen vorbehalten sind oder zu denen Kinder und Jugendlichen keine Meinung haben dürfen. Ich finde es wichtig, dass junge Menschen auch zu Themen wie Stadtplanung, Regionalplanung, Verkehrsplanung, die Aufstellung eines kommunalen Klimaplans einbezogen werden. Angesichts der mangelhaften Berücksichtigung der Kinder- und Jugendperspektiven in Bezug auf die Pandemie, die Klimakatastrophe oder die Verkehrswende muss jetzt insgesamt stärker über Jugendpolitikberatung nachgedacht werden. Ich fordere von der Verwaltung und der Politik daher, betroffene Fokusgruppen junger Menschen routiniert und ernsthaft einzubeziehen. Grundsätzlich sollte es mittlerweile „state of the art“ sein, dass politische Gremien, die kinder- und jugendrelevante Themen bearbeiten oder deren Entscheidungen ihre Belange tangieren, nicht mehr nur für Jugend Politik machen, sondern gemeinsam mit ihnen. Yola Fanroth: Leider spürt man als Jugendlicher auch, dass Erwachsene unsere Meinung schlicht nicht ernst nehmen. Vincent Sipeer: Bei Vorbehalten aufgrund von Kenntnissen und Fähigkeiten ist fraglich, ob es sich nicht auch schlicht um Adultismus handelt. Also Vorurteile gegenüber einer Person aus Gründen des geringeren Alters. Es gibt aber auch Strukturen, die eine Diskriminierung Jugendlicher produzieren und aufrechterhalten. Holger Hofmann: Richtig, ein Beispiel dafür sind einseitige Rahmenplanungen der Kommunen. Nehmen wir die Verkehrsplanung. Für jede größere Kreuzung wird ein Gutachten zum Verkehrsfluss der Autos erstellt. Kinder, die Verkehrssituationen aufgrund ihrer Größe schwerer einschätzen können oder ungeduldig auf Mittelstreifen minutenlang warten müssen, werden dann gar nicht berücksichtigt, geschweige denn, dass man sich wirklich Gedanken darüber macht, wie sie beispielsweise im ländlichen Raum außerhalb der Schulzeiten von A nach B kommen. Bei der Stadtplanung insgesamt muss man festhalten, dass wir auch Rückschritte machen. Einen Rahmenplan für Spiel und Bewegung, in vielen Kommunen auch als Spielleitplanung bezeichnet, gibt es heute nur noch in sehr wenigen Städten und Gemeinden. Rheinland-Pfalz und Nordrhein-Westfalen hatten hier mal eine Vorreiterrolle. Davon ist nicht mehr so viel übriggeblieben. Die Bedürfnisse anderer Bevölkerungsgruppen als die der Kinder und Jugendlichen werden stärker bzw. verbindlicher von den Verwaltungen in den Blick genommen. Vincent Sipeer: Vielleicht die Folge, dass sie nicht wählen und darüber Einfluss ausüben können. Yola, würdest du es begrüßen, wenn auf europäischer Ebene zur Wahl des Europäischen Parlaments oder zur Wahl des Deutschen Bundestages das Wahlalter abgesenkt würde? Yola Fanroth: Ja, das würde ich total begrüßen, weil das ein Schritt zu einer generationengerechteren Gesellschaft wäre und damit Kinder das Selbstvertrauen entwickeln, eine eigene Meinung zu bilden. Ich finde das Argument, Jugendliche hätten noch nicht genügend Reife, nicht richtig. Das Wissen der Jugendlichen über Politik ist durch die Schule oft besser als bei Erwachsenen und wir diskutieren ja auch nicht darüber, ob man mit 90 noch genügend Sachverstand hat. Holger, wird das in dieser Legislatur kommen, was meinst du? Holger Hofmann: Bei der Wahl zum Europäischen Parlament ist eine Wahlaltersabsenkung wahrscheinlich. So steht es im Regierungsprogramm und die Regierung kann das auch allein durchsetzen. Für die Bundestagswahl braucht es auch die Opposition, da hierzu das Grundgesetz geändert werden muss, und da bin ich sehr skeptisch. Yola Fanroth: Gibt es für dich andere Dinge, die sich positiv hinsichtlich der politischen Mitbestimmung von Kindern und Jugendlichen entwickeln? Holger Hofmann: In den letzten Jahrzehnten hat sich eine breite Beteiligungslandschaft in den Kommunen herausgebildet, die durch landesweite Servicestellen in vielen Bundesländern und durch gesetzliche Grundlagen in einigen Bundesländern gestützt wird. Auch die Jugendstrategie der Bundesregierung, die seit einigen Jahren gibt, trägt Früchte. Beispielsweise gibt es jetzt auch in anderen Ministerien außerhalb des Jugendressorts Maßnahmen und Programme zur Jugendbeteiligung. Einen Jugendklimafonds oder jugendgerechte Informationen des Justizministeriums. Das Programm „Demokratie leben!“ des Jugendministeriums stärkt viele Projekte vor Ort, die helfen, Diskriminierung zu überwinden und dabei auf eine aktive Einbeziehung der Kinder und Jugendlichem setzen. Yola Fanroth: Wo meinst du stehen wir in zehn Jahren, auf was kommt es besonders an? Holger Hofmann: Die Beteiligungsangebote auf der kommunalen Ebene müssen bekannter gemacht werden, sie müssen breiter aufgestellt und besser vernetzt sein. Zudem sollten sie eine größere Wirkung nach oben, also in die Landes- und Bundesebene entfalten. Wir müssen dabei nicht unbedingt neue Instrumente erfinden. Wir haben Jugendbeauftragte, auch Vertrauenslehrer_innen an Schulen, wir haben Kinder- und Jugendparlamente oder Jugendverbände. Viele Kinder und Jugendliche kennen diese Angebote aber gar nicht, geschweige denn weiß der Vertrauenslehrer, an wen er eine Schülerin verweisen soll, die sich über ihren gefährlichen Schulweg beschwert oder die die Basketball-AG am Nachmittag nicht nutzen kann, weil kein Bus mehr fährt. Es gibt Kommunen mit einer breiten Palette an Angeboten, in anderen Kommunen gibt es dagegen gar keine Angebote. Dieser Flickenteppich ist nicht tragbar. Auf der Landesebene dünnt es sich noch mehr aus. Die Arbeit deines Landesverbands, Vincent, sticht bundesweit heraus. Durch die massiven Einschränkungen, denen Kinder und Jugendliche durch Corona unterworfen waren, hat sich die Situation an vielen Stellen noch einmal verschärft, beispielsweise auch in der verbandlichen Kinder- und Jugendarbeit. Auch Interessensvertretung durch Erwachsene in Form von Landeskinderbeauftragten gibt es nur in vier Bundesländern, mit teils problematischer Anbindung in den politischen Strukturen, durch die sie kaum Wirkung entfalten können. Auch die Kinderkommission auf Bundesebene ist aufgrund ihrer Ausstattung, Zusammensetzung und Rechte nur ein schwaches Schwert. Da hat der Wehrbeauftragte der Bundesregierung einen viel, viel größeren Arbeitsstab und der Datenschutzbeauftragte mehr Rechte. Vincent Sipeer: Wie sieht bei der gesetzlichen Verankerung von Mitbestimmungsrechten aus? Holger Hofmann: In den letzten drei Jahren haben wir eine positive Entwicklung in Baden-Württemberg, Brandenburg und Hessen gesehen. Dort gibt es konkrete Festlegungen zur Kinder- und Jugendbeteiligung in den Landesverfassungen. Das kann aber nicht über die vielen Leerstellen in anderen Landesverfassungen und dem Grundgesetz hinwegtäuschen. Der letzte Vorschlag zu einer Verfassungsänderung, wie er noch von der großen Koalition auf den Weg gebracht wurde, hat offenbart, dass viele Politikerinnen und Politiker den Jugendlichen vielleicht noch zugestehen, dass sie ihre Meinung sagen dürfen, aber dass es keine verbindlichen Regelungen braucht, wie diese berücksichtigt werden muss. Mit einer Umsetzung entlang dieses Vorschlages wäre im Übrigen nicht nur nichts gewonnen, sondern sie würde in einer Krise, wie wir sie derzeit mit der Covid-19-Pandemie erleben, keine Wirkung entfalten. Die Interessen von Kindern und Jugendlichen würden weiter unter den Tisch fallen, aber viel schlimmer noch, sie würde einen Stand der politischen Mitbestimmung zementieren, den viele Kommunen und manche Bundesländer schon überwunden haben.	0	PMC9734473	NO-CC CODE	2022-12-14 23:28:28	no	Sozial Extra. 2022 Dec 6;:1-5	utf-8	null	null	null	oa_other
==== Front Behav Anal Pract Behav Anal Pract Behavior Analysis in Practice 1998-1929 2196-8934 Springer International Publishing Cham 759 10.1007/s40617-022-00759-9 Editorial Leading the Charge: A Look Inside the Behavior Analysis in Practice Emergency Series of Publications on Systemic Racism and Police Brutality Gingles Denisha [email protected] 1 Watson-Thompson Jomella 2 Anderson-Carpenter Kaston D. 3 Tarbox Jonathan 4 Peterson Stephanie M. [email protected] 5 1 Signature Behavioral Health, Windsor Mill, MD USA 2 grid.266515.3 0000 0001 2106 0692 Applied Behavioral Science, University of Kansas, Lawrence, KS USA 3 grid.17088.36 0000 0001 2150 1785 Department of Psychology, Michigan State University, East Lansing, MI USA 4 grid.42505.36 0000 0001 2156 6853 FirstSteps for Kids, University of Southern California, Los Angeles, CA USA 5 grid.268187.2 0000 0001 0672 1122 Western Michigan University, Kalamazoo, MI USA 5 12 2022 12 2022 15 4 10151022 © 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. This article introduces the “Behavior Analysis in Practice Emergency Series of Publications on Systemic Racism and Police Brutality.” After the murder of George Floyd, the behavior analytic community was charged to respond in the spirit of Dr. Martin Luther King’s challenge to social scientists. The charge of Dr. King was to explain real life phenomena negatively affecting the Black community. This series covered a wide range of topics with the intent of creating solutions that may be used to address remnants of the overarching impact of systemic racism and anti-Blackness. In this editorial, we provide an overview of the major themes of the accepted articles, some personal accounts of the editorial team, context for the special issue, discuss the contributions of the included articles, and a discussion of the areas in need of further work. Keywords Police brutality Systemic racism Racism Behavior analysis issue-copyright-statement© Association for Behavior Analysis International 2022 ==== Body pmcThe field of behavior analysis is currently experiencing a cultural shift that mirrors the larger cultural shift in American society, resulting from the murders of George Floyd and other unarmed Black Americans at the hands of police, and the resulting rise of the #BlackLivesMatter movement. Minneapolis police officers murdered George Floyd on May 25, 2020, and, along with much of U.S. society, the applied behavior analysis (ABA) community demanded immediate action. The Behavior Analysis in Practice editorial team agreed at the time that institutions with power in behavior analysis, especially peer-reviewed journals, needed to do something now. Inaction was unacceptable and even the sometimes years-long traditional peer review process in ABA journals was unacceptable. After being challenged by Black behaviorists to respond to the movement in real-time, the then editor of Behavior Analysis in Practice, Jonathan Tarbox, called upon Denisha Gingles to serve as guest editor and lead a project that would capture the current sociopolitical movement in the scientific literature. Upon her acceptance, the need to embolden fellow Black behaviorists to action was evidenced by seeking out mentors to assist first-time authors with their submissions, recruiting Black behaviorists who were already leading social justice efforts in their local environments, and recruiting a leadership team that reflected those intimately affected by systemic racism and police brutality. It is with this intention that Drs. Jomella Watson-Thompson and Kaston Anderson-Carpenter were asked to join the team as guest associate editors. The acceptance of this role, to the best knowledge of the authors’ knowledge, was the first of its kind. After seeking information from tenured behaviorists, there were no previous records of any Black editor (guest or full), nor was there any confirmation of an all-Black editorial team such as this one. Each editor that made history with their participation in this series, possesses a diverse skill set within the publication process, research on diversity, and grassroots social activism on local, national, and international levels. The team worked to develop a rubric for evaluation that the reviewers could use to ensure the articles were consistent with the call for articles and the mission of the BAP. The call for articles was then released for the series on June 5, 2020, charging authors to provide practitioners in the field of behavior analysis immediately actionable resources that could be used toward creating a more socially just future for the field and beyond. In this introduction to the series, we summarize the broad themes touched by the articles, provide context, discuss the contributions of the articles, and discuss areas in need of further work. The Heart of the Series Before outlining the major themes of the series, it is paramount to address the underlying motivation behind the series and capture the sentiment of each Black editor who chose to respond to the social justice movement on behalf of our science, while also navigating their own reactions to the brutal reminders of historical trauma during this time. The call for Black behaviorists to not only show up for the field but also continue their daily routines, while undergoing imagery of Black pain, is a perspective that will allow readers to tact the significance of this issue decades after this movement has dissipated. Denisha Gingles, Guest Editor The decision to accept the role of guest editor was not an easy one. The deaths of George Floyd, Breonna Taylor, and Elijah McClain reignited intense, yet familiar feelings of the past that I surely was uninterested in reviving. During the 2013 and 2014 movements surrounding Micheal Brown’s, Eric Garner’s, and Trayvon Martin’s deaths, my life drastically changed. Like many Black people during this time, I felt fear for those who looked like me and was scared for the future of my community. This fear compelled me to activate. Although this was not the first time I was compelled to activate, it was the first time I moved to action through protest and organizing. Despite knowing the violence our ancestors experienced during the civil rights movement of the 1960s was possible during this new movement, sleepless nights, and unpaid labor, we tirelessly showed up for a chance at making an impact. The decision to activate for the human rights of your community is not simply a decision of goodwill but a necessary action in line with Skinner’s work on cultural selection. Activism is not only a means to respond to injustice, it is a method of survival. Although the root of my work has been fighting in conjunction with those of intentionally marginalized communities, my professional work prior to this series was not in academia. The editor’s decision to bring a grassroots organizer to the academic space required a level of trust in local organizers to lead in other sectors beyond politics and community engagement and an even greater understanding of the role lived experience plays in academia. Simply because one has not matriculated through the traditional academic system does not make them any less scholarly or qualified, and I hope my participation reminds others of this as well. Acceptance of this role also required a level of trust from myself, as an activist, to believe that, with the help of other leaders in academia, we could reflect the issues of systemic racism and police brutality in a truthful way that did not diminish the perspectives of the Black community and those affected by the American sociopolitical and criminal justice system without the academic publishing system attempting to water down the content. We received many submissions. Some submissions were from people whose work I had previously consumed and thought of as well-respected experts. There were many people whose heart was in the right place but unfortunately got it wrong in terms of historical inaccuracies and understanding the true impact on the Black community. And although I wished we could have accepted most articles, I had to be a responsible steward of the community I represent. I had to make sure, as much as possible, that I was protecting the Black community through academia; a space that has historically violated Black people—a space that taught Black people we were less intelligent, more violent, and overarchingly a place where we were made to think we did not belong. So even although we are discussing the issue of police brutality I had to keep the dignity of our community through this text. This series reveals activism in real time. I was able to use my skills employed when activating social change. I canvassed my community, sought out Black researchers to serve as mentors, reached out to individuals to submit articles, and sought out Black reviewers and associate editors to be part of the team. My work in this unpaid role was intentional. I can only hope I did it justice, but if nothing else I know that I did it with all of my heart in pursuit of justice for not only the criminal justice system, but the academic pipeline that still has much more work to do to address the historical failings of marginalized communities. Jomella Watson-Thompson, Guest Associate Editor As a guest associate editor for this emergency series, my personal and professional experiences, perspectives, and identities were intertwined, which required me to examine my own intersectionality in this role as a Black woman who is also an academician and scholar. Through my professional work, I identify as a behavioral-community psychologist who is also a community-engaged scholar. As a community-engaged scholar, I am committed to positively affecting the communities in which I work, reside, and identify, including as a Black woman. Further, as a behavioral-community psychologist, it is important to contribute to advancing the application of our science to support social change and action. Prior to serving on the guest editorial team, each of us had contributed to addressing social justice issues, including disparities experienced by Black people and other marginalized groups. Through my prior work, I have contributed to addressing disparities experienced by Blacks, particularly in the areas of youth and community violence. In addition, I have examined the systemic and structural determinants or factors that contribute to disparities across systems that have resulted in the many gross injustices experienced by the Black community. Therefore, when provided the opportunity by Denisha Gingles to contribute to the emergency publication series with the aim to advance social justice, including in our field, it was a welcomed invitation to serve in this capacity. I am honored to have served along with Denisha Gingles and Dr. Kaston Anderson-Carpenter as one of the few (if not only) predominantly Black editorial teams for a publication outlet in our field. It should be acknowledged that for those who have never had to ponder if you are the first or only in a space, including in areas of our field, then you are coming from a place of privilege. The demonstrated actions of the white editors of Behavior Analysis in Practice to occasion the opportunity for the Black guest editorial team to respond and lead this special emergency series may serve as an exemplar for advancing systems changes in the areas in which one may have influence. The guest editorial team collaborated for over 2 years in supporting the publication of the emergency special series, which may be a testament to not only our collective commitment but, most important, the prioritized importance of this work. The special emergency series began during the heightened period in which many were experiencing an increased awareness of the injustices and inequities experienced by Black people related to both police brutality coupled with the unparalleled deaths due to COVID-19. When invited to serve as a guest associate editor, it was during a period when I was also seeking to influence systems and practices in areas I could immediately contribute to addressing inequities experienced by people of color, including in our discipline. Like many, I continue to be exhausted by the disproportionate suffering experienced by the Black community, which takes a toll on one's personal, academic, and professional stamina. Yet, I understand the importance of agency. Through the special emergency series, we endeavor to provide an outlet to promote voice and facilitate change in addressing inequities, including in and across our systems that function as a microcosm of our broader society. Kaston Anderson Carpenter, Guest Associate Editor When I received the invitation to serve as a guest associate editor for this emergency series, I knew I had to accept it. That being said, the decision was not easy. The deaths of George Floyd, Sandra Bland, Philando Castile, Alton Sterling, Ahmaud Arbery, and countless other Black people at the hands of law enforcement or racist actions took a physical and mental toll on my well-being. I had to balance that trauma, along with my own as a Black man in America, with the charge that lay before me. I had to come to a place of “understanding the assignment.” Calling upon my ancestors gave me the courage, understanding, and clarity I needed at a personal level to serve effectively as a guest associate editor for this series. Maferefun egun y aṣẹ! A fundamental ethos of my career is to honor and uplift the voices who have been on the margins of society. To that end, I work exclusively with historically marginalized, disempowered, and oppressed communities. I identify as a community-engaged behavioral psychologist, and I believe firmly that applied behavioral science is for everyone. This belief is a constant reminder to me that all of us—regardless of the letters we have (or do not have), our academic pedigree, or any status-related characteristic—have the capacity to occasion positive change. In my career, I have conducted research and engaged in community advocacy in numerous areas, such as underage drinking in rural communities, harm reduction among men who have sex with men and transgender women, behavioral and mental health among Arab Americans, behavioral and psychosocial impacts of COVID-19, and pre-exposure prophylaxis (PrEP) uptake among kink-identified communities. In addition, I have worked in several capacities at local, state, and national levels to support social justice and racial equity for Black people across the African Diaspora. These experiences, despite the horrors of racism on full display in our current society, have sustained my hope not only in the positive impact of ABA but also in humanity. It has been an honor and privilege to serve on the editorial team with Denisha Gingles and Dr. Jomella Watson-Thompson. It is rare to have a Black editorial team for a behavior-analytic journal (or any scientific journal), and we were given a substantial charge to keep. We have supported each other as we dealt with racial battle fatigue and trauma, and we celebrated one another throughout this process. This special series is a product of scholars, implementers, and advocates whose cutting-edge work in addressing police brutality via behavioral principles will, it is hoped, prompt a larger conversation in the field regarding the promise and need for applied behavioral science to address racism and police brutality effectively. The Mind of the Series The journal received 72 requests to submit after the call for articles was published. Of those requests, 63 submissions were received for the special issue, with 31 receiving a final decision of acceptance, thereby yielding an acceptance rate of 50.07% for the series. The guest associate editors served as action editors for submissions, and they assembled a group of 14 peer reviewers to assist authors with finalizing their submissions. The peer reviewers predominantly represented at least one marginalized racial group, however, the most represented racial group in this series was Black/African American. Several editorial choices were made for the series that are still perhaps unconventional for most behavior analytic journals. For example, authors were encouraged, but not required, to include author notes that stated their personal backgrounds and perspectives. Although this may be viewed as “unprofessional” by some academicians, it is a nod to the radical behavioral philosophical foundation of our field that states all behavior, including the behavior of the scientist, is to be included in our science (Skinner, 1945). As we know, it is fallacy, and even mentalistic, to pretend that the scientist’s own background and perspectives are not part of the overall process of scientific writing. To assume that would be to pretend that a scientist is not an organism whose behavior is determined by their history and current circumstances. In particular, considering that one of the major topics addressed in the series is bias, we believed that allowing authors to acknowledge their own potential biases could function as modeling for readers. In addition, it has become common practice for other journals to include acknowledgments of author backgrounds (e.g., Does et al., 2018; Mindlis et al., 2020). The race and ethnicity of the authors, as well as the reviewers, are critical aspects of intersectionality, which is important to both acknowledge and understand. The framework of intersectionality and multidimensionality originated from the work of Dr. Kimberle` Crenshaw (1989) to better understand the experiences and marginalization of Blacks, particularly Black women. The seventh edition of the Publication Manual of the American Psychological Association (APA) offers considerations for acknowledging intersectionality in writing. APA (2020) recommends:When authors write about personal characteristics they should be sensitive to intersectionality—that is, to the way in which individuals are shaped by and identify with a vast array of cultural, structural, sociobiological, economic, and social contexts (Howard & Renfrow, 2014). Intersectionality is a paradigm that addresses the multiple dimensions of identity and social systems as they intersect with one another and relate to inequality. . . . Thus, individuals are located within a range of social groups whose structural inequalities can result in marginalized identities. . . . To address intersectionality in a paper, identify individuals’ relevant characteristics and group memberships (e.g., ability and/or disability status, age, gender, gender identity, generation, historical as well as ongoing experiences of marginalization, immigrant status, language, national origin, race and/or ethnic status, among other variables), and describe how their characteristics and group membership intersect in ways that are relevant to the study. (pp. 148–149) As editors, authors, and reviewers, it was not only critical but necessary to model and demonstrate our intersectionality as contributors shaping this special issue. Through the special issue, it was important to the editorial team to ensure we were modeling and even contributing to shifting systems in the field, including through our own editorial processes. An additional editorial decision that may be viewed as unconventional by some is the intentional use of the word “murder” in many of the manuscripts to refer to the actions of law enforcement in interacting with George Floyd and other unarmed Black people. Some may be concerned about the legal definitions of words such as “murder,” “killing,” “manslaughter,” and so on, while noting that the use of such words in less formal settings, such as social media, may sometimes become highly politicized and detached from conventional standards. Although we acknowledge these potential concerns, we chose to allow authors to use the word “murder” because it is consistent with the use of the word in the broader anti-racism movement. Lastly, another unconventional approach to the series was the decision to decapitalize “white” when contextualizing race and culture. In 2020, the Columbia Journalism Review and Associated Press made the journalistic decision to capitalize “Black” (Daniszewski, 2020; Laws, 2020). Shortly after, they also decided to decapitalize “white.” This stylistic, yet humanistic decision was based on international consultation with other journalists. It was stated the capitalization of “white” perpetuates white supremacy, which in the new emerging social justice movement, is antithetical. This decision also may be viewed as deprioritizing the power of this racial class. Although whiteness is not without power or privilege in a societal context, this symbolic gesture gives credence to those who have literally been decapitalized in society. Although there is certainly a reason to capitalize “white,” such as a journalistic reminder of the power of whiteness, we chose to follow the advice of the aforementioned publications. With the stark contrast of a capitalized “B” and lowercase “w” in press, it has the potential of sparking behavioral responding in the reader. With our overarching goal of producing publications that promote action, the decapitalization of “white” was in line with this goal. These decisions also reflect the intent and purpose of this special series. This series is not intended as a neutral forum for discussions of racism and police brutality. This series was conceived of and executed as an explicitly anti-racist scholarly effort, which demanded using clear and unflinching language, particularly in reference to the murder of unarmed Black people by law enforcement. Readers who are uncomfortable with this stance are encouraged to remember that social change is rarely comfortable and that the position taken by this special series is not the position of the journal more broadly, the association which publishes it (the Association of Behavior Analysis, International), nor of the publishing company that manages it (Springer). The following sections describe and contextualize the articles included in this special issue. We would like to draw one thing to the reader’s attention. Due to the emergency nature of these submissions, articles were handled outside the typical editorial management software. Thus, as articles were accepted, they were not always appropriately tagged as belonging to the special issue. Thus, some of the articles that were intended to be in this special issue have already appeared in previous issues of Behavior Analysis in Practice. Those articles are cited, with full citations in the reference section so the reader can locate them. The articles appearing in this issue do not contain full citations, as they are found later in this issue. All of the articles that were intended to be contained in this special issue will be contained in a special collection, which can be found on the Behavior Analysis in Practice website. Practical Tools for Taking Action Several articles in the special issue were written with the primary intent of providing practical tools that behavior analysts can put into practice immediately to fight racism across the various settings in which we work. For example, Mathur and Rodriguez propose a curriculum for training practicing behavior analysts in cultural responsiveness. The authors reviewed critical race theory and suggested guiding principles for a cross-disciplinary curriculum. Melendez et al. (2021) provided practical guidelines for talking to children with autism about systemic racism. This is the first publication, of which we are aware, that has attempted to address the topic of teaching children with ASD about racism, and it was based on the assumption that if Black children are old enough to be the victims of systemic racism on a daily basis, then all children, including children with ASD, were old enough to learn about it. The authors gave examples using evidenced based practices to actively teach the concept of racism as opposed to solely using didactic methods. Colic et al. further raise the challenge experienced by Black caregivers of children with autism spectrum disorder (ASD) in navigating the health-care system, particularly in regard to obtaining diagnosis and services. The authors provided a practice guideline that is both culturally responsive and context-specific for working with Black caregivers. Baires et al. offer practical skills that can be used to support intercultural communication through effective listening using behavior-analytic methods for examining verbal exchanges. The authors describe how the function of the listener’s behavior is examined in mediating reinforcement for the speaker. Behavior analysts are challenged to develop repertoires of effective listening as a target of behavior change in efforts to address racism and reduce discriminatory practices. Li (2021) discussed the role that non-black people of color in behavior analysis can play in helping our field fight anti-Black racism. Li specifically outlines the ways people of color may advertently or inadvertently perpetuate systemic racism against Black people. The author also discussed the methods people of color could use to combat anti-Black racism within their own cultures. This article provided a necessary commentary on intercultural violence. Gingles provides a framework based on acceptance and commitment training (ACT) for Black behavior analysts to assess and manage their own behavior toward fighting anti-Black racism. Although providing literature best suited to address internalized racism, it is also applicable to those who do not identify as Black. This article has several activities using the six principles of ACT. It also takes a historical viewpoint, paying homage to Black psychology and researchers, while using concepts and traditions that are essential to Black culture. Also addressing the topic of self-managing one’s own social justice-oriented behavior, the article by Machalicek et al. addresses the all-too-common problem of declining activism after the acute period of a social movement has passed. The article provides practical guidelines for how we all can ensure that we continue to engage in overt behaviors oriented toward racial justice well beyond this social justice movement. Najdowski et al. (2021) provided a practical set of guidelines for behavior analysts to put into place for helping graduate programs in behavior analysis take overt anti-racist action. With graduate programs having an immense influence on the career of budding behavior analysts, the ways in which they plan for and respond to racism are critical. The authors took a multilevel approach to address inequity in graduate programs by identifying areas for change in organizational systems and leadership, curriculum, research, and engagement with faculty, students, and staff. Esquierdo-Leal and Houmanfar (2021) challenged us to consider the critical need and related skills necessary to support effective leadership, particularly when facilitating social change. The authors discuss the responsibility of leaders to understand cultural factors contributing to oppression and to be accountable to commitments to address systematic oppression across environments, including in the workplace and other organizational settings. Critiquing the Current Status Quo Several articles in the special issue turn their scope inward, toward analyzing and critiquing the current status of the field of applied behavior analysis, with respect to racial justice. For example, the article by Pritchett et al. is a review and searing critique of applied behavior analytic research, from the perspectives of colonial versus participatory research practices. In the behavior analytic community, turning a blind eye to the evident colonial practices in our field is convenient at best. This article outlines this convenient dissonance and provides a way forward to engage communities in a socially humane way. In their review of applied behavior analytic research, Lovelace et al. reflect on how little research that addresses the needs and perspectives of multiply-marginalized populations (in this case Black autistic girls) has been published. Without a doubt, the lack of understanding of not only the presentation of autism for Black girls, but the differences of experience rooted in intersectionality theory, serves as a disadvantage for the behavior analytic community. The authors provided a detailed literature review to expound on what currently exists and what is yet to be critically studied. Morris and Hollins provide an uncomfortable but critically important comparison between the practices used within the field of developmental disabilities and those used by law enforcement. The authors challenge practitioners to consider the ways in which we may unintentionally cause unnecessary and dangerous behavioral escalation through how we respond to client challenging behavior. Levy et al. discuss the importance of supporting cultural humility and adopting anti-racist practices within the context of systemic racism, including in behavior analysis. Actionable steps to demonstrate cultural humility and anti-racist practices are presented to support a more inclusive and representative field as well as to provide effective delivery of services to those who are Black, Indigenous, People of Color (BIPOC). Feasible practices and actions that may support anti-racist practices for professional organizations/governing bodies, behavior analytic organizations, and individual practitioners are recommended. Sylvain et al. examined the responses to police brutality against Black people by white behavior analytic professionals. In this study, a survey was conducted with Black board certified behavior analysts regarding their experiences following recent police brutality events. The impact of performative allyship by members of the white behavior analysis community was also discussed and the field was challenged to support antiracism by facilitating contingencies that advance equity. Behavioral Conceptual Analyses Practical guidelines for addressing systemic racism are critically important for empowering change immediately; however, it is also important for any science to remain conceptually systematic with the principles upon which it is based (Baer et al., 1968). Several articles in this special issue have attempted subtle and sophisticated behavioral conceptual analyses of the complex behavioral repertoires involved in systemic racism and/or police brutality. For example, Belisle et al. elaborated on a complex, multi-tiered, nested model of racism, that evaluates the contingencies that maintain racism from the level of the individual up to the level of the society. De Sousa et al. examined how behavior analysis can assist in understanding the behavioral variables and processes that support acquiring and maintaining the behaviors of racial aggressors. Hugh-Pennie et al. discuss culturally relevant pedagogy (CRP) as a framework for behavior analysts working in schools to support students in gaining skills in sociopolitical awareness, cultural competence, and academic excellence. The authors propose that the CRP framework is complementary to applied behavior analysis and behavior analysts integrating this framework while working in schools may further reduce the effects of racism experienced by some students of color. Recommendations are provided for how behavior analysts can support the conditions for implementing CRP in schools. Jaramillo and Nohelty discuss the importance of examining implicit bias using behavioral terms and provide actionable steps to support change. These authors suggest that the work supported in other disciplines to study racial implicit bias should inform the field of ABA. Based on the existing literature, recommendations are provided for extending current assessment methods of implicit bias to address racial implicit bias for clinicians in the field. The authors challenge readers to ensure that long-term behavior change is supported when intervening to address implicit bias by targeting measurable behaviors and encouraging self-monitoring by clinicians. There are challenges within and across systems, including in the field of behavior analysis, that perpetuate racism. Rose et al. (2022) operationalized and examined racist behaviors within the context of verbal behavior. Some practical guidance and actionable steps are offered for how to discuss racism, including the selection of terms used and framing that focuses attention on the environmental variables contributing to racist behavior. Behavioral Analysis and Intervention across Systems Multiple articles in the special issue examine how behavior-analytic interventions could contribute to addressing racism across systems, including the educational, health-care, and justice systems, as well as in the field of behavior analysis. The importance of behavior analysts addressing racism across individual, organizational, and cultural levels is also discussed in multiple articles included in the special issue. Some of these focus on systems changes that could be supported in the criminal justice or educational systems. For instance, Ghezzi et al. consider how acceptance and commitment training (ACT) and the prosocial model could be expanded for use with law enforcement agencies. Catagnus et al. discuss how emotions are necessary to consider within behavior analysis and anti-racism work. Parks and Kirby discuss the results of inaction to documented racism in policies and procedures, including with law enforcement, which has reinforced systemic racism and racial disparities. They provide a behavior analytic examination of the history and context of policing that has perpetuated both individual and institutional racism. Carvalho et al. provide a systematic review of empirical articles published across 5 years to examine the relationship between racial prejudice and police stops. They concluded that Black men were more frequently stopped by the police in not only the United States but also some other countries, suggesting that institutional racism in the police force is a rampant problem requiring intervention to reduce individual and collective bias, particularly towards Black men. Machado and Lugo also examine behavior analytic strategies to reduce racial bias in the police force. They conducted a systematic review of use-of-force practices by police and mitigation strategies, such as the use of body-worn cameras (BWC) and implicit bias (IB) training. They found that the effectiveness of these strategies was mixed. Behavior analysis could contribute to the effectiveness of the implementation of these strategies. Machado and Lugo challenge the field to contribute more in the area of use-of-force practices and to inform the implementation of strategies (BWC, IB) used to reduce oppressive policing. In addition, a few of the articles consider how a behavior-analytic approach could inform interventions within and across disciplines, such as in implementing restorative practices. Leland and Stockwell offer a restorative justice approach to behavior change, particularly for historically oppressive systems, including the criminal justice system. They discuss the importance of using the least restrictive procedures and suggest restorative justice aligns with the ethical code for behavior analysis. Likewise, Pavlacic et al. also suggest that restorative justice may support a set of procedures for peacemaking that reduces recidivism and considers reparation for harm. It is suggested that restorative justice is compatible with behavior-analytic principles and behavior science approaches. Restorative practices are appropriate for use by behavior analysts, particularly those working in the education and criminal justice systems. The racial disparities and inequities in school disciplinary practices were raised in multiple articles in the emergency series as contributors to the school-to-prison pipeline. Sevon raised the issue of anti-Black racism in schools and challenged the field of behavior analysis to consider how it can contribute to addressing oppressive systems in education, including the overuse of expulsion and suspension for Black students. Sevon and other authors also raised the challenge of implicit and racial bias as contributing to anti-Black racism. They examined anti-Black racism and student discipline in the schools, within the context of contributing to the school-to-prison pipeline. Likewise, Henry et al. conducted a survey with staff and behavior analytic professionals working in schools. Henry et al. noted there are several behavior analytic interventions that are alternatives to exclusionary discipline practices, including zero-tolerance policies, which contribute to the school-to-prison pipeline. Crowe and Drew (2021) further discussed the challenges experienced by disabled individuals in the schools and justice system. In particular, these authors examined the disparities in incarceration for individuals with differences, including those with intellectual and developmental disabilities. Discussion There has been a clarion call to action issued in the emergency publication series on police brutality and systemic racism. In the call for articles, we requested manuscript submissions that were practical and offered guidance for addressing police brutality and other topics related to systemic racism. Based on the call for articles, the goals of the emergency series were to (1) uplift the voices of BIPOC and other racial minorities, and (2) provide immediately actionable guidance on meaningful steps all citizens can take to address systemic racism, in ourselves, our communities, and beyond. The goals of the emergency series were supported in advancing the work by (1) establishing a guest editorial team of all Black professionals in the field; (2) soliciting articles from a diverse group of authors who submitted and published manuscripts in response to the call; and (3) relying on dedicated service of reviewers who helped shape the manuscripts into the published articles. The overwhelming number of submissions for this special issue was indicative of the gravitas of people across society, including in behavior analysis, interested in contributing to solutions for current, socially significant problems related to recent public demonstrations of police brutality and systemic racism. Indeed, it is an ethical issue if we do not use the best of what we know and can offer through our science to address applied problems of social significance. Yet, as noted in the majority of publications in this special issue, we have a mountain of opportunity ahead of us to climb if we are to address systems of oppression and inequities within the field of behavior analysis. Facing and climbing this mountain, however, will undoubtedly better position us to offer our science to other disciplines and systems in an effort to combat systemic racism. Unfortunately, police brutality was one of the latest and more recently publicized events that happened to be captured on video, which resulted in a deeper awareness of the inequities and systems of oppression experienced by Black people. The time is now for us to act and change the underlying conditions that perpetuate anti-Blackness and racism, including in our systems of influence as a field. Each of us is challenged to consider and accept part of the clarion call to do more and better, which first begins with examining our own behaviors and opportunities for contribution at the individual, organizational, and systems levels to support change. As the editors, reviewers, and authors of the special issue, we invite you to do more than read the content of the series, but, most important, commit and act in ways that are accountable to supporting anti-racism as outlined in the articles. Otherwise, we have been performative, at most, as a field of individual and collective actors, in which case this special issue was done a disservice. The authors thank all contributing peer reviewers. Their dedication to this project was invaluable and helped advance the goals of the series. In alphabetical order, many thanks to the following: Shahla Ala'i-Rosales, Michael Amlung, Marlesha Bell, Carolyn A. Brayko, Kelly Harrison, Jovonnie Esquierdo-Leal, Vincent Francisco, Charlie Greenwood, Nicole Hollins, Bertilde Kamana, Robin Kuhn, Worner Leland, Temple Lovelace, Natalie Parks, Denise E Ross, Mawule Sevon, Shameka McCammon, Malika Pritchett, Alison Szarko, Melody Sylvain, Erica Taylor, Janani Vaidya, and Robin Williams Declarations Ethical Approval This article does not contain any studies with human participants 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 American Psychological Association Publication manual of the American Psychological Association 2020 7 Baer DM Wolf MM Risley TR Some current dimensions of applied behavior analysis Journal of Applied Behavior Analysis 1968 1 1 91 97 10.1901/jaba.1968.1-91 16795165 Crenshaw K Demarginalizing the intersection of race and sex: A black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics University of Chicago Legal Forum 1989 1 8 139 167 10.4324/9781315582924-10 Crowe B Drew C Orange is the new asylum: Incarceration of individuals with disabilities Behavior Analysis in Practice 2021 14 2 387 395 10.1007/s40617-020-00533-9 33643545 Daniszewski, J. (2020). Why we will lowercase white. AP Definitive Source. https://blog.ap.org/announcements/why-we-will-lowercase-white. Accessed 29 Sept 2022 Does S Ellemers N Dovidio JF Norman JB Mentovich A van der Lee R Goff PA Implications of research staff demographics for psychological science American Psychologist 2018 73 5 639 650 10.1037/amp0000199 29494171 Esquierdo-Leal JL Houmanfar RA Creating inclusive and equitable cultural practices by linking leadership to systemic change Behavior Analysis in Practice 2021 14 2 499 512 10.1007/s40617-020-00519-7 33613858 Howard JA Renfrow DG McLeod JD Lawler EJ Schwalbe M Intersectionality Handbook of the social psychology of inequality 2014 Springer 95 121 Laws, M. (2020). Why we capitalize “black” (and not “white”). Columbia Journalism Review. https://www.cjr.org/analysis/capital-b-black-styleguide.php. Accessed 29 Sept 2022 Li A Solidarity: The role of non-Black people of color in promoting racial equity Behavior Analysis in Practice 2021 14 2 549 553 10.1007/s40617-020-00498-9 33101603 Melendez JL Tan IMC Lau JC Leung J Practical resources for talking to children with autism about systemic racism Behavior Analysis in Practice 2021 14 2 451 461 10.1007/s40617-020-00500-4 34150458 Mindlis I Livert D Federman AD Wisnivesky JP Revenson TA Racial/ethnic concordance between patients and researchers as a predictor of study attrition Social Science & Medicine 2020 255 Article 113009 10.1016/j.socscimed.2020.113009 32371270 Najdowski AC Gharapetian L Jewett V Toward the development of antiracist and multicultural graduate training programs in behavior analysis Behavior Analysis in Practice 2021 14 2 462 477 10.1007/s40617-020-00504-0 34150459 Rose JC MacManus C MacDonald J Parry-Cruwys D Mitigating Racial inequity by addressing racism in the criminal justice system: A behavior analytic approach Behavior Analysis in Practice 2022 15 635 641 10.1007/s40617-021-00670-9 35692523 Skinner BF The operational analysis of psychological terms Psychological Review 1945 52 5 270 277 10.1037/h0062535	0	PMC9734474	NO-CC CODE	2022-12-14 23:45:33	no	Behav Anal Pract. 2022 Dec 5; 15(4):1015-1022	utf-8	Behav Anal Pract	2,022	10.1007/s40617-022-00759-9	oa_other
==== Front J Immigr Minor Health J Immigr Minor Health Journal of Immigrant and Minority Health 1557-1912 1557-1920 Springer US New York 36472715 1433 10.1007/s10903-022-01433-6 Brief Communication Online Grocery Shopping Behaviors and Attitudes Among Asian Americans Rummo Pasquale E. [email protected] 1 Ali Shahmir H. 2 Kranick Julie 1 Thorpe Lorna E. 1 Yi Stella S. 1 1 grid.137628.9 0000 0004 1936 8753 Department of Population Health, NYU Grossman School of Medicine, 10016 New York, NY USA 2 grid.137628.9 0000 0004 1936 8753 Department of Social and Behavioral Sciences, New York University School of Global Public Health, New York, NY USA 6 12 2022 19 9 11 2022 © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, 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. How online grocery shopping behaviors differ among Asian American (AA) ethnic subgroups and acculturation level is unknown. From June 9–15, 2020, we administered an online survey to a nationally-derived nonprobability sample of 2,895 AA adults, including 1,737 East, 570 South, and 587 Southeast Asian adults, assessing online grocery shopping (yes/no, frequency, reasons). We used logistic regression to compare responses by subgroup and acculturation score, controlling for sociodemographics. Thirty-percent of participants reported shopping online for groceries in a typical month, with a higher percentage among South (45%) versus East Asian adults (23%). Participants with low (vs. high) acculturation scores were more likely to report a lack of special foods (OR = 0.7; 95% CI: 0.5–0.98) and poor food quality (OR = 0.6; 95% CI: 0.4–0.7) as preventing them from shopping online. Online grocery shopping has the capacity to address inequities in health, potentially via culturally-tailored programs designed for less-acculturated AA adults. Keywords Asian Americans Health Equity Acculturation Online Grocery Cultural Influence http://dx.doi.org/10.13039/100006545 National Institute on Minority Health and Health Disparities U54MD000538 ==== Body pmcIntroduction There is considerable heterogeneity in diet behaviors among Asian American (AA) ethnic subgroups. For example, saturated fat intake is lower and sodium intake is higher among Chinese, Korean, and Japanese adults, relative to non-Hispanic White adults in the U.S.; whereas, intake of saturated fats and refined carbohydrates is higher among Asian Indian adults 1]. Prior research also suggests that acculturation is an important influence on diet among AA adults, such as higher intake of nutrient-poor, energy-dense foods and lower consumption of fruits and vegetables consumption with longer duration in the U.S. [2]. However, the extent to which food purchasing behaviors differ between AA ethnic subgroups or those of different acculturation levels remains unclear. This research is critical for addressing issues related to food purchasing behaviors via intervention (e.g., offering healthy, culturally-specific grocery items in online retail settings). The COVID-19 pandemic and other recent events have fueled an increase in incidents of discrimination among AA adults [3], which may contribute to anxieties about leaving home to shop for groceries and spur AA consumers to shop online. Previous surveys suggest that internet use is higher among English-speaking AA adults relative to other race/ethnic groups [4], but there is a dearth of disaggregated data about online food purchasing behaviors and attitudes from AA adults. In our previous work, we explored changes in diet and food shopping behaviors among AA adults due to COVID-19 [5], but we did not explore online shopping practices in general or differences by acculturation score. For example, online retailers may be an important source for culturally-specific (e.g., produce indigenous to Asia) or culturally-preferred foods (e.g., specific brands of seasonings) among AA adults, but this may differ across the spectrum of acculturation. To address these gaps, we sought to examine differences in online grocery shopping behaviors among AA adults using data disaggregated by AA ethnic subgroup and levels of acculturation. Methods We administered an online survey to a nationally-derived nonprobability sample of 3,084 AA adults from June 9–15, 2020 using Dynata, an online surveying company that recruits volunteer research participants [6]. The sample was recruited to approximately match the distribution of gender and age of Asian adults residing in the U.S. [7] Potential participants completed an online consent, followed by a brief pre-screening questionnaire. Eligibility criteria included identifying as Asian, being age 18 years and older, and being able to read and speak English. Open REDCap, an online survey platform, was used to create and distribute the survey. The survey was designed to assess sociodemographics, health status, diet behaviors, and food shopping behaviors, using questions described elsewhere.[5] A 10-item version of questions adapted from the Marin Short Acculturation Scale was also included in the survey, which yields a total acculturation score ranging from 10 to 50 [8]. All procedures were approved as exempt by the Institutional Review Board. Duplicate responses (n = 37) and implausible skip patterns (n = 29) and participants who identified as more than one of the three ethnic subgroups (defined by country of origin) (n = 124) were excluded. The final sample included 1,737 East Asian, 570 South Asian, and 587 Southeast Asian (n = 2,894) adults. The median completion time for the survey was 17.4 minutes (IQR: 12.1, 25.3). We used logistic regression to compare differences in online grocery shopping (yes/no) by Asian ethnic subgroup (East Asian as referent) and acculturation score (dichotomized using median score). We also examined responses to questions regarding participants’ primary grocery store (type, reasons); shopping for Asian grocery items (yes/no, store type, reasons); shopping for online groceries (yes/no, frequency, location, type of groceries, reasons, intentions), including Asian grocery items; and whether there were other ways they were getting food at home that they were not using before COVID-19 (none, online orders, neighbors, family, food delivery business, food from my child’s school to eat at home). We controlled for age, gender, household income (<$20,000 vs. ≥$20,000), educational attainment (high school or less vs. more than high school), and household size (1, 2, 3, ≥ 4). We used a two-sided alpha of 0.05 as the threshold for statistical significance. Stata version 15.1 (StataCorp LP, College Station, TX) was used for all analyses. Results In our sample, the average age was 43.2 (SD = 16.5) years, 60% reported having at least one foreign-born parent, 39.8% reported being born outside of the U.S., 87.3% reported annual household income above $20,000, and 78.9% of respondents reported having a post-secondary education degree (Table 1). Approximately 29.2% of participants reported shopping online for groceries in a typical month, and 36.4% of those adults shop online for groceries at least once a week (Table 2). About 6% of participants reported shopping online for groceries at iFresh or another type of Asian online retailer (Table 3). A higher percentage of South Asian adults (45.4%) reported shopping online for groceries in a typical month compared to Southeast (30.3%) and East Asian adults (23.4%). Southeast (vs. East) Asian adults had lower odds of buying fresh fruits and vegetables (OR = 0.7; 95% CI: 0.5, 0.97) and higher odds of buying sugar-sweetened beverages (OR = 1.5; 95% CI: 1.02, 2.3) online. Table 1 Sociodemographic characteristics of a nationally-derived nonprobability sample of 2,894 Asian American adults, overall and by Asian American subgroup (n = 2,984) Total East Asian South Asian Southeast Asian N or mean % or SD N or mean % or SD N or mean % or SD N or mean % or SD Total 2894 - 1737 57.6% 570 18.9% 587 19.4% Gender Male 1355 46.8% 820 47.2% 282 49.5% 253 43.1% Female 1536 53.1% 914 52.6% 288 50.5% 334 56.9% Other/Refused/Missing 3 0.1% 3 0.2% 0 0.0% 0 0.0% Age group (years) 18–24 437 15.1% 212 12.2% 110 19.3% 115 19.6% 25–34 604 20.9% 285 16.4% 170 29.8% 149 25.4% 35–44 565 19.5% 305 17.6% 128 22.5% 132 22.5% 45–54 498 17.2% 334 19.2% 66 11.6% 98 16.7% 55–64 367 12.7% 272 15.7% 45 7.9% 50 8.5% 65–99 406 14.0% 323 18.6% 45 7.9% 38 6.5% Refused/Missing 17 0.6% 6 0.3% 6 1.1% 5 0.9% Country of origin, at least one parent United States 1148 39.7% 777 44.7% 186 32.6% 185 31.5% Outside of the United States 1746 60.3% 960 55.3% 384 67.4% 402 68.5% Other/Don’t know/Refused/Missing 0 0.0% 0 0.0% 0 0.0% 0 0.0% Country of origin, self United States 1628 56.3% 1101 63.4% 222 38.9% 305 52.0% Outside of the United States 1152 39.8% 592 34.1% 301 52.8% 259 44.1% Other/Don’t know/Refused/Missing 114 3.9% 44 2.5% 47 8.2% 23 3.9% Educational attainment Less than 9th grade 10 0.3% 5 0.3% 1 0.2% 4 0.7% 9th to 12th grade - No diploma 54 1.8% 26 1.5% 10 1.6% 18 3.0% GED or equivalent 176 5.9% 85 4.8% 32 5.2% 59 9.7% Some college, no degree 364 12.6% 203 11.7% 68 11.9% 93 15.8% Associate’s degree 213 7.4% 129 7.4% 33 5.8% 51 8.7% Bachelor’s degree 1203 41.6% 738 42.5% 220 38.6% 245 41.7% Graduate or Professional degree 866 29.9% 546 31.4% 205 36.0% 115 19.6% Don’t know/Refused/Missing 8 0.3% 5 0.3% 1 0.2% 2 0.3% Household size 1 540 18.7% 392 22.6% 71 12.5% 77 13.1% 2 883 30.5% 592 34.1% 137 24.0% 154 26.2% 3 594 20.5% 343 19.7% 136 23.9% 115 19.6% 4 599 20.7% 299 17.2% 156 27.4% 144 24.5% 5 177 6.1% 76 4.4% 33 5.8% 68 11.6% > 5 96 3.3% 32 1.8% 36 6.3% 28 4.8% Don’t know/Refused/Missing 5 0.2% 3 0.2% 1 0.2% 1 0.2% Household income a Mean (SD) $111,679 $103,685 $116,458 $102,697 $114,621 $112,761 $94,824 $91,022 Refused/Missing/Implausible 510 17.6% 303 17.4% 132 23.2% 151 25.7% Less than $20,000 179 6.2% 86 5.0% 46 8.1% 47 8.0% $20,000 or more 2527 87.3% 1549 89.2% 481 84.4% 497 84.7% Don’t know/Refused/Missing/Implausible 188 6.5% 102 5.9% 43 7.5% 43 7.3% Relationship status Married 1500 51.8% 879 50.6% 340 59.6% 281 47.9% Widowed 40 1.4% 25 1.4% 11 1.9% 4 0.7% Divorced 138 4.8% 103 5.9% 18 3.2% 17 2.9% Separated 33 1.1% 18 1.0% 11 1.9% 4 0.7% Never Married 1018 35.2% 613 35.3% 170 29.8% 235 40.0% Living with Partner 127 4.4% 75 4.3% 15 2.6% 37 6.3% Don’t know/Refused/Missing 38 2.7% 24 2.8% 5 2.2% 9 2.9% Occupational status Working at a job or business 1539 53.2% 946 54.5% 295 51.8% 298 50.8% With a job or business but not at work 153 5.3% 96 5.5% 29 5.1% 28 4.8% Looking for work 258 8.9% 121 7.0% 70 12.3% 67 11.4% Not working at a job or business 615 21.3% 431 24.8% 86 15.1% 98 16.7% Part-time or full-time student 278 9.6% 127 7.3% 70 12.3% 81 13.8% Don’t know/Refused/Missing 51 1.8% 16 0.9% 20 3.5% 15 2.6% Do you or anyone in your household currently get SNAP? Yes 218 7.5% 100 5.8% 61 10.7% 57 9.7% No 2529 87.4% 1578 90.8% 460 80.7% 491 83.6% Don’t know/Refused 147 5.1% 59 3.4% 49 8.6% 39 6.6% Acculturation score (10–50) 36.0 8.0 37.0 7.8 33.2 8.1 35.6 7.8 SNAP = Supplemental Nutrition Assistance Program; East Asian = Chinese/Cantonese, Hmong/Mong, Iwo Jiman, Japanese, Korean, Laohmong, Nipponese, Okinawan, or Taiwanese; South Asian = Asian Indian, Bangladeshi, Bengalese, Bharat, Bhutanese, Burmese, Dravidian, East Indian, Goanese, Maldivian, Nepalese, Pakistani, or Sri Lankan; Southeast Asian = Cambodian, Filipino, Indochinese,Indonesian, Laotian, Madagascar/Malagasy,Malaysian, Siamese, Singaporean, Thai, or Vietnamese aExcludes <$100.0 and >$1000000.5 Table 2 Grocery shopping behaviors and attitudes (%) of a nationally-derived nonprobability sample of 2,894 Asian American adults, overall and by Asian American subgroup (n = 2,984)a Total East Asian South Asian Southeast Asian N or mean % or SD N or mean % or SD N or mean % or SD N or mean % or SD Total 2894 - 1737 60.0% 570 19.7% 587 20.3% Primary store type Supermarket 1733 59.9% 1128 64.9% 266 46.7%* 339 57.8% Online retailer or store 113 3.9% 48 2.8% 51 8.9%* 14 2.4% Small or ethnic grocery store 163 5.6% 85 4.9% 48 8.4% 30 5.1% Convenience store 86 3.0% 29 1.7% 35 6.1%** 22 3.7% Discount or big box store like Target or Walmart 241 8.3% 105 6.0% 56 9.8%* 80 13.6%* Wholesale club like BJs, Costco, or Sam’s Club 509 17.6% 311 17.9% 104 18.2% 94 16.0% Other/Don’t know/Refused 49 1.7% 31 1.8% 10 1.8% 8 1.4% Do you shop for online groceries in a typical month? Yes 844 29.2% 407 23.4% 259 45.4%* 178 30.3%* No 1978 68.3% 1299 74.8% 290 50.9%*** 389 66.3%* Don’t know/Refused/Missing 72 2.5% 31 1.8% 21 3.7% 20 3.4% Do you shop online for Asian grocery items? Yes 392 13.5% 174 10.0% 128 22.5%*** 90 15.3%* No 2428 83.9% 1531 88.1% 415 72.8%*** 482 82.1% Don’t know/Refused 74 2.6% 32 1.8% 27 4.7% 15 2.6% Online grocery shopping, frequency At least once a week 307 36.4% 141 34.6% 113 43.6% 53 29.8% Every other week or less often 524 62.1% 260 63.9% 144 55.6% 120 67.4% Don’t know/Refused/Missing 13 1.5% 6 1.5% 2 0.8% 5 2.8% Online groceries, types of groceries purchased b Fresh produce 493 58.4% 247 60.7% 165 63.7% 81 45.5%* Canned produce 328 38.9% 159 39.1% 91 35.1% 78 43.8% Frozen produce 316 37.4% 151 37.1% 99 38.2% 66 37.1% Dairy products 391 46.3% 186 45.7% 128 49.4% 77 43.3% Soda or other sweetened drinks 234 27.7% 102 25.1% 69 26.6% 63 35.4%* Bottled water 277 32.8% 138 33.9% 77 29.7% 62 34.8% Coffee or tea (unsweetened) 319 37.8% 161 39.6% 82 31.7% 76 42.7% Hot or cold cereals 252 29.9% 121 29.7% 75 29.0% 56 31.5% Bread, rice, or other types of grains 388 46.0% 204 50.1% 112 43.2% 72 40.4% Beans, lentils, or pulses 242 28.7% 96 23.6% 106 40.9%* 40 22.5% Sauces or condiments 269 31.9% 134 32.9% 74 28.6% 61 34.3% Fresh meat, poultry, or fish 288 34.1% 158 38.8% 65 25.1% 65 36.5% Frozen meat, poultry, or fish 240 28.4% 128 31.4% 64 24.7% 48 27.0% Other frozen food 165 19.5% 87 21.4% 48 18.5% 30 16.9% Desserts, snacks, or candy 320 37.9% 159 39.1% 88 34.0% 73 41.0% Other/Don’t know/Refused 51 6.0% 25 6.1% 12 4.6% 14 7.9% Factors motivating participant to shop online for groceries in a typical month c Low prices 353 41.8% 153 37.6% 113 43.6% 87 48.9%* Variety of goods 339 40.2% 156 38.3% 109 42.1% 74 41.6% Variety of special foods 263 31.2% 110 27.0% 90 34.7% 63 35.4% Good quality food 304 36.0% 122 30.0% 118 45.6%*** 64 36.0% Online convenience 485 57.5% 240 59.0% 141 54.4% 104 58.4% Other language options 50 5.9% 15 3.7% 23 8.9% 12 6.7% Other/Don’t know/Refused 47 5.6% 32 7.9% 10 3.9% 5 2.8% Factors preventing participant from online grocery shopping in a typical month d High prices 609 30.8% 391 30.1% 97 33.4%* 121 31.1% Lack of variety of goods 323 16.3% 197 15.2% 63 21.7% 63 16.2% Lack of variety of special foods 331 16.7% 191 14.7% 58 20.0% 82 21.1% Poor quality food 233 11.8% 134 10.3% 51 17.6% 48 12.3% Lack of social interaction 139 7.0% 74 5.7% 29 10.0% 36 9.3% Lack of interaction with food itself 775 39.2% 531 40.9% 98 33.8%* 146 37.5% No loyalty/frequent shopping program 149 7.5% 92 7.1% 28 9.7% 29 7.5% Extra delivery fee for frozen/fresh foods 682 34.5% 443 34.1% 90 31.0%* 149 38.3% Other language options 21 1.1% 10 0.8% 7 2.4% 4 1.0% Other/Don’t know/Refused 448 22.6% 325 25.0% 49 16.9% 74 19.0% Factors that would motivate participant to shop online for groceries d Free shipping 1217 61.5% 804 61.9% 157 54.1%* 256 65.8% Lower prices 1078 54.5% 717 55.2% 140 48.3%* 221 56.8% Accepts EBT 81 4.1% 39 3.0% 18 6.2% 24 6.2% Greater variety of goods 652 33.0% 412 31.7% 109 37.6%** 131 33.7% Greater variety of special foods 349 17.6% 215 16.6% 58 20.0%* 76 19.5% Higher quality food 686 34.7% 442 34.0% 114 39.3%** 130 33.4%* Other language options 39 2.0% 21 1.6% 10 3.4% 8 2.1% Other/Don’t know/Refused 467 23.6% 338 26.0% 59 20.3% 70 18.0% *P < 0.05; P < 0.01; P < 0.001 aWe used logistic regression to compare differences in survey responses by Asian ethnic subgroup (East Asian as referent group) and acculturation score (dichotomized using median score), controlling for age, gender, household income (<$20,000 vs. ≥$20,000), educational attainment (High School or less vs. more than High school), and household size (1, 2, 3, ≥ 4). We used a two-sided alpha of 0.05 as the threshold for statistical significance. bParticipants could select more than one response, so the denominator reflects unique non-missing responses. cAmong participants who reported shopping online for groceries. dAmong participants who did not reported shopping online for groceries Table 3 Grocery shopping behaviors and attitudes (%) of a nationally-derived nonprobability sample of 2,894 Asian American adults, overall and by Asian American subgroup Total East Asian South Asian Southeast Asian N or mean % or SD N or mean % or SD N or mean % or SD N or mean % or SD Total 2894 - 1737 60.0% 570 19.7% 587 20.3% Are you the main person responsible for food shopping in your household? Yes 1957 67.6% 1197 68.9% 379 66.5% 381 64.9% No 478 16.5% 285 16.4% 77 13.5% 116 19.8% No one person is responsible 411 14.2% 239 13.8% 91 16.0% 81 13.8% Don’t know/Refused 48 1.7% 16 0.9% 23 4.0% 9 1.5% 81.8% Are you the main person responsible for food preparation in your household? Yes 1708 59.0% 1062 61.1% 316 55.4% 330 56.2% No 778 26.9% 451 26.0% 161 28.2% 166 28.3% No one person is responsible 365 12.6% 208 12.0% 77 13.5% 80 13.6% Don’t know/Refused 43 1.5% 16 0.9% 16 2.8% 11 1.9% 71.6% Transportation mode to primary store? Drive own car 2316 80.0% 1443 83.1% 393 68.9% 480 81.8% Use someone else’s car 54 1.9% 20 1.2% 15 2.6% 19 3.2% Someone else drives me or I use a ridesharing app 84 2.9% 37 2.1% 22 3.9% 25 4.3% Walk 205 7.1% 134 7.7% 44 7.7% 27 4.6% Bus 57 2.0% 31 1.8% 15 2.6% 11 1.9% Metro 27 0.9% 9 0.5% 15 2.6% 3 0.5% Taxi 11 0.4% 2 0.1% 7 1.2% 2 0.3% Ride bicycle 8 0.3% 5 0.3% 1 0.2% 2 0.3% Other/Don’t know/Refused/Missing 132 4.6% 56 3.2% 58 10.2% 18 3.1% Minutes to get to primary store Between 0 and 15 min 1691 58.4% 1083 62.3% 267 46.8% 341 58.1% Between 15–30 min 724 25.0% 431 24.8% 143 25.0% 150 25.6% Between 30–45 min 177 6.1% 93 5.3% 50 8.7% 34 5.8% Between 45 min − 1 h 97 3.4% 43 2.5% 30 5.3% 24 4.1% Between 1–2 h 55 1.9% 22 1.3% 19 3.3% 14 2.4% Between 2–3 h 14 0.5% 7 0.4% 4 0.7% 3 0.5% More than 3 h 7 0.2% 4 0.2% 2 0.4% 1 0.2% Don’t Know/Refused/Missing 129 4.5% 54 3.1% 55 9.6% 20 3.4% Miles from home to primary store Less than 0.5 miles 237 8.2% 144 8.3% 46 8.1% 47 8.0% Between 0.5–1 miles 423 14.6% 259 14.9% 74 13.0% 90 15.3% Between 1–2 miles 517 17.9% 319 18.4% 98 17.2% 100 17.0% Between 2–3 miles 447 15.4% 284 16.4% 84 14.7% 79 13.5% Between 3–4 miles 352 12.2% 209 12.0% 67 11.8% 76 12.9% Between 5–10 miles 540 18.7% 321 18.5% 114 20.0% 105 17.9% Between 10–15 miles 114 3.9% 62 3.6% 21 3.7% 31 5.3% Between 15–20 miles 53 1.8% 33 1.9% 6 1.1% 14 2.4% More than 20 miles 27 0.9% 15 0.9% 4 0.7% 8 1.4% Don’t know/Refused/Missing 184 6.4% 91 5.2% 56 9.8% 37 6.3% Reasons for shopping at primary store? a Low Prices 1247 43.1% 720 41.5% 239 41.9% 288 49.1% Produce Selection 1083 37.4% 670 38.6% 213 37.4% 200 34.1% Meat Department 660 22.8% 395 22.7% 95 16.7% 170 29.0% Variety of foods 1298 44.9% 786 45.3% 241 42.3% 271 46.2% Variety of special foods 622 21.5% 337 19.4% 148 26.0% 137 23.3% Close to home 1475 51.0% 915 52.7% 239 41.9% 321 54.7% Loyalty/Frequent Shopper Program 578 20.0% 344 19.8% 100 17.5% 134 22.8% Online convenience 188 6.5% 75 4.3% 75 13.2% 38 6.5% Other/Don’t know/Refused/Missing 81 2.8% 48 2.8% 17 3.0% 16 2.7% Do you shop for Asian grocery items? Yes 2316 80.0% 1393 80.2% 446 78.2% 477 81.3% No 508 17.6% 314 18.1% 101 17.7% 93 15.8% Don’t know/Refused 70 2.4% 30 1.7% 23 4.0% 17 2.9% Primary store type for Asian grocery items a Supermarket 610 26.3% 419 30.1% 62 13.9% 129 27.0% Online retailer or store 25 1.1% 9 0.6% 13 2.9% 3 0.6% Convenience store 37 1.6% 8 0.6% 15 3.4% 14 2.9% Discount or big box store like Target or Walmart 19 0.8% 8 0.6% 7 1.6% 4 0.8% Wholesale club like BJs, Costco, or Sam’s Club 29 1.3% 18 1.3% 6 1.3% 5 1.0% Small or ethnic grocery store 747 32.3% 421 30.2% 165 37.0% 161 33.8% Other/Don’t know/Refused 1427 61.6% 854 61.3% 302 67.7% 271 56.8% Primary store for Asian grocery items, location a My primary grocery store/market 814 35.1% 486 34.9% 169 37.9% 159 33.3% Other grocery store/market 1484 64.1% 898 64.5% 274 61.4% 312 65.4% Don’t know/Refused 596 25.7% 353 25.3% 127 28.5% 116 24.3% Primary store for Asian grocery items, frequency a 1 time last month 564 24.4% 346 24.8% 107 24.0% 111 23.3% 2–3 times last month 553 23.9% 326 23.4% 104 23.3% 123 25.8% 1 time per week 288 12.4% 174 12.5% 53 11.9% 61 12.8% 2 times per week 51 2.2% 28 2.0% 11 2.5% 12 2.5% 3–4 times per week 18 0.8% 10 0.7% 2 0.4% 6 1.3% 5–6 times per week 8 0.3% 7 0.5% 0 0.0% 1 0.2% Don’t know/Refused/Missing 1412 61.0% 846 60.7% 293 65.7% 273 57.2% Minutes to get to primary store for Asian grocery items? a Between 0 and 15 min 524 22.6% 324 23.3% 80 17.9% 120 25.2% Between 15–30 min 599 25.9% 364 26.1% 117 26.2% 118 24.7% Between 30–45 min 219 9.5% 134 9.6% 40 9.0% 45 9.4% Between 45 min − 1 h 74 3.2% 39 2.8% 18 4.0% 17 3.6% Between 1–2 h 40 1.7% 23 1.7% 7 1.6% 10 2.1% Between 2–3 h 4 0.2% 2 0.1% 1 0.2% 1 0.2% More than 3 h 1 0.0% 1 0.1% 0 0.0% 0 0.0% Don’t know/Refused 1433 61.9% 850 61.0% 307 68.8% 276 57.9% Miles from home to primary store for Asian grocery items? a Less than 0.5 miles 63 2.7% 37 2.7% 9 2.0% 17 3.6% Between 0.5–1 miles 115 5.0% 66 4.7% 23 5.2% 26 5.5% Between 1–2 miles 189 8.2% 111 8.0% 38 8.5% 40 8.4% Between 2–3 miles 171 7.4% 99 7.1% 32 7.2% 40 8.4% Between 3–4 miles 157 6.8% 97 7.0% 33 7.4% 27 5.7% Between 5–10 miles 362 15.6% 228 16.4% 63 14.1% 71 14.9% Between 10–15 miles 164 7.1% 99 7.1% 30 6.7% 35 7.3% Between 15–20 miles 96 4.1% 61 4.4% 14 3.1% 21 4.4% More than 20 miles 109 4.7% 67 4.8% 17 3.8% 25 5.2% Don’t know/Refused 1468 63.4% 872 62.6% 311 69.7% 285 59.7% Reasons for purchasing Asian grocery items from primary store? a It has the best prices 437 18.9% 247 17.7% 90 20.2% 100 21.0% It has the best food quality 459 19.8% 262 18.8% 107 24.0% 90 18.9% It is clean 260 11.2% 133 9.5% 71 15.9% 56 11.7% It is easy to find the items and brands that I like/need 740 32.0% 436 31.3% 148 33.2% 156 32.7% It carries items and brands that I like to buy 1061 45.8% 655 47.0% 186 41.7% 220 46.1% It is near or on the way to a place I frequently go to 327 14.1% 181 13.0% 75 16.8% 71 14.9% It is convenient, for example, it saves me time or offers delivery services 208 9.0% 103 7.4% 52 11.7% 53 11.1% It is not crowded 172 7.4% 84 6.0% 37 8.3% 51 10.7% The staff are friendly or I know them 148 6.4% 59 4.2% 49 11.0% 40 8.4% It has good service 164 7.1% 79 5.7% 51 11.4% 34 7.1% Another reason 29 1.3% 14 1.0% 6 1.3% 9 1.9% Don’t know/Refused 19 0.8% 12 0.9% 2 0.4% 5 1.0% Online grocery shopping, retailer type b Amazon Prime Pantry 510 60.4% 232 57.0% 170 65.6% 108 60.7% Peapod 89 10.5% 32 7.9% 38 14.7% 19 10.7% Fresh Direct 113 13.4% 46 11.3% 42 16.2% 25 14.0% iFresh 38 4.5% 15 3.7% 13 5.0% 10 5.6% Other 261 30.9% 148 36.4% 62 23.9% 51 28.7% Online grocery shopping, language used on website b English 821 97.3% 394 96.8% 251 96.9% 176 98.9% Language other than English 18 2.1% 12 2.9% 5 1.9% 1 0.6% Don’t know/Refused 5 0.6% 1 0.2% 3 1.2% 1 0.6% Online groceries, mostly delivery or pick-up b Home 527 62.4% 264 64.9% 164 63.3% 99 55.6% Physical store location 298 35.3% 135 33.2% 89 34.4% 74 41.6% Don’t know/Refused/Missing 19 2.3% 8 2.0% 6 2.3% 5 2.8% Order groceries online for other people who do not live in your household? b Yes 268 31.8% 106 26.0% 99 38.2% 63 35.4% No 549 65.0% 290 71.3% 151 58.3% 108 60.7% Don’t know/Refused/Missing 27 3.2% 11 2.7% 9 3.5% 7 3.9% Online groceries, how likely over next 6 months? Very likely 490 16.9% 242 13.9% 144 25.3% 104 17.7% Somewhat likely 607 21.0% 298 17.2% 178 31.2% 131 22.3% Neither likely nor unlikely 597 20.6% 347 20.0% 111 19.5% 139 23.7% Somewhat unlikely 432 14.9% 293 16.9% 58 10.2% 81 13.8% Very unlikely 768 26.5% 557 32.1% 79 13.9% 132 22.5% Other ways you are getting food at home that you were not using before COVID-19? None 146 25.3% 83 27.6% 39 24.4% 24 20.9% Online orders 160 27.8% 81 26.9% 49 30.6% 30 26.1% Neighbors 47 8.2% 19 6.3% 20 12.5% 8 7.0% Family 69 12.0% 27 9.0% 26 16.3% 16 13.9% Food delivery business 72 12.5% 39 13.0% 15 9.4% 18 15.7% Food from my child’s school to eat at home 20 3.5% 10 3.3% 6 3.8% 4 3.5% Other/Refused 24 4.2% 11 3.7% 8 5.0% 5 4.3% aAmong participants who reported shopping for Asian grocery items. bAmong participants who reported shopping online for groceries. cAmong participants who reported that they or anyone from their household had not gone to the store for food in the last week. Among participants who did not shop online for groceries, South Asian adults had lower odds of reporting high prices (OR = 0.7; 95% CI: 0.6, 0.9) and extra delivery fees (OR = 0.6; 95% CI: 0.4, 0.7) as preventing them from starting; and lower odds of reporting free shipping (OR = 0.5; 95% CI: 0.4, 0.6) and lower prices (OR = 0.5; 95% CI: 0.4, 0.6) as what would motivate them to start shopping online for groceries (Table 3). The pattern was opposite for the variety and quality of goods online, with a higher prevalence of South Asian adults reporting them as potential motivating factors. We also observed higher odds of reporting shopping online for groceries among those who reported getting food via online orders (OR = 4.8; 95% CI: 3.0, 7.8) and food delivery businesses (OR = 3.0; 95% CI: 1.7, 5.2) for the first time since COVID-19. The acculturation score of the overall sample was moderate [36.0 (SD = 8.0)]. Compared to East Asian adults [37.0 (SD = 7.8)], we observed a lower acculturation score among Southeast [35.6 (SD = 7.8)] and South [33.2 (SD = 8.1)] Asian adults (Table 1). Participants with a high (vs. low) acculturation score had higher odds of reporting shopping online for groceries due to low prices (OR = 1.6; 95% CI: 1.1, 2.3) and lower odds of reporting shopping online for groceries due to the variety of special foods, such as Asian grocery items (OR = 0.7; 95% CI: 0.5, 0.98) (Table 3). Participants with a high acculturation score also had lower odds of reporting poor quality foods as preventing them from shopping online for groceries (OR = 0.6; 95% CI: 0.4, 0.7). Participants with a high acculturation score were not less likely to buy groceries on the internet, but they had lower odds of reporting shopping online for Asian groceries specifically (OR = 0.7; 95% CI: 0.5, 0.9). Discussion We found that almost 30% of AA adults shop online for groceries in a typical month, which is slightly lower than the 39% of participants in the Nielsen National Consumer Panel who reported ever shopping online for groceries in July 2020 [9]. Our results also suggest that less-acculturated AA adults are more likely to shop for Asian grocery items online, potentially due to the availability of special foods. South Asian adults were the least acculturated subgroup in our sample, for example, and they were also more likely than other subgroups to purchase Asian grocery items online. It is possible that less-acculturated South Asian adults may be able to find culturally-preferred foods (e.g., spices) in popular online retail websites, whereas, less-acculturated East and Southeast Asian adults may be able to find preferred items in their primary brick-and-mortar stores (e.g., soy sauce, fish sauce). The variety and quality of goods were more likely to motivate South Asian adults to start shopping online for groceries; whereas, Southeast Asian adults, who reported lower income levels, were more likely to shop online due to low prices. Southeast Asian adults were also more likely to report buying sugar-sweetened beverages online, and less likely to report buying fresh fruits and vegetables online. Taken as a whole, these findings suggest that initiatives and changes designed to promote healthy food purchases among AA adults should be tailored to the food shopping considerations of ethnic subgroups. These findings echo our recent report examining changes in diet and food shopping behaviors among AA adults due to COVID-19, wherein a higher percentage of Southeast Asian adults reported not having sufficient financial resources to safely acquire an adequate supply of food compared to other AA adults [5]. Our work has important limitations. Because we recruited participants from an online panel, responses may be more generalizable to adults who are proficient at using the internet. Respondents also had higher levels of income and education than the total AA population in the U.S., so results are less generalizable to those with lower food budgets. However, we observed wide variation in generational status and acculturation score in our sample, and we had a large sample size overall. We also collected data disaggregated by country of origin, which allowed us to compare responses by ethnic subgroup. Given the capacity for online grocery shopping to address racial/ethnic inequities in health, including mitigating disparities in food access and protecting individuals during emergencies [10], our work highlights which issues related to online shopping could be addressed by interventions and for whom these changes would have the greatest impact. For example, offering high-quality and culturally-specific grocery items may increase the use of online grocery shopping among less-acculturated AA adults, similar to other culturally-tailored programs. Our findings also strengthen the argument that disaggregating responses by racial/ethnic subgroup is an essential next step in survey research. Future researchers may want to explore differences in the availability of culturally-preferred food items and brands in online versus brick-and-mortar stores, and how the availability of Asian-specific online retailers influences online food purchasing behaviors. Acknowledgements This project was supported by the NYU Center for the Study of Asian American Health under the NIH National Institute on Minority Health & Health Disparities grant award # U54MD000538-15. We would like to thank Stella Chong, Lily Divino, Mary Joy Garcia, Alka Kanaya, Simona Kwon, Stephanie Liu, Binh Lu, Deborah Min, Rhea Naik, Chorong Park, MD Taher, Sameer Talegawkar, Kosuke Tamura, Tracy Vo, and Jennifer Wong for their feedback on our survey questions related to acculturation, diet, and grocery shopping. Authors Contribution PR, LT, and SY conceived of the study design. PR analyzed and interpreted the data. SA and JK were involved in the literature search and development of survey questions. All authors were involved in writing the paper and had final approval of the submitted and published versions. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Van Horn L Carson JAS Appel LJ Recommended dietary pattern to achieve adherence to the American Heart Association/American College of Cardiology (AHA/ACC) guidelines: a scientific statement from the American Heart Association Circulation 2016 134 22 e505-e29 27789558 2. Satia-Abouta J Patterson RE Neuhouser ML Dietary acculturation: applications to nutrition research and dietetics J Am Diet Assoc 2002 102 8 1105 18 10.1016/S0002-8223(02)90247-6 12171455 3. Ruiz NG, Horowitz JM, Tamir C. Many Black. Asian Americans Say They Have Experienced Discrimination Amid Coronavirus. https://www.pewsocialtrends.org/2020/07/01/many-black-and-asian-americans-say-they-have-experienced-discrimination-amid-the-covid-19-outbreak/. Pew Research Center. Accessed March, 2021. 4. Perrin A. English-speaking Asian Americans stand out for their technology use. https://www.pewresearch.org/fact-tank/2016/02/18/english-speaking-asian-americans-stand-out-for-their-technology-use/. Pew Research Center. 2016. Access March, 2021. 2016. 5. Rummo PE, Naik R, Thorpe LE, et al. Changes in diet and food shopping behaviors among asian-american adults due to COVID-19. Obes Sci Pract. 2021. 6. Dynata. https://www.dynata.com. Accessed April, 2020. 7. U.S. Census Bureau. (2020). (Sex by Age (Asian Alone), 2014–2018 American Community Survey 5-year estimates.) https://data.census.gov/cedsci/table?d=ACS%205-Year%20Estimates%20Detailed%20Tables&tid=ACSDT5Y2018.B01001D. Accessed August, 2020. 8. Marin G Sabogal F Marin BV Development of a short Acculturation Scale for Hispanics Hispanic J Behav Sci 1987 9 2 183 205 10.1177/07399863870092005 9. Duffy EW, Lo A, Hall MG, et al. Prevalence and demographic correlates of online grocery shopping: results from a nationally representative survey during the COVID-19 pandemic. Public Health Nutr. 2022:1–7. 10. Rummo PE Bragg MA Yi SS Supporting Equitable Food Access during National Emergencies—The Promise of Online Grocery Shopping and Food Delivery Services JAMA Health Forum 2020 1 3 e200365-e 10.1001/jamahealthforum.2020.0365 36218602	36472715	PMC9734475	NO-CC CODE	2022-12-14 23:28:28	no	J Immigr Minor Health. 2022 Dec 6;:1-9	utf-8	J Immigr Minor Health	2,022	10.1007/s10903-022-01433-6	oa_other
==== Front J Gen Intern Med J Gen Intern Med Journal of General Internal Medicine 0884-8734 1525-1497 Springer International Publishing Cham 36474004 7891 10.1007/s11606-022-07891-w Original Research Buprenorphine Treatment Episodes During the First Year of COVID: a Retrospective Examination of Treatment Initiation and Retention http://orcid.org/0000-0003-1544-458X Stein Bradley D. MD, PhD [email protected] 1 Landis Rachel K. MPP 23 Sheng Flora MBBS, MPH 3 Saloner Brendan PhD 4 Gordon Adam J. MD, MPH 5 Sorbero Mark MS 1 Dick Andrew W. PhD 6 1 grid.34474.30 0000 0004 0370 7685 RAND Corporation, 4570 Fifth Avenue, Suite 600, Pittsburgh, PA 15213 USA 2 grid.253615.6 0000 0004 1936 9510 George Washington University Trachtenberg School of Public Policy, Washington, DC USA 3 grid.34474.30 0000 0004 0370 7685 RAND Corporation, Arlington, VA USA 4 grid.21107.35 0000 0001 2171 9311 Department of Health Policy and Management, Johns Hopkins University, Baltimore, MD USA 5 grid.280807.5 0000 0000 9555 3716 VA Salt Lake City Health Care System and University of Utah School of Medicine, Salt Lake City, UT USA 6 grid.34474.30 0000 0004 0370 7685 RAND Corporation, Boston, MA USA 6 12 2022 15 14 4 2022 26 10 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. Background During the COVID pandemic, overall buprenorphine treatment appeared to remain relatively stable, despite some studies suggesting a decrease in patients starting buprenorphine. There is a paucity of empirical information regarding patterns of buprenorphine treatment during the pandemic. Objective To better understand the patterns of buprenorphine episodes during the pandemic and how those patterns compared to pre-pandemic patterns. Design Pharmacy claims representing approximately 92% of all prescriptions filled at retail pharmacies in all 50 US states and the District of Columbia. Participants Individuals filling buprenorphine prescriptions indicated for treatment of opioid use disorder. Main Measures The number of active, starting, and ending buprenorphine treatment episodes March 13 to December 1, 2020, and the expected number of such episodes in 2020 based on the growth in treatment episodes from March 13 to December 1, 2019. Key Results The observed number of active buprenorphine episodes in December 2020 was comparable to the expected number, but new treatment episodes starting between March 13 and December 1, 2020, were 17.2% fewer than expected based on the 2019 experience. Similarly, the number of episodes that ended between March 13 and December 1, 2020, was 16.0% fewer than expected. Decreases from expected episode starts and ends occurred throughout the period but were greatest in the 2 months after the declaration of the public health emergency. Conclusions and Relevance Beneath the apparent stability of buprenorphine patient numbers during the pandemic, the flow of individuals receiving buprenorphine treatment changed substantially. Our findings shed light on how policy changes meant to support buprenorphine prescribing influenced prescribing dynamics during that period, suggesting that while policy efforts may have been successful in maintaining existing patients in treatment, that success did not extend to individuals not yet in treatment. KEY WORDS COVID buprenorphine opioid use disorder ==== Body pmcIntroduction The isolation, anxiety, and psychological stress wrought by the COVID-19 pandemic resulted in greater rates of drug use and misuse,1–3 mental health and substance use disorders,1, 4 emergency medical system calls, emergency department visits for opioid overdose,5, 6 and more fatal overdoses in 2020 than in any prior year.7 There were also substantial disruptions in the delivery of outpatient care in the USA: outpatient medical visits declined more than 50% in the months following the declaration of the public health emergency on March 13, 2020,8–10 and there have also been concerns about disruptions in treatment among those receiving medication treatment for opioid use disorder (MOUD).11–15 Studies of buprenorphine utilization during the pandemic suggest that overall buprenorphine utilization during the pandemic did not follow the patterns seen for other outpatient care, and the number of filled buprenorphine prescriptions and the number of individuals filling prescriptions were relatively unchanged from 2019 numbers.16–20 However, although the total number of individuals receiving buprenorphine changed only slightly, some studies have suggested that the number of individuals initiating buprenorphine treatment and MOUD during the pandemic decreased, although findings are mixed,17, 18, 21–23 potentially reflecting differences in populations or time periods studied. Examining patterns of buprenorphine care alongside the total number of patients is needed, as a focus just on the total number of individuals receiving buprenorphine can be misleading and obscure important gaps in care that may require focused policy attention. To better understand the dynamics of buprenorphine episodes during the pandemic, and how those patterns compared to pre-pandemic patterns, we used national retail pharmacy claims to examine how many buprenorphine treatment episodes were initiated, how many episodes ended in the months following the declaration of the public health emergency in 2020, and how those numbers differed from what was observed in the analogous period in 2019. Methods We used 2019 and 2020 de-identified pharmacy claims from the IQVIA Real World Data–Longitudinal Prescriptions24 to identify buprenorphine treatment episodes. These data capture an estimated 92% of all prescriptions dispensed at retail pharmacies in all 50 US states and the District of Columbia. The study was approved with a waiver of consent from the corresponding author’s Institutional Review Board, and STROBE Guidelines for reporting observational studies were followed. We used the IQVIA data to define the number of active treatment episodes, the number of treatment episodes starting and the number of treatment episodes ending. An active treatment episode started on the date of the first dispensed buprenorphine prescription (formulated with an FDA-approved indication for OUD treatment; e.g., sublingual buprenorphine, buprenorphine/naloxone, extended-release buprenorphine) after a 30-day buprenorphine-free period in which the days’ supply from a previously dispensed prescription was exhausted, and ended when the days’ supply of buprenorphine was exhausted, followed by at least 30 days with no new buprenorphine prescription dispensed. We conducted a sensitivity analysis requiring a 60-day buprenorphine-free period to start the episode, and 60 days with no new buprenorphine prescriptions to end the episode. The results were not meaningfully different than the analysis using a 30-day buprenorphine-free period. During a pre-COVID-19 time period (3/13/2019 to 12/1/2019) and a COVID-19 time period (3/13/2020 to 12/1/2020), we identified all treatment episodes that were active, new treatment episodes that started during the period, and treatment episodes that ended during the period. We divided each of the two periods into three stages: spring (first stage) (March 13 to June 13), summer (second stage) (June 14 to September 13), and fall (third stage) (September 14 to December 1). We chose March 13 as the start date as it coincided with the federal declaration of the COVID public health emergency in 2020. We calculated the percentage change in the total number of active episodes during each stage and over the entire period. We also calculated the percentage change in 2020 compared to 2019 during each stage and over the total period in the number of active episodes, the number of episodes starting, and the number of episodes ending. We used the growth in active treatment episodes during 2019 to generate expected levels of active treatment episodes at the end of each stage during the pandemic in 2020. To generate the expected number of starting and ending episodes in each 2020 stage, we used the ratios of active treatment episodes in 2019 to the number of starting and ending episodes in each time period in 2019 and applied those ratios to the number of active treatment episodes at the beginning of each stage in 2020 (see Table 2 footnotes). Essentially, this enabled us to calculate what would have been expected in 2020 overall and for each time period if the rate of growth in buprenorphine use after March 13, 2020, was the same as the rate of growth after March 13, 2019. Results The number of active buprenorphine episodes increased by 11.1% from March 13, 2019 (n=622,220), to March 13, 2020 (n=691,292), and by 8.5% from December 1, 2019 (n=668,139), to December 1, 2020 (n=725,189) (Table 1). Compared to the expected number of active episodes on December 1, 2020 (n=742,308), the observed number of episodes was only 2.3% lower. The results by stage during the pandemic were similar: the observed number of active episodes in 2020 fell short of expected by 0.3%, 1.6%, and 2.3% by June 14, September 14, and December 1, 2020, respectively. Table 1 Active buprenorphine episodes in 2019 to 2020 Active Episodes 2019 actual 2020 expected* 2020 actual Raw % change % above expected March 13 622,220 n/a 691,292 11.1 n/a June 14 643,081 714,469 712,639 10.8 −0.3% September 14 656,308 729,164 717,165 9.3 −1.6% December 1 668,139 742,308 725,189 8.5 −2.3% Expected number of active episodes in 2020 are based on the actual number of active episodes on March 13, 2020, and the growth rates in active treatment episodes from March 13, 2019 to each of the subsequent dates in 2019 Although the total number of active episodes in 2020 was similar to the expected number, we found that substantially fewer new treatment episodes started between March 13 and December 1, 2020, than expected. Specifically, the observed 600,080 episodes starting during the 2020 period was 17.2% fewer than expected based on the 2019 experience. The greatest discrepancy was in the first stage during 2020, when the number of episodes starting was 19.0% below the expected number of new episodes starting. The summer and fall stages during 2020 also had substantially fewer episodes starting than expected (16.6% and 14.0%, respectively) (Table 2). Table 2 Changes in buprenorphine episodes starting and ending from 2019 to 2020 2019 actual 2020 expected 2020 actual Raw % change % above expected Overall (March 14 to December 1) New treatment episodes starting 652,477 724,908 600,080 −8.0% −17.2% Treatment episodes ending 606,558 673,891 566,183 −6.7% −16.0% Stage 1 (March 14 to June 14) New treatment episodes starting 235,502 261,645 211,813 −10.1% −19.0% Treatment episodes ending 214,641 238,468 190,466 −11.3% −20.1% Stage 2 (June 14 to September 14) New treatment episodes starting 227,549 252,162 210,226 −7.6% −16.6% Treatment episodes ending 214,322 237,504 205,700 −4.0% −13.4% Stage 3 (September 14 to December 1) New treatment episodes starting 189,426 206,991 178,041 −6.0% −14.0% Treatment episodes ending 177,595 194,063 170,017 −4.3% −12.4% *Expected number of episode ending from March 14, 2020, until December 1, 2020 = (Active treatment episodes on March 13, 2020) × (Episodes ending from March 14, 2019, until December 1, 2019 / Active treatment episodes on March 13, 2019) Expected number of new episode starts from March 14, 2020, until December 1, 2020 = (Active treatment episodes on March 13, 2020) × (New episode starts from March 14, 2019, until December 1, 2019 / Active treatment episodes on March 13, 2019) In 2020, we found similarly large shortfalls in the observed number of episodes that ended compared to expected. The number of treatment episodes that ended from March 14 to December 1, 2020, was 566,183, a 6.7% reduction for the number of episodes that ended during the same time period in 2019 but a 16.0% decrease from the number expected to end based on 2019 patterns. As with episodes starting, the biggest difference in the number of episodes that ended compared with the number expected was during the first stage of the pandemic (20.1% decrease), with 13.4% and 12.4% decrease from expected during the summer and fall stages, respectively. Discussion We compared buprenorphine treatment episodes in the period following the declaration of the public health pandemic in 2020 to the analogous period in 2019. We found that the number of individuals receiving buprenorphine treatment in 2020 increased slightly compared to 2019, despite the fact that fewer individuals started treatment. This pattern is very different from that observed for many aspects of outpatient healthcare including trends of overall prescriptions filled at retail pharmacies, but consistent with other studies that found relatively little change in buprenorphine prescribing after the COVID public health emergency declaration.8–10, 16, 25 Our findings suggest that beneath this apparent stability, the flow of individuals receiving buprenorphine treatment changed substantially. The number of individuals starting buprenorphine treatment decreased dramatically; others have also found decreases.17, 18 However, these decreases were compensated for by substantial decreases in the number of individuals ending their buprenorphine treatment episodes compared to what would have been expected, based on the comparable period in 2019. We also found that the greatest changes from 2019 occurred in the months immediately following the declaration of the public health emergency, a period during which there were also substantial decreases to substance use disorder treatment facilities. 26 We do not know what factors contributed to the changing dynamics in treatment initiation or retention. After declaration of the public health emergency, federal and state governments relaxed regulations regarding buprenorphine treatment,27, 28 including those related to telehealth,27–29 such as no longer requiring in-person visits to start buprenorphine treatment and enabling payment parity for telehealth services. There were also changes in regulations related to substance use disorder treatment more generally, including changes in prior authorization policies, relaxation of requirements related to drug testing of individuals in treatment, and waiving of copayments for some services. 27, 28 Furthermore, clinicians appear to have been providing more days’ supply for each dispensed prescription,18 and many states relaxed urine drug testing requirements for individuals being treated for substance use disorders resulting in a decline in such testing.27, 30 These changes may have helped many individuals stay in treatment who might otherwise have been unable to attend in-person appointments. That pattern may have been reinforced by stay-at-home orders and changes in the economy that decreased competing demands on time for many patients. Federal policy changes during the pandemic were intended to decrease the number of individuals losing Medicaid coverage31 and may have also decreased the number ending buprenorphine treatment episodes due to lack of coverage. There were also substantial policy changes related to methadone, and we have no information regarding how methadone treatment during the pandemic may have influenced the outcomes we examine, although recent studies suggest there may not have been much substitution.32 Furthermore, many patients may have also sought to continue treatment to address additional psychological stressors during the pandemic,1, 4 and it is possible that clinicians concerned about having fewer patients in their practice may have been less likely to discharge individuals from treatment. We also do not know why the number of patients starting treatment dropped so dramatically. Given the many disruptions and changes during the pandemic, there are multiple possibilities. One possibility is simply that fewer individuals sought treatment. However, this seems inconsistent with other studies finding increases in fatal overdoses and harms from opioid misuse during the pandemic 1–6 and the fact that existing patients were able to access treatment. It is also possible that there was little change in the overall number of buprenorphine prescribing clinicians or that fewer individuals seeking treatment had commercial insurance coverage to pay for such treatment. Prior research suggests that the monthly patient census of buprenorphine prescribers is relatively stable after the first 2 years,33 and existing buprenorphine prescribers may have been reluctant to increase the number of individuals they are treating. In such situations, existing patients may be more likely to receive appointments than new patients.34 A better grasp of these dynamics, how they may vary across prescribers and influence different patient populations, and how they were influenced by policies implemented during the pandemic is essential to assessing how policies implemented during the pandemic could support future access to MOUD. Our findings must be considered in the context of study limitations. We observed only buprenorphine prescriptions dispensed at retail pharmacies; we have no information about prescriptions written but not filled, prescriptions dispensed at other types of pharmacies, such as hospital pharmacies, nor about other services individuals may have received. We have no information about an individual’s clinical status or about more general changes in clinical practice that might influence a clinician’s prescribing behavior. We restricted our analysis to FDA-approved formulations for treating opioid use disorder, but those formulations may be used off-label to treat pain, and we do not know that individuals are receiving the medication for opioid use disorder treatment. As we discuss above, we did not directly observe the factors contributing to the changing dynamics in treatment initiation or retention, nor do we know if the lower rates of leaving treatment have persisted. Finally, IQVIA data capture approximately 92% of prescriptions filled at retail pharmacies; we do not know if our results generalize to prescriptions filled in pharmacies not captured by IQVIA. Despite these limitations, our findings contribute to the understanding of MOUD since the start of the pandemic, potentially shedding light on how policy changes meant to support buprenorphine prescribing influenced prescribing dynamics during that period. Our findings suggest that while policy efforts may have been successful in maintaining existing patients in treatment, that success did not extend to individuals not yet in treatment. Future focused efforts are needed to enhance access to and ongoing engagement in treatment for individuals seeking to start buprenorphine treatment, including expanding low threshold and interim care strategies. Acknowledgements The authors thank Mary Vaiana Ph.D., Courtney Kase, MPH, and Hilary Peterson B.A., all of the RAND Corporation, for their feedback and editorial assistance on earlier versions of the manuscript. Funding This work was supported by the Foundation for Opioid Response Efforts (FORE) and the National Institute on Drug Abuse (NIDA) through R01DA045800-01 R01 (Stein, PI). Declarations Conflict of Interest None. Prior Presentations: None. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Czeisler ME Lane RI Petrosky E Mental Health, Substance Use, and Suicidal Ideation During the COVID-19 Pandemic - United States, June 24-30, 2020 MMWR Morb Mortal Wkly Rep 2020 69 32 1049 1057 10.15585/mmwr.mm6932a1 32790653 2. Taquet M Luciano S Geddes JR Harrison PJ Bidirectional associations between COVID-19 and psychiatric disorder: retrospective cohort studies of 62 354 COVID-19 cases in the USA Lancet Psychiatry. 2021 8 2 130 140 10.1016/S2215-0366(20)30462-4 33181098 3. Panchal N, Kamal R, Orgera K, et al. The Implications of COVID-19 for Mental Health and Substance Use. Kaiser Family Foundation. . https://www.kff.org/coronavirus-covid-19/issue-brief/the-implications-of-covid-19-for-mental-health-and-substance-use/ 4. Centers for Disease Control and Prevention. Early release of selected mental health estimates based on data from the January–June 2019 National Health Interview Survey. . https://www.cdc.gov/nchs/data/nhis/earlyrelease/ERmentalhealth-508.pdf 5. Friedman J Beletsky L Schriger DL Overdose-Related Cardiac Arrests Observed by Emergency Medical Services During the US COVID-19 Epidemic JAMA Psych 2021 78 5 562 564 10.1001/jamapsychiatry.2020.4218 6. Soares WE, 3rd, Melnick ER, Nath B, et al. Emergency Department Visits for Nonfatal Opioid Overdose During the COVID-19 Pandemic Across Six US Health Care Systems. Ann Emerg Med. 2021;10.1016/j.annemergmed.2021.03.013 7. Ahmad F, Rossen L, Sutton P. Provisional drug overdose death counts. National Center for Health Statistics; 2020. 8. Mehrotra A, Chernew M, Linetsky D, Hatch H, Cutler D. The impact of the COVID-19 pandemic on outpatient visits: A rebound emerges. The Commonwealth Fund; 2020. Accessed December 16, 2020. https://www.commonwealthfund.org/publications/2020/apr/impact-covid-19-outpatient-visits 9. Epic Health Research Network. Expansion of Telehealth during COVID-19 Pandemic. . https://ehrnprd.blob.core.windows.net/wordpress/pdfs/Expansions-in-Telehealth-COVID-19 10. Martin K, Kurowski D, Given P. The Impact of COVID-19 on the Use of Preventive Health Care. Health Care Cost Institute. . https://healthcostinstitute.org/hcci-research/the-impact-of-covid-19-on-the-use-of-preventive-health-care 11. Alexander GC Stoller KB Haffajee RL Saloner B An Epidemic in the Midst of a Pandemic: Opioid Use Disorder and COVID-19 Ann Intern Med 2020 173 1 57 58 10.7326/M20-1141 32240283 12. White A, Lundahl B, Bryan MA, et al. Pregnancy and the Opioid Crisis: Heightened Effects of COVID-19. J Addict Med. 2021;10.1097/ADM.0000000000000822 13. Cochran G Bruneau J Cox N Gordon AJ Medication treatment for opioid use disorder and community pharmacy: Expanding care during a national epidemic and global pandemic Subst Abus. 2020 41 3 269 274 10.1080/08897077.2020.1787300 32697171 14. Gustavson AM Gordon AJ Kenny ME Response to coronavirus 2019 in Veterans Health Administration facilities participating in an implementation initiative to enhance access to medication for opioid use disorder Subst Abus. 2020 41 4 413 418 10.1080/08897077.2020.1809609 32936695 15. Kelley AT Dungan MT Gordon AJ Barriers and Facilitators to Buprenorphine Prescribing for Opioid Use Disorder in the Veterans Health Administration During COVID-19 J Addict Med 2021 15 5 439 440 10.1097/ADM.0000000000000786 33323694 16. Nguyen TD, Gupta S, Ziedan E, et al. Assessment of Filled Buprenorphine Prescriptions for Opioid Use Disorder During the Coronavirus Disease 2019 Pandemic. JAMA Intern Med. 2020;10.1001/jamainternmed.2020.7497 17. Chalasani R, Shinabery JM, Goetz CT, et al. Buprenorphine Dispensing in Pennsylvania During the COVID-19 Pandemic, January to October 2020. J Gen Intern Med. 2021;10.1007/s11606-021-07083-y 18. Currie JM Schnell MK Schwandt H Zhang J Prescribing of Opioid Analgesics and Buprenorphine for Opioid Use Disorder During the COVID-19 Pandemic JAMA Netw Open 2021 4 4 e216147 10.1001/jamanetworkopen.2021.6147 33856474 19. Cantor J Dick AW Haffajee R Use of buprenorphine for those with employer-sponsored insurance during the initial phase of the COVID-19 pandemic J Subst Abuse Treat 2021 129 108384 10.1016/j.jsat.2021.108384 34080552 20. Jones CM Guy GP Jr Board A Comparing actual and forecasted numbers of unique patients dispensed select medications for opioid use disorder, opioid overdose reversal, and mental health, during the COVID-19 pandemic, United States, January 2019 to May 2020 Drug Alcohol Depend 2021 219 108486 10.1016/j.drugalcdep.2020.108486 33421802 21. Cance JD Doyle E Changes in Outpatient Buprenorphine Dispensing During the COVID-19 Pandemic JAMA 2020 324 23 2442 2444 10.1001/jama.2020.22154 33320215 22. Wang L Weiss J Ryan EB Waldman J Rubin S Griffin JL Telemedicine increases access to buprenorphine initiation during the COVID-19 pandemic J Subst Abuse Treat 2021 124 108272 10.1016/j.jsat.2020.108272 33771276 23. Samuels EA, Khatri UG, Snyder H, Wightman RS, Tofighi B, Krawczyk N. Buprenorphine Telehealth Treatment Initiation and Follow-Up During COVID-19. J Gen Intern Med. 2022;10.1007/s11606-021-07249-8 24. IQVIA: Real World Data and Insights. Accessed July 13, 2020. https://www.iqvia.com/solutions/real-world-evidence/real-world-data-and-insights 25. Saloner B, Krawczyk N, Solomon K, et al. Experiences with Substance Use Disorder Treatment During the COVID-19 Pandemic: Findings from a Multistate Survey. International Journal of Drug Policy. 2021;10.1016/j.drugpo.2021.103537 26. Cantor J Kravitz D Sorbero M Trends in visits to substance use disorder treatment facilities in 2020 J Subst Abuse Treat 2021 127 108462 10.1016/j.jsat.2021.108462 34134879 27. Andraka-Christou B, Bouskill K, Haffajee RL, et al. Common themes in early state policy responses to substance use disorder treatment during COVID-19. Am J Drug Alcohol Abuse 2021:1-11. 10.1080/00952990.2021.1903023 28. Pessar SC, Boustead A, Ge Y, Smart R, Pacula RL. Assessment of State and Federal Health Policies for Opioid Use Disorder Treatment During the COVID-19 Pandemic and Beyond. JAMA Health Forum. 2021;2(11)10.1001/jamahealthforum.2021.3833 29. Long K, Manz J, Mette E. States rapidly build their telehealth capacity to deliver opioid use disorder treatment. Covid-19 State Action Center. National Academy for State Health Policy; 2020. April 13. Accessed March 29, 2022. https://www.nashp.org/states-rapidly-build-their-telehealth-capacity-to-deliver-opioid-use-disorder-treatment/ 30. Tilhou AS, Dague L, Saloner B, Beemon D, Burns M. Trends in Engagement With Opioid Use Disorder Treatment Among Medicaid Beneficiaries During the COVID-19 Pandemic. JAMA Health Forum. 2022;3(3)10.1001/jamahealthforum.2022.0093 31. H.R.6201 - Families First Coronavirus Response Act, Public Law No: 116-127 (2020). 32. Chen AY, Powell D, Stein BD. Changes in Buprenorphine and Methadone Supplies in the US During the COVID-19 Pandemic. JAMA Network Open. 2022;5(7)10.1001/jamanetworkopen.2022.23708 33. Cabreros I Griffin BA Saloner B Gordon AJ Kerber R Stein BD Buprenorphine prescriber monthly patient caseloads: An examination of 6-year trajectories Drug Alcohol Depend 2021 228 109089 10.1016/j.drugalcdep.2021.109089 34600259 34. Yee CA Legler A Davies M Prentice J Pizer S Priority access to health care: Evidence from an exogenous policy shock Health Econ. 2020 29 3 306 323 10.1002/hec.3982 31999884	36474004	PMC9734477	NO-CC CODE	2022-12-14 23:28:28	no	J Gen Intern Med. 2022 Dec 6;:1-5	utf-8	J Gen Intern Med	2,022	10.1007/s11606-022-07891-w	oa_other
==== Front Dtsch Dermatolog Deutsche Dermatologie 2731-7692 2731-7706 Springer Medizin Heidelberg 5689 10.1007/s15011-022-5689-0 Industrieforum Nachrichten Meldungen, Studien und Produktneuheiten aus der Industrie Facharztmagazine Redaktion Springer Medizin Verlag GmbH, Aschauer Str. 30, 81549 München, Deutschland 9 12 2022 2022 70 12 990995 © Berufsverband der Deutschen Dermatologen e.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© Berufsverband der Deutschen Dermatologen e.V. 2022 ==== Body pmcEffektiv gegen Schuppen und Juckreiz Akut oder als Erhaltungstherapie bei seborrhoischer Dermatitis - das neue medizinische Shampoo Ketoconazol Klinge® wirkt gegen Schuppen, lindert Juckreiz und Entzündungen [Shuttleworth D et al. J Dermatolog Treat 1998]. Es bekämpft Schuppenpilze effektiv bereits ab der ersten Anwendung und ist für alle Haartypen geeignet. Das Präparat setzt auf den Wirkstoff Ketoconazol und enthält als einziges Shampoo mit diesem Antimykotikum zusätzlich den pflegenden Hilfsstoff Dexpanthenol. Schuppende Kopfhaut ist eines der häufigsten Hautprobleme in der dermatologischen Praxis: etwa 3−10 % der Bevölkerung leiden an seborrhoischer Dermatitis, die in der Regel chronisch verläuft. Mit den typisch gelblich-fettigen Schuppen, der geröteten Kopfhaut und dem Juckreiz sind die Symptome für Betroffene zwar nicht gefährlich, aber unangenehm und können unbehandelt bis zum Haarausfall führen. Bei diesem meist chronischen Krankheitsbild hat sich der Einsatz des Antimykotikums Ketoconazol bewährt. Die Kombination mit Dexpanthenol bekämpft nicht nur die Schuppen, lindert Juckreiz und Entzündungen, sondern kann darüber hinaus Kopfhaut und Haare auch mit Feuchtigkeit versorgen und so zu ihrer Regeneration beitragen. Das medizinischen Shampoo ist zur Anwendung sowohl bei akuten Symptomen als auch zur Prophylaxe geeignet. Im Akutfall werden die Haare über zwei bis vier Wochen zweimal pro Woche gewaschen. Nach Abklingen der Symptome beugt eine Anwendung alle ein bis zwei Wochen dem Wiederauftreten der Pilzinfektion vor. red Nach Informationen von Klinge Pharma Schnelle und anhaltende Verbesserung bei Psoriasis Wesentliche Treiber der Psoriasis sind erhöhte Konzentrationen von IL-17A und IL-17F. Mit dem humanisierten monoklonalen IgG1- Antikörper Bimekizumab (Bimzelx®), der selektiv die beiden proinflammatorischen Zytokine IL-17A und IL-17F hemmt, können ein schnelles Therapieansprechen und eine Verbesserung der klinischen Symptome bei Betroffenen mit mittelschwerer bis schwerer Plaque-Psoriasis, die für eine systemische Therapie infrage kommen, erzielt werden. Die Zulassung von Bimekizumab basiert auf den Ergebnissen der drei publizierten Phase-III-Studien BE VIVID [Reich K et al. Lancet 2021], BE READY [Gordon KB et al. Lancet 2021] und BE SURE [Warren RB et al. N Engl J Med 2021], in denen die Wirksamkeit und Sicherheit des Mittels bei 1.480 Erwachsenen mit mittelschwerer bis schwerer Plaque-Psoriasis untersucht wurde. Die Phase-IIIb-Studie BE RADIANT analysierte außerdem Bimekizumab im Vergleich zu dem IL-17A-Inhibitor Secukinumab [Reich K et al. N Engl J Med 2021]. In Woche 16 wurde mit Bimekizumab bei einem signifikant größeren Anteil von Betroffenen (61,7 %) der primäre Endpunkt (PASI100-Ansprechen) erreicht, verglichen mit Secukinumab (48,9 %). In Woche 48 betrug der Anteil der PASI-100-Ansprechenden im Bimekizumab-Arm 67 % (vs. Secukinumab: 46,2 %). "Dieses Ansprechen hielt bis zu einem Jahr an", berichtete Prof. Diamant Thaçi, Lübeck. "Zu den relevantesten Behandlungszielen aus Betroffenensicht zählen ein schnelles Therapieansprechen, eine möglichst erscheinungsfreie Haut sowie dauerhaft erscheinungsfrei zu bleiben", so Dr. Nina Magnolo, Münster. Ihrer Erfahrung nach hat Bimekizumab das Potenzial, diesen Wünschen gerecht zu werden. Es sei das erste Arzneimittel, dem ein Zusatznutzen gegenüber der zweckmäßigen Vergleichstherapie Secukinumab zugesprochen wurde und es könne budgetneutral verordnet werden. Martina Eimer Post-EADV-Pressewebcast und Symposium "EADV Kompakt: Psoriasis-Management aktuell", 19.9.2022; Veranstalter: UCB Pharma Regeneration der epidermalen Barriere bei atopischer Dermatitis Läsionen und Juckreiz belasten Betroffene mit atopischer Dermatitis (AD) schwer. Eine Normalisierung der Hautbarriere ist wichtig, um den Teufelskreis aus Kratzen und voranschreitender Hautschädigung zu durchbrechen. Die AD ist durch eine gesteigerte Typ-2-Immunantwort, unter anderem mit einer Überexpression der proinflammatorischen Zytokine IL-4 und IL-13, gekennzeichnet. Dabei wird die Hautbarriere durch eine Kombination verschiedener komplexer Defekte gestört, wie Dr. Anna De Benedetto, Rochester, USA, erläuterte. Dysfunktionen treten ihr zufolge unter anderem in der Lipid- zusammensetzung und den Proteinen der Epidermis auf, betreffen den pH-Wert sowie das Mikrobiom der Haut. Prof. Michael J. Cork, Sheffield, Großbritannien, erinnerte daran, dass dadurch der transepidermale Wasserverlust (TEWL) bei AD-Erkrankten deutlich erhöht ist. In einer prospektiven Beobachtungsstudie wurde die Auswirkung verschiedener Therapien auf den Hautbarriereparameter untersucht [Montero-Vilchez T et al. J Clin Med 2022]. Behandelt wurden die Betroffenen entweder mit topischen Kortikosteroiden, Ciclo-sporin A oder Dupilumab (Dupixent®). Nach 16 Wochen wurde bei Patienten mit Haut-läsionen der TEWL am stärksten durch Dupi- lumab gesenkt (Abb. 1). Im Vergleich zum Ausgangswert sank der TEWL-Wert unter Dupilumab um 18,93 g/m²/h. Mit topischen Kortikosteroiden oder Ciclosporin A betrug die Abnahme nur 10,43 g/m²/h beziehungsweise 7,73 g/m²/h. Wie die Regulierung molekularer Mechanismen, die auf die Typ-2-Inflammation abzielen, die Barrierefunktion während einer Therapie mit Dupilumab verbessert, erläuterte Cork anhand weiterer Studiendaten. Sechs bis acht Wochen nach Beginn einer Dupilumab-Therapie konnte bei Betroffenen eine vermehrte Expression von antimikrobiellen Peptiden (HBD-3), Protease-Inhibitoren (LEKTI) sowie des Strukturproteins Filaggrin festgestellt werden [Rohner MH et al. Allergy 2021]. Die Blockade der IL-4- und IL-13-Signalwege führt so zu einer Erhöhung der Strukturproteinlevel in läsionaler und nicht läsionaler Haut. Eine Therapie mit Dupilumab sei somit nicht bloß antiinflammatorisch, sondern wirke auch normalisierend auf die Hautbarriere, so Cork. Ingo Schroeder Satellitensymposium "Inside Atopic Dermatitis: Type 2 Inflammation Underlies Skin Barrier Dysfunction and Neuroimmune Dysregulation", 31. EADV, Mailand, 8.9.2022; Veranstalter: Sanofi Genzyme/Regeneron Patienten und Ärzte schätzen Krankheitsaspekte unterschiedlich ein Wie stark sind von Psoriasis betroffene Menschen rund um den Globus in ihrer Lebensqualität beeinträchtigt? Dieser Frage ging die multinationale UPLIFT-Studie [Lebwohl M et al. Dermatol Ther 2022] nach, die die Versorgungssituation und Therapieziele beleuchtete. An der Studie nahmen 3.806 erwachsene Psoriasiserkrankte und 473 Dermatologinnen und Dermatologen aus Kanada, USA, Frankreich, Deutschland, Italien, Spanien, UK und Japan teil. Die meisten Patientinnen und Patienten (67 %) waren ausschließlich an Psoriasis erkrankt, fast ein Drittel (28 %) war an Psoriasis und Psoriasis-Arthritis (PsA) erkrankt und einige (5 %) litten ausschließlich unter PsA. Zentrale Erkenntnis der Studie ist, dass Betroffene sowie Ärztinnen und Ärzte verschiedene Aspekte der Erkrankung und ihrer Behandlung unterschiedlich wahrnehmen. So zeigte sich mit Blick auf die Patienten, dass sie ihre Erkrankung als schwerer einschätzten als es die Ärzte taten: Bei 78 % der Betroffenen war laut Diagnose lediglich eine begrenzte Hautfläche von Psoriasis-Plaques betroffen. Dennoch gaben 58 % von ihnen an, dass sie ihre Erkrankung als mittelschwer bis schwer einschätzten. Insbesondere, wenn Patienten unter quälendem Juckreiz litten und/oder besonders sensible, oft sichtbare Hautpartien wie Kopfhaut oder Nägel betroffen waren, fühlten sie sich stark in ihrer Lebensqualität beeinträchtigt - trotz limitiertem Hautbefall. Bei Patienten mit Gesichtsbeteiligung war der Leidensdruck am größten. Auch der Einfluss verschiedener Krankheitsmerkmale auf die Lebensqualität wird unterschiedlich wahrgenommen: So nahm für Patienten die Art der Psoriasis-Symptome - hier insbesondere der Juckreiz - Platz eins unter den Faktoren ein (23,7 %), die zur Schwere der Erkrankung beitragen. Bei den Ärzten rangierte der Juckreiz jedoch nur auf Platz drei (12,2 %). Zudem waren die Erkrankungsdauer und die Lokalisation der Hautläsionen mit einer Nennung von jeweils 10,6 % unter den Top-drei-Faktoren der Erkrankten zu finden, nahmen bei den Ärzten jedoch keinen vergleichbaren Stellenwert ein. Diese gaben auf Position eins und zwei das Ausmaß der Einschränkung der Lebensqualität (21,4 %) sowie die Ausprägung der betroffenen Hautfläche (19,8 %) an. Dabei glaubten mehr als 80 % der Patienten und Ärzte an eine Übereinstimmung ihrer Behandlungsziele, tatsächlich wurde diesen aber unterschiedliche Bedeutung beigemessen. Vor diesem Hintergrund erklären sich auch die unterschiedlichen Attribute, die von einer "idealen Therapie" erwartet werden: Für Betroffene war es die Verbesserung der Symptome, für Behandelnde die Langzeiteffektivität der Therapie. red Nach Informationen von Amgen Anhaltende Wirksamkeit über zwei Jahre Neue Zwei-Jahres-Daten der Extensionsstudie POETYK PSO-LTE zeigen, dass die klinische Wirksamkeit unter fortgesetzter Gabe von Deucravacitinib bei der Behandlung erwachsener Patienten mit mittelschwerer bis schwerer Plaque-Psoriasis anhielt. In der aktuellen Analyse wurden die Ergebnisse der an der zulassungsrelevanten Studie POETYK PSO-1 teilnehmenden Patientinnen und Patienten ausgewertet, die nach einem Jahr in die Verlängerungsstudie POETYK PSO-LTE aufgenommen wurden. Nach 112 Wochen betrugen die unter Verwendung einer modifizierten Non-Responder-Imputation (mNRI) ermittelten Ansprechraten 82,4 % für PASI 75, 55,2 % für PASI 90 sowie 66,5 % gemäß des static Physicians Global Assessment (sPGA) 0/1 (klares Hautbild/fast klares Hautbild) [Lebwohl M et al. EADV 2022]. "Menschen mit Psoriasis und ihre behandelnden Dermatologinnen und Dermatologen benötigen weiterhin zusätzliche orale Behandlungsoptionen, die noch wirksamer und gut verträglich sind, da Psoriasis eine chronische, systemische, immunvermittelte Erkrankung ist, die mit schweren Komorbiditäten einhergeht", sagte Dr. Mark Lebwohl, New York/USA. "Diese neuen Langzeit- ergebnisse zeigen unter fortgesetzter Behandlung mit Deucravacitinib eine anhaltende Wirksamkeit über einen Zeitraum von bis zu zwei Jahren. Die Daten untermauern damit das Potenzial, das dieser einmal täglich einzunehmende Wirkstoff für die Behandlung von Menschen mit mittelschwerer bis schwerer Plaque-Psoriasis hat, für die ein therapeutischer Bedarf an wirksameren oralen Behandlungsoptionen besteht." Von den 262 mit Deucravacitinib behandelten Patienten, die in der Analyse ausgewertet wurden, hatten 171 nach 16 Wochen der Studie POETYK PSO-1 einen PASI 75 erreicht. Bei ihnen konnte ein Ansprechen bis zu Woche 112 aufrechterhalten werden. Die Ansprechraten betrugen für PASI 75 100 % nach Woche 16, 90,1 % nach Woche 52 sowie 91,0 % nach Woche 112. Für PASI 90 lagen die Werte nach Woche 16 bei 62,6 %, nach Woche 52 bei 64,9 % und nach Woche 112 bei insgesamt 63,0 %. Der sPGA 0/1 betrug 84,2 % nach Woche 16 sowie 73,7 % nach Woche 52 und 73,5 % nach einem Behandlungszeitraum von insgesamt 112 Wochen. red Nach Informationen von Bristol-Myers Squibb Unterversorgung mit Nährstoffen bei Haarausfall Hinter diffusem Haarausfall kann eine Unterversorgung der Haarwurzeln mit wichtigen Nährstoffen stecken, sodass die Wachstumsphase der Haare verkürzt ist und sie vorzeitig ausfallen. Eine Supplementierung brachte in einer Anwendungsbeobachtung beim Großteil der Betroffenen eine Besserung [Schwichtenberg U et al. EADV 2022]. In der Haarsprechstunde werden viele mögliche Ursachen für Haarausfall betrachtet. "Der geschulte Blick des Arztes ist wichtig bei der Frage, ob eine Haardichteminderung vorliegt und ob sie umschrieben oder diffus ist. Mit der Vergrößerung im Dermato- beziehungsweise Trichoskop lässt sich schon eine Verdachtsdiagnose stellen", erläuterte Dr. Andreas Finner, Berlin. Wichtige Fragen sind Beginn, Verlauf und Lokalisation des Haarausfalls und mögliche Triggerfaktoren vor Beginn des Effluviums. Dazu zählen Erkrankungen, Medikamente, Ernährungsumstellungen oder Diäten. Bei androgenetischer Alopezie bietet Alfatradiol 0,25 mg/ml (Pantostin®) als Lösung zum Auftragen auf die Kopfhaut ei- ne evidenzbasierte Behandlungsoption [Schaart FM. Haut 2000]. Im Gegensatz dazu zeigen diffuse Effluvien keine Muster, so Dr. Uwe Schwichtenberg, Bremen. Akute Telogeneffluvien können nach Entbindungen, Crash-Diäten oder Infektionen auftreten, wie aktuell häufig nach COVID-19-Infektionen [Nguyen B et al, JAAD Int 2022]. Schwichtenbergs Praxis war an einer prospektiven Anwendungsbeobachtung beteiligt, in der die Supplementierung mit einem Kombinationspräparat aus Cystin, Thiamin, Pantothensäure, Folsäure, Biotin, Eisen und Zink (Pantovigar® vegan) untersucht wurde. 31 Patientinnen mit diffusem Haarausfall nahmen das Lebensmittel für besondere medizinische Zwecke (bilanzierte Diät) drei Monate lang ein. Die Frauen waren da- nach subjektiv deutlich weniger oder nicht mehr mit Haarausfall belastet, berichtete Schwichtenberg. Die ärztliche Bewertung fiel bei knapp drei Vierteln der Betroffenen "besser" oder "sehr viel besser" aus: Der Haarzugtest war erfolgreicher und die Haardichte nahm in der Trichoskopie numerisch klinisch relevant zu. Die Supplementierung kann mit einer Haarpflege mit Pantovigar® Shampoo und Pantovigar® Tonic für Frauen noch ergänzt werden. Martina Freyer Mittagsseminar "Haarausfall erkennen und Haarwachstum anregen - Woran kann es liegen und wie geht man vor?", 28. FOBI, München, 14.7.2022; Veranstalter: Merz Neue Leitlinie bestätigt Wirksamkeit bei Nagelpilz Die Onychomykose ist eine Pilzinfektion des Nagelorgans, die sowohl Zehen- als auch Fingernägel betrifft. Über zwei Drittel der Nagelpilzinfektionen werden durch Dermatophyten verursacht, hier ist vor allem Trichophyton rubrum als Haupterreger zu nennen. Die Infektion ist unbehandelt progredient und besitzt keine Selbstheilungstendenz. In der lokalen Behandlung einer leichten bis mäßig ausgeprägten Onychomykose gilt antimykotischer Nagellack, etwa Ciclopoli® gegen Nagelpilz, als Mittel der Wahl. Basierend auf dem klinischen Bild und Erregernachweis erfolgt die Therapie, entweder topisch antimykotisch oder zusätzlich mit oral applizierbaren Antimykotika. Bei leichten oder mäßig ausgeprägten Nagelinfektionen, das heißt, maximal 40 % der Nagel- oberfläche und maximal drei von zehn Zehennägeln sind betroffen, empfiehlt die jüngst veröffentlichte Leitlinie die Lokaltherapie mit antimykotischem Nagellack [Nenoff P et al. S1-Leitlinie Onychomykose 2022]. Ebenso weist die Leitlinie darauf hin, dass laut internationaler Konsensuskonferenz auch noch bei einem Befall des Nagels bis 50 % die Lokalbehandlung zu empfehlen ist. In manchen Fällen, etwa bei Kindern oder bei älteren Patientinnen und Patienten mit Komedikation, können Ärztinnen und Ärzte ihre Therapiefreiheit nutzen und auch bei einem über 50 % hinausgehenden Befall zunächst einen lokalen Therapieversuch erwägen. Zur Lokalbehandlung kommen Amorolfin oder Ciclopirox als wasserunlösliche Acryllacke oder Ciclopirox als wasserlöslicher HPCH(Hydroxypropyl-chitosan)-Lack, der einen elastischen Film auf dem Nagel bildet, zur Anwendung. Die Kombination aus einer wasserlöslichen HPCH-Lackgrundlage und Ciclopirox 8 % zeichnet Ciclopoli® aus. Im Gegensatz zu wasserfesten Nagellacken verbindet sich die hydrophile Lackgrundlage mit dem Keratin des Nagels und befördert den Wirkstoff tief in den Nagel. In-vitro-Studien belegen, dass die Wirkstoffpermeation in den Nagel durch die HPCH-Lackgrundlage beschleunigt wird [Monti D et al. J Drugs Dermatol 2014; Monti D et al. Br J Dermatol 2010]. Zudem wurde gezeigt, dass die Permeation für die wasserlösliche Ciclopirox-Formulierung signifikant höher war als für eine wasserfeste Acrylformulierung [Mon- ti D et al. Drug Dev Ind Pharm 2005]. Die Leitlinie bestätigt diesen Vorteil. Klinische Head-to-Head Stu- dien zeigen, dass Ciclopoli® gegen Nagelpilz gegenüber anderen antimykotischen Lacken beim Behandlungserfolg punktet. Im Vergleich mit wasserfestem Ciclopirox 8-%-Lack und auch wasserfestem Amorolfin 5-%-Lack ist Ciclopoli® bezüglich Therapieerfolg und Komplettheilung statistisch signifikant überlegen [Baran R et al. J Eur Acad Dermatol Venerol 2009]. red Nach Informationen von Almirall Hermal Hoher UV-Schutz für die Kopfhaut Chronische UV-Exposition ist ein Hauptrisikofaktor für die Entstehung von aktinischen Keratosen (AK), die vor allem bei hellhäutigen Menschen sowie überwiegend an besonders lichtexponierten Hautstellen auftreten. Um diesen sonneninduzierten Hautschädigungen vorzubeugen, ist ein konsequenter Lichtschutz notwendig. Die besondere Formulierung des Dermasence Solvinea Liquid AK LSF 50+ wird dem Anspruch nach einem hohen UV-Schutz der empfindlichen und besonders sonnenexponierten Kopfhaut gerecht. Sie ist speziell zur Prävention und Therapiebegleitung bei aktinischen Keratosen entwickelt. Neben dem hohen UV-B- und UV-A-Schutzniveau zeichnet sich das Mittel durch seine besonders leichte, trans-parente und nicht fettende Textur aus. Es schützt behaarte sowie unbehaarte Areale der Kopfhaut - ohne einen Film zu hinterlassen oder die Haare zu verkleben. Gleichzeitig unterstützt es mit Panthenol die Hautregeneration. Es ist parfumfrei und enthält kein Octocrylene. red Nach Informationen von P&M Cosmetics Nicht invasives Laser-Facelifting Fotona4D® ist eine neue Möglichkeit der Laserbehandlung im Gesicht, die in vier Schrit- ten die Haut ohne operativen Eingriff strafft und sie frisch und jugendlich erscheinen lässt − ein schmerzfreies nicht invasives Laser-Facelifting mit minimalen Nebenwirkungen und ohne Ausfallzeit. Durchgeführt wird sie mit dem dafür speziell konzipierten Gerät Fotona Timewalker 4D Pro. Dabei wird die Haut nicht nur von der Außen-, sondern auch von der Innenseite des Mundes aus behandelt wird. Die Anwendung nutzt mit tief eindringenden thermischen Reizen die Selbstheilungskraft der Haut. Die Neubildung von Kollagen verbessert die Elastizität des Gewebes und gleicht Falten aus. Auch Marionettenfalten, die mürrisch wirkende Mundwinkel verursachen, können gemildert, sowie leicht herabgesunkene Lider gehoben werden. Das Ergebnis insgesamt: Falten sind geglättet, die Haut ist gestrafft, der Teint wirkt ebenmäßig. Gesicht, Kinnlinie, Hals und Dekolleté erhalten ein frisches, jugendlicheres Aussehen. Schritt eins: SmoothLiftinTM. Geeignet für den Ausgleich der Nasolabialfalte und die Straffung der Kinnpartie durch eine nicht ablative interorale Er:YAG-Behandlung, die mit einer kontrollierten und sanften Erwärmung der Mundhöhle die Neubildung von Kollagen stimuliert und das Gewebe strafft. Schritt zwei: FRAC3. Ergänzt die Wirkung der intraoralen Behandlung und stellt durch kurze, hochenergetische Nd:YAG-FRAC3-Pulse ein jugendliches Hautbild her. Schritt drei: PIANO. Erzielt mit Wärme einen Straffungseffekt durch einen ultralangen Nd:YAG-Pulsmodus. Dabei wird Energiezufuhr subkutan konzentriert und so die Neubildung von Kollagenfasern anregt. Schritt vier: SupErficialTM. Glättet und verbessert das Erscheinungsbild der Haut mittels einer sanften Ablation durch ultrakur- ze Er:YAG-Pulse. Die Behandlung wirkt als oberflächliches Laser-Peeling, das die Hornschicht entfernt. red Nach Informationen von Fotona Erfolgreiches Langzeitmanagement durch Basistherapie bei Akne Eine Basistherapie als Monotherapie oder Begleitpflege spielt in der Behandlung von Akne eine zentrale Rolle und sollte dabei auf die Bedürfnisse der Haut angepasst werden. Die Wirksamkeit der Inhaltsstoffe, entweder als Begleitpflege neben einer medikamentösen Therapie oder auch als Basistherapie, kann zur Wiederherstellung eines intakten Mikrobioms sowie einer gestärkten Hautbarriere beitragen und der Entstehung weiterer Entzündungsprozesse entgegenwirken. "Wir reden heute über ein völlig verändertes Verständnis der Bedeutung von Bakterien für die Akneentstehung" betonte PD Dr. Thomas Jansen, Köln. Entgegen früheren Vorstellungen zeigen neue Untersuchungen des Mikrobioms der Haut, dass sich Cutibacterium acnes (C. acnes, früher: Propionibacterium acnes) nicht wie gedacht vermehrt in den Follikeln, sondern in unterschiedlicher Dichte auf der gesamten Haut des Menschen befindet [Grice EA et al. Science 2009]. Kommensale Mikroorganismen wie C. acnes interagieren dabei konstant mit dem angeborenen Immunsystem und sorgen für eine Abwehr von pathogenen Keimen sowie den Aufbau einer Immuntoleranz [Byrd AL et al. Nat Rev Microbiol 2018]. Bei einer Akneerkrankung ist die Barrierefunktion der Haut gestört. Unabhängig von der Art und dem Schweregrad der Akne kann der Einsatz von Pharmakotherapeutika die von der Akne ausgelöste Beeinträchtigung der Hautbarriere verstärken. Eine gestörte Barrierefunktion äußert sich zum Beispiel durch Rötung, Juckreiz, Trockenheit und Brennen. "Laut Patientenbefragungen liegt die Therapieadhärenz nur bei 35 bis 51 Prozent", so Jansen. Experten fordern deshalb ein Umdenken in der Behandlung von Akne. "Nur durch einen verantwortungsvollen Um- gang mit Antibiotika, die Vermeidung von Monotherapien sowie durch die zeitlich begrenzte Verwendung dieser Medikamente, kann die anhaltende und nebenwirkungsarme Wirksamkeit erhalten bleiben", betonte Prof. Petra Staubach-Renz, Mainz. Um die Therapieadhärenz sicherzustellen, sind die Stärkung der Hautbarriere sowie die Minderung von Nebenwirkungen essenziell. Dafür eignet sich eine Basistherapie als Monotherapie oder Begleitpflege. "Moderne Dermatokosmetika modulieren das Hautmikrobiom und stärken die Hautbarriere", erläuterte Jansen. Zudem verfolgen sie einen modernen Therapieansatz, indem harmlose Bakterien gegen entzündungsfördernde Cutibakterien eingesetzt werden. Dazu wird das Bakterium Vitreoscilla filiformis kultiviert und als Lysat der Begleitpflege beigesetzt. "Aqua Posae Filiformis ist belegbar immunologisch aktiv, aktiviert das angeborene Immunsystem und kann auf vielfache Weise immunmodellierend die Aknetherapie unterstützen", so Jansen [Mahe YF et al. Clin Cosmet Investig Dermatol 2013]. In einer Studie wurden der Einfluss eines Antibiotikums und eines Dermatokosmetikums auf die Mikrobiota der Haut über 28 Tage verglichen [Dréno B et al. Exp Dermatol 2017]. Dabei wurde eine vergleichbare klinische Wirksamkeit festgestellt. Im Vergleich zu einem Antibiotikum moduliert das untersuchte Dermatokosmetikum jedoch das Mikrobiom der Haut. Es verbindet in seiner Zusammensetzung sich ergänzende und verstärkende Wirkstoffe ohne das Risiko einer bakteriellen Resistenzentwicklung und kann damit von Therapiebeginn an dauerhaft eingesetzt werden. Essenziell ist, dass die Basistherapie auf den Schweregrad der Akne, den Zustand und die Bedürfnisse der Haut angepasst ist. Zur Begleitung einer topischen Aknetherapie oder auch als Monotherapie bei milderen Formen von Akne eignen sich Reinigungs- und Pflegeprodukte, die sich positiv auf das Mikrobiom auswirken, Unreinheiten und Entzündungen entgegenwirken und Rückfällen vorbeugen. Niacinamid besitzt antiinflammatorische Eigenschaften, mildert Rötungen und wirkt beruhigend auf die Haut. Das Ceramid Procerad® hilft, die Schutzbarriere der Haut wiederherzustellen und Pickelmalen vorzubeugen. Die zusätzliche Anwendung von Aqua Posae Filiformis trägt zur Bekämpfung entzündungsfördernder Bakterien bei und fördert das Gleich- gewicht des Mikrobioms. Salicylsäure und Lipohydroxysäure unterstützen die Epidermis für ein ebenmäßigeres Hautbild. red Nach Informationen von La Roche Posay	0	PMC9734478	NO-CC CODE	2022-12-14 23:28:28	no	Dtsch Dermatolog. 2022 Dec 9; 70(12):990-995	utf-8	null	null	null	oa_other
==== Front J Immigr Minor Health J Immigr Minor Health Journal of Immigrant and Minority Health 1557-1912 1557-1920 Springer US New York 36472714 1435 10.1007/s10903-022-01435-4 Brief Communication COVID-19 in Patients with a Primary Refugee-Associated Language in a Kentucky Emergency Department During 2020 Hamm Joel [email protected] 1 Duncan Meredith S. [email protected] 2 Robertson Nicole M. [email protected] 3 Keck James W. [email protected] 4 http://orcid.org/0000-0003-0767-9426 Crabtree Katherine [email protected] 5 1 grid.266539.d 0000 0004 1936 8438 Department of Emergency Medicine, University of Kentucky, Lexington, KY USA 2 grid.266539.d 0000 0004 1936 8438 Department of Biostatistics, University of Kentucky, Lexington, KY USA 3 grid.266539.d 0000 0004 1936 8438 University of Kentucky College of Medicine, Lexington, KY USA 4 grid.266539.d 0000 0004 1936 8438 Department of Family and Community Medicine, University of Kentucky, Lexington, KY USA 5 grid.266539.d 0000 0004 1936 8438 Department of Internal Medicine, University of Kentucky, 800 Rose St, Lexington, KY 40536 USA 6 12 2022 15 24 4 2022 20 11 2022 24 11 2022 © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, 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. COVID-19 has heavily impacted the refugee population in the United States due to exposure risks, living and working conditions, and healthcare access, but little is known about outcomes. We reviewed emergency department visits to a Kentucky hospital among 2163 patients from March-December 2020, studying incidence of COVID-19 diagnosis for patients with a primary refugee-associated language compared to English speakers, and outcomes after diagnosis including hospitalization, length of stay, and in-hospital mortality. Patients in the population of interest had higher odds of COVID-19 diagnosis in the hospital (OR = 12.31, 95% CI 7.80–19.40), but, among those with COVID-19, lower odds of hospital admission (OR = 0.58, 95% CI 0.37–0.90) and shorter median length of stay (4.1 vs. 10.5 days) compared to English speakers. The study corroborates reports of comparatively higher COVID-19 incidence in patients speaking a primary refugee-associated language, but implies milder illness severity, possibly reflecting this population’s baseline health. Keywords Refugee COVID-19 Emergency department Foreign language http://dx.doi.org/10.13039/100006108 National Center for Advancing Translational Sciences UL1TR001998 KL2TR001996 ==== Body pmcIntroduction The COVID-19 pandemic has had a global impact, and the world’s 82.4 million refugees comprise one of the most vulnerable populations [1]. However, information on COVID-19 cases and mortality in refugees is limited. Refugees in the US may be at greater risk for COVID-19 illness than the general population due to living and working conditions, access to healthcare, and comorbidities [2]. This is pertinent to Kentucky, which ranked 5th in the U.S. for refugee resettlement in 2019 [3]. The Kentucky Office for Refugees reports that 30,800 refugees resettled in Kentucky since 1994, equaling 0.6% of the state’s population [4]. In Minnesota, a study found that foreign-born individuals had twice the age- and sex-adjusted COVID-19 mortality rate compared to U.S.-born individuals [5]. A California study found that foreign-born non-Hispanic participants had a COVID-19 proportionate mortality ratio 10.7 times higher than U.S.-born non-Hispanic participants [6]. We hypothesize that patients with a primary refugee-associated language in Kentucky experienced COVID-19 disparities during 2020. We evaluated SARS-CoV-2 infection incidence and outcomes in patients seeking care at an academic medical center emergency department (ED), using primary language as a proxy for refugee status. METHODS Participants This is a retrospective cohort study of patients 18 years or older seen in the ED at the University of Kentucky hospital system between March 1, 2020, and December 31, 2020. COVID-19 testing during this period was not applied differentially based on language or immigration status. The ED initiated PCR-based COVID-19 testing on March 16, 2020 for any patient with fever plus cough or shortness of breath and expanded testing criteria on March 31, 2020, to include sore throat, myalgias, chills, loss of taste or smell, or unresponsiveness. As refugee status is not specified in hospital records, we used documented primary language and hospital location in Kentucky to presumptively identify non-Hispanic refugee patients. Kentucky Office for Refugees reports that Swahili, Arabic, Kinyarwanda, Ukrainian, Bembe, Lingala, Somali, Haitian Creole, Chin, Tigrinya, Mai Mai, Bantu, Pashto, Dari/Farsi, Karen, Kurdish and “Other” are spoken among refugees relocated to Lexington and Louisville, Kentucky [7, 8]. The study population included patients with these documented primary languages who are referred to in the paper as “patients with a primary refugee-associated language.” The method of using language, local resettlement data, and geographic location, has been validated for prediction of refugee status [9]. Data Collection Patients with a primary refugee-associated language were frequency matched on age, sex, and insurance status to the general population of ED users during the same period. Our base sample included data from multiple visits (n = 14,447) during the study period made by the 2163 patients in our sample. We excluded 638 visits from Spanish speakers as their immigration status could encompass multiple categories; 4 visits from incarcerated individuals, 67 visits with erroneous admitting service listed, and patients with mean ED presentation greater than 4 times per month. After exclusions, our sample included data from 11,573 unique encounters among 2024 patients (1631 English-speakers). Monthly ED patient volume was obtained via patient registration data from March 1, 2020, to December 31, 2020, and from March 1, 2019, to December 31, 2019. Measures Covariates included age, sex, language, insurance status, Charlson comorbidity index, body mass index (BMI) at time of visit, self-reported smoking status, self-reported alcohol consumption, and admission diagnosis. COVID-19 illness as determined by ED documented ICD-10 code was the primary outcome. Patients with a positive COVID-19 PCR during their stay had this applied to their chart consistently due to need for triage to COVID-19 quarantine units of the hospital. Other outcomes included hospital admission, intensive care unit (ICU) stay, length of stay, and in-hospital mortality. Analysis Multivariate imputation by chained equations generated twenty complete data sets to handle missing covariate data while maintaining correlation structure. For continuous variables, regression-based predictive mean matching produced biologically plausible imputed values; discriminant function with a noninformative Jeffrey’s prior imputed categorical variables [10]. Results across datasets were combined according to Rubin’s rules [11]. Descriptive statistics and group comparison used χ2 tests for categorical variables and t-tests for continuous variables. Logistic regression with generalized estimating equations (GEE) and a compound symmetry correlation structure to account for multiple observations per person generated odds ratios to measure the associations with COVID-19 diagnosis. These models included primary refugee-associated language, age, sex, race, Charlson Index, BMI, smoking status, and insurance status as independent variables. Changes in ED volume were compared using one-tailed t-test. Subgroup analysis was limited to patients with COVID-19. We used logistic regression with GEE and a compound symmetry correlation structure to account for multiple observations per person, adjusting for comorbidities using Charlson Index. Analyses were performed in SAS 9.4 (Cary, NC). A p-value < 0.05 was considered statistically significant. Results Patients with a primary refugee-associated language were significantly younger than English-speaking patients, more likely to be male, and more likely to identify as Black or Asian (Table 1). These patients were more likely to be insured with Medicaid and English-speaking patients with Medicare. BMI, Charlson Index and current smoking rates were significantly higher in English speaking groups. A larger percentage of patients with a primary refugee-associated language were admitted to the hospital and ICU but had shorter length of stay in both (Table 1). Table 1 Demographics Characteristic Refugee (N = 870) English speaker (N = 10,703) p-value Age, years 40.6 (16.3) 47.7 (17.2) < 0.0001 Male Sex 291 (33.5) 2507 (23.4) < 0.0001 Race White 266 (30.6) 8582 (80.2) < 0.0001 Black 433 (49.8) 1864 (17.4) Asian 157 (18.1) 170 (1.6) Other 14 (1.6) 87 (0.8) Ethnicity Hispanic 5 (0.6) 305 (2.8) < 0.0001 Non-Hispanic 807 (92.8) 9903 (92.5) Unknown 58 (6.7) 495 (4.6) Language English 0 (0.0) 10,703 (100.0) < 0.0001 Nepali 224 (25.8) 0 (0.0) Swahili 217 (24.9) 0 (0.0) French 151 (17.4) 0 (0.0) Arabic 143 (16.4) 0 (0.0) Kinyarwanda 54 (6.2) 0 (0.0) Russian 51 (5.9) 0 (0.0) Lingala 12 (1.4) 0 (0.0) Farsi 7 (0.8) 0 (0.0) Ukrainian 4 (0.5) 0 (0.0) Kirundi 1 (0.1) 0 (0.0) Amharic 5 (0.6) 0 (0.0) Pushto 1 (0.1) 0 (0.0) Insurance Private 193 (22.2) 3417 (31.9) < 0.0001 Medicare 47 (5.4) 2537 (23.7) Medicaid 520 (59.8) 3941 (36.8) Other 34 (3.9) 557 (5.2) None 76 (8.7) 210 (2.0) Unknown 0 (0.0) 41 (0.4) BMI, kg/m2, a 27.6 (6.1) 31.7 (9.1) < 0.0001 Smoking Status Never 521 (59.9) 4752 (44.4) < 0.0001 Former 51 (5.9) 2189 (20.5) Current 60 (6.9) 2454 (22.9) Unknown 238 (27.4) 1308 (12.2) Alcohol Use Disorder 8 (0.9) 105 (1.0) 0.8592 COVID-19 71 (8.2) 64 (0.6) < 0.0001 Admitted to Hospital 103 (11.8) 420 (3.9) < 0.0001 Length of Stay, days b, c 2.8 (1.9, 3.9) 19.6 (3.2, 29.4) < 0.0001 In-hospital Death 7 (0.8) 23 (0.2) 0.0010 Charlson Indexa 1.2 (2.1) 2.5 (2.9) < 0.0001 Admitted to ICU 20 (2.3) 127 (1.2) 0.0048 Length of ICU Stayb, d 3.0 (1.0, 4.0) 4.0 (2.0, 10.0) 0.0004 Values are presented as Mean (SD) or N (%) for continuous and categorical variables, respectively, unless otherwise noted aContinuous variables with missing values: BMI (189 for refugees, 1022 for English speakers); Charlson Index (0 for refugees, 572 for English-speakers) bMedian (Q1, Q3) cLimited to those admitted to the hospital dLimited to those admitted to the ICU COVID-19 was diagnosed in 16.5% (65/393) of patients with a primary refugee-associated language and in 3.1% (50/1,631) of English-speaking patients. Primary refugee-associated language status (OR = 12.31, 95% CI 7.80–19.40), age (OR per 10 years = 1.27, 95% CI 1.09–1.48), BMI (OR per 5 kg/m2 = 1.23, 95% CI 1.11–1.36), and Black (OR = 1.65, 95% CI 1.08–2.52) or Asian (OR = 2.15, 95% CI 1.16–3.99) race compared to White race were associated with greater odds of COVID-19 diagnosis (Table 2). Table 2 Predictors of a positive COVID-19 test Variable Odds ratio [95% CI] p-value Refugee statusa 12.31 [7.80, 19.40] < 0.0001 Age, per 10 years 1.27 [1.09, 1.48] 0.0026 Female vs. male sex 0.73 [0.49, 1.09] 0.1280 Black race vs. white 1.65 [1.08, 2.52] 0.0195 Asian race vs. white 2.15 [1.16, 3.99] 0.0155 Other race vs. white 1.95 [0.44, 8.60] 0.3789 Charlson index, per 1 point 0.91 [0.81, 1.01] 0.0860 BMI, per 5 kg/m2 1.23 [1.11, 1.36] 0.0001 Former vs. never smoking 0.74 [0.41, 1.31] 0.3010 Current vs. never smoking 0.81 [0.45, 1.47] 0.4867 Medicare vs. private insurance 0.77 [0.40, 1.48] 0.4254 Medicaid vs. private insurance 0.80 [0.52, 1.22] 0.3035 Other insurance vs. private insurance 0.70 [0.27, 1.81] 0.4649 No insurance vs. private insurance 0.42 [0.14, 1.20] 0.1051 aBased on language and location in Lexington, KY—speakers of primary refugee-associated language Among patients with COVID-19, those who spoke a primary refugee-associated language had lower rates of hospital admission (14.1% vs. 38.5%, p = 0.0009) and ICU admission (7.0% vs. 23.4%, p = 0.0074, Table 3) compared to English speaking patients. After adjustment for comorbidities via Charlson Index, the association between language status and hospital admission held (OR = 0.58, 95% CI 0.37–0.90), but the association with ICU admission was attenuated (OR = 0.56, 95% CI 0.30–1.06). Among admitted patients with COVID-19, English speakers had a greater median length of stay in the hospital (10.5 vs. 4.1 days) and in the ICU (11.0 vs. 4.0 days). In-hospital mortality among English speaking patients with COVID-19 was higher than that of patients with a primary refugee-associated language during the 9-month period of the study, although not statistically significant (6.3% versus 1.4%, p = 0.19). (Table 3). Table 3 Outcomes in COVID-19 positive patients Summary statistics Characteristic Refugee (N = 71) English speaker (N = 64) p-value Age, years 44.7 (16.4) 48.4 (17.9) 0.2207 Male Sex 29 (40.9) 14 (21.9) 0.0182 Racea < 0.0001 White 19 (26.8) 42 (65.6) Black 33 (46.5) 21 (32.8) Asian 17 (23.9) 1 (1.6) Other 2 (2.8) 0 (0.0) BMI, kg/m2,b 29.3 (5.9) 34.9 (10.7) 0.0008 Admitted to Hospital 10 (14.1) 25 (38.5) 0.0009 Length of Stay, daysc,d 4.1 (2.3, 9.0) 10.5 (5.9, 24.5) 0.0028 In-hospital Death 1 (1.4) 4 (6.3) 0.1897 Charlson Index 1.2 (2.0) 2.1 (2.3) 0.0149 Admitted to ICU 5 (7.0) 15 (23.4) 0.0074 Length of ICU Stayc,e 4.0 (3.0, 6.0) 11.0 (3.0, 25.0) 0.0114 Model Results Variable Outcome Odds ratio [95% CI] p-value Refugee (vs. English-Speaker) Hospital admission 0.58 [0.37, 0.90] 0.0150 Admission to ICU 0.56 [0.30, 1.06] 0.0766 aDue to low proportion of non-white proportions and the small number of COVID-positive patients, all non-white races were combined for analyses limited to those with a diagnosis of COVID-19 b Missing values: BMI (13 for refugees, 6 for English speakers) cMedian (Q1, Q3) d Limited to those admitted to the hospital e Limited to those admitted to the ICU Mean patient volume at the University of Kentucky ED decreased from 7290 patients per month during March-December 2019 to 5532 patients per month during March-December 2020 (24% decrease, p = 0.0002). For patients with a primary refugee-associated language, mean patient volume in the ED changed from 92.4 patients per month during March-December 2019 to 89.7 patients per month during March-December 2020 (2.9% decrease, p = 0.74). Limitations Our study is retrospective and observational, and likely has misclassified some patients’ refugee status because this is not designated in the record. Some participants are likely to be misclassified into both groups. The findings from our study may not generalize to all refugees given the heterogeneity of refugee communities and different access to care across the US. Our study does not comment on outcomes in Spanish-speaking patients due to overlap of Spanish-speaking refugee and migrant worker populations. We were unable to measure some potential confounding factors, including if ED providers offered more frequent COVID testing to non-English speakers. We were unable to measure reasons for high prevalence of COVID-19 in the refugee population. This is a single-institution study which limits generalizability of the results. Discussion In the 9-month study period from March 1, 2020, to December 31, 2020, patients with a primary refugee-associated language seeking care in the ED were more likely to receive a COVID-19 diagnosis; however, COVID-19 disease severity may have been lower in this group, as they were less likely to be admitted to the hospital and had shorter lengths of stay than English-speaking patients. Admission to the ICU among those with COVID-19 was lower in patients with a primary refugee-associated language compared to English-speakers, but upon adjustment for comorbidities, this association was attenuated, suggesting that patients with a primary refugee-associated language benefited from better baseline health status. These findings clarify the results of studies performed in Minnesota, California, and Massachusetts [5, 6, 12]. The California and Minnesota studies used death certificate data and found higher mortality from COVID-19 in all foreign-born groups but particularly Hispanic men. In Massachusetts, Spanish-speaking patients with COVID-19 had a higher hospitalization rate. The opposite was true in our study, as patients with a primary refugee-associated language had lower hospitalization and mortality rate than English speakers. Overall patient volumes in the ED decreased during the study period, while patients with a primary refugee-associated language utilized the ED at a similar rate. Other studies have shown that disadvantaged groups use the ED when lacking access to care, one of many potential reasons for higher rates and lower acuity of COVID-19 in our study population [13, 14]. In summary, our study suggests a high burden of COVID-19 in patients with a primary refugee-associated language. Additional research is needed to characterize ongoing disparities in access to prevention, diagnosis, and treatment of COVID-19 in refugees. Acknowledgements Nicholas Jewell PhD assisted in the early planning of this project. Sources of Support include UK Center for Health Equity and Transformation, UK Center for Clinical and Translational Science, the NIH National Center for Advancing Translational Sciences through grant number UL1TR001998 and the National Center for Advancing Translational Sciences, National Institutes of Health, through grant number KL2TR001996. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. Declarations Conflict of interest The authors have nothing to disclose. The views expressed in the submitted article are the authors’ own and not an official position of the institution. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. UNHCR. Refugee Statistics. https://www.unhcr.org/refugee-statistics/. Accessed 1 Nov 2021. 2. Clark E Fredricks K Woc-Colburn L Bottazzi ME Weatherhead J Disproportionate impact of the COVID-19 pandemic on immigrant communities in the United States PLOS Neglected Trop Dis 2020 14 7 e0008484 10.1371/journal.pntd.0008484 3. Krogstad JM. Key facts about refugees to the U.S. Pew Research Center. https://www.pewresearch.org/fact-tank/2019/10/07/key-facts-about-refugees-to-the-u-s/. Accessed 1 Nov 2021. 4. Kentucky Office for Refugees. Refugee Resettlement in Kentucky. https://www.kentuckyrefugees.org/refugees-in-kentucky/. Accessed 17 Nov 2022. 5. Horner KM, Wrigley-Field E, Leider JP. A first look: disparities in COVID-19 mortality among US-Born and Foreign-Born Minnesota residents [published online ahead of print, 2021 Aug 2]. Popul Res Policy Rev. 2021;1–14. 6. Garcia E Eckel SP Chen Z Li K Gilliland FD COVID-19 mortality in California based on death certificates: disproportionate impacts across racial/ethnic groups and nativity Ann Epidemiol 2021 58 69 75 10.1016/j.annepidem.2021.03.006 33746033 7. Kentucky Office for Refugees. Arrival Data by City, Lexington. https://www.kentuckyrefugees.org/refugees-in-kentucky/lexington/. Accessed 1 Sep 2022. 8. Kentucky Office for Refugees: Kentucky Arrival Data by City, Louisville. https://www.kentuckyrefugees.org/refugees-in-kentucky/louisville/. Accessed 1 Sep 2022. 9. Morrison M Nobles V Johnson-Agbakwu CE Bailey C Liu L Classifying refugee status using common features in EMR Chem Biodivers 2022 1 e202200651 10. Little R. Missing data adjustments in large surveys. J Bus Econ Stat. 1988;6:287–329. 11. Rubin DB. Multiple imputation for nonresponse in surveys. New Jersey: John Wiley and Sons; 1987. 12. Smati H Cohen PA Nagda DV Saravanan Y Kalugin P Li C Ranker L Risk factors for hospitalization among patients with COVID-19 at a community ambulatory clinic in Massachusetts during the initial pandemic surge J Immigr Minor Health 2021 23 1110 5 10.1007/s10903-021-01189-5 33772419 13. Hanchate AD Dyer KS Paasche-Orlow MK Banerjee S Baker WE Lin M Xue WD Feldman J Disparities in emergency department visits among collocated racial/ethnic medicare enrollees Ann Emerg Med 2019 73 3 225 35 10.1016/j.annemergmed.2018.09.007 30798793 14. Naouri D Ranchon G Vuagnat A Schmidt J El Khoury C Yordanov Y Factors associated with inappropriate use of emergency departments: findings from a cross-sectional national study in France BMJ Qual Saf 2020 29 6 449 64 10.1136/bmjqs-2019-009396	36472714	PMC9734481	NO-CC CODE	2022-12-14 23:29:57	no	J Immigr Minor Health. 2022 Dec 6;:1-5	utf-8	J Immigr Minor Health	2,022	10.1007/s10903-022-01435-4	oa_other
==== Front Mamm Genome Mamm Genome Mammalian Genome 0938-8990 1432-1777 Springer US New York 36481846 9972 10.1007/s00335-022-09972-x Article Telemedical monitoring in patients with inborn cardiac disease – experience of a tertiary care centre http://orcid.org/0000-0003-4870-9863 Westphal Dominik S. [email protected] 12 Federle David 1 Steger Alexander 13 Vodermeier Tanja 1 Scheiper-Welling Stefanie 45 Jenewein Tina 45 Beckmann Britt-Maria 4 Kauferstein Silke 4 Martens Eimo 1 Hahn Franziska 1 1 grid.6936.a 0000000123222966 Department of Internal Medicine I, Klinikum rechts der Isar, School of Medicine & Health, Technical University of Munich, Munich, Germany 2 grid.6936.a 0000000123222966 Institute of Human Genetics, Klinikum rechts der Isar, School of Medicine & Health, Technical University of Munich, Munich, Germany 3 grid.452396.f 0000 0004 5937 5237 DZHK (German Centre for Cardiovascular Research), Partner Site Munich Heart Alliance, Munich, Germany 4 Institute of Legal Medicine, Goethe University, University Hospital Frankfurt, Frankfurt, Germany 5 grid.411088.4 0000 0004 0578 8220 German Red Cross Blood Center, Institute of Transfusion Medicine and Immunohaematology, University Hospital Frankfurt, Frankfurt, Germany 8 12 2022 18 29 7 2022 29 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 number of cardiologically relevant genetic findings will continue to increase. This is due to the use of high-throughput sequencing techniques and the critical role of incidental findings in cardiac disease genes. Telemedicine can be a useful diagnostic tool to monitor the heart rhythm of patients with inborn cardiac diseases. Methods Patients were screened once they had been referred to our outpatient department for rare cardiac diseases between January 2020 and May 2022. Those patients who underwent genetic testing and were consequently diagnosed with a genetic disorder were included in this study. Their medical records were evaluated regarding implanted cardiac electronic devices and findings in the telemedical monitoring. Results 304 patients were seen in our outpatient department for rare cardiac diseases in the mentioned period. In 100 cases, genetic testing was performed. 10 patients (10%) with an identified inborn cardiac disease were monitored via telemedicine until the end of May 2022. 4 patients were monitored by implantable loop recorders (ILR), 4 patients were monitored by Implantable Cardioverter Defibrillators (ICD), and 2 patients received both devices. Clinical relevant arrhythmias making medical intervention necessary were identified in 4 cases. In two cases, data interpretation was hampered by sinus tachycardia caused by physical exercise. Discussion Telemonitoring of the heart rhythm by medical devices is beneficial for patients with monogenic heart diseases. Especially, when the indication for an ICD is not clear, implantation of a telemonitored ILR can be a suitable choice. However, rhythm analysis can be challenging in young patients who are physically active. Supplementary Information The online version contains supplementary material available at 10.1007/s00335-022-09972-x. http://dx.doi.org/10.13039/501100020027 Dr. Rolf M. Schwiete Stiftung ==== Body pmcBackground During the last years, molecular diagnostics experienced a boost by the growing use of next-generation sequencing. Exome sequencing enables the complete analysis of the coding regions of the human DNA. The use of this technique has become a popular method to identify a growing number of disease genes. Eleven different genes have shown a definitive disease-causing association with non-syndromic dilated cardiomyopathy (DCM) (Hershberger and Jordan 1993), whilst variants in nine genes, partly overlapping with the aforementioned, were proven to cause non-syndromic hypertrophic cardiomyopathy (Cirino and Ho 1993). The importance of cardiac disease genes causing cardiomyopathies and/or primary arrhythmias is strengthened by their role as so called “actionable genes”. Even in asymptomatic individuals, pathogenic variants in these genes should be reported as incidental findings which can occur using high-throughput technologies (Green et al. 2013). This list of genes was recently extended by TTN (Miller et al. 2021) since certain truncating variants in this gene are associated with an increased risk for the development of a DCM phenotype (Haggerty et al. 2019). Hence, the number of patients with rare and complex cardiac diseases in need of a close follow-up by experienced cardiologists will be further increasing. This in turn requires a personalized risk assessment as well as individual prophylactic and therapeutic interventions in order to reduce the number of sudden cardiac deaths in the young caused by preventable cardiac arrhythmias. Telemedicine facilitates outpatient monitoring and heart rhythm analysis. It can be combined with cardiac implantable devices such as the Implantable Cardioverter Defibrillators (ICD) that can terminate a life-threatening arrhythmia. Another used device in telemedicine are implantable loop recorders (ILR) that solely record the heart rhythm without any possibility of intervention (Jamal et al. 2021). It was shown that telemonitoring reduces hospitalization in patients with heart failure (Winkler et al. 2021). Therefore, its use is recommended in those patients by the guidelines of the European Society of Cardiology (Theresa AM et al. 2022). Especially during the COVID-19 pandemic, telemedicine played a significant role in patient care (Monaghesh E, Hajizadeh A 2020). The use of telemedicine is not restricted to the distant monitoring of biosignals. It was also shown that patients referred for cardiogenetic visits benefited from the use of telemedicine for genetic counselling (Liang LW et al. 2022). In this study, we provide an overview of our experiences with telemonitoring of patients with inherited cardiac diseases in our tertiary centre. The aim is to point out the advantages and limitations of telemonitoring in these rare diseases. Methods Study design Patients were screened once they had been referred to our outpatient department for rare cardiac diseases (“Zentrum für seltene Erkrankungen”, Department of Internal Medicine I, Klinikum rechts der Isar, Technical University Munich, Germany) between January 2020 and May 2022. Those patients who underwent genetic testing and were consequently diagnosed with a genetic disorder were included in this study. Their medical records were evaluated regarding implanted cardiac electronic devices and findings in the telemedical monitoring. The study was performed according to the declaration of Helsinki. Written consent was obtained from all patients. Telemedical surveillance Telemedically monitored devices were screened on a daily basis by our telemedicine centre for one week. In cases of automatically detected arrhythmias, all alerts were manually reviewed by medical experts. If an arrhythmia was confirmed, the respective patient was informed by telephone or by video-visit and, if needed, invited to our outpatient department or the cardiological ward, respectively, for further evaluation. Results Most patients were affected by dilated cardiomyopathy 304 patients were seen in our outpatient department for rare cardiac diseases in the mentioned period of time. Genetic testing was performed in 91 of these cases because of a suspected monogenic disease. In nine additional cases, genetic testing had already been performed. The average age at presentation of these combined 100 patients was 47 years (minimum: 16 years, maximum: 80 years) with a 50:50 sex ratio. The most common referral diagnosis of patients was “dilated cardiomyopathy” (DCM) followed by “hypertrophic cardiomyopathy” (HCM). Four patients were clinically unaffected relatives that were predictively tested for a known familial variant (Fig. 1, overview of all patients: Supplementary Table).Fig. 1 Referral diagnosis of patients that received genetic testing in our outpatient department or had already been diagnosed with a genetic disease. ACM arrhythmogenic cardiomyopathy, BrS Brugada syndrome, CTD connective tissue disorder, DCM dilated cardiomyopathy, HCM hypertrophic cardiomyopathy A genetic variant that was at least classified as a variant of an unknown significance (VUS) or (likely) pathogenic according to the recommendations published by the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG) (Richards S et al. 2015) was identified in a total of 31 patients. Two (likely) benign variants were detected in a mother of a deceased index case (patient ID 66, for case description see Supplementary case reports). In 59 out of the 91 cases that received genetic testing in our centre (57.1%), neither a disease-causing variant nor a VUS could be detected (Table 1).Table 1 Number of identified variants in our patients Identified variants Number of patients (genetic testing in our centre) Number of patients (external genetic testing) Pathogenic 13 5 Likely pathogenic 9 3 VUS 7 1 Likely pathogenic + VUS 1 none Likely pathogenic + likely benign + benign 1 none Likely benign + benign 1 none No variant detected 59 none ∑ 91 9 VUS variant of unknown significance Telemedical monitoring identified clinically relevant arrhythmias in four out of ten cases Ten patients (10%) with an inborn cardiac disease of known genetic origin were monitored telemedically until the end of May 2022 (Table 2, clinical description of all cases provided as supplementary case reports). In a further case, inclusion was planned (patient ID: 83). Implanted and monitored devices were as follows: four patients were monitored by an ILR, four patients were monitored by an Implantable Cardioverter Defibrillator (ICD), and two patients received both devices. In these cases, an ILR was used combined with a subcutaneous ICD (sICD) since rhythm disturbances with a cycle length out of the therapy zone such as slow ventricular tachycardias (slow VT) cannot be monitored with the sICD-algorithm. On average, patients were monitored for a period of 11.5 months [2–31 months]. Clinically relevant arrhythmias with the necessity of medical intervention were identified in four cases (patient IDs 28, 63, 92, 100).Table 2 Telemonitored patients with identified genetic variants Patient ID Diagnosis Genetic variant Implanted device Time period of telemonitoring Detected arrhythmia 2 BrS c.934G > T, p.(Glu312), heterozygous in SCN5A (NM_198056.3) sICD 21 months none 15 HCM c.3130C > T, p.(Gln1044), heterozygous in MYBPC3 (NM_000256.3) ILR 16 months sinus tachycardia due to physical exercising 28a ACM c.(1510 + 1_1511-1)_(1688 + 1_1689-1)del, heterozygous in PKP2 (NM_004572.3) sICD/ILR 2 months sinus bradycardia, nsVTs 29a ACM c.(1510 + 1_1511-1)_(1688 + 1_1689-1)del, heterozygous in PKP2 (NM_004572.3) sICD 6 months none 43 BrS c.2804A > T, p.(Glu935Val), heterozygous in MYH7 (NM_000257.4), incidental finding ICD 11 months none 63b DCM c.21304C > T, p.(Arg7102), heterozygous in TTN (NM_003319.4), c.1673 A > G, p.(His558Arg), heterozygous and c.1715 C > A, p.(Ala572Asp), heterozygous in SCN5A (NM_198056.3) ILR/sICD 13 months nsVT 65b DCM c.21304C > T, p.(Arg7102), heterozygous in TTN (NM_003319.4) ILR 4 months none 92 Survived SCD c.433 T > C, p.(Phe145Leu), heterozygous in NKX2-5 (NM_004387.3) ICD 3 months several nsVTs 97 ACM c.4789G > T, p.(Glu1597*), heterozygous in DSP (NM_004415.4) ILR 8 months sinus tachycardia due to physical exercising 100 FA 800 ± 100 and 900 ± 100 GAA repeats in FXN (NM_000144.5) ILR 31 months sinus tachycardia, SVES, recurrent AF ACM arrhythmogenic cardiomyopathy, AF atrial fibrillation, BrS Brugada Syndrome, FA Friedreich Ataxia, HCM hypertrophic cardiomyopathy, ILR implantable loop recorder, sICD subcutaneous implantable cardioverter-defibrillator, nsVT non-sustained ventricular tachycardia, SVES supraventricular extrasystole, PVC premature ventricular contraction, SCD sudden cardiac death aMother and son, bFather and son In patient ID 28, a 20-year-old male diagnosed with arrhythmogenic cardiomyopathy (ACM), sICD monitoring identified an episode of VT with a heart rate between 249 and 251/min. An ILR was implanted subsequently for detection of arrhythmias beneath the monitor zone. The patient contracted SARS-CoV-2 and a total of four VTs (cycle length 245–265 ms) could be monitored at that time (Fig. 2). He was admitted to a hospital until recovery and betablocker therapy was intensified. Indication for ablation therapy will be evaluated if intensification of pharmaceutical therapy is insufficient to reduce the arrhythmic burden.Fig. 2 Episode of a ventricular tachycardia (VT) in the telemonitored implantable loop recorder. The arrow marks the beginning of the VT. Right before the beginning, four ventricular extrasystoles occurred as bigeminus (asterisks) The brother of patient ID 65 died of sudden cardiac death (SCD) at the age of 20 years. The patient himself showed signs of DCM in cardiac MRI. Genetic workup identified two proarrhythmic polymorphisms in SCN5A as well as the heterozygous nonsense variant in TTN (NM_003319.4) in the patient’s and the deceased brother’s DNA. A telemedically monitored ILR was implanted before genetic testing. Two months later, a sICD was implanted because of a detected nsVT. The 40-year-old female patient ID 92 was affected by recurrent idiopathic ventricular fibrillation. An ICD was implanted subsequently and shocked repeatedly because of polymorphic VTs and VF. Previously performed genetic analysis had identified the heterozygous polymorphism c.433 T > C, p.(Phe145Leu) in NKX2-5 (NM_004387.3) which is associated with an elevated odds ratio for VTs and SCD next to structural heart defects (Sveinbjornsson G et al. 2018). Since recurrent VTs were detected in the monitoring of the ICD (Fig. 3), a therapy of mexiletine was started.Fig. 3 Telemedically detected ventricular tachycardia (VT). The detected VT was terminated by the ICD using six beats of antitachycardia pacing. Arrow = start VT, Lightning = therapy with antitachycardiac pacing Patient ID 100 was diagnosed with Friedreich’s ataxia caused by biallelic GAA repeats in FXN. The patient was affected by recurrent atrial fibrillation (AF) and flutter, respectively, and received several ablations and cardioversion therapies. An ILR was implanted at the age of 22 years. From the age of 28 years, the patient noted episodes of tachycardia which could be correlated with episodes of AF as well as sinus tachycardia in the telemonitoring. Implantation of a Cardiac Resynchronization Therapy (CRT) pacemaker and ablation of the atrioventricular node was discussed with the patient. Sinus tachycardia induced by physical exercising is monitored telemedically In two further cases, data interpretation was hampered by sinus tachycardia caused by physical exercising (patient IDs 15, 97). Patient ID 15 is a 39-year-old male affected by a HCM caused by a pathogenic heterozygous nonsense variant in the MYBPC3 gene (NM_000256.3) Since the HCM Risk-SCD (O’Mahony C et al. 2014) was below the threshold of 5%, there was no clear indication for ICD implantation according to the guideline of the European Society of Cardiology (ESC) (Authors TF, M. et al. 2014). An ILR was implanted instead. Sinus tachycardia caused by regular physical exercising was repeatedly recorded. Patient ID 97’s older sister was severely affected by left ventricular ACM caused by a heterozygous nonsense variant in DSP with multiple episodes of electrical storm, resuscitation, and performed VT ablations (Westphal DS et al. 2022). Patient ID 97 is her 25-year-old brother who also inherited the variant in DSP. He is asymptomatic; however, cardiac MRI revealed left ventricular LGE-positive areas. Based on the positive family history, an ILR was implanted. The patient was recommended to reduce physical exercise because of his extensive training. Nevertheless, telemedical monitoring was overlaid by episodes of sinus tachycardia caused by continuous training (Fig. 4).Fig. 4 Monitored sinus tachycardia with 180 beats per minute (bpm). The implanted implantable loop recorder detected several episodes of tachycardia. Analysis revealed episodes of sinus tachycardia is correlated with physical exercising Discussion From a total of 100 patients with suspected genetic diseases, 10 patients (10%) with an inborn cardiac disease of known origin were monitored via telemedicine in our centre. Although the time period of monitoring was relatively short with an average of 11.5 months [minimum: 2 months, maximum: 31 months], clinically relevant arrhythmia could be detected in 40% of the cases (4/10 cases). In patient ID 28, nsVTs during a SARS-CoV-2 infection led to admission to hospital. Because the patient did not intend to visit the hospital despite his infectious disease, these nsVTs would not have been noticed until the next control of the implanted devices in best case. Patient ID 63 received a sICD after detection of an nsVT in the telemonitoring of the ILR, whilst monitored nsVTs in patient ID 98 led to a prompt change in medication and the start of mexiletine. AF could be differentiated from benign sinus tachycardia which was repeatedly noted by patient ID 100, diagnosed with Friedreich’s ataxia. These cases illustrate the relevance of the monitored arrhythmias and the direct consequence on the further therapy in patients with diagnosed inborn arrhythmias or cardiomyopathies. There are recommendations regarding implantation of an ICD for different genetic diseases (O’Mahony C et al. 2014; Towbin JA et al. 2019; Wilde AAM, Amin AS, Postema PG 2022). Despite recommendations, indication is not always clear. In ACM, for example, detection of nsVTs is a major criterion that can be decisive for the recommendation of an ICD implantation (Towbin JA et al. 2019). These nsVTs can escape Holter ECGs that only monitors a very limited period of time. In these cases, implantation of a telemonitored ILR may prevent losing precious time which could be vital for the patient. Although all described patients in this study were affected by known genetic diseases, telemonitoring in patients with an unsolved disease origin can be useful too. For example, implantation of an ILR can be a helpful tool if an ICD is not indicated and genetic testing identified a VUS in a proarrhythmic gene. The identification of a VUS can have challenging impacts, not only on the clinical management but also on the psychological outcome of the affected patients (Mighton C, Shickh S, Uleryk E, Pechlivanoglou P, Bombard Y 2021). Although an ILR is not able to intervene, the knowledge of a monitored heart rhythm can give the feeling of security to a certain degree (Leppert F et al. 2021). However, there are limitations that can hamper telemonitoring. In two of the reports (patient IDs 15 and 97), analysis of the telemonitoring was disturbed by recurrent sinus tachycardia that were caused by physical exercising. Especially in patient 95, physical exercise is performed despite the recommended restriction due to the genetic predisposition for ACM. These episodes did not only make the analysis a difficult task but also led to full storages of the ILR. In principle, recordings of such benign arrhythmias can be avoided by changing the programming. However, less slow ventricular arrhythmias are not recorded if this is done. In addition, the monitoring of excessive sporting activity is also useful in individual cases for advising patients. These problems mostly arise in young patients who regular participate in sport. Considering that the average age of the patients in our centre for rare cardiac diseases is 47 years [16–80 years] which is lower than the age of usual adult cardiological patients and that the acceptance for telemonitoring is decreasing with age (Siggemann BG, C., Mensing, M., Classen, T., Hornberg, C. & Terschuren, C. 2013), this problem might play an essential role in using telemedicine in cardiogenetic patients. Apart from that, there are also general ethical issues that should be considered when implementing telemedicine such as data privacy and the obligation to continue the patient care despite the distance (Chaet D et al. 2017). Conclusion Cardiac arrhythmias and technical problems can be diagnosed in a short time followed by appropriate therapy in patients with a monogenic heart disease. Indication for ICD therapy can be substantiated by documented arrhythmias. Although sporting activity and sinus tachycardia bedevil telemonitoring, it can be useful for advising patients. Remote monitoring should be recommended not only for patients with severe heart failure—as in the ESC Guidelines—but also for patients with rare diseases. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 36 KB) Acknowledgements We would like to thank the patients for participating in this study. Author Contribution DSW, EM, and FH—Drafting the manuscript. SSW, TJ, BB, and SK—Genetic analysis. DSW—Genetic counselling. DF, AS, TV, EM, and FH—Telemonitoring. All authors—Critical reviewing the manuscript and final approval. Funding Britt Beckmann received funding by the “Dr. Rolf Schwiete Stiftung”. Data Availability The data that support the findings of this study are available from the corresponding author upon reasonable request. Declarations Conflict of interest The authors declare that there is no conflict of interest. Consent for publication Written informed consent to publish the data was obtained from all patients. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Eimo Martens and Franziska Hahn have contributed equally to this work. ==== Refs References Authors TF m. ESC Guidelines on diagnosis and management of hypertrophic cardiomyopathy: the task force for the diagnosis and management of hypertrophic cardiomyopathy of the European Society of Cardiology (ESC) Eur Heart J 2014 35 2733–2779 2014 10.1093/eurheartj/ehu284 Chaet D Ethical practice in Telehealth and Telemedicine J Gen Intern Med 2017 32 1136 1140 10.1007/s11606-017-4082-2 28653233 Cirino, A. L. & Ho, C. in GeneReviews((R)) (eds M. P. Adam et al.) (1993). Green RC ACMG recommendations for reporting of incidental findings in clinical exome and genome sequencing Genet Med 2013 15 565 574 10.1038/gim.2013.73 23788249 Haggerty CM Genomics-first evaluation of heart disease associated with titin-truncating variants Circulation 2019 140 42 54 10.1161/CIRCULATIONAHA.119.039573 31216868 Hershberger RE Jordan E Adam MP Dilated cardiomyopathy overview GeneReviews((R)) 1993 Seattle University of Washington Jamal NE Abi-Saleh B Isma'eel H Advances in telemedicine for the management of the elderly cardiac patient J Geriatr Cardiol 2021 18 759 767 10.11909/j.issn.1671-5411.2021.09.004 34659382 Leppert F The INFluence of Remote monitoring on Anxiety/depRession, quality of lifE, and Device acceptance in ICD patients: a prospective, randomized, controlled, single-center trial Clin Res Cardiol 2021 110 789 800 10.1007/s00392-020-01667-0 32417952 Liang LW The use of telemedicine in cardiogenetics clinical practice during the COVID-19 pandemic Mol Genet Genomic Med 2022 10 e1946 10.1002/mgg3.1946 35388985 Mighton C Shickh S Uleryk E Pechlivanoglou P Bombard Y Clinical and psychological outcomes of receiving a variant of uncertain significance from multigene panel testing or genomic sequencing: a systematic review and meta-analysis Genet Med 2021 23 22 33 10.1038/s41436-020-00957-2 32921787 Miller DT ACMG SF v3.0 list for reporting of secondary findings in clinical exome and genome sequencing: a policy statement of the American College of Medical Genetics and Genomics (ACMG) Genet Med 2021 23 1381 1390 10.1038/s41436-021-01172-3 34012068 Monaghesh E Hajizadeh A The role of telehealth during COVID-19 outbreak: a systematic review based on current evidence BMC Public Health 2020 20 1193 10.1186/s12889-020-09301-4 32738884 O'Mahony C A novel clinical risk prediction model for sudden cardiac death in hypertrophic cardiomyopathy (HCM risk-SCD) Eur Heart J 2014 35 2010 2020 10.1093/eurheartj/eht439 24126876 Richards S Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology Genet Med 2015 17 405 424 10.1038/gim.2015.30 25741868 Siggemann BG C., Mensing, M., Classen, T., Hornberg, C. & Terschuren, C. Specific health status has an impact on the willingness to use telemonitoring: data from a 2009 health survey in north rhine-westphalia, Germany Telemed J E Health 2013 19 692 698 10.1089/tmj.2012.0214 23906307 Sveinbjornsson G Variants in NKX2-5 and FLNC cause dilated cardiomyopathy and sudden cardiac death Circ Genom Precis Med 2018 11 e002151 10.1161/CIRCGEN.117.002151 30354339 Theresa AM et al. (2022) ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: Developed by the Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC) With the special contribution of the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 24, 4–131 Towbin JA 2019 HRS expert consensus statement on evaluation, risk stratification, and management of arrhythmogenic cardiomyopathy Heart Rhythm 2019 16 e301 e372 10.1016/j.hrthm.2019.05.007 31078652 Westphal DS Myocarditis or inherited disease? - The multifaceted presentation of arrhythmogenic cardiomyopathy Gene 2022 827 146470 10.1016/j.gene.2022.146470 35381313 Wilde AAM Amin AS Postema PG Diagnosis, management and therapeutic strategies for congenital long QT syndrome Heart 2022 108 332 338 10.1136/heartjnl-2020-318259 34039680 Winkler S Is 24/7 remote patient management in heart failure necessary? Results of the telemedical emergency service used in the TIM-HF and in the TIM-HF2 trials ESC Heart Fail 2021 8 3613 3620 10.1002/ehf2.13413 34182596	36481846	PMC9734484	NO-CC CODE	2022-12-14 23:28:29	no	Mamm Genome. 2022 Dec 8;:1-8	utf-8	Mamm Genome	2,022	10.1007/s00335-022-09972-x	oa_other
==== Front Appl Spat Anal Policy Appl Spat Anal Policy Applied Spatial Analysis and Policy 1874-463X 1874-4621 Springer Netherlands Dordrecht 9493 10.1007/s12061-022-09493-9 Article Linking Historical Discriminatory Housing Patterns to the Contemporary Alcohol Environment http://orcid.org/0000-0002-5130-8561 Sadler Richard Casey [email protected] 1 Wojciechowski Thomas Walter [email protected] 2 Trangenstein Pamela [email protected] 3 Harris Alan [email protected] 1 Buchalski Zachary [email protected] 1 Furr-Holden Debra [email protected] 1 1 grid.17088.36 0000 0001 2150 1785 Michigan State University, 200 E 1st St., Flint, MI 48502 USA 2 grid.17088.36 0000 0001 2150 1785 Michigan State University, East Lansing, MI USA 3 grid.417853.c 0000 0001 2106 6461 Alcohol Research Group, Emeryville, CA USA 6 12 2022 121 9 12 2021 16 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. Research on alcohol outlet density consistently shows greater disparities in exposure in disinvested communities. Likewise, structural racism via discriminatory housing practices has created many of the issues that beset contemporary disinvested neighborhoods. Little work, however, has examined the relationship between housing practices and alcohol outlet disparities. The central premise of our work is that these discriminatory and inequitable practices create distinctions in the alcohol environment, and that such disparities have implications for work on alcohol policy. Here we link alcohol outlet density with a spatial database examining redlining, blockbusting, and gentrification in Baltimore, Maryland, and Flint, Michigan (two cities with common experiences of urban disinvestment over the last 50 years). Standard measures are used to account for the impacts of neighborhood racial, socioeconomic, and housing composition in a multilevel model. Our findings highlight that gentrification and redlining are strongly associated with alcohol outlet density, while blockbusting is not. Gentrification and redlining also frequently co-occur in inner-urban areas, while the more suburban phenomenon of blockbusting rarely overlaps with either. These findings further contextualize nascent work on structural racism in housing that illustrates important disparities along the lines of these distinct practices. Future work should consider how legacy impacts of discriminatory housing patterns impact our communities today. Keywords Alcohol Food environments GIS Gentrification Blockbusting Redlining http://dx.doi.org/10.13039/100000027 National Institute on Alcohol Abuse and Alcoholism R21AA026674 Sadler Richard Casey ==== Body pmcBackground Alcohol outlet density (AOD) and the quality of the housing environment are correlated, especially insofar as alcohol retailers often target denser and poorer neighborhoods. This inequality—and the undue harms associated with higher AOD—represents a health equity issue (Marmot et al., 2008; Roche et al., 2015; Northridge & Freeman, 2011). Zoning is a tool to address health equity (Furr-Holden et al., 2020; Ransom et al., 2011; Hartnett, 1993), but historically has been used for the opposite purpose of perpetuating inequality (Manville et al., 2020; Massey & Rugh, 2017). Places that have experienced deliberate disinvestment in housing also frequently have higher AOD as a function of loosely enforced zoning codes and lack of political power to oppose such stores (Trangenstein et al., 2020a; Raleigh & Galster, 2015). Abandonment of homes and lower property values can drive many kinds of businesses out of neighborhoods, leaving small stores that make most of their sales from alcohol as the only remaining retailers (Singleton et al., 2017). This can be problematic for such neighborhoods, as higher AOD is also correlated to higher rates of addiction (Corburn, 2015), child abuse (Morton et al., 2014), robberies (Snowden & Freiburger, 2015), and, of course, alcohol consumption (Slutske et al., 2019; Schonlau et al., 2008) and alcohol-related harms (Connor et al., 2011). Despite community-level disinvestment being closely linked to AOD, to date no one has explored the connections between historical forms of structural racism via housing discrimination and AOD. In this paper, we link measures of the historical and contemporary housing environment to the contemporary alcohol environment. While redlining has become of increased interest to urban scholars (including in one study by Trangenstein et al., 2020b), ours is the first study to explicitly link multiple forms of housing discrimination to AOD. Below we outline several types of structural racism in housing that may influence AOD, building on our earlier work showing connections to food access (Sadler et al., 2021a). We then introduce our study communities of Baltimore, Maryland, and Flint, Michigan, and the measures we use here to indicate AOD. Housing Discrimination Impoverished, immigrant, and minoritized groups have been systematically isolated in under-resourced neighborhoods with environmental justice issues since long before formal techniques such as redlining existed (Cohn, 2009; Zunz, 1977; Deskins, 1972). As housing and the urban development process became more formalized and mass-produced in the early 20th century, discriminatory practices too became more formal. Redlining was one of the first explicit legal tools used by lenders to separate housing based on race and other characteristics. In particular, lenders used it to exclude racial minorities and other groups from obtaining loans in their neighborhoods and prevent them from living elsewhere (i.e. in predominately White neighborhoods) (Sugrue, 2014). Redlining exacerbated segregation and generated artificial neighborhood decline long before American cities were affected by deindustrialization (Aaronson et al., 2021; Swope & Hernandez, 2019; Hoalst-Pullen et al., 2011; Squires & Kubrin, 2005); indeed, it likely helped support the self-fulfilling prophecy of the neighborhood life cycle (Lang, 2000; Metzger, 2000)—the idea that neighborhoods have a predictable and inevitable tendency to decline over time, as new housing becomes available and older housing ‘filters’ to less and less in-demand land uses. Even today, areas that were previously redlined have lower property values and more problems with abandonment when compared to similar (but not redlined) neighborhoods (Appel & Nickerson, 2016), hence why scholars have studied its association to contemporary outcomes. A national Fair Housing Act was passed in 1968 that outlawed formal housing discrimination (Callies & Simon, 2017). But in its wake, other practices—such as blockbusting, racial steering, and exclusionary zoning—emerged to continue to uphold racial segregation in housing (Highsmith, 2009; Liebmann, 1990; Kmiec, 1986). Although it was illegal to directly discriminate against people based on their race, many remained effectively immobile, as these practices sought to reestablish racial dividing lines (Braveman et al., 2018). Blockbusting and White flight characterized the abandonment of previously all-White neighborhoods, but the movement of people back to earlier disinvested neighborhoods occurred simultaneously. This process—known as gentrification—contrasts with standard redevelopment because it displaces populations and alters a neighborhood’s social fabric (Zuk et al., 2015; Smith, 1982; Pattison, 1977). Gentrification in the US dovetails neatly with the progression of housing discrimination into new forms, whereas in other countries the effects of White flight, urban renewal, and other policies have been less pronounced (Ley, 1986). As noted above, practices such as redlining, blockbusting, and gentrification have been described as reflecting real consumer demands and used to justify the neighborhood life cycle theory (Metzger, 2000). And yet these deliberate practices undertaken by real estate agents, banks, and the government bely the idea that these processes of decline were inevitable (Aalbers, 2014; Lang, 2000). As we relate in our earlier work (Sadler et al., 2021a, b, quoting Aalbers, 2014, p. 527): “public and private actors…actively and passively structure the process of neighborhood decline, e.g. by producing maps that not only describe but also prescribe neighborhood decline”. That neighborhood decline has been intentionally provoked nationwide is as important as the recognition that AOD has likewise been shaped by deliberate actors seeking to maximize profits and prey on low-income and minority consumers. Alcohol Outlet Density Structural racism has had a long and enduring impact on neighborhoods via the housing market, but the inequality resulting from such practices as redlining, blockbusting, and gentrification also has interconnected impacts on the alcohol environment. In particular, there is good reason to believe that links exist between high concentrations of alcohol outlets and areas that had been redlined. Even so, research has not yet interrogated links between historical patterns of structural racism and AOD. Peripherally, much work has been done on the types of environments where AOD is highest. For example, lottery outlets are more common in poorer neighborhoods, with the authors noting commercial disinvestment may have been responsible for the shift in retailer type (Wiggins et al., 2010). In another paper, the authors explored how AOD moderated the relationship between crime and blight—two features well-known to exist in disinvested neighborhoods (Kajeepata et al., 2020). It’s also been made clear that patterns of disinvestment driving inequalities in AOD are the result of structural racism (Bieretz & Schilling, 2019). This disinvestment drives down economic opportunity for residents (Pietila, 2012), and the inverse of moving to neighborhoods with better conditions has been shown to improve similar outcomes (Linton et al., 2016). High AOD, therefore, is an added burden residents must face in their efforts to build healthy communities. Study Context and Rationale Our study communities are Baltimore, Maryland, and Flint, Michigan, two cities that have experienced considerable disinvestment and depopulation, and which therefore make for interesting case studies in studying links between patterns of housing discrimination and contemporary AOD. Baltimore was the first American city to enact a ‘segregation law’ over 100 years ago (Power, 1983), while Flint was long the nation’s most segregated city (Highsmith, 2009). Their shared experiences with blockbusting point to the potential existence of a pattern in that phenomenon in many cities; namely, that areas proximate to redlined neighborhoods were blockbusted first and worst, and that the impacts spread into suburban locales over time (Orser, 2015; Sadler et al., 2021a, b; Sadler & Lafreniere, 2017). And while Baltimore saw considerable gentrification starting in the 1990s (Wyly & Hammel, 2004), redevelopment in Flint has lagged more, leaving most neighborhoods still in a state of stagnation or decline. In Baltimore, attempts to address inequalities via an inclusionary housing ordinance have not yet yielded positive results (Brown, 2016). Meanwhile, Flint’s recently adopted master plan aims to redevelop Flint in an equitable manner, but may not be able to do enough to overcome years of White flight to surrounding suburban municipalities (Sadler et al., 2021). Both cities remain two of America’s most segregated (Massey & Tannen, 2015; Sadler & Highsmith, 2016). Inequalities in AOD has been of increasing interest in Flint, but has been studied extensively in Baltimore. Earlier work established inequalities in areas with high AOD, including higher rates of pedestrian injury (Nesoff et al., 2018, 2019), higher propensity for violent crime (Trangenstein et al., 2018; 2020c) and strong association with historical redlining (Trangenstein et al., 2020b). Policy-oriented work has shown that changes to the zoning code will potentially reduce AOD in the poorest neighborhoods (Hippensteel et al., 2019; Furr-Holden et al., 2020). The body of work in Flint, meanwhile, is older and less well developed, but still offers important insights into the environment there. In Goldstick et al. (2015), the authors found greater harms associated with both (a) off-premise alcohol outlets when compared to on-premise outlets and (b) greater density of any type of alcohol outlet. Earlier work by Resko et al. (2010), meanwhile, found that greater AOD was associated with higher odds of violent behavior among Flint youth. Our paper builds on work from multiple directions, including Sadler et al.’s (2021a, b) examination of food access and discriminatory housing practices, Furr-Holden et al.’s (2020, 2019) work on AOD in Baltimore, and the broader body of work on the effect of the built environment on contemporary health and economic outcomes. We also build on Trangenstein et al.’s (2020a) recommendation to use spatial access measures when defining AOD. As with our paper examining food access and discriminatory housing practices, this work presents new opportunities for those studying AOD. Researchers are well aware of the ways that the contemporary environment influences AOD. But here we elaborate on how these ‘legacy’ or long-term effects of structural racism may influence contemporary built environments. Our core pursuit here is to investigate how historical and contemporary aspects of structural racism can influence differential rates of AOD. Some new work has explored the association between structural racism and AOD (Scott et al., 2020; Trangenstein et al., 2020b), but more work is necessary to demonstrate specific links between different types of discriminatory practices and AOD. Our hypothesis is that AOD will be highest in blockbusted neighborhoods owing to patterns of disinvestment, while associations in gentrifying and redlined neighborhoods will be weaker but potentially still significant. Data Our data includes three measures of structural racism and/or inequitable housing practices—redlining, blockbusting, gentrification—as well as AOD. Redlining is from the University of Richmond’s Mapping Inequality Project (Nelson et al., 2020). The creators of this project digitized every 1930s era redlining map, including Baltimore and Flint. Although we acknowledge potential gaps between Home Owners’ Loan Corporation (HOLC) maps and the Federal Housing Administration’s (FHA’s) actual implementation of them, we still believe these maps should be interrogated for potential associations in the built environment (Fishback et al., 2020). Our blockbusting variable is measured by calculating the change in racial composition at the census tract level from one decade to another. We created this method in Flint and adapted more recently for Baltimore (Sadler & Lafreniere, 2017; Sadler et al., 2021a, b). Gentrification was defined by the National Community Reinvestment Coalition (NCRC) (Richardson et al., 2019), which is also based on changes in population from US census data. The data for the AOD measure comes from the state licensing databases for Michigan and Maryland. All outlets that prospectively sold alcohol were geocoded and categorized by type. Final geocoded shapefiles of all layers were then combined for further spatial analysis, to enable exploration of the alcohol landscapes in Flint and Baltimore. Methods Our central motivation was to use geographic information systems to connect spatial measures of AOD and housing discrimination. We used parcel centroids as our geographic and statistical unit of analysis, with the intention of determining the association between AOD and housing discrimination, controlling for age of housing, neighborhood socioeconomic distress, and neighborhood racial composition. We ran a series of diagnostic tests to ensure that our variables signified unique constructs and were acceptable for our multivariate model. Our multicollinearity check yielded acceptable VIFs (all < 2.5). Residuals were normally distributed, with a reasonable amount of homoscedasticity. We then used multilevel mixed effects modeling to assess the relevance of redlining, gentrification, and blockbusting for predicting AOD net of relevant sociodemographic controls. This method was chosen because of the nested nature of parcels within census tracts and the measurement of key characteristics at the census tract level of aggregation (blockbusting, gentrification, percent White, socioeconomic distress). Census tract intercepts were modeled as random effects to account for this nesting, whereas all other characteristics were modeled at the fixed effects level. The ‘identity’ option was chosen for modeling random effects, as this applied a shared variance parameter for random effects and no covariance parameter given that only intercepts were modeled for census tracts as random effects. Poisson regression was utilized because of the right-skewed nature of the AOD dependent variable. Coefficients were then interpreted as the expected effect that a one-unit change in a given independent variable has on the log of the expected count value on the AOD scale. Two models were estimated: one for Baltimore city and one for Genesee County. This was done because none of the parcels in the Genesee County data met the study definition for gentrification. As such, the models were estimated separately. We used Stata MP 16.1 to conduct these analyses. Alcohol Outlet Density After geocoding the location of all alcohol outlets, we ran the kernel density analysis tool in ArcGIS; each alcohol outlet serving as an unweighted point. Kernel density analysis creates a raster whose values are based on the density of features (in this case, alcohol outlets) around each cell. Given its continuous surface, it is also useful for visualizing areas that have relatively lower or higher AOD. We linked this raster surface to our unit of analysis using the Extract Values to Points tool, which performs a similar function to what a spatial join would do with vector data. Housing As noted, our housing variable (residential parcels) doubles as our unit of analysis. Our parcel datasets from the Baltimore City Open GIS (2020) and City of Flint (2016) included land use type and year of construction. For this paper, we isolated only the residential parcels, and appended AOD scores from above and the housing variables introduced below to their records. Redlining Redlining is based on the HOLC’s designation of neighborhoods in one of 4 classes (green = best, blue = still desirable, yellow = declining, red = hazardous). Since mortgages were restricted in redlined ‘hazardous’ neighborhoods, we focus especially on these areas. This resulted in a binary variable which delineated redlined and non-redlined neighborhoods (0 = not redlined; 1 = redlined). The model coefficients pertaining to this variable would then be indicative of whether or not individual housing parcels in redlined vs. non-redlined neighborhoods report greater average AOD scores. By using parcels as our unit of analysis and delineating the exact boundary of redlined neighborhoods, we avoid some of the methodological issues found in other papers; namely, the spatial mismatch and data quality reduction seen when joining redlining data to census tracts, ZIP codes, or other polygonal features (Li & Yuan, 2022; Li et al., 2021; Lynch et al., 2021; Nardone et al., 2020a, b; Rutan & Glass, 2018). Blockbusting We adapt the framework created by Sadler et al. (2021a, b) to define neighborhoods where blockbusting likely occurred. This entails calculating the percent change in the White population between census periods for the 3 decades from 1950 to 1980. Census tracts experiencing declines in the White population of more than 75% were deemed ‘extreme blockbusting’ and given a score of 3. Tracts with White population decline of between 50% and 74% were deemed ‘high blockbusting’ and given a score of 2. Neighborhoods where White flight was between 25% and 49% were deemed ‘moderate blockbusting’. These scores were summed across the three time periods, though no neighborhood saw such an extreme in more than two. The maximum sum blockbusting score was 3; neighborhoods were designated as having very high, high, or some blockbusting (scores of 3, 2, or 1). Gentrification Gentrification was derived using a method developed by the National Community Reinvestment Coalition (NCRC) (Richardson et al., 2019) measuring economic gentrification and population displacement. Eligible tracts fall below the 40th percentile in median home value for the region. Gentrified tracts have a median home value and percent of college-educated residents at or above the 60th percentile regionally, and an increase in median household income. Tracts are said to have experienced displacement if they lose 5% or more of any non-White racial or ethnic group. Sociodemographic Characteristics We are focused here on identifying the ongoing effects of historical structural racism, but we are also controlling for contemporary socioeconomic and racial composition. To simplify the influence of material and social deprivation, we have included a census block group-level socioeconomic distress index (originated by Pampalon et al., 2009) used in Flint and Baltimore (Sadler & Lafreniere, 2017; Sadler et al., 2021a, b). This index is calculated from the unweighted sum of the z-scores of 4 census variables: low educational attainment, low income (i.e. living below the poverty line), unemployment, and lone parent families. We also computed the percent of White population for each CBG in 2010 (since racial disparities in both housing and AOD persist), and control for the era of construction by using the age of the residence measured in the year the building was constructed (from the parcel dataset). Results In our results when speaking about Baltimore, we will refer to Baltimore Community Statistical Area names as contained in Baltimore’s Neighborhood Health Profile Reports (Baltimore City Health Department, 2017) (Fig. 1). Flint does not have formal neighborhood names, but we have appended a map highlighting neighborhood groups and colloquial names (Fig. 2). Fig. 1 Alcohol Outlet Density, Baltimore, Maryland Fig. 2 Alcohol Outlet Density, Flint & Genesee County, Michigan AOD Figures 1 and 2 highlight the results of the AOD kernel density analysis in Baltimore and Flint (Figs. 1 and 2). In Baltimore, the highest scores strongly follow the ‘White L’ of gentrifying and predominately White neighborhoods (e.g. Canton, Fells Point, Downtown, Federal Hill, Midtown, Greater Charles Village, Hampden), with some dense areas in more suburban areas (e.g. Southwest Baltimore, Sandtown-Winchester, Highlandtown, Pilimco/Arlington/Hilltop). Lower scores are generally found in outlying suburban neighborhoods, but also in the highly distressed neighborhoods of Jonestown/Oldtown and Perkins/Middle East. In Flint, the highest scores are downtown and in inner-urban neighborhoods on the northeast (Kearsley Park, Columbia Heights), south (Lincoln Park, Southside), and southwest (Westside) sides of town. Lower scores are generally found on the edge of the city limits and into the suburbs. Housing Given our central interest in the relationship of AOD and housing discrimination to other built environment characteristics, we also include variables for age of housing, socioeconomic distress, and racial composition. For both cities, we found an expected pattern of newer housing on the cities’ fringes. Distinct patterns of socioeconomic distress are also visible: in Flint, higher distress spreads through the middle and northern sections of the city. In Baltimore, meanwhile, two swaths are visible: one throughout most of west Baltimore, the other in a more concentrated area in east-central Baltimore. Racial composition follows a somewhat similar pattern to distress in both cities, and de facto ‘color lines’ exist separating African American from White populations in many places. To illustrate locations of the discriminatory housing variables, redlining, blockbusting, and gentrification are shown in Figs. 3 and 4. And to visually link these processes to AOD, we have also overlaid alcohol outlets by their primary serving method (on- or off-premise). Fig. 3 Discriminatory housing practices and alcohol outlets, Baltimore, Maryland Fig. 4 Discriminatory housing practices and alcohol outlets, Flint & Genesee County, Michigan Redlining In Baltimore and Flint alike, the oldest inner-most parts of the cities were redlined. Only the wealthiest, most well-established arewas areas received green or blue ratings. In Figs. 3 and 4, redlined areas can be found near industrial uses, railroads, rivers, and other less desirable areas (where investment was restricted until redlining’s abolition via the Fair Housing Act). Blockbusting Blockbusting was run for the time period 1950 to 1980 (Figs. 3 and 4). In Baltimore, blockbusting began in inner-suburban areas closer to the core, and followed toward the northwest and eventually to outer-suburban areas. In Flint, blockbusting followed a northwesterly direction, likewise starting closer to redlined neighborhoods and heading toward the northwest city limits (passing through Park Heights, King, Garfield Bunche, Flint Park Blvd, Merrill, and Manley Village/Flint Park Lake). In Baltimore, blockbusted neighborhoods make up most of west Baltimore (including Dorchester/Ashburton, Southern Park Heights, Greater Rosemont, and Edmonson Village), as well as parts of east and northeast Baltimore (including Clifton-Berea, Midway/Coldstream, Northwood, and Claremont/Armistead). Our general hypothesis is that AOD could be higher in these areas because of the recency and severity of disinvestment characterized by blockbusting and White flight. This delineation is important for understanding the impacts of White flight, as not every predominately African American neighborhood was blockbusted (nor did every neighborhood see White flight), thus the ongoing experience of structural racism could be felt differently depending on the neighborhood. Gentrification Our final housing related variable was gentrification (in yellow on Fig. 3, only for Baltimore). To accord with the NCRC definition, we computed areas where displacement occurred and median income increased. Gentrified areas nearly perfectly mirror the ‘White L’, from Hampden in the north, into downtown, and east toward Canton. We also expect gentrified areas to have higher AOD scores. Housing Dynamics Figures 3 and 4 reveal some key patterns about housing dynamics. Notably, few neighborhoods in either city have experienced two or more of these phenomena. Blockbusted neighborhoods are almost never already gentrified, and redlined areas were almost never blockbusted (owing to the long time horizon involved in disinvestment and reinvestment). Redlined areas in Baltimore, by contrast, have seen gentrification in recent years. A critical point to consider, therefore, is that disparities may exist between (a) neighborhoods that were redlined but are now gentrifying and (b) neighborhoods that were blockbusted and still experience disinvestment. Linking the Alcohol Environment Results from the Baltimore city data (Table 1) indicated that parcels located within redlined areas had significantly greater AOD compared to parcels not located within redlined areas (Coefficient = 0.224). Parcels in gentrified areas also had significantly greater AOD (Coefficient = 0.872). Blockbusting scores at the census tract level did not significantly predict AOD. Census tracts with greater socioeconomic distress scores and greater proportions of White residents also had significantly greater AOD in the Baltimore city data. Age of housing was not a significant predictor of AOD in this model. Table 1 Mixed effects poisson regression modeling of covariate effects on Alcohol Outlet Density (AOD) nested within census tracts: Baltimore City (N = 193,147) Coefficient 95% Confidence interval p-Value Fixed effects model Redlined (0 = No; 1 = Yes) 0.224 0.157 0.291 p < .001 Blockbusting 0.023 − 0.163 0.208 0.811 Gentrification 0.872 0.528 1.215 p < .001 Parcel year of construction >-0.001 >-0.001 < 0.001 0.568 Socioeconomic distress 0.191 0.117 0.265 p < .001 Percentage of white residents 2.546 1.618 3.473 p < .001 Constant -4.126 -4.540 -3.713 p < .001 Random effects model Constant variance 1.418 Results from the Genesee County data (Table 2) indicated that parcels within redlined areas had significantly greater AOD (Coefficient = 0.214). Blockbusted neighborhoods were not significant predictors of AOD. Census tracts with greater socioeconomic distress and lower proportions of White residents also had significantly greater AOD in the Genesee County data. Age of housing was not a significant predictor of AOD score in this model either. Table 2 Mixed effects poisson regression modeling of covariate effects on Alcohol Outlet Density (AOD) nested within census tracts: Genesee County (N = 149,196) Coefficient 95% Confidence Interval p-Value Fixed effects model Redlined (0 = No; 1 = Yes) 0.214 0.184 p < .001 0.244 Blockbusting − 0.165 − 0.608 0.465 0.278 Parcel year of construction >-0.001 >-0.001 0.270 < 0.001 Socioeconomic distress 0.122 0.035 0.006 0.208 Percentage of white residents -1.260 -2.433 0.035 − 0.087 Constant − 0.218 -1.151 0.647 0.716 Random effects model Constant variance 1.133 Sensitivity analyses omitted the gentrification variable from the Baltimore City model to see if findings were robust. This was done because gentrification data was unavailable for Genesee County, so this additional model estimation would provide indication of how omission of these data may have influenced findings. These results were robust, essentially indicating analogous findings from the main analyses for Baltimore City even when the gentrification variable was omitted. Discussion While Trangenstein et al. (2020b) offered the first glimpse into the association between redlining and AOD, the need to examine further spatial patterns of historical and contemporary housing discrimination in relation to AOD remains. Our study is the first to use multiple established designations to investigate the association to AOD, and builds on the growing body of work linking historical structural racism to contemporary health determinants and outcomes. Literature from the urban development and health equity nexus has recently worked more in the realm of redlining or environmental racism more broadly (Beyer et al., 2016; Krieger et al., 2020; Nardone et al., 2020a, b). One recent review highlighted 12 studies examining redlining and health, and conveyed a number of important findings, including an increased risk of pre-term birth across 3 studies alongside health disparities in domains of asthma, cancer, COVID-19, gun-related injuries, chronic diseases, and heat-related diseases (Lee et al., 2021). Yet even that study oversimplified the complex and ongoing process of historical and contemporary structural racism in housing, jumping from the passage of the Fair Housing Act to contemporary health studies. The continued use of Sadler and Lafreniere’s (2017) blockbusting measure is therefore an important continued contribution, as it has only been used in the context of food environments (Sadler et al., 2021a, b). We discovered no such association between blockbusting and AOD, suggesting a potential unintended positive of living in disinvested communities: that overexposure to AOD may be less pronounced in these generally suburban locales. While blockbusting measures the impact of more recent disinvestment patterns, it may also signify places that were never as dense as redlined communities, meaning exposure to deleterious business uses may well be lower. More importantly, this is the first study to explicitly link these patterns with AOD. By linking these important housing measures to a tangible health determinant (e.g. alcohol outlet exposure), we bring to bear the potential negative health outcomes of structural racism in housing. Repeating such work in other cities and with additional forms of historical structural racism are important future directions for research. Our key major findings include: (1) redlining and gentrification are key examples of housing discrimination that shapes higher AOD; (2) as with our previous work, blockbusting and gentrification are often mutually exclusive, and gentrifying neighborhoods therefore often have higher AOD; (3) blockbusting was not a major predictor of high AOD, possibly owing to the recency of business disinvestment, including alcohol outlets. The overall pattern of higher AOD in redlined neighborhoods is noteworthy, and may reflect coincidence with gentrification (i.e. recent reinvestment in these places and the proliferation of bars and nightclubs). Conversely, the historical injustices perpetrated on redlined neighborhoods that are not gentrifying may yet be negatively impacting low-income and minoritized populations (i.e. in the form of higher rates of liquor stores). Continued advocacy to address predatory lending and other insidious forms of housing discrimination and structural racism remain important given these new findings (Mock, 2015). Policy Implications Our findings should influence advocacy around alcohol outlet exposure and health equity. The processes that have made obtaining adequate and fair housing difficult or impossible for minoritized populations are a key determinant of contemporary health disparities (Bailey et al., 2017; Ramaswamy & Kelly, 2015; Osypuk & Acevedo-Garcia, 2010). Redlining has been hugely important in shaping African American health disparities over the last hundred years, and our study suggests that such neighborhoods may still be most disadvantaged when it comes to AOD. By not giving enough consideration to how neighborhoods were tragically harmed by way of disinvestment, the modern process of gentrification may wind up repeating the disinvestment patterns seen elsewhere in our cities, thereby making equitable redevelopment even more difficult. We continue to advocate that these structural racism-related variables be used in future research. Doing so would enable decision-makers and advocates to make more well-informed decisions based on how disinvestment patterns might negatively impact the proliferation of the alcohol environment. Coordination of efforts around limiting alcohol outlet exposure and providing new evidence such as that presented here is essential to effectively translating research findings into more equitable urban development patterns. Limitations We acknowledge a few limitations. First, we highlighted just two cities, so it is not clear whether these patterns would exist or persist elsewhere. We continue to recommend further inquiry into these patterns across the US by replicating these methods, including with larger datasets for which health determinants or outcomes are available (i.e. the CDC’s PLACES Project) (CDC, 2021). Second, we acknowledge the limitations with using only a kernel-density weighted measure of AOD, as we cannot infer the price or availability of alcohol types within the stores. Future work should seek to link objective, in-store assessments of the alcohol environment to our housing measures. Third, we acknowledge that age of residents could be a relevant sociodemographic characteristic to consider as a confound. Prior research has indicated that young adults are at elevated risk for alcohol use (Grucza et al., 2018; White, 2020). As such, communities with larger proportions of individuals in this age range may have greater AOD density to satiate increased demand for alcohol. That said, data pertaining to this characteristic was unavailable, thus, these analyses were beyond the scope of this study. This indicates the need for future research to investigate the potential that average resident age within neighborhoods may act as a confounder and determine the robustness of these results. Conclusion With this and ongoing work, we are identifying, measuring, and raising up the influence of a variety of forms of structural racism in the built environment (i.e. beyond redlining). We hope this work will be valuable for a range of researchers and policy advocates, including those working toward eliminating disparities in housing and alcohol environments. As each of these are closely linked to issues of health equity, we likewise hope that future attention is given to additional aspects of the built environment that could be linked to these practices; that is, how multiple forms of housing discrimination can continue to have deleterious effects on communities. We believe that blockbusting, as well as other measurable forms of discrimination that are heretofore not commonly measured, form important next steps for researchers to consider. As with our earlier work on food access, we continue to advocate for the consideration of this and other variables in the study of contemporary alcohol environments as well as other topics. We hope this propels health equity research meaningfully forward into studying more underlying determinants beyond the standard measures. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Aalbers, M. B. (2014). Do maps make geography? Part 1: Redlining, planned shrinkage, and the places of decline. ACME: An International Journal for Critical Geographies, 13(4), 525–556. Aaronson D Hartley D Mazumder B The effects of the 1930s HOLC” redlining” maps American Economic Journal: Economic Policy 2021 13 4 355 392 Appel, I., & Nickerson, J. (2016). Pockets of poverty: The long-term effects of redlining. Available at SSRN2852856. Bailey ZD Krieger N Agénor M Graves J Linos N Bassett MT Structural racism and health inequities in the USA: evidence and interventions The Lancet 2017 389 10077 1453 1463 10.1016/S0140-6736(17)30569-X Baltimore City (2020). Baltimore City Open GIS. Retrieved from: http://gis-baltimore.opendata.arcgis.com/. Accessed 30 Oct 2021. Baltimore City Health Department (2017). Neighborhood health profile reports. Retrieved from: https://health.baltimorecity.gov/neighborhoods/neighborhood-health-profile-reports. Accessed 30 Oct 2021. Beyer KM Zhou Y Matthews K Bemanian A Laud PW Nattinger AB New spatially continuous indices of redlining and racial bias in mortgage lending: links to survival after breast cancer diagnosis and implications for health disparities research Health & Place 2016 40 34 43 10.1016/j.healthplace.2016.04.014 27173381 Bieretz, B., & Schilling, J. (2019). Pay for success and blighted properties. Urban Institute, 22. Braveman P Arkin E Orleans T Proctor D Acker J Plough A What is health equity? Behavioral Science & Policy 2018 4 1 1 14 10.1353/bsp.2018.0000 Brown, L. (2016). Two Baltimores: The White L Vs. The Black Butterfly, In the Baltimore Sun, June 28. Retrieved from: https://www.baltimoresun.com/citypaper/bcpnews-two-baltimores-the-white-l-vs-the-black-butterfly-20160628-htmlstory.html. Accessed 30 Oct 2021. Callies DL Simon DB Fair housing, discrimination and inclusionary zoning in the United States Journal of International & Comparative Law 2017 4 39 Centers for Disease Control and Prevention (CDC) (2021). PLACES project. Local data for better health. Retrieved from: https://www.cdc.gov/places City of Flint (2016). City parcel shapefile. Flint, Michigan. Cohn A Fair housing testing The Urban Lawyer 2009 41 273 Connor JL Kypri K Bell ML Cousins K Alcohol outlet density, levels of drinking and alcohol-related harm in New Zealand: a national study Journal of Epidemiology and Community Health 2011 65 10 841 846 10.1136/jech.2009.104935 20947871 Corburn J City planning as preventive medicine Preventive Medicine 2015 77 48 51 10.1016/j.ypmed.2015.04.022 25937591 Deskins DR Jr Race, residence, and workplace in Detroit, 1880 to 1965 Economic Geography 1972 48 1 79 94 10.2307/143017 Fishback, P. V., LaVoice, J., Shertzer, A., & Walsh, R. (2020). Race, risk, and the emergence of Federal Redlining. National Bureau of Economic Research. No. w28146. Furr-Holden CDM Milam AJ Nesoff ED Linton S Reboussin B Sadler RC Leaf PJ Using zoning as a public health tool to reduce alcohol outlet oversaturation, promote compliance, and guide future enforcement: a preliminary analysis of Transform Baltimore Journal of Urban Health 2020 97 4 568 582 10.1007/s11524-020-00453-7 32632795 Furr-Holden CDM Nesoff ED Nelson V Milam AJ Smart M Lacey K Leaf PJ Understanding the relationship between alcohol outlet density and life expectancy in Baltimore City: the role of community violence and community disadvantage Journal of Community Psychology 2019 47 1 63 75 10.1002/jcop.22099 30506926 Goldstick JE Brenner AB Lipton RI Mistry R Aiyer SM Reischl TM Zimmerman MA A spatial analysis of heterogeneity in the link between alcohol outlets and assault victimization: differences across victim subpopulations Violence and Victims 2015 30 4 649 662 10.1891/0886-6708.VV-D-13-00187 26159914 Grucza RA Sher KJ Kerr WC Krauss MJ Lui CK McDowell YE Bierut LJ Trends in adult alcohol use and binge drinking in the early 21st-century United States: a meta‐analysis of 6 National Survey Series Alcoholism: Clinical and Experimental Research 2018 42 10 1939 1950 10.1111/acer.13859 30080258 Hartnett, J. J. (1993). Affordable housing, exclusionary zoning, and American apartheid: Using title VIII to foster statewide racial integration. New York University Law Review, 68, 89. Highsmith AR Demolition means progress: urban renewal, local politics, and state-sanctioned ghetto formation in Flint, Michigan Journal of Urban History 2009 35 3 348 368 10.1177/0096144208330403 Hippensteel CL Sadler RC Milam AJ Nelson V Furr-Holden CD Using zoning as a public health tool to reduce oversaturation of alcohol outlets: an examination of the effects of the New “300 foot Rule” on packaged goods stores in a Mid-Atlantic City Prevention Science 2019 20 6 833 843 10.1007/s11121-018-0947-9 30284159 Hoalst-Pullen N Patterson MW Gatrell JD Empty spaces: Neighbourhood change and the greening of Detroit, 1975–2005 Geocarto International 2011 26 6 417 434 10.1080/10106049.2011.585439 Kajeepeta S Theall KP Kondo MC Branas CC Wallace ME Jacoby SF Morrison CN The association between blighted property remediation and domestic crime by alcohol availability Health & Place 2020 64 102364 10.1016/j.healthplace.2020.102364 32838890 Kmiec, D. W. (1986). Exclusionary zoning and purposeful racial segregation in housing: Two wrongs deserving separate remedies. The Urban Lawyer, 18(2), 393–422. Krieger, N., Wright, E., Chen, J. T., Waterman, P. D., Huntley, E. R., & Arcaya, M. (2020). Cancer stage at diagnosis, historical redlining, and current neighborhood characteristics: Breast, cervical, lung, and colorectal cancer, Massachusetts, 2001–2015. American Journal of Epidemiology, 189(10), 1065–1075. Lang RE Did neighborhood life-cycle theory cause urban decline? Housing Policy Debate 2000 11 1 1 6 10.1080/10511482.2000.9521358 Lee, E. K., Donley, G., Ciesielski, T. H., Yamoah, O., Roche, A., Martinez, R., & Freedman, D. A. (2021). Health outcomes in redlined versus non-redlined neighborhoods: A systematic review and meta-analysis. Social Science & Medicine, 294, 114696. Ley D Alternative explanations for inner-city gentrification: a canadian assessment Annals of the Association of American Geographers 1986 76 4 521 535 10.1111/j.1467-8306.1986.tb00134.x Li, D., Newman, G. D., Wilson, B., Zhang, Y., & Brown, R. D. (2021). Modeling the relationships between historical redlining, urban heat, and heat-related emergency department visits: An examination of 11 Texas cities. Environment and Planning B: Urban Analytics and City Science, 49(3), 933–952. Li M Yuan F Historical redlining and resident exposure to COVID-19: a study of New York City Race and Social Problems 2022 14 2 85 100 10.1007/s12552-021-09338-z 34178163 Liebmann GW Suburban zoning-two modest proposals Real Property Probate and Trust Journal 1990 25 1 Linton SL Cooper HL Luo R Karnes C Renneker K Haley DF Rothenberg R People and places: relocating to neighborhoods with better economic and social conditions is associated with less risky drug/alcohol network characteristics among african american adults in Atlanta, GA Drug and Alcohol Dependence 2016 160 30 41 10.1016/j.drugalcdep.2015.11.036 26781062 Lynch EE Malcoe LH Laurent SE Richardson J Mitchell BC Meier HC The legacy of structural racism: Associations between historic redlining, current mortgage lending, and health SSM-Population Health 2021 14 100793 10.1016/j.ssmph.2021.100793 33997243 Manville M Monkkonen P Lens M It’s time to end single-family zoning Journal of the American Planning Association 2020 86 1 106 112 10.1080/01944363.2019.1651216 Marmot M Friel S Bell R Houweling TA Taylor S Commission on Social Determinants of Health Closing the gap in a generation: health equity through action on the social determinants of health The Lancet 2008 372 9650 1661 1669 10.1016/S0140-6736(08)61690-6 Massey, D. S., & Rugh, J. S. (2017). Zoning, affordable housing, and segregation in US metropolitan areas. Ch. 14 (pp. 245–264) in The fight for fair housing: Causes, consequences, and future implications of the 1968 Federal Fair Housing Act, ed. G.D. Squires. Routledge. Massey DS Tannen J A research note on trends in black hypersegregation Demography 2015 52 3 1025 1034 10.1007/s13524-015-0381-6 25791615 Metzger JT Planned abandonment: the neighborhood life-cycle theory and national urban policy Housing Policy Debate 2000 11 1 7 40 10.1080/10511482.2000.9521359 Mock, B. (2015). Redlining is Alive and Well—and Evolving. CityLab. Retrieved from: https://www.citylab.com/equity/2015/09/redlining-is-alive-and-welland-evolving/407497/. Accessed 30 Oct 2021. Morton CM Simmel C Peterson NA Neighborhood alcohol outlet density and rates of child abuse and neglect: moderating effects of access to substance abuse services Child Abuse & Neglect 2014 38 5 952 961 10.1016/j.chiabu.2014.01.002 24529493 Nardone, A. L., Casey, J. A., Rudolph, K. E., Karasek, D., Mujahid, M., & Morello-Frosch, R. (2020a). Associations between historical redlining and birth outcomes from 2006 through 2015 in California. PLoS One, 15(8), e0237241. Nardone A Chiang J Corburn J Historic redlining and urban health today in US cities Environmental Justice 2020 13 4 109 119 10.1089/env.2020.0011 Nelson, R. K., Winling, L., Marciano, R., Connolly, N., et al. (2020). Mapping inequality, redlining in New Deal America. Retrieved from: https://dsl.richmond.edu/panorama/redlining/. Accessed 30 Oct 2021. Nesoff ED Milam AJ Branas CC Martins SS Knowlton AR Furr-Holden DM Alcohol outlets, neighborhood retail environments, and pedestrian injury risk Alcoholism: Clinical and Experimental Research 2018 42 10 1979 1987 10.1111/acer.13844 30102415 Nesoff ED Milam AJ Pollack KM Curriero FC Bowie JV Knowlton AR Furr-Holden DM Neighbourhood alcohol environment and injury risk: a spatial analysis of pedestrian injury in Baltimore City Injury Prevention 2019 25 5 350 356 10.1136/injuryprev-2018-042736 29588410 Northridge ME Freeman L Urban planning and health equity Journal of Urban Health 2011 88 3 582 597 10.1007/s11524-011-9558-5 21365355 Orser, W. E. (2015). Blockbusting in Baltimore: the Edmondson Village Story. University Press of Kentucky. Osypuk TL Acevedo-Garcia D Beyond individual neighborhoods: a geography of opportunity perspective for understanding racial/ethnic health disparities Health & Place 2010 16 6 1113 1123 10.1016/j.healthplace.2010.07.002 20705500 Pampalon R Hamel D Gamache P Raymond G A deprivation index for health planning in Canada Chronic Diseases in Canada 2009 29 4 178 191 10.24095/hpcdp.29.4.05 19804682 Pattison, T. J. (1977). The process of neighborhood upgrading and gentrification: an examination of two neighborhoods in the Boston metropolitan area (Doctoral dissertation, Massachusetts Institute of Technology). Pietila, A. (2012). Not in my neighborhood: how bigotry shaped a great american city. Rowman & Littlefield. Power G Apartheid Baltimore style: the residential segregation ordinances of 1910–1913 Maryland Law Review 1983 42 289 Raleigh E Galster G Neighborhood disinvestment, abandonment, and crime dynamics Journal of Urban Affairs 2015 37 4 367 396 10.1111/juaf.12102 Ramaswamy M Kelly PJ Institutional racism as a critical social determinant of health Public Health Nursing 2015 32 4 285 286 10.1111/phn.12212 26199054 Ransom MM Greiner A Kochtitzky C Major KS Pursuing health equity: zoning codes and public health The Journal of Law Medicine and Ethics 2011 39 1_suppl 94 97 10.1111/j.1748-720X.2011.00576.x Resko SM Walton MA Bingham CR Shope JT Zimmerman M Chermack ST Cunningham RM Alcohol availability and violence among inner-city adolescents: a multi-level analysis of the role of alcohol outlet density American Journal of Community Psychology 2010 46 3 253 262 10.1007/s10464-010-9353-6 20857328 Richardson, J., Mitchell, B., & Franco, J. (2019). Shifting neighborhood: Gentrification and cultural displacement in American Cities. National Community Reinvestment Coalition. Retrieved from: https://ncrc.org/gentrification/. Accessed 30 Oct 2021. Roche A Kostadinov V Fischer J Nicholas R O’Rourke K Pidd K Trifonoff A Addressing inequities in alcohol consumption and related harms Health Promotion International 2015 30 suppl_2 ii20 ii35 10.1093/heapro/dav030 26420810 Rutan DQ Glass MR The lingering effects of neighborhood appraisal: evaluating redlining’s legacy in Pittsburgh The Professional Geographer 2018 70 3 339 349 10.1080/00330124.2017.1371610 Sadler RC Bilal U Furr-Holden CD Linking historical discriminatory housing patterns to the contemporary food environment in Baltimore Spatial and Spatio-temporal Epidemiology 2021 36 100387 10.1016/j.sste.2020.100387 33509435 Sadler, R. C., Furr-Holden, D., Greene-Moton, E., Larkin, B., Timlin, M., Walling, D., & Wyatt, T. (2021b). Right-Sizing Flint’s infrastructure in the wake of the Flint water crisis would constitute an additional environmental injustice. Journal of the American Planning Association, 87(3), 424–432. Sadler RC Highsmith AR Rethinking Tiebout: the contribution of political fragmentation and racial/economic segregation to the Flint water crisis Environmental Justice 2016 9 5 143 151 10.1089/env.2016.0015 Sadler RC Lafreniere DJ Racist housing practices as a precursor to uneven neighborhood change in a post-industrial city Housing Studies 2017 32 2 186 208 10.1080/02673037.2016.1181724 Schonlau M Scribner R Farley TA Theall KP Bluthenthal RN Scott M Cohen DA Alcohol outlet density and alcohol consumption in Los Angeles county and southern Louisiana Geospatial Health 2008 3 1 91 10.4081/gh.2008.235 19021112 Scott J Danos D Collins R Simonsen N Leonardi C Scribner R Herd D Structural racism in the built environment: segregation and the overconcentration of alcohol outlets Health & Place 2020 64 102385 10.1016/j.healthplace.2020.102385 32755812 Singleton CR Li Y Duran AC Zenk SN Odoms-Young A Powell LM Food and beverage availability in small food stores located in healthy food financing initiative eligible communities International Journal of Environmental Research and Public Health 2017 14 10 1242 10.3390/ijerph14101242 29057794 Slutske WS Deutsch AR Piasecki TM Neighborhood alcohol outlet density and genetic influences on alcohol use: evidence for gene–environment interaction Psychological Medicine 2019 49 3 474 10.1017/S0033291718001095 29730997 Smith N Gentrification and uneven development Economic Geography 1982 58 2 139 155 10.2307/143793 Squires GD Kubrin CE Privileged places: race, uneven development and the geography of opportunity in urban America Urban Studies 2005 42 1 47 68 10.1080/0042098042000309694 Snowden AJ Freiburger TL Alcohol outlets, social disorganization, and robberies: accounting for neighborhood characteristics and alcohol outlet types Social Science Research 2015 51 145 162 10.1016/j.ssresearch.2015.01.011 25769858 Sugrue, T. J. (2014). The Origins of the Urban Crisis: race and inequality in Postwar Detroit. Princeton University Press. Swope CB Hernández D Housing as a determinant of health equity: a conceptual model Social Science & Medicine 2019 243 112571 10.1016/j.socscimed.2019.112571 31675514 Trangenstein PJ Curriero FC Webster D Jennings JM Latkin C Eck R Jernigan DH Outlet type, access to alcohol, and violent crime Alcoholism: Clinical and Experimental Research 2018 42 11 2234 2245 10.1111/acer.13880 30256427 Trangenstein, P. J., Sadler, R. C., Morrison, C. N., & Jernigan, D. H. (2020a). Looking back and moving forward: the evolution and potential opportunities for the future of alcohol outlet density measurement. Addiction Research & Theory, 29(2), 117–128. Trangenstein PJ Gray C Rossheim ME Sadler R Jernigan DH Alcohol outlet clusters and population disparities Journal of Urban Health 2020 97 1 123 136 10.1007/s11524-019-00372-2 31264024 Trangenstein PJ Eck RH Lu Y Webster D Jennings JM Latkin C Jernigan DH The violence prevention potential of reducing alcohol outlet access in Baltimore, Maryland Journal of Studies on Alcohol and Drugs 2020 81 1 24 33 10.15288/jsad.2020.81.24 32048598 White, A. M. (2020). Gender differences in the epidemiology of alcohol use and related harms in the United States. Alcohol Research: Current Reviews, 40(2), 1–13. Wiggins L Nower L Mayers RS Peterson NA A geospatial statistical analysis of the density of lottery outlets within ethnically concentrated neighborhoods Journal of Community Psychology 2010 38 4 486 496 10.1002/jcop.20376 Wyly EK Hammel DJ Gentrification, segregation, and discrimination in the american urban system Environment and Planning A 2004 36 7 1215 1241 10.1068/a3610 Zuk, M., Bierbaum, A. H., Chapple, K., Gorska, K., Loukaitou-Sideris, A., Ong, P., & Thomas, T. (2015). Gentrification, displacement and the role of public investment: a literature review. In Federal Reserve Bank of San Francisco (Vol. 79). Zunz O The organization of the american city in the late nineteenth century: ethnic structure and spatial arrangement in Detroit Journal of Urban History 1977 3 4 443 466 10.1177/009614427700300405	0	PMC9734485	NO-CC CODE	2022-12-14 23:28:29	no	Appl Spat Anal Policy. 2022 Dec 6;:1-21	utf-8	Appl Spat Anal Policy	2,022	10.1007/s12061-022-09493-9	oa_other
==== Front Pharm Chem J Pharm Chem J Pharmaceutical Chemistry Journal 0091-150X 1573-9031 Springer US New York 2782 10.1007/s11094-022-02782-0 Article Ethionamide and Prothionamide Based Coumarinyl-Thiazole Derivatives: Synthesis, Antitubercular Activity, Toxicity Investigations and Molecular Docking Studies Imran Mohd [email protected] grid.449533.c 0000 0004 1757 2152 Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Northern Border University, Rafha, Saudi Arabia 7 12 2022 111 1 6 2021 © 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 goal of this research work was to prepare and evaluate the antitubercular (anti-TB) activity of ethionamide (ETH) and prothionamide (PTH) based coumarinyl-thiazole derivatives. ETH and PTH were reacted with coumarin intermediates (3a-3e) to provide the target compounds (4a-4e and 4f-4j, respectively). Spectral studies confirmed the assigned structures of 4a-4j. The Microplate Alamar Blue Assay was utilized to evaluate the anti-TB activity of compounds 4a-4j against Mycobacterium tuberculosis H37Rv strain in comparison to ETH, PTH, isoniazid (INH), and pyrazinamide (PYZ) as standard drugs. The cytotoxicity studies were carried out versus HepG2 and Vero cell lines. In addition. molecular docking studies of 4a-4j concerning the DprE1 enzyme and the in-silico evaluation of physicochemical and pharmacokinetic parameters were performed. Compounds 4a, 4b, 4f, and 4g displayed equal minimum inhibitory concentration (MIC) values in comparison to INH (3.125 μg/ml) and PYZ (3.125 μg/ml), whereas 4c-4e and 4h-4j displayed better MIC values (1.562 μg/mL) than INH and PYZ. All compounds presented better anti-TB potential than ETH (6.25 μg/mL) and PTH (6.25 μg/mL). The studies of toxicity revealed that 4a-4j were safe up to 300 μg/mL concentration versus Vero and HepG2 cell lines. The molecular docking studies suggested that 4a-4j could possess anti-TB activity through the inhibition of the DprE1 enzyme. The in silico studies showed that 4a-4j followed Lipinski’s rule (drug-likeliness) and exhibited better gastrointestinal absorption than BTZ043 and macozinone. In conclusion, the ETH and PTH-based coumarinyl-thiazole template can help developing selective DprE1 enzyme inhibitors as potent anti-TB agents. Keywords synthesis coumarin thiazole antitubercular activity cytotoxicity molecular docking ==== Body pmcIntroduction Tuberculosis (TB), a global concern, is instigated through Mycobacterium tuberculosis (Mtb) bacteria. This communicable illness is among the top ten bases of mortality, and the principal reason of mortality from this particular contagious microorganism. The World Health Organization (WHO) global tuberculosis report of 2020 declared that about ten million people suffered from TB in 2019, and about 1.2 million and 0.208 million HIV-negative and HIV-positive people, respectively, died in 2019. The recent report also states that the number of TB patients is expected to increase by 0.2 – 0.4 million in 2020 due to COVID-19 [1]. Conventional treatment of TB comprises the use of isoniazid (INH), rifampin, ethambutol, pyrazinamide (PYZ), pretomanid, and bedaquiline along with other anti-tubercular (anti-TB) drugs for at least 6 – 9 months. However, many TB treatments can cause hepatotoxicity in chronic use. The rise of multi-drug resistant TB (MDR-TB), totally drug-resistant TB (TDR-TB), and extensive drug-resistant TB (XDR-TB) is also posing questions to TB treatment. This situation necessitates additional research into new molecular frameworks that may tackle these difficulties with insignificant adverse impacts [1, 2]. In the pursuit of innovative and safe anti-TB treatment, researchers from all over the world have experimented with several molecular scaffolds of various designs [3]. The research work on coumarin [4] and thiazole [5] scaffolds has provided some new promising anti-TB agents. Coumarinthiazole based compounds (Fig. 1) have also been reported as potent anti-TB agents [6, 7]. Ethionamide (ETH) and prothioanamide (PTH) are thioamide groups containing second-line anti-TB drugs, which are utilized as a replacement for INH and rifampin in the case of MDR-TB. The structure of ETH and PTH is analogous to INH (Fig. 2). However, ETH and PTH are relatively weak anti-TB agents as compared to INH. This leads to an increase in the treatment duration (up to two years) and the probability of increased side effects [8]. The literature teaches that the thioamide group of ETH and PTH can cyclize to thiazole derivatives after reaction with phenyl acyl bromide, for example, with 3-coumarinylacetyl bromide [6, 7, 9]. Therefore, in continuation of our endeavor to develop anti-TB compounds [10, 11] and because of the challenges associated with TB [1, 2], the author decided to develop ETH and PTH-based coumarinylthiazole derivatives as anti-TB agents.Fig. 1. Design of the compounds A and B: The structure of coumarin-thiazole based compounds as anti-TB agents reported in reference [6, 7], respectively. 4a-4j: The structure of the synthesized compounds as anti-TB agents. Fig. 2. Chemical structures of isoniazid (INH), ethionamide (ETH), and prothionamide (PTH). Experimental Chemical Part Chemicals and Instruments Compounds ETH, PTH, INH, and PYZ were purchased from Sigma Aldrich (USA). Analytical grade solvents (Sigma, Spectrochem, and Merck) were used during the synthesis of target compounds. A mixture of formic acid, ethyl acetate, and toluene (1:4:5) was utilized to establish the Rf values. The uncorrected melting points (m.p.) in °C (Gallenkamp apparatus), FTIR spectra in KBr (Shimadzu 440 spectrometer), 1H and 13C NMR spectra in DMSO-d6 (Varian Gemini 500/125 MHz spectrometer), mass spectra in m/z (GCMS/QP 1000 Ex mass spectrometer, 70 eV), and elemental analyses (Vario El Elementar apparatus) were recorded on the instruments stated in parentheses. Synthesis of Intermediates (3a-3e) The preparation of intermediates (3a-3e) from salicyldehydes (1a-1e) and ethyl acetoacetate (2) has been reported in our earlier publication [9] (Scheme 1).Scheme 1. Synthesis of intermediates (3a-3e) Synthesis of 3-(2-(2-Ethylpyridin-4-yl)thiazol-4-yl)-2H-chromen-2-one (4a) Equimolar amounts of 3a (0.01 mol) and ETH (0.01 mol) in absolute ethanol (40 mL) were mixed and refluxed for 2 h. A solid precipitate formed during reflux was filtered without cooling to obtain intermediate 4a (yellowish-brown crystal). Yield: 80%; m.p. 185 – 187°C; Rf 0.79; IR (νmax, cm-1): 1718 (C=O), 1568 (C=N), 1527 (C=C), 1258 (C–O–C), 1111 (C–S); 1H NMR (, ppm): 1.28 (t, J = 7 Hz, 3H, -CH3), 3.38 (q, J = 4.5 Hz, 2H, -CH2-), 7.36 – 7.42 (m, 2H, Ar-H), 7.58 – 7.61 (dd, 1H, Ar-H), 7.84 (d, J = 8 Hz, 1H, Ar-H), 8.11 (s, 1H, C4-coumarin), 8.23 (d, J = 8 Hz, 1H, C5-pyridine), 8.31 (s, 1H, C3-pyridine), 8.40 (s, 1H, C5-thiazole), 8.78 (d, J = 8 Hz, 1H, C6-pyridine); 13C-NMR (DMSO-d6, 125 MHz, , ppm): 13.2 (-CH3), 30.1 (-CH2-), 107.2, 111.6, 115.0, 119.8, 123.1, 124.3, 126.8, 127.2, 128.3, 136.7, 143.1, 145.1, 148.3, 150.5, 152.5, 159.2, 160.8 (C=O); Mass (m/z): 334 (M+); Elemental analysis (EA) for C19H14N2O2S [Calcd. (Found)]: C, 68.25 (68.14); H, 4.22 (4.18); N, 8.38 (8.32). Compounds 4b-4j were prepared in a similar manner (Scheme 2).Scheme 2. Synthesis of ethionamide and prothionamide based coumarinyl-thiazole compounds (4a-4j) 3-(2-(2-Ethylpyridin-4-yl)thiazol-4-yl)-6-fluoro-2Hchromen-2-one (4b). The reaction of 3b with ETH provided 4b (yellow crystals). Yield:80%; m.p. 188 – 189°C; Rf 0.82; IR (νmax, cm-1): 1716, 1562, 1525, 1255, 1112; 1H NMR (δ, ppm): 1.27 (t, J = 7 Hz, 3H, -CH3), 3.37 (q, J = 4.5 Hz, 2H, -CH2-), 7.13 (d, J = 8 Hz, 1H, Ar-H), 7.31 (dd, 2H, Ar-H), 8.10 (s, 1H, C4-coumarin), 8.21 (d, J = 8.5 Hz, 1H, C5-pyridine), 8.30 (s, 1H, C3-pyridine), 8.40 (s, 1H, C5-thiazole), 8.78 (d, J = 8 Hz, 1H, C6-pyridine); 13C-NMR: 13.2 (-CH3), 30.1 (-CH2-), 107.2, 111.6, 113.5, 114.1, 122.7, 123.3, 124.0, 128.3, 136.7, 143.0, 145.1, 147.5, 148.3, 152.3, 158.5, 159.3, 160.7 (C=O); Mass: 352 (M+); EA for C19H13FN2O2S: C, 64.76 (64.66); H, 3.72 (3.68); N, 7.95 (7.88). 6-Chloro-3-(2-(2-ethylpyridin-4-yl)thiazol-4-yl)-2Hchromen-2-one (4c). The reaction of 3c with ETH provided 4c(golden crystals). Yield: 85%; m.p. 174 – 176°C; Rf 0.75; IR (νmax, cm-1): 1718, 1567, 1529, 1255, 1110; 1H-NMR: 1.26 (t, J = 7 Hz, 3H, -CH3), 3.35 (q, J = 4.5 Hz, 2H, -CH2-), 7.27 (d, J = 7.5 Hz, 1H, Ar-H), 7.42 (d, J = 7.5 Hz, 1H, Ar-H), 7.99 (s, 1H, Ar-H), 8.09 (s, 1H, C4-coumarin), 8.19 (d, J = 7.5 Hz, 1H, C5-pyridine), 8.28 (s, 1H, C3-pyridine), 8.39 (s, 1H, C5-thiazole), 8.77 (d, J = 7.5 Hz, 1H, C6-pyridine); 13C NMR: 13.2 (-CH3), 30.3 (-CH2-), 107.1, 111.5, 117.1, 122.5, 123.0, 125.7, 128.3, 128.4, 132.0, 136.7, 143.0, 145.1, 148.3, 150.1, 152.3, 159.2, 160.8 (C=O); Mass: 368 (M+), 369 (M++1), 370 (M++1); EA for C19H13ClN2O2S: C, 61.87 (61.79); H, 3.55 (3.49); N, 7.60 (7.54). 6-Bromo-3-(2-(2-ethylpyridin-4-yl)thiazol-4-yl)-2Hchromen-2-one (4d). The reaction of 3d with ETH provided 4d (amber crystals).Yield: 85%; m.p. 195 – 197°C; Rf 0.82; IR (νmax, cm-1): 1717, 1566, 1528, 1258, 1113; 1H-NMR: 1.28 (t, J = 7 Hz, 3H, -CH3), 3.36 (q, J = 4 Hz, 2H, -CH2-), 7.20 (d, J = 8 Hz, 1H, Ar-H), 7.55 (d, J = 8 Hz, 1H, Ar-H), 8.01 (s, 1H, Ar-H), 8.10 (s, 1H, C4-coumarin), 8.19 (d, J = 8 Hz, 1H, C5-pyridine), 8.28 (s, 1H, C3-pyridine), 8.40 (s, 1H, C5-thiazole), 8.78 (d, J = 8 Hz, 1H, C6-pyridine); 13C NMR: 13.2 (-CH3), 30.1 (-CH2-), 107.3, 111.7, 117.2, 118.7, 123.0, 123.4, 128.2, 129.2, 133.1, 136.7, 143.0, 145.0, 148.3, 151.0, 152.4, 159.3, 160.9 (C=O); Mass: 412 (M+), 413 (M++1), 414 (M++2); EA for C19H13BrN2O2S: C, 55.22 (55.15); H, 3.17 (3.10); N, 6.78 (6.75). 3-(2-(2-Ethylpyridin-4-yl)thiazol-4-yl)-6-iodo-2H-chromen-2-one (4e). The reaction of 3e with ETH provided 4e (brown crystals). Yield: 80%; m.p. 169 – 171°C; Rf 0.81; IR (νmax, cm-1): 1719, 1566, 1527, 1256, 1112; 1H NMR: 1.27 (t, J = 7 Hz, 3H, -CH3), 3.38 (q, J = 4.5 Hz, 2H, -CH2-), 7.08 (d, J = 7.5 Hz, 1H, Ar-H), 7.79 (d, J = 7.5 Hz, 1H, Ar-H), 7.99 (s, 1H, Ar-H), 8.11 (s, 1H, C4-coumarin), 8.21 (d, J = 7.5 Hz, 1H, C5-pyridine), 8.30 (s, 1H, C3-pyridine), 8.42 (s, 1H, C5-thiazole), 8.79 (d, J = 7.5 Hz, 1H, C6-pyridine); 13C NMR: 13.3 (-CH3), 30.3 (-CH2-), 91.8, 107.1, 111.4, 119.2, 122.6, 123.0, 128.2, 133.1, 136.1, 136.7, 143.0, 145.0, 148.3, 150.8, 152.3, 159.2, 160.8 (C=O); Mass: 460 (M+), 461 (M++1), 462 (M++1); EA for C19H13IN2O2S: C, 49.58 (49.52); H, 2.85 (2.81); N, 6.09 (6.03). 3-(2-(2-Propylpyridin-4-yl)thiazol-4-yl)-2H-chromen-2-one (4f). The reaction of 3a with PTH provided 4f (yellow crystals). Yield: 75%; m.p. 181 – 183°C; Rf 0.77; IR (νmax, cm-1): 1715, 1565, 1525, 1256, 1115; 1H NMR: 0.95 (t, J = 7 Hz, 3H, -CH3), 1.79 (m, 2H, -CH2-Me), 3.03 (t, J = 7 Hz, 2H, -CH2-), 7.36 – 7.42 (m, 2H, Ar-H), 7.58 (dd, 1H, Ar-H), 7.83 (dd, 1H, Ar-H), 8.08 (s, 1H, C4-coumarin), 8.21 (d, J = 8 Hz, 1H, C5-pyridine), 8.28 (s, 1H, C3-pyridine), 8.41 (s, 1H, C5-thiazole), 8.79 (d, J = 8 Hz, 1H, C6-pyridine); 13C NMR: 13.3 (-CH3), 22.4 (-CH2-Me), 41.1 (-CH2-), 108.2, 113.3, 115.1, 119.8, 123.0, 124.3, 126.8, 127.2, 128.3, 136.6, 143.1, 145.0, 148.3, 152.0, 152.6, 159.5, 160.7 (C=O); Mass: 348 (M+); EA for C20H16N2O2S: C, 68.95 (68.88); H, 4.63 (4.55); N, 8.04 (8.01). 6-Fluoro-3-(2-(2-propylpyridin-4-yl)thiazol-4-yl)-2Hchromen-2-one (4g). The reaction of 3b with PTH provided 4g (reddish yellow crystals). Yield: 85%; m.p. 173 – 175°C; Rf 0.80; IR (νmax, cm-1): 1718, 1564, 1526, 1257, 1110; 1H NMR: 0.96 (t, J = 7 Hz, 3H, -CH3), 1.81 (m, 2H, -CH2-Me), 3.03 (t, J = 7 Hz, 2H, -CH2-), 7.10 (d, J = 8 Hz, 1H, Ar-H), 7.31 (dd, 2H, Ar-H), 8.08 (s, 1H, C4-coumarin), 8.22 (d, J = 8 Hz, 1H, C5-pyridine), 8.29 (s, 1H, C3-pyridine), 8.41 (s, 1H, C5-thiazole), 8.79 (d, J = 8 Hz, 1H, C6-pyridine); 13C NMR: 13.2 (-CH3), 22.5 (-CH2-Me), 41.0 (-CH2-), 108.1, 113.3, 113.8, 114.5, 122.7, 123.0, 124.0, 128.3, 136.6, 143.2, 145.1, 147.5, 148.3, 152.3, 158.1, 159.4, 160.7 (C=O); Mass: 366 (M+); EA for C20H15FN2O2S: C, 65.56 (65.50); H, 4.13 (4.08); N, 7.65 (7.60). 6-Chloro-3-(2-(2-propylpyridin-4-yl)thiazol-4-yl)-2Hchromen-2-one (4h). The reaction of 3c with PTH provided 4h (orange crystals). Yield: 70%; m.p. 193 – 195°C; Rf 0.77; IR (νmax, cm-1): 1721, 1566, 1529, 1256, 1111; 1H NMR: 0.96 (t, J = 7 Hz, 3H, -CH3), 1.79 (m, 2H, -CH2-Me), 3.04 (t, J = 7 Hz, 2H, -CH2-), 7.27 (d, J = 8 Hz, 1H, Ar-H), 7.40 (d, J = 8 Hz, 1H, Ar-H), 7.98 (s, 1H, Ar-H), 8.09 (s, 1H, C4-coumarin), 8.19 (d, J = 8 Hz, 1H, C5-pyridine), 8.28 (s, 1H, C3-pyridine), 8.39 (s, 1H, C5-thiazole), 8.78 (d, J = 8 Hz, 1H, C6-pyridine); 13C NMR: 13.4 (-CH3), 22.6 (-CH2-Me), 41.1 (-CH2-), 108.1, 113.2, 117.1, 122.4, 123.2, 126.7, 128.2, 128.5, 130.0, 136.6, 143.2, 145.1, 148.3, 150.0, 152.3, 159.5, 160.7 (C=O); Mass: 382 (M+), 383 (M++1), 384 (M++2); EA for C20H15ClN2O2S: C, 62.74 (62.68); H, 3.95 (3.92); N, 7.32 (7.30). 6-Bromo-3-(2-(2-propylpyridin-4-yl)thiazol-4-yl)-2Hchromen-2-one (4i). The reaction of 3d with PTH provided 4i (creamy crystals). Yield: 80%; m.p. 178 – 180°C; Rf 0.75; IR (νmax, cm-1): 1720, 1566, 1530, 1256, 1112; 1H NMR: 0.95 (t, J = 7 Hz, 3H, -CH3), 1.79 (m, 2H, -CH2-Me), 3.03 (t, J = 7 Hz, 2H, -CH2-), 7.19 (d, J = 8 Hz, 1H, Ar-H), 7.55 (d, J = 8 Hz, 1H, Ar-H), 8.01 (s, 1H, Ar-H), 8.08 (s, 1H, C4-coumarin), 8.18 (d, J = 8 Hz, 1H, C5-pyridine), 8.29 (s, 1H, C3-pyridine), 8.39 (s, 1H, C5-thiazole), 8.77 (d, J = 8 Hz, 1H, C6-pyridine); 13C NMR: 13.3 (-CH3), 22.4 (-CH2-Me), 41.0 (-CH2-), 108.1, 113.3, 117.1, 118.7, 123.1, 123.5, 128.4, 129.2, 133.1, 136.6, 143.2, 145.1, 148.3, 151.0, 152.3, 159.5, 160.8 (C=O); Mass: 426 (M+), 427 (M++1), 428 (M++2); EA for C20H15BrN2O2S: C, 56.22 (56.15); H, 3.54 (3.50); N, 6.56 (6.53). 6-Iodo-3-(2-(2-propylpyridin-4-yl)thiazol-4-yl)-2Hchromen-2-one (4j). The reaction of 3e with PTH provided 4j (yellow crystals). Yield: 85%; m.p. 195 – 197°C; Rf 0.84; IR (νmax, cm-1): 1721, 1566, 1529, 1256, 1112; 1H NMR: 0.95 (t, J = 7 Hz, 3H, -CH3), 1.81 (m, 2H, -CH2-Me), 3.02 (t, J = 7 Hz, 2H, -CH2-), 7.08 (d, J = 8 Hz, 1H, Ar-H), 7.77 (d, J = 8 Hz, 1H, Ar-H), 7.96 (s, 1H, Ar-H), 8.10 (s, 1H, C4-coumarin), 8.19 (d, J = 8 Hz, 1H, C5-pyridine), 8.28 (s, 1H, C3-pyridine), 8.40 (s, 1H, C5-thiazole), 8.76 (d, J = 8 Hz, 1H, C6-pyridine); 13C NMR: 13.4 (-CH3), 22.6 (-CH2-Me), 41.0 (-CH2-), 91.9, 108.4, 115.2, 119.5, 122.7, 123.1, 128.2, 133.1, 136.1, 136.7, 143.2, 145.1, 148.3, 150.8, 152.5, 159.6, 160.7 (C=O); Mass: 474 (M+), 475 (M++1), 476 (M++2); EA for C20H15IN2O2S: C, 50.65 (50.60); H, 3.19 (3.15); N, 5.91 (5.88). Experimental Biological Activity Part In-Vitro Anti-TB Activity Study The study was accomplished by employing the Microplate Alamar Blue Assay (MABA) method versus the Mycobacterium tuberculosis (Mtb H37Rv) strain [10, 12]. This assay is based on the color change of resazurin (blue), wherein the blue color changes to pink with microbial growth. All the chemicals and reagents were made corresponding to the detailed method specified in the literature [12]. The dilutions (0.781 – 50 μg/ml) of 4a-4j and standard drugs (ETH, PTH, INH, and PYZ) were prepared in sterile DMSO. The microplate absorbance was read at 530 and 590 nm (excitation and emission wavelengths, respectively), and the minimum inhibitory concentration (MIC) of the compounds was calculated (see Table 1 below).Table 1. Anti-TB Activity and Cytotoxicity Data and In-Silico Studies (PC and PK parameters) of Compounds 4a-4j Compound MIC (μg/mL) against Mtb H37Rv [CI 99%] MTT assay data (CC50 in μg/mL) Selectivity Index TPSA Log P Aqueous solubility Druglikeness (Lipinski’s rule violations) Calculated absorption (%) BBB permeant P-gp substrate Mutagenicity HCL VCL 4a 3.125 ± 0.0* [3.125 ± 0.0] > 300 > 300 > 96 84.23 4.01 Poor Yes (0) 79.94 No No No 4b 3.125 ± 0.0* [3.125 ± 0.0] > 300 > 300 > 96 84.23 4.35 Poor Yes (0) 79.94 No No Yes 4c 1.562 ± 0.0* [1.562 ± 0.0] > 300 > 300 > 192 84.23 4.61 Poor Yes (0) 79.94 No No No 4d 1.562 ± 0.0* [1.562 ± 0.0] > 300 > 300 > 192 84.23 4.61 Poor Yes (0) 79.94 No No No 4e 1.562 ± 0.0* [1.562 ± 0.0] > 300 > 300 > 192 84.23 4.65 Poor Yes (0) 79.94 No No - 4f 3.125 ± 0.0* [3.125 ± 0.0] > 300 > 300 > 96 84.23 4.35 Poor Yes (0) 79.94 No No No 4g 3.125 ± 0.0* [3.125 ± 0.0] > 300 > 300 > 96 84.23 4.62 Poor Yes (0) 79.94 No No No 4h 1.562 ± 0.0* [1.562 ± 0.0] > 300 > 300 > 192 84.23 4.84 Poor Yes (0) 79.94 No No No 4i 1.562 ± 0.0* [1.562 ± 0.0] > 300 > 300 > 192 84.23 4.93 Poor Yes (0) 79.94 No No No 4j 1.562 ± 0.0* [1.562 ± 0.0] 300 300 > 192 84.23 4.95 Poor Yes (0) 79.94 No No - BTZ043 ND ND ND ND 125.72 2.89 Moderate Yes (0) 65.62 No Yes Yes Macozinone ND ND ND ND 110.50 3.86 Moderate Yes (0) 70.87 No No Yes ETH 6.25 ± 0.0* [6.25 ± 0.0] > 150 > 150 > 24 71.0 1.47 Soluble Yes (0) 84.50 No No - PTH 6.25 ± 0.0* [6.25 ± 0.0] > 150 > 150 > 24 71.0 1.84 Soluble Yes (0) 84.50 Yes No - INH 3.125 ± 0.0* [3.125 ± 0.0] > 200 > 200 > 64 68.01 -0.35 Soluble Yes (0) 85.53 No No Yes PYZ 3.125 ± 0.0* 3.125 ± 0.0] > 200 > 200 > 64 68.87 -0.37 Soluble Yes (0) 85.23 No No No ND: not determined; * p < 0.05 (SPSS, version 20; n = 3); CI 99% = confidence interval 99% (SPSS, version 2.0; n = 3). MTT Assay of 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium Bromide The toxicity profile of 4a-4j was evaluated against HepG2 cell (HCL) and Vero cell (VCL) lines by the MTT assay [13, 14]. The test is based on living cells (HCL, VCL, etc.) exhibiting dehydrogenase enzyme activity that converts MTT to formazan crystals. The intensity of formazan color (purple) is measured colorimetrically, and the %viability of cells is calculated. The HCL (5 × 103 cells/well) and VCL (104 cells/well) were placed in well plates and incubated (37°C) for 24 hours. The working solutions of the sample and standard (50, 100, 150, 200, 250, and 300 μg/ml) were prepared in Dulbecco’s Modified Eagle’s Medium (DMEM). The blank solution without standard/test compounds was also prepared. The working/standard/blank solutions were added to wells comprising HCL and VCL. The HCL and VCL well plates were incubated for 24 hours and 72 hours, respectively. The MTT reagent (50 μl, 2 mg/ml) was added to well plates and incubated for 4 hours. The sterile dimethyl sulfoxide (50 μl) was added to each well to dissolve the formed crystals of formazan. The optical density (OD) of the wells was measured at 540 nm utilizing an Elisa reader. The %cell viability (OD of test × 100 / OD of blank), and %cell inhibition (100 - %cell viability) were calculated. The CC50 values (minimum concentration needed for 50% cell death) were determined by the curve fitting program, and the selectivity index (SI = CC50/MI) was also determined (Table 1). Determination of Physicochemical (PC) and Pharmacokinetic (PK) Parameters The Swiss web server [15] was used to determine the PC and PK parameters of the compounds, whereas the mutagenicity was found by the test software [16]. The SMILES of the compounds were generated, incorporated into the software, and the data were collected (Table 1). The percentage absorption was calculated according to the following formula [17]:%ABS=109−0.345×tPSA, where tPSA is the topological polar surface area. Molecular Docking Studies It was performed by Molecular Operating Environment (MOE) 2019.0102 (Chemical Computing Group Inc., Canada). The chain A of various MtB proteins (PDB IDs: 6HEZ, 4NCR, and 4F4Q) were used for the docking purpose [18 – 20]. Chain A was purified by employing the Quickprep functionality of the software. The ligand structures (4a-4j and the standard drugs) were also prepared and stored as mdb files. The docking was done by the default docking setting of the software with 10 poses. The docking score (DS) and the root mean square deviation (RMSD) of the docked molecules are provided in Table 2.Table 2. Molecular Docking Results and Anti-TB Activity of Compounds 4a-4j against Mtb Compound 6HEZ (Chain A) 4NCR (Chain A) 4F4Q (Chain A) DS [CI 99%] RMSD [CI 99%] Main interacting amino acids DS [CI 99%] RMSD [CI 99%] Main interacting amino acids DS [CI 99%] RMSD [CI 99%] Main interacting amino acids 4a -6.46 ± 0.03* [-6.505 to -6.415] 0.99 ± 0.01* [0.975 to 1.005] Lys418 -6.22 ± 0.04* [-6.279 to -6.161] 0.78 ± 0.03* [0.735 to 0.825] Lys418, Tyr415, Lys134 -6.63 ± 0.04* [-6.689 to -6.571] 1.24 ± 0.05* [1.166 to 1.314] Val372, Gly124 4b -6.94 ± 0.02* [-6.970 to -6.910] 1.20 ± 0.03* [1.155 to 1.245] Lys418, Tyr60 -7.08 ± 0.05* [-7.154 to -7.006] 1.40 ± 0.04* [1.341 to 1.459] Tyr415 -6.41 ± 0.01* [-6.425 to -6.395] 1.16 ± 0.04* [1.101 to 1.219] Lys425 4c -7.01 ± 0.01* [-7.025 to -6.995] 0.87 ± 0.02* [0.840 to 0.900] Lys418, Tyr60 -6.65 ± 0.05* [-6.724 to -6.576] 0.82 ± 0.04* [0.761 to 0.879] Lys134 -6.74 ± 0.01* [-6.755 to -6.725] 0.95 ± 0.04* [0.891 to 1.009] Val372, Gly124, Lys425 4d -6.86 ± 0.03* [-6.905 to -6.815] 1.27 ± 0.03* [1.225 to 1.315] Tyr415 -6.75 ± 0.02* [-6.780 to -6.720] 1.14 ± 0.03* [1.095 to 1.185] Leu363, Gly117 -6.60 ± 0.03* [-6.645 to -6.555] 1.48 ± 0.03* [1.435 to 1.525] - 4e -6.36 ± 0.05* [-6.434 to -6.286] 0.83 ± 0.04* [0.771 to 0.889] Lys418, Val365 -6.57 ± 0.04* [-6.629 to -6.511] 1.31 ± 0.05* [1.236 to 1.384] Leu363, Gly117, Lys134 -7.02 ± 0.04* [-7.079 to -6.961] 1.08 ± 0.05* [1.006 to 1.154] - 4f -7.07 ± 0.04* [-7.129 to -7.011] 1.03 ± 0.05* [0.956 to 1.104] - -6.62 ± 0.05* [-6.694 to -6.546] 1.29 ± 0.01* [1.275 to 1.305] His132, Gly117 -6.73 ± 0.05* [-6.804 to -6.656] 1.33 ± 0.02* [1.300 to 1.360] Gly124 4g -7.03 ± 0.02* [-7.060 to -7.000] 1.30 ± 0.03* [1.255 to 1.345] Lys418 -6.76 ± 0.03* [-6.805 to -6.715] 1.33 ± 0.04* [1.271 to 1.389] Lys418, Gly117, Lys134 -6.11 ± 0.03* [-6.155 to -6.065] 1.25 ± 0.04* [1.191 to 1.309] Pro123 4h -7.21 ± 0.01* [-7.225 to -7.195] 0.95 ± 0.04* [0.891 to 1.009] - -6.57 ± 0.02* [-6.600 to -6.540] 1.02 ± 0.01* [1.005 to 1.035] His132 -6.95 ± 0.04* [-7.009 to -6.891] 1.43 ± 0.03* [1.385 to 1.475] His139, Gly124 4i -6.27 ± 0.03* [-6.315 to -6.225] 1.29 ± 0.04* [1.231 to 1.349] Lys418, Arg58, Gly117 -6.90 ± 0.03* [-6.945 to -6.855] 1.29 ± 0.02* [1.260 to 1.320] - -7.02 ± 0.04* [-7.079 to -6.961] 0.64 ± 0.03* [0.595 to 0.685] Gly124 4j -7.53 ± 0.04* [-7.589 to -7.471] 1.36 ± 0.03* [1.315 to 1.405] Lys418, Tyr60 -6.94 ± 0.02* [-6.970 to -6.910] 0.99 ± 0.02* [0.960 to 1.020] Ly418 -7.10 ± 0.04* [-7.159 to -7.041] 1.33 ± 0.04* [1.271 to 1.389] Gly124 BTZ043 -6.82 ± 0.03* [-6.865 to -6.775] 1.11 ± 0.05* [1.036 to 1.184] Cy387, Val365, Lys134 -6.70 ± 0.03* [-6.745 to -6.655] 1.17 ± 0.01* [1.155 to 1.185] Gly133 -5.90 ± 0.03* [-5.945 to -5.855] 1.27 ± 0.04* [1.211 to 1.329] Cys394, Val372, Lys141 Macozinone -6.76 ± 0.03* [-6.805 to -6.715] 1.23 ± 0.02* [1.200 to 1.260] - -7.14 ± 0.04* [-7.199 to -7.081] 1.44 ± 0.04* [1.381 to 1.499] Cys387, Lys134 -7.26 ± 0.04* [-7.319 to -7.201] 1.24 ± 0.05* [1.166 to 1.314] Gly124 ETH -4.71 ± 0.03* [-4.755 to -4.665] 1.22 ± 0.03* [1.175 to 1.265] Lys418, Lys134 -4.98 ± 0.04* [-5.039 to -4.921] 1.36 ± 0.04* [1.301 to 1.419] Lys134 -5.09 ± 0.01* [-5.105 to -5.075] 1.27 ± 0.05* [1.196 to 1.344] His139 PTH -4.78 ± 0.02* [-4.810 to -4.750] 0.78 ± 0.01* [0.765 to 0.795] - -5.13 ± 0.03* [-5.175 to -5.085] 0.62 ± 0.05* [0.546 to 0.694] Lys134 -5.09 ± 0.02* [-5.120 to -5.060] 1.30 ± 0.04* [1.241 to 1.359] - INH -4.38 ± 0.01* [-4.395 to -4.365] 1.25 ± 0.02* [1.220 to 1.280] Lys134, Leu115 -4.63 ± 0.04* [-4.689 to -4.571] 1.49 ± 0.04* [1.431 to 1.549] Lys134, Leu115 -4.37 ± 0.03* [-4.415 to -4.325] 1.34 ± 0.04* [1.281 to 1.399] Leu122, Gly124 PYZ -4.23 ± 0.03* [-4.275 to -4.185] 1.42 ± 0.04* [1.361 to 1.479] Lys418, His132 -4.04 ± 0.02* [-4.070 to -4.010] 1.07 ± 0.01* [1.055 to 1.085] Leu115, Gly117 -4.18 ± 0.04* [-4.239 to -4.121] 1.38 ± 0.03* [1.335 to 1.425] - *p < 0.05 (SPSS, version 2.0; n = 3); CI 99% = confidence interval 95% (SPSS, version 2.0; n = 3). Statistical Analysis All results were statistically processed. The statistical analysis of results was done utilizing SPSS software (version 20, Chicago, IL, USA). The data is represented as mean±standard deviation. The p-value < 0.05 (number of experiments = 3) denotes statistically significant result, whereas CI 99% represents confidence interval 99%. Results and Discussion Chemistry Compounds (4a-4j) were synthesized corresponding to Schemes 1 and 2. The substituted salicyl aldehydes (1a-1e) were reacted with ethyl acetoacetate (2) to get 6-substituted-3-acetylcoumarins, which were treated with bromine to provide compounds 3a-3e (Scheme 1). The intermediates (3a-3e) were reacted with ETH and PTH to obtain 4a-4e and 4f-4j, respectively (Scheme 2). The novelty of 4a-4j was revealed by an accurate structural search in the Sci-Finder database. The IR spectrum of 4a-4j showed characteristic peaks for C=O (1715 – 1721 cm-1), C=N (1562 – 1568 cm-1), C=C (1525 – 1530 cm-1), C–O–C (1255 – 1258 cm-1), and C–S (1110 – 1115 cm-1). The 1H NMR spectra of 4a-4j displayed characteristic singlets for the proton of C4-coumarin (ä 8.08 – 8.11 ppm), C3-pyridine (δ 8.28 – 8.31), and C5-thiazole (δ 8.39 – 8.42 ppm). The proton at C6-pyridine was the most deshielded and appeared as a doublet at δ 8.76 – 8.79 ppm. The methyl proton of 4a-4e appeared as a triplet at δ 1.26 – 128 ppm, whereas the methyl proton of 4f-4j appeared at δ 0.95 – 0.96 ppm due to the shielding effect. Similar effects were observed for the –CH2- group adjacent to the -CH3 group (quartet at δ 3.35 – 3.38 ppm for 4a-4e and multiplet at δ 1.79 – 1.81 ppm for 4f-4j). The -CH2- group of 4f-4j attached to the pyridine ring appeared at δ 3.02 – 3.04 ppm. The 13C NMR spectra of 4a-4j displayed a characteristic peak for C=O (δ 160.7 – 160.9) ppm, and the methyl carbon (δ 13.2 – 13.4 ppm). The –CH2- group carbon adjacent to the –CH3 group appeared at δ 30.1 – 30.3 ppm for 4a-4e, whereas it appeared at δ 22.4 – 22.6 ppm for 4f-4j due to the shielding effect. The -CH2- group of 4f-4j attached to the pyridine ring appeared at δ 41.0 – 41.1 ppm due to the deshielding effect. The mass spectra and the elemental analysis were in concurrence with the assigned structures of 4a-4j. Anti-TB Activity and Toxicity Studies The anti-TB activity of compounds 4a-4j unveiled the potency of synthesized derivatives (Table 1). It was surprising to observe that all these compounds displayed equal or better anti-TB activity than INH and PYZ against Mtb. All these compounds also displayed better anti-TB activity than ETH and PTH. Compounds 4a, 4b, 4f, and 4g displayed equal MIC values in comparison to INH and PYZ, whereas 4c-4e and 4h-4j displayed better MIC values than INH and PYZ (Table 1, Fig. 3). The MIC values were statistically significant (p-values and CI 99% values). This observation indicates that the synthesized coumarin-thiazole-pyridine nucleus is a promising pharmacophore to develop potent anti-TB drugs. The presence of -Cl, -Br, and -I (4c-4e and 4h-4j) in this nucleus provides more potent compounds than the fluorine substituted nucleus (4b and 4f). This effect might be because of the higher lipophilic character of 4c-4e and 4h-4j (Table 1). The literature has reported molecules similar to 4a-4j (compounds A and B of Fig. 1) [6, 7]. Compound A of Fig. 1 stated MIC values in the range of 15 to >663 μg/ml against Mtb H37Rv, whereas compound B testified MIC between 6.25 – 25 μg/mL. The MIC values expressed by compounds A and B were more than the MIC values of 4a-4j. The 4a-4j possess a pyridine ring at C-2 of the thiazole ring, whereas the pyridine ring is not the structural part of compounds A and B. Therefore, the author trusts that 4a-4j displayed higher potency than compounds A and B because of the presence of the pyridine ring in the structure of 4a-4j. The MTT assay of 4a-4j against HCL and VCL demonstrated non-toxic behavior of 4a-4j up to 300 μg/ml concentration. The selectivity index of 4a-4j was also higher than that of clinically used drugs (ETH, PTH, INH, and PYZ) (Table 1, Fig. 3).Fig. 3. MIC and selectivity index values of compounds 4a-4j, ETH, PTH, INH, and PYZ. Molecular Docking Studies The author was surprised to observe the potent anti-TB activity displayed by compounds 4a-4j. Accordingly, the author also performed the molecular docking of compounds using various proteins of Mtb (Table 2) to identify the possible mechanism of action and the reason behind the potency of the synthesized compounds. For this purpose 6HEZ protein [18] and 4F4Q protein [19] of DprE1 enzyme complexed with BTZ043, and 4NCR protein [20] of DprE1 enzyme complexed with macozinone were employed. The DprE1 enzyme is a new validated target to develop novel anti-TB agents [2]. BTZ043 and macozinone are DprE1 inhibitors, which are in a clinical trial [2, 21]. Like MIC value data, the DS and the RMSD values were also statistically significant (p-values and CI 99% values). The docking results revealed that the interacting pattern of BTZ043 and macozinone with 6HEZ/4F4Q(Cy387 and Lys134) and 4NCR (Cy394, Lys141) proteins were as per the reported literature [2, 18 – 20] (Figs. 4–6). A higher negative value of the docking score (DS) is an indicator of the potency of a compound, whereas the RMSD value < 2 represents good binding with the protein receptor. Compounds 4b-4d, 4f-4h, and 4j displayed a higher negative DS than BTZ043 and maozinone with 6HEZ. Compounds 4b, 4c, 4g, and 4j displayed interaction with Lys418 and other amino acids of 6HEZ. This interaction might be the reason for the better DS of these compounds. Compounds 4b, 4d, 4g, 4i, and 4j displayed a higher negative DS than BTZ043 with 4NCR. However, the interaction patterns of these compounds with 4NCR were different from each other. Compounds 4a-4j displayed a higher negative DS than BTZ043 with 4F4Q. The interaction of 4a, 4c, 4f, and 4h-4j with Gly124 of 4F4Q might be responsible for their higher negative DS. Compound 4j displayed the highest negative DS with 6HEZ (Fig. 7) in addition to 4F4Q (Fig. 8), and the second maximum negative DS with 4NCR (Fig. 9). The molecular docking data of compounds 4a-4j suggest that they are inhibitors of DprE1 (Fig. 10).Fig. 4. Interaction of BTZ043 with Chain A of 6HEZ. Fig. 5. Interaction of BTZ043 with Chain A of 4F4Q. Fig. 6. Interaction of macozinone with Chain A of 4NCR. Fig. 7. Interaction of 4j with Chain A of 6HEZ. Fig. 8. Interaction of 4j with Chain A of 4F4Q. Fig. 9. Interaction of 4j with Chain A of 4NCR. Fig. 10. The docking scores of compounds 4a-4j and other drugs involving DprE1 proteins (6HEZ, 4NCR, and 4F4Q). PC and PK Parameters The physicochemical (PC) properties of a compound are determinants of its pharmacokinetic (PK) parameters and behavior [22]. According to the PC values of compounds 4a-4j and standard drugs, the LogP values of compounds were higher than those of standard drugs. Compounds 4a-4j, BTZ043, macozinone, ETH, INH, and PYZ did not exhibit BBB permeation property (except for PTH). All compounds obeyed Lipinski’s rule (drug likeliness). The calculated gastrointestinal absorption of 4a-4j was higher than that of BTZ043 and macozinone. None of the compounds displayed P-gp substrate inhibitory property. This indicates that their PK behavior will not be affected by drugs that induce or inhibit these enzymes. Further, only compound 4b displayed mutagenicity potential. All these parameters of 4a-4j indicate their promising PK and safety profile in comparison to the existing anti-TB drugs [2, 23]. In conclusion, the ETH and PTH-based coumarinyl-thiazole derivatives (4a-4j) displayed outstanding anti-TB activity against Mtb H37Rv and did not demonstrate any toxicity against HCL and VCL. The molecular docking studies demonstrate that 4a-4j are DprE1 inhibitors. The in silico studies revealed that 4a-4j followed Lipinski’s rule (drug-likeliness) and exhibited higher gastrointestinal absorption than BTZ043 and macozinone. These observations indicate that the synthesized nucleus is a good template for developing selective DprE1 enzyme inhibitors and potent anti-TB agents. Accordingly, further structure-activity relationship studies are recommended. Conflict of Interest The author declares that he has no conflicts of interest. Funding The author extends his appreciation to the Deputyship for Research & Innovation, Ministry of Education, Saudi Arabia, for funding this research work through the project number IF-2020-NBU-209. ==== Refs References 1. WHO Global Tuberculosis Report 2020, World Health Organization: Geneva (2020) (available at https://apps.who.int/iris/bitstream/handle/10665/336069/9789240013131-eng.pdf; Accessed on June 1, 2021). 2. M. Imran, A. S. Alshrari, H. K. Thabet, et al., Expert Opin. Ther. Pat., (2021). 10.1080/13543776.2021.1902990. 3. Alghamdi S Rehman SU Shesha NT Molecules 2020 25 23 1 15 10.3390/molecules25235685 4. Reddy DS Kongot M Kumar A Tuberculosis 2021 127 102050 10.1016/j.tube.2020.102050 33540334 5. Gümüş M Yakan M Koca I Future Med. Chem. 2019 11 15 1979 1998 10.4155/fmc-2018-0196 31517529 6. Jadhav BS Yamgar RS Kenny RS Curr. Comput. Aided Drug Des. 2020 16 5 511 522 10.2174/1386207322666190722162100 31438831 7. Arshad A Osman H Bagley MC Eur. J. Med. Chem. 2011 46 9 3788 3794 10.1016/j.ejmech.2011.05.044 21712145 8. Vale N Gomes P Santos HA Curr. Drug Metab. 2013 14 1 151 158 10.2174/1389200211309010151 23215813 9. Ansari MI Khan SA Med. Chem. Res. 2017 26 7 1481 1496 10.1007/s00044-017-1855-4 10. Asif M Imran M Curr. Bioact. Compd. 2021 17 3 261 266 10.2174/1573407216999200507122605 11. Imran M Bawadekji A Ali MA Res. J. Pharm. Bio. Chem. Sci. 2018 9 5 2413 2419 12. Cho S Lee HS Franzblau S Methods Mol. Biol. 2015 1285 281 292 10.1007/978-1-4939-2450-917 25779323 13. Kamiloglu S Sari G Ozdal T Capanoglu E Food Front. 2020 1 332 349 10.1002/fft2.44 14. N. C. Ammerman, M. Beier-Sexton, and A. F. Azad, Curr. Protoc. Microbiol., Appendix 4, Appendix 4E (2008). 10.1002/9780471729259.mca04es11. 15. Daina A Michielin O Zoete V Sci. Rep. 2017 7 1 13 10.1038/srep42717 28127051 16. Sripriya N Kumar MR Karthick NA Drug Chem. Toxicol. 2019 21 1 7 10.1080/01480545.2019 17. Imran M Molbank 2020 2020 3 M1155 10.3390/M1155 18. Richter A Rudolph I Möllmann U Sci. Rep. 2018 8 1 1 12 10.1038/s41598-018-31316-6 29311619 19. J. Neres, F. Pojer, E. Molteni, et al., Sci. Transl. Med., 4(150), 150ra121(2012). 10.1126/scitranslmed.3004395. 20. Makarov V Lechartier B Zhang M Mol. Med. 2014 6 3 372 383 10.1002/emmm.201303575 21. Stephanie F Saragih M Tambunan USF Pharmaceutics 2021 13 5 1 21 10.3390/pharmaceutics13050592 22. Hurst S Loi CM Brodfuehrer J El-Kattan A Expert Opin. Drug Metab. Toxicol. 2007 3 4 469 489 10.1517/17425225.3.4.469 17696800 23. Umumararungu T Mukazayire MJ Mpenda M Indian J. Tuberc. 2020 67 4 539 559 10.1016/j.ijtb.2020.07.017 33077057	0	PMC9734486	NO-CC CODE	2022-12-14 23:28:29	no	Pharm Chem J. 2022 Dec 7; 56(9):1215-1225	utf-8	Pharm Chem J	2,022	10.1007/s11094-022-02782-0	oa_other
==== Front Dig Dis Sci Dig Dis Sci Digestive Diseases and Sciences 0163-2116 1573-2568 Springer US New York 36478314 7781 10.1007/s10620-022-07781-5 Original Article Ameliorating Role of Hydrogen-Rich Water Against NSAID-Induced Enteropathy via Reduction of ROS and Production of Short-Chain Fatty Acids Akita Yoshihiro 12 http://orcid.org/0000-0002-3206-4055 Higashiyama Masaaki [email protected] 1 Kurihara Chie 1 Ito Suguru 1 Nishii Shin 1 Mizoguchi Akinori 1 Inaba Kenichi 1 Tanemoto Rina 1 Sugihara Nao 1 Hanawa Yoshinori 1 Wada Akinori 1 Horiuchi Kazuki 1 Okada Yoshikiyo 1 Narimatsu Kazuyuki 1 Komoto Shunsuke 1 Tomita Kengo 1 Takei Fumie 3 Satoh Yasushi 4 Saruta Masayuki 2 Hokari Ryota 1 1 grid.416614.0 0000 0004 0374 0880 Department of Internal Medicine, National Defense Medical College, 3-2 Namiki, Tokorozawa, Saitama 359-8513 Japan 2 grid.411898.d 0000 0001 0661 2073 Division of Gastroenterology and Hepatology, Department of Internal Medicine, The Jikei University School of Medicine, Tokyo, Japan 3 grid.416614.0 0000 0004 0374 0880 Department of Chemistry, National Defense Medical College, Saitama, Japan 4 grid.416614.0 0000 0004 0374 0880 Department of Biochemistry, National Defense Medical College, Saitama, Japan 7 12 2022 111 24 3 2022 28 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 Nonsteroidal anti-inflammatory drug (NSAID)-induced enteropathy, the mechanism of which is involved in oxidative stress, can be lethal due to hemorrhage. Thus, we aimed to investigate the effect of hydrogen-rich water (HRW), in terms of oxidative stress, on intestinal mucosal damage as well as changes in the gut microbiome and the short-chain fatty acids (SCFAs) content in feces. Methods Hydrogen-rich water was orally administered for 5 days to investigate the effectiveness of indomethacin-induced enteropathy in mice. Small intestinal damage and luminal reactive oxygen species (ROS) were evaluated to investigate the ameliorating effects of hydrogen. Then, components of the gut microbiome were analyzed; fecal microbiota transplantation (FMT) was performed using the cecal contents obtained from mice drinking HRW. The cecal contents were analyzed for the SCFAs content. Finally, cells from the macrophage cell line RAW264 were co-cultured with the supernatants of cecal contents. Results Hydrogen-rich water significantly ameliorated IND-induced enteropathy histologically and reduced the expression of IND-induced inflammatory cytokines. Microscopic evaluation revealed that luminal ROS was significantly reduced and that HRW did not change the gut microbiota; however, FMT from HRW-treated animals ameliorated IND-induced enteropathy. The SCFA content in the cecal contents of HRW-treated animals was significantly higher than that in control animals. The supernatant had significantly increased interleukin-10 expression in RAW264 cells in vitro. Conclusion Hydrogen-rich water ameliorated NSAID-induced enteropathy, not only via direct antioxidant effects but also via anti-inflammatory effects by increasing luminal SCFAs. These results suggest that hydrogen may have therapeutic potential in small intestinal inflammatory diseases. Supplementary Information The online version contains supplementary material available at 10.1007/s10620-022-07781-5. Keywords Hydrogen NSAIDs-induced enteropathy Short-chain fatty acids Antioxidants IL-10 National Defense Medical Collegehttp://dx.doi.org/10.13039/100008695 Japan Foundation for Applied Enzymology http://dx.doi.org/10.13039/501100003478 Ministry of Health, Labour and Welfare ==== Body pmcIntroduction Nonsteroidal anti-inflammatory drugs (NSAIDs), the most frequently used drugs worldwide to control pain, are known to cause damage to the gastrointestinal tract. They damage the small intestine more frequently than the stomach [1]. Although proton pump inhibitors and histamine H2-receptor antagonists are clinically applied to prevent NSAID-induced damage to the gastrointestinal tract, the drugs are not effective against small intestinal damage. Some drugs and materials, including rebamipide [2], misoprostol [3], and sulforaphane [4], have been reported to reduce small intestinal damage [5]. Besides, antioxidants are also expected to be effective in the prevention and treatment of NSAID-induced enteropathy [6], because reactive oxygen species (ROS) are assumed to be involved in the pathogenesis of this disease [7, 8]. Recently, Ohsawa et al. reported that molecular hydrogen acted as a therapeutic antioxidant by selectively reducing cytotoxic oxygen radicals [9]. Since then, many hydrogen-related studies have been reported for several diseases, including ischemic tissue injury [9, 10], metabolic syndrome [11], sepsis [12], and novel coronavirus 2019-induced pneumonia [13]. In gastrointestinal diseases, hydrogen was reported to ameliorate dextran sodium sulfate-induced colitis, an inflammatory bowel disease (IBD) model [14], and aspirin-induced gastric mucosal damage [15]. In addition, gut microbiota, which is also involved in the pathogenesis of NSAID-induced enteropathy [5], was altered by hydrogen [16, 17]. Controlling luminal oxidative stress using hydrogen exerted protective effects for gut/systemic immunity [18]. From these reports, we hypothesized that hydrogen may have potential therapeutic effects against NSAID-induced enteropathy through not only its antioxidant effect but its modulation of gut microbiota by inducing microbiome-friendly anaerobic condition in enteropathy. In addition, there are some merits that hydrogen-rich water (HRW) can be easily generated by the machine and available in every home and taken with fewer side effects than drugs. Hydrogen can be administered in several ways, such as inhalation of hydrogen gas, oral intake of hydrogen-rich water (HRW), or intravenous administration of hydrogen-rich saline. Within them, the concentration of hydrogen in the intestinal lumen increases to the highest level by oral intake than by other ways [19]. Therefore, in this study, we aimed to clarify whether oral administration of hydrogen-rich water (HRW) could ameliorate NSAID-induced enteropathy using a murine experimental model, with respect to antioxidative stress and modulation of gut microbiota. Materials and Methods Preparation of Hydrogen-Rich Water Hydrogen-rich water (HRW, 1.0 mg/L) was produced through the electrolysis of water using Active BIO II (TAKAOKA CHEMICAL Co., LTD., Aichi, Japan). Some mice were allowed to drink HRW (HRW mice) for 5 days based on the previous study [20]. Given that hydrogen is volatile, HRW was kept in an aluminum bottle to maintain the concentration of hydrogen and was changed daily. Other mice were allowed to drink normalized water (NW) produced by leaving HRW for a week to volatilize hydrogen (NW mice). Induction of Small Intestinal Damage Indomethacin (IND, Wako Pure Chemical Industries, Ltd., Osaka, Japan) was dissolved in dimethyl sulfoxide 20 ng/mL) to a concentration of 1 mg/mL [21]. To induce small intestinal damage, some mice were injected IND (10 mg/kg) or phosphate-buffered saline (PBS; 10 mg/kg) intraperitoneally on day 4. Macroscopic and Histological Evaluation of Enteropathy To measure the ulcer area, 1% Evans blue (Wako Pure Chemical Industries, Ltd.) was injected intravenously 30 min before sacrifice, as done in a previous study [22]. The small intestine was removed and opened along the antimesenteric attachment. The blue-stained depressed areas were measured using a grid sheet. The middle part of the removed small intestine without an obvious ulcer area was selected for measuring villous height and crypt depth [23, 24]. Measurement of Myeloperoxidase Activity We selected a 3-cm portion of the small intestine located between 10 and 15 cm from the terminal ileum [21]. The tissue was homogenized, and myeloperoxidase activity was measured using the QuantiChromTM Peroxidase Assay Kit (BioAssay Systems, CA, USA) in a colorimetric assay using the supernatant [24]. Results were expressed as international units per gram tissue protein. Quantitative Reverse Transcription (RT)-Polymerase Chain Reaction (PCR) for Intestinal mRNA Expression The degree of mRNA expression of tumor necrosis factor α (TNF-α), interleukin (IL)-1β, IL-6, and IL-10 in the small intestine, and RAW264 cells was measured as previously described [25]. The samples of the small intestine were taken from a 1-cm area that did not have an obvious ulcer, located between 10 and 15 cm from the terminal ileum. Total mRNA was extracted using the RNeasy Mini Isolation Kit (Qiagen, Valencia, CA, USA). The primers and probes were purchased from Applied Biosystems: TNFα (Mm00443258), IL-1β (Mm01336189), IL-6 (Mm00446190), and IL-10 (Mm00439616). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a housekeeping gene. Reverse transcription polymerase chain reaction (RT-PCR) was performed in triplicate for each sample using the ABI Prism 7900HT sequence detection system (Applied Biosystems, Foster City, CA, USA). Intestinal Mucosal Permeability by Fluorescein Isothiocyanate-Conjugated (FITC)-Dextran Intestinal permeability was measured using 4-kDa FITC-dextran (Sigma, USA) as done in a previous study [23]. Briefly, the mice were gavaged with FITC-dextran 6 h after fasting. After 1 h, blood samples were collected from the portal vein under anesthesia. Plasma was collected through centrifugation of the samples. The FITC-dextran concentration in the plasma was analyzed using a fluorescence spectrophotometer (Gemini EM microplate reader, Molecular Devices, CA, USA) at an excitation wavelength of 485 nm and an emission wavelength of 535 nm. Evaluation of ROS on the Intestinal Mucosa The ROS were visualized as described in a previous study [24, 26]. Briefly, the mice were anesthetized, and the ileum was extracted through a small incision in the peritoneum. To visualize the mucosa of the ileum through microscopic observation, a 1- to 2-cm long incision along the long axis at the antimesenteric attachment was made in the mucosa. Given that one of the fluorescence probes for detecting ROS, namely, aminophenyl fluorescein (APF, 50 μmol/L, Goryokayaku Hokkaido, Japan), has been developed for evaluating the ROS in the mucosa [24, 26, 27], 20 μL of APF was poured onto the mucosa after wiping away the intraluminal contents using a paint brush. Fifteen minutes after APF was poured, formation of ROS was observed on the mucosa with an intravital confocal laser scanning microscope (CLSM; A1R+; Nikon, Tokyo, Japan). The fluorescence intensity of the APF images was measured using NIS Elements software, version 4.50 (Nikon Co., Tokyo, Japan), per field of vision (approximately 0.3 mm2). Effect of Fecal Microbiota Transplantation (FMT) on Enteropathy Antibiotics and the microbial contents were administered according to a protocol used in previous studies [28, 29]. The mice drank the antibiotic cocktail (1-g/L ampicillin, 500-mg/L vancomycin, 1-g/L neomycin sulfate, and 1-g/L metronidazole) ad libitum for 28 days to deplete all detectable intestinal microbiota. The cecal microbial contents from donor mice (NW mice or HRW mice) were harvested, diluted in PBS (1.5 mL/0.2 mL of the cecal contents), and agitated. The cecal contents (0.2 mL) were gavaged to the mice once a day for 5 days. Indomethacin (10 mg/kg) was administered intraperitoneally 24 h after the final dose of cecal contents was administered. The intestinal damage (macroscopic and histological evaluation and quantitative RT-PCR for intestinal mRNA expression) was then compared between the mice that received cecal contents from the NW mice or the HRW mice. High-Throughput Sequencing of Gut Microbiota We evaluated the gut microbiota in the cecal contents of NW and HRW mice according to a previous study [24]. Measurement of Short-Chain Fatty Acids (SCFAs) in the Small Intestinal and Cecal Contents SCFA levels were measured according to a partially modified method using gas chromatography (GC) [30]. Small intestinal and cecal contents from NW or HRW mice were frozen immediately after collection and stored at − 80 °C. Samples mixed with 10% isobutanol were pretreated with a homogenizer (MagNA Lyser, Roche, Germany) twice at 6000 rpm for 20 s and centrifuged at 20,000×g for 5 min. The supernatants were collected and 125 μL of 20-mM NaOH and 3-methyl pentanoate were added as internal standards. After centrifugation in the same way, we added 100 μL of pyridine and 80 μL of isobutanol to the supernatants and then derivatized with isobutyl chloroformate. The SCFAs were then extracted with hexane. The concentrations of SCFAs (butyrate, acetate, and propionate) were measured by gas chromatography with a flame ionization detector (GC-2014, SHIMADZU Co., Kyoto, Japan) by injecting 1 μL of sample into the capillary column (DB-1: 30 m × 0.25 mm i.d., 0.25-μm film thickness, Agilent Technologies Japan, Ltd., Tokyo, Japan). Analyses were performed using GC solution (version 2.44). Preparation of Cecal Content Supernatant (CCS) Cecal contents from NW or HRW mice were diluted in PBS (0.4 μL/1 mg of the cecal contents), homogenized twice at 6,000 rpm for 20 s, and centrifuged at 20,000×g for 5 min under the same conditions for the measurement of SCFAs. The CCS was cultured in Minimum Essential Medium (MEM; Thermo Fisher Scientific, Waltham, MA, USA) and observed under microscope to confirm there was no contamination of bacteria. The CCS was collected and stored at − 80 ℃. Effect of Cecal Content Supernatant on mRNA Expression in RAW264 Cells RAW264 cells were purchased from RIKEN BRC Cell Bank (Ibaraki, Japan). RAW264 cells were grown in MEM supplemented with 10% fetal bovine serum (Thermo Fisher Scientific), 0.1-mM MEM non-essential amino acids solution (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan), and 1% penicillin–streptomycin (Thermo Fisher Scientific). The cells were seeded at a density of 1 × 105 cells/well in 12-well plates and cultured for 3 days at 37 ℃ in 5% CO2. The cells were then stimulated with 1% CCS from the NW or HRW mice for 6 h. The degree of mRNA expression of TNF-α, IL-6, and IL-10 was measured. Statistical Analysis This study was performed using JMP PRO 15 software (SAS Institute Inc., Cary, NC, USA). All results are expressed as mean ± standard error of the mean. For comparative analysis in each group, homoscedasticity was examined using Levene’s test for equality of variance. Differences between two groups were examined using Student’s t test with equal variance and Welch’s t test with unequal variance. Tukey’s honest significant difference was used between the multiple groups. Statistical significance was set at P < 0.05. Results Oral Intake of HRW Ameliorated IND-Induced Enteropathy We treated mice with HRW orally to evaluate the ameliorating effects of IND-induced enteropathy. First, the length and ulcer area of the small intestine were measured (Fig. 1A–C). Indomethacin significantly induced small intestinal ulcers mainly from middle part to ileum and shortened their length. Treatment with HRW for 5 days (IND was administered on day 4) significantly suppressed IND-induced shortening of the small intestine (25.8 ± 0.3 cm vs. 27.7 ± 0.6 cm, P < 0.01) and reduced the IND-induced ulcer area (22.8 ± 2.3 mm2vs. 8.0 ± 0.6 mm2, P < 0.01). Then, we evaluated the histological damage in the middle part of the removed small intestine by H&E staining (Fig. 1D, E). Indomethacin induced elevation of the crypt villus ratio (CVR). This elevation was ameliorated by HRW pretreatment (0.60 ± 0.04 vs. 0.39 ± 0.02, P < 0.01). Myeloperoxidase activity in the small intestinal mucosa was significantly increased by IND treatment. This increase was suppressed by pretreatment with HRW (Fig. 1F) (21.4 ± 2.3 U/g protein vs. 10.0 ± 2.2 U/g protein, P < 0.05). Next, we compared the mRNA expression of TNF-α, IL-1β, and IL-6 between NW and HRW mice with or without IND (Fig. 1G). The mRNA expression levels of TNF-α, IL-1β, and IL-6 were increased in NW-treated animals after IND treatment. All inflammatory cytokines increased by IND were significantly reduced by pretreatment with HRW (TNF-α: 2.71 ± 0.48 vs. 1.38 ± 0.35, P < 0.05. IL-1β: 12.2 ± 1.78 vs. 5.34 ± 1.31, P < 0.05. IL-6: 105 ± 14.2 vs. 57.8 ± 11.0, P < 0.05). However, anti-inflammatory cytokine IL-10 was not reduced significantly (data not shown).Fig. 1 Effects of hydrogen-rich water (HRW) on indomethacin-induced enteropathy. a Representative macroscopic features of the small intestine. b Small intestinal length (n = 8 per group). c Macroscopic ulcer area (n = 4 per group). d Representative histological features with hematoxylin eosin staining. e Ratio of crypt depth to villous height in the small intestine (n = 8 per group). f Myeloperoxidase activity in the small intestine (n = 4 per group). g mRNA expression of inflammatory cytokines in the small intestine (n = 5 per group). P < 0.05, P < 0.01. NW normalized water, HRW hydrogen-rich water HRW Attenuated Intestinal Hyperpermeability Induced by IND To evaluate intestinal hyperpermeability, FITC-dextran was gavaged, and plasma concentrations were measured (Fig. 2). Indomethacin treatment increased the plasma FITC-dextran concentration, which was significantly reduced by HRW.Fig. 2 Intestinal permeability analyzed by plasma FITC-dextran concentration (n = 4 per group). P < 0.05. NW normalized water, HRW hydrogen-rich water HRW Decreased IND-Induced ROS on the Small Intestinal Mucosa Given that ROS is considered one of the pathophysiological factors in IND-induced enteropathy, the ROS detector APF was used to visually evaluate the effect of HRW on IND-induced small intestinal mucosal damage. Indomethacin treatment increased the intensity of APF in the mucosa. The increased fluorescence intensity by IND was decreased significantly in HRW-treated animals (Fig. 3A, B). The results indicated that HRW reduced the ROS induced by IND in the mucosa.Fig. 3 Visual analysis of reactive oxygen species (ROS) in small intestinal mucosa. a Representative fluorescence images of each group. b Fluorescence intensity in the images of each group (n = 4 per group). P < 0.05. NW normalized water, HRW hydrogen-rich water Fecal Microbiota Transplantation (FMT) from HRW-Treated Animals Ameliorated the IND-Induced Small Intestinal Damage Given that it was reported that hydrogen changed the gut microbiota [16] and increased the diversity of the microbiota [17], we speculated that alteration of the gut microbiota by oral intake of HRW could ameliorate IND-induced enteropathy. Therefore, the effects of FMT from NW and HRW mice on IND-induced enteropathy were compared (Fig. 4). There was little amount of intestinal juice in the ileum for FMT, and colonic microbiota is largely different from that of small intestine. We collected the cecal feces for FMT because of its proximity to small intestine. The mice treated with the FMT from HRW mice were significantly less susceptible to IND-induced small intestinal damage than those treated with the FMT from NW mice. Specifically, the shortened small intestine (28.0 ± 0.8 cm vs. 31.1 ± 0.6 cm, P < 0.01), induced ulcers (12.4 ± 1.2 mm2 vs. 6.1 ± 0.6 mm2, P < 0.01), elevated CVR (0.45 ± 0.02, 0.34 ± 0.02, P < 0.01), and mRNA expression of TNF-α (2.32 ± 0.42 vs. 0.90 ± 0.12, P < 0.05), IL-1β (1.31 ± 0.31 vs. 0.43 ± 0.11, P < 0.05), and IL-6 (2.32 ± 0.70 vs. 0.46 ± 0.13, P < 0.05) by IND were significantly suppressed by FMT from HRW mice. The cecum microbiota composition was not assessed before or after FMT.Fig. 4 Effects of fecal microbiota transplantation from hydrogen-rich water (HRW) mice on indomethacin-induced enteritis. a Small intestinal length (n = 8 per group). b Macroscopic ulcer area (n = 8 per group). c Representative histological features with hematoxylin eosin staining. d Ratio of crypt depth to villous height in the small intestine (n = 6 per group). e mRNA expression of inflammatory cytokines in the small intestine (n = 6 per group). P < 0.05. FMT fecal microbiota transplantation, NW normalized water, HRW hydrogen-rich water Effect of HRW on the Gut Microbiota Composition Given that the FMT from HRW mice was shown to ameliorate IND-induced enteropathy, we speculated that HRW altered gut microbiota composition through which enteropathy was ameliorated. However, our analysis showed that α-diversity and β-diversity did not change significantly, and the composition of the microbiota did not change significantly between NW and HRW mice (Supplementary figure). HRW Increased the SCFA Levels in the Small Intestinal and Cecal Contents Given that the FMT from HRW ameliorated IND-induced enteropathy and HRW did not alter gut microbiota composition, we speculated that ingredients of HRW-treated feces other than microbiota were responsible for ameliorating the effect of enteropathy. We then focused on the amount of SCFAs in the feces; butyrate, acetate, and propionate, representative metabolites produced by gut microbiota, which were favorable for gut immunity [31–33]. In cecal contents, the concentrations of acetate, propionate, and butyrate were greater in HRW-treated mice than in NW-treated mice (acetate: 22.4 ± 1.78 mmol/mg feces vs. 34.4 ± 1.46 mmol/mg feces, P < 0.01. propionate: 2.77 ± 0.23 mmol/mg feces vs. 4.10 ± 0.23 mmol/mg feces, P < 0.01. butyrate: 3.72 ± 0.78 mmol/mg feces vs. 8.17 ± 0.50 mmol/mg feces, P < 0.01) (Fig. 5A). The concentration of butyrate in the small intestine was significantly higher in HRW-treated mice than in treated NW mice (1.00 ± 0.16 vs. 1.62 ± 0.11, P < 0.05) (Fig. 5B).Fig. 5 Concentration of short-chain fatty acids (SCFAs) in the enteral contents obtained from the small intestine or cecum of mice taking hydrogen-rich water (HRW). a Small intestine (n = 5 per group). b Cecum (n = 8 per group). P < 0.05, P < 0.01. NW normalized water, HRW hydrogen-rich water, ND not detected Effects of Microbiome-Excluded Cecal Contents on Inflammatory Cells In Vitro To show that cecal contents, including SCFAs, but not microbiota itself, are responsible for ameliorating enteropathy, we investigated the direct effect of microbiome-excluded cecal contents on inflammatory cells in vitro. The supernatant (CCS) was obtained by diluting and centrifuging the cecal contents to remove microbiota. Given that macrophages play pivotal roles in IND-induced enteropathy [34], CCS from each group was added to RAW264 cells, and the mRNA expression levels of TNF-α, IL-6, and IL-10 were compared (Fig. 6). There was no difference in the degree of mRNA expression of inflammatory cytokines, such as TNF-α (1.56 ± 0.12 vs. 1.59 ± 0.12) and IL-6 (2.30 ± 0.32 vs. 2.25 ± 0.22), but the degree of IL-10 expression was significantly increased in RAW264 cells exposed to CCS from HRW mice (1.30 ± 0.16 vs. 1.88 ± 0.08, P < 0.05). These results suggest that CCS from HRW mice had anti-inflammatory effects by promoting the production of IL-10 by macrophages.Fig. 6 The mRNA expressions in RAW264 cells stimulated by cecal contents supernatant from hydrogen-rich water (HRW) mice (n = 6 per group). P < 0.05. CCS cecal content supernatant, NW normalized water, HRW hydrogen-rich water Discussion This is the first study to investigate the efficacy of hydrogen on IND-induced enteropathy. In this study, we showed that IND-induced inflammatory cytokines, myeloperoxidase activity, mucosal permeability, and ROS generation in the small intestinal mucosa were suppressed by the oral intake of HRW. The production of ROS from the mitochondrial damage of small intestinal epithelial cells is considered to be one of the main mechanisms of NSAID-induced enteropathy [7]. In this study, AFP was used to evaluate ROS production according to the previous one [26, 35]. In Fig. 3A, AFP was detected around the cell membrane and inside the cells as dots. Considering ROS is produced from mitochondria, dots in the cells are supposed to be mitochondria. Hydrogen penetrates the cell membrane and diffuses rapidly [36]. Therefore, direct elimination of ROS by diffusely spreading into mitochondria might be one of the ameliorating mechanisms as well as elimination of ROS around cell membrane. Interleukin-1β derived from the nucleotide-binding domain-like receptor protein 3 (NLRP3) inflammasome in immune cells such as macrophages also plays a crucial role in NSAID-induced enteropathy [34], and hydrogen suppresses NLRP3 inflammasome activation by targeting mitochondrial ROS in vitro [37]; thus, decrease in IND-induced elevation of IL-1β might be through suppression of NLRP3 activation by hydrogen. Although it has been reported that HRW changed gut microbiota previously [16, 17, 38], no significant differences were observed in our study. This difference was probably due to the difference in the administration period of hydrogen (3 weeks to 2 months in previous studies). We administered HRW for 5 days based on the previous report investigating ameliorating effect of HRW on radiation gastrointestinal toxicity through MyD88’s effects, which also showed that there was no difference of abundance of enteric bacteria [20]. Collectively, HRW could exert protective effect on inflamed intestine without alteration of microbiota, suggesting that alteration of microbiota in previous studies mentioned above might be secondary effect after long-term administration of hydrogen. However, FMT from HRW significantly ameliorated enteropathy in our study. The concentrations of acetate, propionate, and butyrate in the cecal contents were significantly higher, and the concentration of butylate in the small intestine was significantly higher in HRW-treated animals. In addition, microbiome-excluded cecal contents increased the expression of IL-10 in macrophage cell lines. These results suggested that HRW did not change the composition of microbiota but changed the metabolism of microbiota to produce more SCFAs. Taking into consideration that a longer duration of HRW treatment changed the composition of microbiota from previous reports, a short period of HRW changes the function of metabolism before compositional changes. Luminal SCFAs exist in the lower small intestine and large intestine as a consequence of bacterial fermentation of non-digestible fibers, and it is possible that oral-administered SCFAs are rapidly absorbed from SCFA transporter expressed in duodenum [39]. Therefore, direct alteration of luminal concentration of SCFAs is difficult but enema. In this situation, we hypothesized that HRW might alter intestinal condition preferable for SCFA-producing bacteria. The most numerous butyrate-producing bacteria found in feces are highly oxygen-sensitive anaerobes belonging to the Clostridial clusters IV and XIVa [40], whose biological function decreases under oxygen because they are unable to generate the enzyme needed to metabolize ROS generated by the reduction of oxygen. Faecalibacterium prausnitzii, which is sensitive to oxygen and belongs to Clostridium cluster IV, is reported to be able to remain alive in ambient air with antioxidants [41]. In addition, spore-forming bacteria like Clostridium genus might not be evaluated properly in NGS due to surrounding condition of oxygen. Under anaerobic condition induced by hydrogen, they can be activated and possibly regain their SCFA-producing ability. Therefore, although we did not directly examine the composition of microbiota in the small intestine, we speculated that the suppression of ROS production by hydrogen may restore anaerobic environment in the intestine and increase the metabolic function of obligate anaerobic bacteria to produce more SCFAs. In the intestinal mucosa, acetate, propionate, and butyrate have beneficial effects on intestinal epithelial cells and immune cells. In particular, butyrate has been reported to suppress lipopolysaccharide (LPS)-induced activation of nuclear factor kappa B (NF-κB) via GPR109A [42] and promote the differentiation of IL-10-producing regulatory T cells [31, 43]. In a previous study, SCFAs suppressed the expression of TNF-α, IL-6, and inducible nitric oxide synthase (iNOS) and enhanced IL-10 expression in RAW 264.7 [44]. Taken together, in addition to a direct antioxidant effect by hydrogen, there seemed to be another anti-inflammatory effect by increasing butyrate and promoting macrophages to produce IL-10. In the near future, the clinical application of hydrogen in several diseases is expected. Molecular hydrogen has not been reported to exert toxic effects. Oral intake of HRW has been reported to improve lipid and glucose metabolism in patients with type 2 diabetes or glucose intolerance [45] and reduce liver fat accumulation and improve liver enzymes in patients with non-alcoholic fatty liver disease [46]. It will be also presumably preferable in many types of inflamed gastrointestinal diseases, such as IBD. Because hydrogen is easily volatile from water, delivering high concentration of hydrogen to lower intestinal lumen was an issue to be solved. To overcome the limitations of HRW, nanocapsules designed to sustain hydrogen release have also been investigated, which was proven to be effective in a mouse model of fatty liver diseases [47]. As a limitation of this study, positive control was not employed. Materials which show similar pharmacokinetics to hydrogen are desirable. However, several drugs including misoprostol [3] or NO-donors [48] show quite different pharmacokinetics to hydrogen. NO-donating NSAIDs [49] is fascinating but not freely available. In addition, these drugs had some non-specific effects other than antioxidant effects. To our knowledge, there is no antioxidant with similar pharmacokinetics to hydrogen and desirable positive control could not be employed. In conclusion, our study showed that hydrogen ameliorates IND-induced enteropathy through not only antioxidant direct effects but also the production of SCFAs. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (JPG 548 kb) (A) Representation of the major genus (genera < 1% are included in others) in the NW and HRW mice. (B) α-Diversity of the ileal microbiota in each group, as indicated by the values of the Chao1 index. (C) β-Diversity of the ileal microbiota in each group, represented by unweighted UniFrac distances. (D) β-Diversity of the ileal microbiota in each group, represented by a two-dimensional graph obtained using principal coordinate analysis (PCoA) of weighted UniFrac distances. NW, normalized water; HRW, hydrogen-rich water. This research was supported by grants from the National Defense Medical College and from the Japan Foundation for Applied Enzymology and by a Health and Labor Sciences research grant into research on intractable diseases, from the Ministry of Health, Labor and Welfare, Japan. We would like to thank Mr. Takanori Konuma (TAKAOKA TOKO CO., LTD.), Mr. Shunichi Suzuki (TAKAOKA TOKO CO., LTD.), and Mr. Takeshi Kawashima (TAKAOKA CHEMICAL CO., LTD.) for their help on preparation of hydrogen-rich water, Ms. Hanae Tsuchihashi (Meiji Co., Ltd.) for the technical support to analyze the microbiota, and Editage (www.editage.com) for English language editing. Author’s Contribution YA, MH, CK, and RH designed the experiments. YA, MH, CK, SI, SN, AM, KI, RT, NS, YH, AW, KH, YO, KN, SK, KT, FT, YS, MS, and RH were involved in the interpretation of data. YA, MH, and RH drafted the manuscript. Declarations Conflict of interest The authors declare that they have no conflicts of interest. Ethical approval Eight-week-old male C57BL/6J mice were purchased from CLEA Japan (Tokyo, Japan). They were cared for as per the guidelines of the animal facility at the National Defense Medical College (NDMC). This study was approved by the Animal Research Committee of the NDMC (No. 20011). Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Graham DY Opekun AR Willingham FF Qureshi WA Visible small-intestinal mucosal injury in chronic NSAID users Clin Gastroenterol Hepatol. 2005 3 1 55 59 10.1016/S1542-3565(04)00603-2 15645405 2. Niwa Y Nakamura M Ohmiya N Maeda O Ando T Itoh A Efficacy of rebamipide for diclofenac-induced small-intestinal mucosal injuries in healthy subjects: a prospective, randomized, double-blinded, placebo-controlled, cross-over study J Gastroenterol. 2008 43 4 270 276 10.1007/s00535-007-2155-4 18458842 3. Watanabe T Sugimori S Kameda N Machida H Okazaki H Tanigawa T Small bowel injury by low-dose enteric-coated aspirin and treatment with misoprostol: a pilot study Clin Gastroenterol Hepatol. 2008 6 11 1279 1282 10.1016/j.cgh.2008.06.021 18995219 4. Yanaka A Sato J Ohmori S Sulforaphane protects small intestinal mucosa from aspirin/NSAID-induced injury by enhancing host defense systems against oxidative stress and by inhibiting mucosal invasion of anaerobic enterobacteria Curr Pharm Des. 2013 19 1 157 162 22950492 5. Wang X Tang Q Hou H Zhang W Li M Chen D Gut Microbiota in NSAID Enteropathy: New Insights From Inside Front Cell Infect Microbiol. 2021 11 679396 10.3389/fcimb.2021.679396 34295835 6. Omatsu T Naito Y Handa O Mizushima K Hayashi N Qin Y Reactive oxygen species-quenching and anti-apoptotic effect of polaprezinc on indomethacin-induced small intestinal epithelial cell injury J Gastroenterol. 2010 45 7 692 702 10.1007/s00535-010-0213-9 20174833 7. Higuchi K Umegaki E Watanabe T Yoda Y Morita E Murano M Present status and strategy of NSAIDs-induced small bowel injury J Gastroenterol. 2009 44 9 879 888 10.1007/s00535-009-0102-2 19568687 8. Kim YJ Kim EH Hahm KB Oxidative stress in inflammation-based gastrointestinal tract diseases: challenges and opportunities J Gastroenterol Hepatol. 2012 27 6 1004 1010 10.1111/j.1440-1746.2012.07108.x 22413852 9. Ohsawa I Ishikawa M Takahashi K Watanabe M Nishimaki K Yamagata K Hydrogen acts as a therapeutic antioxidant by selectively reducing cytotoxic oxygen radicals Nat Med. 2007 13 6 688 694 10.1038/nm1577 17486089 10. Shigeta T Sakamoto S Li XK Cai S Liu C Kurokawa R Luminal injection of hydrogen-rich solution attenuates intestinal ischemia-reperfusion injury in rats Transplantation. 2015 99 3 500 507 10.1097/TP.0000000000000510 25539463 11. Kamimura N Nishimaki K Ohsawa I Ohta S Molecular hydrogen improves obesity and diabetes by inducing hepatic FGF21 and stimulating energy metabolism in db/db mice Obesity (Silver Spring). 2011 19 7 1396 1403 10.1038/oby.2011.6 21293445 12. Ikeda M Shimizu K Ogura H Kurakawa T Umemoto E Motooka D Hydrogen-Rich Saline Regulates Intestinal Barrier Dysfunction, Dysbiosis, and Bacterial Translocation in a Murine Model of Sepsis Shock. 2018 50 6 640 647 10.1097/SHK.0000000000001098 29293174 13. Guan WJ, Chen RC, Zhong NS. Strategies for the prevention and management of coronavirus disease 2019. Eur Respir J. 2020;55. 14. Kajiya M Silva MJ Sato K Ouhara K Kawai T Hydrogen mediates suppression of colon inflammation induced by dextran sodium sulfate Biochem Biophys Res Commun. 2009 386 1 11 15 10.1016/j.bbrc.2009.05.117 19486890 15. Zhang JY Wu QF Wan Y Song SD Xu J Xu XS Protective role of hydrogen-rich water on aspirin-induced gastric mucosal damage in rats World J Gastroenterol. 2014 20 6 1614 1622 10.3748/wjg.v20.i6.1614 24587639 16. Higashimura Y Baba Y Inoue R Takagi T Uchiyama K Mizushima K Effects of molecular hydrogen-dissolved alkaline electrolyzed water on intestinal environment in mice Med Gas Res. 2018 8 1 6 11 10.4103/2045-9912.229597 29770190 17. Sha JB Zhang SS Lu YM Gong WJ Jiang XP Wang JJ Effects of the long-term consumption of hydrogen-rich water on the antioxidant activity and the gut flora in female juvenile soccer players from Suzhou China. Med Gas Res. 2018 8 4 135 143 10.4103/2045-9912.248263 30713665 18. Iddir M, Brito A, Dingeo G, et al. Strengthening the Immune System and Reducing Inflammation and Oxidative Stress through Diet and Nutrition: Considerations during the COVID-19 Crisis. Nutrients. 2020;12. 19. Liu C Kurokawa R Fujino M Hirano S Sato B Li XK Estimation of the hydrogen concentration in rat tissue using an airtight tube following the administration of hydrogen via various routes Sci Rep. 2014 4 5485 10.1038/srep05485 24975958 20. Xiao HW Li Y Luo D Dong JL Zhou LX Zhao SY Hydrogen-water ameliorates radiation-induced gastrointestinal toxicity via MyD88's effects on the gut microbiota Exp Mol Med. 2018 50 1 e433 10.1038/emm.2017.246 29371696 21. Higashiyama M Hokari R Kurihara C Ueda T Watanabe C Tomita K Indomethacin-induced small intestinal injury is ameliorated by cilostazol, a specific PDE-3 inhibitor Scand J Gastroenterol. 2012 47 8–9 993 1002 10.3109/00365521.2012.690043 22934593 22. Watanabe T Higuchi K Kobata A Nishio H Tanigawa T Shiba M Non-steroidal anti-inflammatory drug-induced small intestinal damage is Toll-like receptor 4 dependent Gut. 2008 57 2 181 187 10.1136/gut.2007.125963 17639086 23. Yoshikawa K Kurihara C Furuhashi H Takajo T Maruta K Yasutake Y Psychological stress exacerbates NSAID-induced small bowel injury by inducing changes in intestinal microbiota and permeability via glucocorticoid receptor signaling J Gastroenterol. 2017 52 1 61 71 10.1007/s00535-016-1205-1 27075753 24. Wada A, Higashiyama M, Kurihara C, et al. Protective Effect of Luminal Uric Acid Against Indomethacin-Induced Enteropathy: Role of Antioxidant Effect and Gut Microbiota. Dig Dis Sci. 2021;67:121–133. 25. Ishida T Miki I Tanahashi T Yagi S Kondo Y Inoue J Effect of 18beta-glycyrrhetinic acid and hydroxypropyl gammacyclodextrin complex on indomethacin-induced small intestinal injury in mice Eur J Pharmacol. 2013 714 1–3 125 131 10.1016/j.ejphar.2013.06.007 23792039 26. Tomita T Sadakata H Tamura M Matsui H Indomethacin-induced generation of reactive oxygen species leads to epithelial cell injury before the formation of intestinal lesions in mice J Physiol Pharmacol. 2014 65 3 435 440 24930516 27. Setsukinai K Urano Y Kakinuma K Majima HJ Nagano T Development of novel fluorescence probes that can reliably detect reactive oxygen species and distinguish specific species J Biol Chem. 2003 278 5 3170 3175 10.1074/jbc.M209264200 12419811 28. Gregory JC Buffa JA Org E Wang Z Levison BS Zhu W Transmission of atherosclerosis susceptibility with gut microbial transplantation J Biol Chem. 2015 290 9 5647 5660 10.1074/jbc.M114.618249 25550161 29. Wang Z Klipfell E Bennett BJ Koeth R Levison BS Dugar B Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease Nature. 2011 472 7341 57 63 10.1038/nature09922 21475195 30. Furuhashi T Sugitate K Nakai T Jikumaru Y Ishihara G Rapid profiling method for mammalian feces short chain fatty acids by GC-MS Anal Biochem. 2018 543 51 54 10.1016/j.ab.2017.12.001 29221706 31. Furusawa Y Obata Y Fukuda S Endo TA Nakato G Takahashi D Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells Nature. 2013 504 7480 446 450 10.1038/nature12721 24226770 32. Sun M Wu W Liu Z Cong Y Microbiota metabolite short chain fatty acids, GPCR, and inflammatory bowel diseases J Gastroenterol. 2017 52 1 1 8 10.1007/s00535-016-1242-9 27448578 33. Parada Venegas D De la Fuente MK Landskron G González MJ Quera R Dijkstra G Short Chain Fatty Acids (SCFAs)-Mediated Gut Epithelial and Immune Regulation and Its Relevance for Inflammatory Bowel Diseases Front Immunol. 2019 10 277 10.3389/fimmu.2019.00277 30915065 34. Higashimori A Watanabe T Nadatani Y Takeda S Otani K Tanigawa T Mechanisms of NLRP3 inflammasome activation and its role in NSAID-induced enteropathy Mucosal Immunol. 2016 9 3 659 668 10.1038/mi.2015.89 26349656 35. Wada A Higashiyama M Kurihara C Ito S Tanemoto R Mizoguchi A Protective Effect of Luminal Uric Acid Against Indomethacin-Induced Enteropathy: Role of Antioxidant Effect and Gut Microbiota Dig Dis Sci. 2022 67 1 121 133 10.1007/s10620-021-06848-z 33569665 36. Ohta S Recent progress toward hydrogen medicine: potential of molecular hydrogen for preventive and therapeutic applications Curr Pharm Des. 2011 17 22 2241 2252 10.2174/138161211797052664 21736547 37. Ren JD Wu XB Jiang R Hao DP Liu Y Molecular hydrogen inhibits lipopolysaccharide-triggered NLRP3 inflammasome activation in macrophages by targeting the mitochondrial reactive oxygen species Biochim Biophys Acta. 2016 1863 1 50 55 10.1016/j.bbamcr.2015.10.012 26488087 38. Tanaka Y Kiuchi M Higashimura Y Naito Y Koyama K The effects of ingestion of hydrogen-dissolved alkaline electrolyzed water on stool consistency and gut microbiota: a double-blind randomized trial Med Gas Res. 2021 11 4 138 144 10.4103/2045-9912.318858 34213495 39. Kaji I Iwanaga T Watanabe M Guth PH Engel E Kaunitz JD SCFA transport in rat duodenum Am J Physiol Gastrointest Liver Physiol. 2015 308 3 G188 G197 10.1152/ajpgi.00298.2014 25394661 40. Pryde SE Duncan SH Hold GL Stewart CS Flint HJ The microbiology of butyrate formation in the human colon FEMS Microbiol Lett. 2002 217 2 133 139 10.1111/j.1574-6968.2002.tb11467.x 12480096 41. Khan MT van Dijl JM Harmsen HJ Antioxidants keep the potentially probiotic but highly oxygen-sensitive human gut bacterium Faecalibacterium prausnitzii alive at ambient air PLoS One. 2014 9 5 e96097 10.1371/journal.pone.0096097 24798051 42. Thangaraju M Cresci GA Liu K Ananth S Gnanaprakasam JP Browning DD GPR109A is a G-protein-coupled receptor for the bacterial fermentation product butyrate and functions as a tumor suppressor in colon Cancer Res. 2009 69 7 2826 2832 10.1158/0008-5472.CAN-08-4466 19276343 43. Rubtsov YP Rasmussen JP Chi EY Fontenot J Castelli L Ye X Regulatory T cell-derived interleukin-10 limits inflammation at environmental interfaces Immunity. 2008 28 4 546 558 10.1016/j.immuni.2008.02.017 18387831 44. Park JS, Lee EJ, Lee JC, Kim WK, Kim HS. Anti-inflammatory effects of short chain fatty acids in IFN-gamma-stimulated RAW 264.7 murine macrophage cells: involvement of NF-kappaB and ERK signaling pathways. Int Immunopharmacol. 2007;7:70–7. 45. Kajiyama S Hasegawa G Asano M Hosoda H Fukui M Nakamura N Supplementation of hydrogen-rich water improves lipid and glucose metabolism in patients with type 2 diabetes or impaired glucose tolerance Nutr Res. 2008 28 3 137 143 10.1016/j.nutres.2008.01.008 19083400 46. Korovljev D Stajer V Ostojic J LeBaron TW Ostojic SM Hydrogen-rich water reduces liver fat accumulation and improves liver enzyme profiles in patients with non-alcoholic fatty liver disease: a randomized controlled pilot trial Clin Res Hepatol Gastroenterol. 2019 43 6 688 693 10.1016/j.clinre.2019.03.008 30982748 47. Jin Z Sun Y Yang T Tan L Lv P Xu Q Nanocapsule-mediated sustained H2 release in the gut ameliorates metabolic dysfunction-associated fatty liver disease Biomaterials. 2021 276 121030 10.1016/j.biomaterials.2021.121030 34298442 48. Porras M Martín MT Terán E Mourelle M Vergara P The nitric oxide donor LA-419 [S-(6-Nitro-oxi-hexahydro-furo[3,2-b]furan-3-1-il)thioacetate] prevents intestinal dysmotility, bacterial translocation, and inflammation in a rat model of enteritis J Pharmacol Exp Ther. 2008 324 2 740 748 10.1124/jpet.107.129189 17959746 49. Bandarage UK Janero DR Nitric oxide-releasing nonsteroidal anti-inflammatory drugs: novel gastrointestinal-sparing drugs Mini Rev Med Chem. 2001 1 1 57 70 10.2174/1389557013407160 12369991	36478314	PMC9734488	NO-CC CODE	2022-12-14 23:28:29	no	Dig Dis Sci. 2022 Dec 7;:1-11	utf-8	Dig Dis Sci	2,022	10.1007/s10620-022-07781-5	oa_other
==== Front Educ Inf Technol (Dordr) Educ Inf Technol (Dordr) Education and Information Technologies 1360-2357 1573-7608 Springer US New York 11489 10.1007/s10639-022-11489-4 Article What is needed to build a personalized recommender system for K-12 students’ E-Learning? Recommendations for future systems and a conceptual framework Zayet Tasnim M. A. [email protected] 1 Ismail Maizatul Akmar [email protected] 1 Almadi Sara H. S. [email protected] 1 Zawia Jamallah Mohammed Hussein [email protected] 1 Mohamad Nor Azmawaty [email protected] 2 1 grid.10347.31 0000 0001 2308 5949 Faculty of Computer Science and Information Technology, Universiti Malaya, 50603 Kuala Lumpur, Malaysia 2 grid.10347.31 0000 0001 2308 5949 Faculty of Education, Universiti Malaya, 50603 Kuala Lumpur, Malaysia 5 12 2022 122 22 3 2022 28 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. Online learning has significantly expanded along with the spread of the coronavirus disease (COVID-19). Personalization becomes an essential component of learning systems due to students’ different learning styles and abilities. Recommending materials that meet the needs and are tailored to learners’ styles and abilities is necessary to ensure a personalized learning system. The study conducted a systematic literature review (SLR) of papers on recommendation systems for e-learning in the K12 setting published between 2017 and 2021 and aims to identify the most important component of a personalized recommender system for school students’ e-learning. Recommendations for later studies were proposed based on the identified components, namely a personalized conceptual framework for providing materials to school students. The proposed framework comprised four stages: student profiling, material collection, material filtering, and validation. Keywords e-learning Personalization Recommendation systems School Systematic review http://dx.doi.org/10.13039/501100004386 Universiti Malaya IIRG001B-19SAH Ismail Maizatul Akmar ==== Body pmcIntroduction The importance of e-learning has grown since 2020 due to the COVID-19 pandemic. Many countries were compelled to undergo complete lockdown during the crisis, including comprehensive movement control operations, and forcing educational institutions to shift from face-to-face to online learning (Radha et al., 2020; Su et al., 2021). The peak of the pandemic caused school closures that affected approximately 1.6 billion students in 190 countries and regions, which resulted in the sudden expansion of e-learning (Tadeo, 2021). The sudden change forced academic institutions to choose between available and affordable tools, such as social media platforms (Facebook and WhatsApp), video conferencing tools (Zoom, WebEx, and MS Teams), and learning management systems, namely Moodle, Blackboard, and Google Classroom. The transformation was poorly received by instructors and students due to the complexity of transitioning from traditional teaching and learning methods to technology (Almaiah et al., 2020; Hong et al., 2022). Furthermore, teachers received insufficient training. In traditional teaching, students’ facial expressions are indicators used by teachers to gauge students’ understanding of certain topics (Klašnja-Milićević et al., 2018; Zhu et al., 2020). Conversely, online tools complicate follow-up with students, specifically in large classrooms. E-learning has overcome specific issues, such as following up with teachers and equipping students with the necessary knowledge through greater access to resources at any time and location (Almaiah et al., 2020). Nonetheless, students possess varying levels of learning ability, learning styles, and behaviors, which might lead to varying performance levels despite using the same materials and taught with the same approach (Premlatha et al., 2016). Students may become distracted by the vast amount of information available online. Additionally, students may be unable to choose acceptable materials or sequences to employ those items (Venkatesh et al., 2020). Several studies have examined the challenges students encounter in e-learning environments (Ali et al., 2018; Almaiah, Al-Khasawneh et al., 2020), specifically lack of content interaction, adaptation to students’ requirements, and content relevancy to students’ needs and performance levels. Teachers were also unable to provide specialized instruction and tailored materials to each student due to inadequate time. Consequently, the need to offer materials to students that suit their requirements and level of performance has emerged (Rahman et al., 2018). Personalized recommender systems are thus an essential aspect of e-learning systems (Sarwar et al., 2019) to enhance student performance and to be in the hands of teachers. Personalization has become increasingly popular in recommendation systems and services. Personalization guarantees that the quality of the service provided improves based on consumer satisfaction. Personalization results from suggestions based on user preferences that tend to satisfy their needs, which has been incorporated into various recommendation systems, including healthcare (Rohani et al., 2020), tourism (Missaoui et al., 2019), e-commerce (Dixit et al., 2020), and transportation (Borodinov et al., 2019). Thus, incorporating personalization into e-learning systems can provide learners or students with personalized resources tailored to their requirements and learning styles for high performance (Li et al., 2019). The study conducted a five-year systematic review of personalized recommendation systems for e-learning in the school environment (PRS-ES) from 2017 to 2021 to determine the PRS-main ES components. The components were identified and the study provided recommendations for future PRS-ES. A conceptual framework was also presented based on the recommendations and addressed the following research questions to determine the main components of PRS-ES: Q1. What are the “must exist” modules in PRS-ES? Q2. What are the personalization features that can be used to ensure personalization? The study contributed to PRS-ES research as follows: Presenting a five-year systematic review focusing on schools. Identifying the primary elements of PRS-ES systems for school dedicated systems. Identifying the personalization features that should be used to ensure personalization and the measurement methods. Proposing a conceptual framework for developing PRS-ES systems for schools. A tailored framework for proposing materials to school students that is based on a methodical analysis of the works already published has effects on the students, the teachers and the system. (i) Students can increase their productivity, performance level, and knowledge while also developing their self-managed learning style. (ii) It can primarily save teachers’ time and effort. (iii) It might increase usage and efficiency for the system. (iv) The foundation of student profiling is personalization characteristics, which the framework employs to assure customisation. The study comprised five sections. The first section presented an introduction while the remaining are organized as follows: Sect. 2 elaborates on the most recent systematic reviews with a comparison, Sect. 3 discusses the methodology used in conducting the systematic review, Sect. 4 presents the results and discussion, Sect. 5 proposes the conceptual framework, and the final section concludes the study. Related work Several systematic review articles on personalized e-learning recommender systems have been published in recent years, which differ in purpose, the covered range of years, digital libraries (DLs), and queries. This section discusses some of the most recent reviews. The systematic study aims to identify the main components and features of PRS-ES that ensure personalization and prioritizes school students’ characteristics and preferences for personalization. Bernacki et al., (2021) conducted a systematic review on personalized learning (PL). The current study identified 376 studies that investigated one or more PL design aspects using the ERIC, PsychInfo, and IEEE DLs published from 2010 to 2018. The study compiled a list of the various PL definitions to guide implementation in education and reviewed key educational theories that facilitate design and implementation. However, the study did not focus enough on the learner’s characteristics or knowledge level. The collected papers targeted k12 and higher education learners with varying learning needs and preferences. Although the search query includes words, such as “personalization,” “personalized learning,” and “personalized instruction”, future studies should further examine keyword approaches to capture personalization and adaptivity as they involve human subject research in the learning process and its outcomes to manage challenges with the many relevant keywords. Raj et al. (2021) examined the customized content recommenders in PL environments in 52 journal papers from the Science Citation Index (SCI) and Scopus-indexed journals. The main goal was to examine and describe the research in PL environments between 2015 and 2020 and identify the various e-learning content recommendation strategies, personalization parameters, models, algorithms, and evaluation measures. Nevertheless, the study did not examine a specific type of learners, such as school students or the specific student characteristic that plays a major role in creating a highly personalized e-learning system. One of the students’ features that ensure personalization is the learning path, which presents the sequence of materials that students consider through the learning process. MacHado et al. (2021) reviewed learning path recommendations over five decades of studies ranging from 1971 to 2021 but most included papers were from 2014 to 2020. Xie et al.’s (2019) systematic review of learners from elementary and higher education involved many topics, including PL parameters, learning aids, learning outcomes, subjects, participants, and hardware. The study collected journal papers from 2007 to 2017, while the concept of personalized e-learning changed during that period. The collected papers were only from one index–the Web of Science, which only included the most reputable journal articles. Sajjad et al. (2021) presented a systematic review of the recommender systems for massive open online courses (MOOCs), which were web-based distance learning programs for large groups of students that were geographically dispersed. Khanal et al. (2019) focused on the approaches used in the recommendation process and developed machine learning (ML) based recommendation systems for e-learning to develop adaptive or personalized e-learning systems. The study identified 35 papers from 2016 to 2018 from Q1 and Q2 journals and obtained 10 papers as a final set. Nonetheless, none of the aforementioned reviews emphasised K-12 school students. School students differ from higher education students due to different learning systems in most countries with different needs, preferences, abilities, and goals (Emanuel et al., 1992; Tüysüz et al., 2010). Western Sydney University and the University of Adelaide mentioned that universities and schools differ in many aspects. The timetable of school students is fixed while the timetable in universities is flexible as students can control and choose the courses. Additionally, teachers provide regular homework to school students who are often guided towards task completion. Meanwhile, university assignments will be known in the first week and the students are responsible for completing and submitting them within the stipulated time. Regarding the student-teacher relationship, school teachers provide regular and direct guidance and feedback to students while university students obtain feedback through assignments. The study specifically targeted school students from all levels: elementary, primary, and secondary students. The studies were retrieved from five DLs (ACM, Web of Science, Scopus, Science Direct, and Springer) from 2017 to 6th November 2021. Methodology The SLR followed Kitchinham guidelines (Kitchenham et al., 2010). Initially, the SLR has followed a comprehensive review protocol that includes various stages to minimize the likelihood of bias in the literature. First, the SLR research questions and DLs were identified and utilized to retrieve studies. Subsequently, the study specified the search procedure, inclusion and exclusion criteria, and quality assessment criteria to filter the studies most relevant to the phenomenon of interest. Finally, the required data were extracted from the selected studies to address the SLR research questions. The following sections clarify the methodology applied in conducting the SLR. The DLs and keywords Five leading databases were chosen for the SLR as they contained studies on state-of-the-art personalized e-learning recommendations. The DLs are ScienceDirect, Springer, Web of Science, ACM, and Scopus. The study identified relevant keywords to retrieve research from 2017 to 2021 with the Boolean operators AND/OR used interchangeably on the keywords (see Table 1). Table 1 Keywords and digital libraries Year 2017–2021 Search terms ((personaliz* OR personalis* OR customiz* OR customis) AND (“e-learning” OR “online learning” OR “distance learning” OR “virtual learning” OR “web-based learning” OR “internet-based learning”) AND recomm AND (“secondary school” OR “elementary school” OR “primary school” OR “k12”)) Digital libraries ScienceDirect, Springer-Link, Web of Science, ACM, Scopus Inclusion and exclusion criteria The study initialized different inclusion and exclusion criteria to determine the relevant studies within the research boundaries. The study applied the identified inclusion and exclusion criteria for retrieving English publications from peer-reviewed conferences and journal articles. Each article was scanned based on the inclusion and exclusion criteria. The article was included if it matched all the inclusion terms and none of the exclusion terms. The retrieved articles were related to computer science, engineering, and educational technology domains. Moreover, the duplicated articles or book chapters, discussion notes, or reports were excluded from the study. Table 2 lists the eligibility criteria applied in the study. Table 2 Inclusion and exclusion criteria Inclusion criteria IC1: Publication date 2017 to 2021 (both years inclusive). IC2: Conference proceedings AND Peer-reviewed journal articles. IC3: In English and accessible. IC4: The study proposes PRS. IC5: The study targets primary school, secondary school, or elementary school students. Exclusion criteria EC1: Gray literature. EC2: Not in the English language. EC3: The study that proposes a recommendation system BUT does not focus on personalization or e-learning. EC4: The subject is personalization BUT in fields other than recommendation systems. EC5: The study that targets populations other than school students, such as teachers, training, disability students, group students, postgraduate or undergraduate students, and researchers. EC6: A duplicated study published in different venues (reporting similar results). EC7: Conference papers that extended to journal papers. Study selection and data analysis The study performed various steps to select the most relevant studies aligned with the SLR objectives using the above-mentioned inclusion and exclusion criteria. Figure 1 depicts the steps undertaken to select related studies. First, 780 articles were retrieved from the specified DLs using the identified keywords. Subsequently, independent researchers meticulously scanned the article titles and abstracts. Some articles were irrelevant to the state-of-the-art and excluded from the SLR (724 studies), thus minimizing the number of articles to 35. The small number of articles was due to most articles being related to the field of recommendation systems for e-learning dedicated to universities or MOOCs. The study critically scanned the full content of each included article in the second step of filtration. Thus, three studies were categorized as irrelevant and included 32 studies with strong relevance to the SLR objectives. Finally, the study applied quality assessment criteria to assess the quality of each selected article and obtained 23 selected articles. Figure 1 illustrates the filtration of articles and selection procedure. Fig. 1 The SLR phases and study selection Quality Assessment The quality assessment stage is crucial as it assessed the included studies to analyze the findings and interpretations (Kitchenham et al., 2010; Nidhra et al., 2013). The study identified Zayet and Al-Madi’s five quality assessment (QA) criteria to assess the relevant studies. QA1: Are the study objectives and goals clearly defined? QA2: Does the study clearly state the research methodology? QA3: Are the study contributions and limitations clearly stated? QA4: Are the data collection procedures and results clearly explained? QA5: Does the study mention how the personalized recommendation system is built? The quality assessment procedure was conducted through three quality rankings: “high”, “medium”, and “low” and applied to each QA criterion (Nidhra et al., 2013). A score of 1 is given to the study that comprehensively satisfied the quality criterion. Similarly, 0.5 is assigned to a quality criterion that partially satisfied the study. A score of 0 is assigned to the quality criterion that has not been satisfied. Thus, 5 is considered the highest score, while 0 score is the lowest. Depending on the coding scheme, the assessed study with a score of > 4 is considered high quality. The assessed study with a score of < 3.5 to > 2.5 is considered medium quality, while the study is considered low quality if the score is < 2.5. Table 3 presents various examples of quality assessment results for seven studies. Ultimately, 32 studies were high, medium, and low quality, while nine studies were excluded for being low quality. Table 3 Quality assessment criteria Study ID QA1 QA2 QA3 QA4 QA5 Total Include/exclude 1 1 1 1 1 1 5 Include 2 1 1 1 1 1 5 Include 3 1 1 0.5 1 1 4.5 Include 4 0.5 1 0.5 1 1 4 Include 5 1 0.5 0.5 0.5 1 3.5 Include 6 1 1 1 1 1 1 Include 7 1 0.5 0.5 0 0.5 2.5 exclude Data extraction The data extraction stage extracted the required data from the selected studies. The study created a form to record the data extraction of 23 articles for data collection completeness (Kitchenham et al., 2010). Several critical elements were identified for data extraction: study ID, types of system modules listed in the study, types of personalization features, students’ characteristics, and type of recommended items or context. Finally, the content of the remaining studies was carefully reviewed and analyzed to accurately extract the data for each identified element. Results and discussion The systematic review findings are presented and discussed in this section with recommendations for further study and the development of recommendation systems. Table 4 summarizes the final collection of papers and the proposed system and personalization feature that was used. Each personalization feature is associated with the students’ characteristics to ensure its usage and measurement. The discussion focused on the main modules and features employed to ensure personalization following the study theme. Section 4.1 and 4.2 addressed the research questions (Q1 and Q2) mentioned in the introduction section, respectively. Section 4.1 presents the identified primary modules of the PRS-ES system and Sect. 4.2 demonstrates the identified personalization features used in the articles. Improvement issues concerning future systems were identified and presented as suggestions for later systems during the analysis process in Sect. 4.3. Figure 2 displays the trend of final papers set over the last five years. Fig. 2 The trend in the selected publications over the past five years Table 4 The final set of papers Reference Modules Personalization features Students’ characteristics Recommended Item (Kopeinik et al., 2017) material repository, tags repository, frequent tags extractor, domain modeling, tags recommender tagging used tags by the student suitable tags for the uploaded materials (Mutahi et al., 2017) user manager, content manager, attention manager, context manager, notification manager performance content interaction patterns, comments, questions, affective state resource and activity (Wongwatkit et al., 2017) learning diagnostics module, learning style diagnostics module, mastery learning-based guided-inquiry learning mechanism module learning problems, learning styles current understanding learning activities (Gong et al., 2018) knowledge component recognition, knowledge graph, exercise generation proficiency level time spent on exercises, score, the ability of memory exercises with a suitable degree of difficulty (Hongthong et al., 2018) mobile application with four main modules, including interfacing, content repository, student assessment, and student feedback response modules performance and preferences score guidance to cyber security awareness (Klašnja-Milićević, Ivanović, et al., 2018) learner module, domain module, application module, adaptation module, the recommendation module interests and knowledge needs and previously acquired knowledge learning content (Klašnja-Milićević, Vesin, et al., 2018) learner-system interaction module, recommendation module [tags recommendation - recommendation of resources - reports generator], data storage module [tag repository - learner model] educational goals, learning history used tags learning resource, tags (Perišić et al., 2018) learning object module, student module, user interface module, adaptation module, visualization module, and reporting module learning style general information (name and surname, date of birth, email, interest), learning progress (average grade, learning style, time spent in the course, action), information about the student’s actions (viewed, loaded, deleted, graded, submitted, posted), duration of the sessions, learning object attendance, time spent on the learning object, and number of visits of the learning object learning material, semantic report (Lee et al., 2018) contents registration, management, and recommendations learning history video contents data, types of similar contents, sharing subjects, contents log, satisfaction, and comment data learning video contents (Guan et al., 2019) personal information management, learning course plan management, course selection, assessment, achievement management, learning ability identification learning ability knowledge points, length of course learning, number of the learned courses, course credits, specialty personalized curriculum (Troussas et al., 2019) students’ repository, students’ modeling, materials generator, recommender, hints, and trophies repository knowledge level, learning style age, (pre-existing knowledge on a domain, current knowledge level, knowledge level on previous concepts (scores and concepts)), preferred learning styles and techniques individualized hints, possible collaborators, learning material, trophies (Mimis et al., 2019) students’ repository, students’ modeling, rank prediction, the recommendation module performance level score (national baccalaureate score, first-year score, score of class council of the second year), students ranking in accordance to other students, quarterly rank in each subject, (age, social motivation) guidance to a career path (Jagušt et al., 2019) communication (server communication, lesson delivery, group work delivery, and progress monitoring modules), central (database, multimedia content repository, event log), adaptivity and aggregate data calculation, (lesson authoring and conducting, and lesson management (for teachers)) performance level, knowledge level relative score to other students, time spent on tasks, activities solved activity, visual representation of a lesson, suitable time to finish an activity (Ch et al., 2019) sentence reformation, summarization, factual sentence identification, trial test generation and evaluation, identification of the less confident portion performance level score (of the provided trial exams) sections to revise (Bhaskaran et al., 2019) System interaction module, Off-line modeling, Recommendation engine learning style and knowledge level personal data, preferences, dominant meaning words, behavior courses (Fakooa et al., 2019) student ontology, English verb ontology, admin panel, ANN module learning style, level of knowledge A score of quizzes, time spent on the quiz, text, and visual contents quizzes and verb ontology (Segal et al., 2019) difficulty ranking module, the recommendation module student performance grades, number of retries, and time spent solving questions. suitable problem sets or exams to student’s ability, topics to strengthen (Nian et al., 2019) recognition module of expression information, the personalized recommendation module performance, emotions score, expression courses (Troussas et al., 2019b) students’ module, domain knowledge adaptation module, assessment adaptation module, advice provider module knowledge level and preferences scores personalized guidance and questions (Saito et al., 2020) clustering module, prediction module, the recommendation module submission history, ability chart scores, current knowledge, goal learning path recommendation (Ma et al., 2020) advanced automated assessment module, peer tutor recommender module learning performance scores peer tutor (Nurzaman et al., 2021) students modeling, learning style identification, material repository, assessment, recommender performance level, learning style score (from teacher and systems) learning resource (Y. Zhang, 2021) students’ repository, resources repository, model generator, recommender Students’ history Students’ evaluation score for each resource auxiliary English teaching resources The PRS-ES: main modules Each system is formed from modules that are in charge of a set of duties usually tied to one of the system actors. Three primary elements (see Table 4) are needed to ensure personalization in the e-learning recommendation system: student profiling, collection and processing of materials, and recommendation generator. I. Student profiling module: The module is considered the most crucial as it determines which items are recommended. Students’ attributes are defined and assessed in the student profiling module, which describes students’ interests, needs, performance level, knowledge points and level, learning style, and other individualized aspects. The qualities were utilized to suggest appropriate materials or learning paths to each student to improve their performance and understanding. The module is present throughout the publication, as presented in Table 4. II. Material collection and processing: The resources for the module are the recommended objects for the students gathered from various sources, including teachers, the internet, and the students. Teachers will usually supply pupils with at least the most basic materials, such as textbooks, syllabuses, notes, and previous tests. The system can automatically collect resources from the internet using scraping techniques and students may share such resources with their peers. The material collector sub-module is where the collection takes place. Subsequently, the materials were treated and examined. The scraped and shared resources must be screened and validated as the internet provides a great number of resources and students may upload or share irrelevant resources, thus necessitating the use of a sub-module known as material validation. Finally, material profiling determines the material topic. The material profile may also include information, such as the number of views, average rating, average time spent on the material, frequently asked questions about the material, the average difficulty level of the material, the performance level of students who typically view it, a summary of the material, and sub-subjects of the material. The information improves individualized recommendations for each student in the future. III. Recommendation generator: The stage determined which items were recommended to each student based on their needs and preferences, hence making it a “must” in the recommendation system (see Table 4). The step involved the use of data mining (DM) and ML techniques and one or more recommendation approaches, such as collaborative filtering, content-based filtering, context-based filtering, knowledge-based filtering, and hybrid approaches. The PRS-ES: personalization features Personalization features are features that the system uses to ensure personalization and are the core of student profiling. The system determined the most suitable material or task for the specific student based on the personalization features. Figure 3 displays the trend of the main used personalization features over the reviewed literature. Fig. 3 The distribution of the personalized features over the publications I. Performance level: The degree of students’ performance is a typical personalization characteristic utilized in e-learning recommendation systems as the primary goal of the systems is to improve students’ performance. Students’ scores are a frequent individualized feature utilized to assess their performance level. Nonetheless, utilizing scores as a single indicator of students’ performance is inaccurate due to the student’s psychological state during the tests (Mudenda et al., 2020). Other personalized characteristics should be used to obtain a more precise performance levels measurement, such as students’ content interaction patterns, comments, questions, affective state (Mutahi et al., 2017), time spent on activities or materials, number of solved materials (Jagut et al., 2019), and students’ expression (Nian et al., 2019). II. Learning style: Learners’ preferred learning strategies and styles often vary. The type of content (visual, textual, and aural) is usually related to learning style where some students prefer to learn through images, hearing, demonstrating, or a combination of the methods (Troussas et al., 2019a). Two approaches determine a student’s learning style (Fakooa et al., 2019): manual and automatic. Students submit input on their learning style via a form or questionnaire in the manual technique, while the system updates the students’ learning style based on their learning behavior in the automatic method. Researchers have employed a human method to determine the preferred learning style of the initial students and used an automatic method to update and confirm the original learning styles (Bhaskaran et al., 2019; Perišić et al., 2018). Others relied solely on the automated way to obtain the information (Fakooa et al., 2019). III. Learning ability: Multiple elements, such as overall performance level, knowledge level, and achievement rate or level interfere with determining the learning ability (Guan et al., 2019). Hence, the number of items learned over time is a significant determinant. The PRS-ES: recommendations for later Systems The study provided suggestions for developing new recommendation systems to ensure the delivery of more tailored resources based on two categories: recommendations on student profiling and recommendations on material processing. Recommendations on student profiling level Considering that student profile is the most important aspect of the recommendation system, the feature requires more improvements to ensure that students are provided with the most appropriate materials. The aspect has potential for development. Many proposals have been researched but require further efforts and improvements as the followings: I. Identifying the difficulty level of each task or material by each student: The difficulty level of each task or material cannot be generalized due to the diversity of students’ abilities, performance levels, and knowledge levels. The difficulty level should be tailored to each learner instead of generalization (Segal et al., 2019; Yaqian Zhang et al., 2021). II. Identifying the performance level by considering more factors other than scores: As stated in Sect. 4.2, scores may not be a precise indicator of a student’s performance level as the score can be influenced by anxiety and other psychological issues. Therefore, other factors should be used to measure student performance, such as knowledge level, time spent on a task or material, number of views of the material, rating of the material, and students’ questions and comments (Mutahi et al., 2017; Perii et al., 2018; Fakooa et al., 2019; Chang et al., 2022). III. Diagnosing the students’ ability: Ability is a broad term that refers to critical thinking ability, ability to comprehend a topic, and ability to memorize (Burin et al., 2021; Supriyatno et al., 2020; Yaniawati et al., 2020). Thus, recommendation systems detect students’ talents automatically and make individualized suggestions to improve them (Gong et al., 2018; Saito et al., 2020). IV. Diagnosing student learning style: Various materials, including text, audio, interactive games, and video are available. Different students prefer to obtain information through various types of materials. “Automatically” identifying the desired and appropriate materials for each student improves the recommended resources. Hence, the process enhances students’ performance and knowledge levels (Rasheed et al., 2021). Recommendations on the material processing level Material processing prepares resources recommended to pupils. Material profiling is the most important process in material processing. The following suggestions are made to improve material profiling and recommendation results: I. Creating the materials graph: Gaining information through learning is an accumulative process, hence many materials can be a prerequisite for others. The relationships between the topics of the materials should be recognized (Su et al., 2020) and presented in the form of a directed graph. II. Assigning the general difficulty level of the material: Identifying the material difficulty level determines the target group of students for material recommendations. Teachers can manually determine the general difficulty level or automatically via systems. Comments, reviews, ratings, number of views, content, inquiries, and other factors can be used to automatically assign a difficulty level. III. Using augmented reality technology: Many courses and classes require practical experience, specifically science-related ones, such as physics and chemistry. Students cannot undertake experiments in most e-learning environments given that experiments require specialized equipment and usually involve experimenting without the teacher’s direct supervision, which could be unsafe. Students can conduct experiments safely and interactively using augmented reality (El Kabtane et al., 2018; Marienko et al., 2020; Rongting et al., 2016; Wu et al., 2019). Proposal of material recommendation system architecture The study proposed the overall design of the material recommendation system for the school by considering the prior proposals. The technology personalizes the learning experience for students by offering appropriate content depending on their performance and needs. The suggestion procedure is semi-automated as the process involves teacher monitoring. Figure 4 depicts the overall module architecture as follows: I. Course repository: data provided by teachers, such as course syllabus, basic materials, exams, and grades. II. Students profiling module: The module provides student information, such as material history, performance level (updated by the system regularly), material rank, and grade. Part of the information was provided by students while others by teachers. III. Material repository: The module contains the materials retrieved from the internet and recommended to the students. The module also contains reviews and ranks from other reviewers and students. IV. Recommendation module: The module is responsible for producing the recommended materials for each student using DM and artificial intelligence approaches. V. Validation module: Validation of the recommended materials to each student by teachers. Fig. 4 General Architecture of the proposed system Conceptual framework of the proposed architecture The previously presented architecture conceptual framework involved four primary stages: student profiling, material collection, material filtering, and material validation. The student profiling stage oversees the generation of information about students’ needs, performance levels, and academic history using ML and DM approaches. In the material collection stage, DM was used to produce the keywords of courses. The keywords were used to create the queries that were utilized in the material search process. Subsequently, all of the retrieved materials were gathered in a repository. The DM and ML play the most important role during the material screening stage. The ML techniques were utilized primarily to filter the materials in the repository at the stage to select the most appropriate materials for each learner. The chosen materials were validated by teachers who determined whether they benefit the identified students. Each stage is explained in detail in the sub-sections below. Students profiling The stage is primarily responsible for profiling information on students, such as their level of performance, subjects chosen, and material history. Some data were entered by students or teachers, while others were identified through DM and ML techniques. The three types of student information are presented as follows: I. Personal information: Includes students’ gender, name, and grade (level of study). The students entered the information through a separate form or read from the school database. II. Session information: The recommendation system heavily relies on session information. The system provides information about any activity that the student engages in, such as the materials that each student has used, ratings, comments, and the number of views. Summarily, the system contains information on the student’s history. III. Performance information: Students’ performance information was divided into two parts: explicit and learned. The explicit part was entered by teachers or the system through the generated exams and quizzes and includes data, such as students’ marks for each question, exam, and course. Meanwhile, the learned performance utilized the session information apart from the data of the studied materials to learn the student performance through DM and ML approaches. Material collection The materials indicated each course-related item including visual materials, such as videos, reading materials, such as reports, articles, and books, or interactive materials, such as games. The materials were suggested to the students and gathered from the following two sources: I. Materials provided by teachers: Teachers provided materials to students in the study framework, namely basic materials, such as textbooks, lecture notes, additional questions, and exam samples. II. Materials collected from the internet: The teachers’ materials and course information were used to generate queries, which were used to search the internet for similar resources. Finally, a web scrapper was used to collect the materials and their associated data and placed in the materials database. The DM techniques were used to extract keywords from the material provided to develop queries. III. Materials generated by the system: The exams, quizzes, and other materials were generated by the system to suit each student. Material filtering The filtering stage entailed the production of the recommendation. The outcome of the stage is a list of recommended materials for each student, which contains four modules as follows: I. The content-based module: The module is responsible for analyzing the contents of the materials and representing each material with a set of keywords and assigning them to topics and courses. II. The collaborative module: The module used the ratings, reviews, and number of views of the materials in the student’s history. III. The contextual module: The module used the students’ marks and level of performance. IV. The serendipity module: The module used the publicity of the materials and their reviews in the material database. The DM and ML approaches were used in all of the above-mentioned courses. The first three modules shaped students’ study habits and performance and generate sequential study patterns for each student. The sequential patterns included a list of materials that the student should consider. Validation of the recommended materials Teachers recognized students’ weaknesses and strengths and the material filtering stage is useful in finding suitable materials for individual student use by checking the materials beforehand. Therefore, students can view only the materials approved by teachers. Teacher approval of the materials was considered feedback to the system to periodically enhance the recommendation list. Conclusion The study conducted an SLR on PRS-ES. The study identified the system primary components in providing recommendations for developing individualized e-learning recommendation systems. The review was based on articles published between 2017 and 2021 with a focus on publications related to the school setting. The total number of papers reported in the study was 23 based on the screening and quality assessment of the papers. The study suggested a personalized conceptual framework to recommend materials to school students based on the proposed recommendations. The framework operates in a semi-automated mode with certain activities requiring human intervention and others being completed automatically. The four primary stages of the framework are student profiling, material gathering, material filtering, and result validation. The proposed personalized framework can improve student engagement, performance, and knowledge as student behavior, requirements, preferences, background, learning style, and ability are considered. Furthermore, the study focused on school students and presented recommendations for future research directions, hence paving the way for more research. Future research should adopt and test the proposed framework with the aid of teachers and students in Malaysian high schools. The goal of the implementation is to determine the effectiveness of the proposed framework in assisting students’ e-learning. The results are limited based on the review of past literature, thus the study proposed data collection using survey forms and interviews with students and teachers to improve the proposed framework. Additionally, the technique will provide an avenue to identify real needs and preferences and understand the real situation in teaching and learning systems. Acknowledgements This work is supported by Universiti Malaya, Impact Oriented Interdisciplinary Research Grant, grant number IIRG001B-19SAH. Data availability “The data that support the findings of this study are available from the corresponding author upon reasonable request”. Declarations Conflict of interest The authors have no conflicts of interest to declare. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Ali, S., Uppal, M. A., Gulliver, S. R., J., I. T., & People (2018).A conceptual framework highlighting e-learning implementation barriers. Almaiah MA Al-Khasawneh A Althunibat A Exploring the critical challenges and factors influencing the E-learning system usage during the COVID-19 pandemic Education and Information Technologies 2020 25 5261 5280 10.1007/s10639-020-10219-y 32837229 Almaiah, M. A., Al-Khasawneh, A., Althunibat, A. J. E., & Technologies, I. (2020). Exploring the critical challenges and factors influencing the E-learning system usage during the COVID-19 pandemic. 25, 5261–5280. Bernacki, M. L., Greene, M. J., & Lobczowski, N. G. (2021). A Systematic Review of Research on Personalized Learning: Personalized by Whom, to What, How, and for What Purpose(s)? Educational Psychology Review. doi:10.1007/s10648-021-09615-8 Bhaskaran S Santhi B An efficient personalized trust-based hybrid recommendation (tbhr) strategy for e-learning system in cloud computing Cluster Computing 2019 22 1 1137 1149 10.1007/s10586-017-1160-5 Borodinov, A., & Myasnikov, V. (2019). Analysis of the preferences of public transport passengers in the task of building a personalized recommender system. Paper presented at the CEUR Workshop Proc. Burin DI González FM Martínez M Marrujo JG Expository multimedia comprehension in E-learning: presentation format, verbal ability and working memory capacity Journal of Computer Assisted Learning 2021 37 3 797 809 10.1111/jcal.12524 Chang HT Lin CY Wang LW Tseng FC How students can effectively choose the right courses: building a recommendation system to assist students in choosing Courses adaptively Educational Technology & Society 2022 25 1 61 74 Ch DR Saha SK RemedialTutor: a blended learning platform for weak students and study its efficiency in social science learning of middle school students in India Education and Information Technologies 2019 24 3 1925 1941 10.1007/s10639-018-9813-4 Dixit, V. S., & Gupta, S. (2020). Personalized recommender agent for e-commerce products based on data mining techniques. Intelligent Systems, Technologies and Applications (pp. 77–90). Springer. El Kabtane H El Adnani M Sadgal M Mourdi Y Toward an occluded augmented reality framework in e-learning platforms for practical activities Journal of Engineering Science and Technology 2018 13 2 394 408 Emanuel RC Potter WJ Do students’ style preferences differ by grade level, orientation toward college, and academic major? Research in Higher Education 1992 33 3 395 414 10.1007/BF00992267 Fakooa, B., Banon, M. B. D., & Gobin-Rahimbux, B. (2019). A Smart Mobile Application for Learning English Verbs in Mauritian Primary Schools. Information Systems Design and Intelligent Applications (pp. 399–409). Springer. Gong, T. J., Yao, X., Ma, W. GRE: An Adaptive and Personalized Exercise Model for K-12 Online Education. Paper presented at the Proceedings of the 2018 2nd International Conference on Education and, & E-Learning (2018). Bali, Indonesia. https://doi-org.ezproxy.um.edu.my/10.1145/3291078.3291118 Guan, J., Ma, Z., Liu, Q., Chen, H., & Wang, S. (2019). Design of Personal Learning Management Platform Based on Information Technology. Paper presented at the Proceedings of the 2019 3rd International Workshop on Education, Big Data and Information Technology. Hong, J. C., Liu, X., Cao, W., Tai, K. H., & Zhao, L. (2022). Effects of Self-Efficacy and Online Learning Mind States on Learning Ineffectiveness during the COVID-19 Lockdown. Educatio{Hong, 2022 #92}nal Technology & Society, 25 (1),142–154. Hongthong, T., & Temdee, P. (2018). Personalized mobile learning for digital literacy enhancement of Thai youth. Paper presented at the 2018 International Workshop on Advanced Image Technology (IWAIT). Jagušt T Botički I Mobile learning system for enabling collaborative and adaptive pedagogies with modular digital learning contents Journal of Computers in Education 2019 6 3 335 362 10.1007/s40692-019-00139-3 Kitchenham B Pretorius R Budgen D Brereton OP Turner M Niazi M Linkman S Systematic literature reviews in software engineering–a tertiary study Information and software technology 2010 52 8 792 805 10.1016/j.infsof.2010.03.006 Klašnja-Milićević A Ivanović M Vesin B Budimac Z Enhancing e-learning systems with personalized recommendation based on collaborative tagging techniques Applied Intelligence 2018 48 6 1519 1535 10.1007/s10489-017-1051-8 Klašnja-Milićević A Vesin B Ivanović M Social tagging strategy for enhancing e-learning experience Computers & Education 2018 118 166 181 10.1016/j.compedu.2017.12.002 Kopeinik, S., Lex, E., Seitlinger, P., Albert, D., & Ley, T. (2017). Supporting collaborative learning with tag recommendations: a real-world study in an inquiry-based classroom project. Paper presented at the Proceedings of the Seventh International Learning Analytics & Knowledge Conference, Vancouver, British Columbia, Canada. https://doi-org.ezproxy.um.edu.my/10.1145/3027385.3027421 Lee DH Cho SH Kim Y A design and development of the learning contents management based on the personalized online learning International Journal on Advanced Science Engineering and Information Technology 2018 8 4 1321 1326 10.18517/ijaseit.8.4.5724 Li, H., Li, H., Zhang, S., Zhong, Z., Cheng, J. J. N. C., & Applications (2019). Intelligent learning system based on personalized recommendation technology. 31(9), 4455–4462. Ma ZH Hwang WY Shih TK Effects of a peer tutor recommender system (PTRS) with machine learning and automated assessment on vocational high school students’ computer application operating skills Journal of Computers in Education 2020 7 3 435 462 10.1007/s40692-020-00162-9 MacHado, G. M., & Boyer, A. (2021). 2021). Learning Path Recommender Systems: A Systematic Mapping. Paper presented at the UMAP 2021 - Adjunct Publication of the 29th ACM Conference on User Modeling, Adaptation and Personalization. Marienko, M., Nosenko, Y., & Shyshkina, M. (2020). Personalization of learning using adaptive technologies and augmented reality. arXiv preprint arXiv:2011.05802. Mimis M El Hajji M Es-Saady Y Guejdi AO Douzi H Mammass D A framework for smart academic guidance using educational data mining Education and Information Technologies 2019 24 2 1379 1393 10.1007/s10639-018-9838-8 Missaoui, S., Kassem, F., Viviani, M., Agostini, A., Faiz, R., Pasi, G. J. P., & Computing, U. (2019). LOOKER: a mobile, personalized recommender system in the tourism domain based on social media user-generated content. 23(2),181–197. Mudenda S Zulu A Phiri MN Ngazimbi M Mufwambi W Kasanga M Banda M Impact of coronavirus disease 2019 (COVID-19) on college and university students: a global health and education problem Aquademia 2020 4 2 ep20026 10.29333/aquademia/8494 Mutahi, J., Kinai, A., Bore, N., Diriye, A., & Weldemariam, K. (2017). Studying engagement and performance with learning technology in an African classroom. Paper presented at the Proceedings of the Seventh International Learning Analytics & Knowledge Conference, Vancouver, British Columbia, Canada. https://doi.org/10.1145/3027385.3027395 Nian LH Wei J Yin CB The promotion role of mobile online education platform in students’ self-learning International Journal of Continuing Engineering Education and Life Long Learning 2019 29 1–2 56 71 10.1504/IJCEELL.2019.099244 Nidhra S Yanamadala M Afzal W Torkar R Knowledge transfer challenges and mitigation strategies in global software development—A systematic literature review and industrial validation International journal of information management 2013 33 2 333 355 10.1016/j.ijinfomgt.2012.11.004 Nurzaman, A. F., Tedja, R. T., Kusuma, L., Kurniawan, Y., Bhutkar, G., & Johan, J. (2021). LAISES – Learning AI Integrating System for Elementary Students. Paper presented at the 2021 the 5th International Conference on Graphics and Signal Processing, Nagoya, Japan. https://doi-org.ezproxy.um.edu.my/10.1145/3474906.3474913 Perišić J Milovanović M Kazi Z A semantic approach to enhance moodle with personalization Computer Applications in Engineering Education 2018 26 4 884 901 10.1002/cae.21929 Premlatha, K., Dharani, B., & Geetha, T. J. I. L. E. (2016). Dynamic learner profiling and automatic learner classification for adaptive e-learning environment. 24(6),1054–1075. Radha, R., Mahalakshmi, K., Kumar, V. S., & Saravanakumar, A. (2020). J. I. j. o. c., & automation. E-Learning during lockdown of Covid-19 pandemic: A global perspective. 13(4), 1088–1099. Rahman, M. M., Abdullah, N. A., & J. I., A. (2018). A personalized group-based recommendation approach for Web search in E-learning. 6, 34166–34178. Raj NS Renumol VG A systematic literature review on adaptive content recommenders in personalized learning environments from 2015 to 2020 Journal of Computers in Education 2021 10.1007/s40692-021-00199-4 Rasheed F Wahid A Learning style detection in E-learning systems using machine learning techniques Expert Systems with Applications 2021 174 114774 10.1016/j.eswa.2021.114774 Rohani, D. A., Quemada Lopategui, A., Tuxen, N., Faurholt-Jepsen, M., Kessing, L. V., & Bardram, J. E. (2020). MUBS: A personalized recommender system for behavioral activation in mental health. Paper presented at the Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems. Rongting, Z., Yiran, S., Tongliang, H., & Asmi, F. (2016). Applying Augmented Reality Technology to E-Learning: Science Educational AR Products as an Example. Paper presented at the 2016 IEEE 13th International Conference on e-Business Engineering (ICEBE). Saito T Watanobe Y Learning path recommendation system for programming education based on neural networks International Journal of Distance Education Technologies (IJDET) 2020 18 1 36 64 10.4018/IJDET.2020010103 Sajjad IU Imran AS Muhammad K Fayyaz N A systematic mapping review on MOOC Recommender Systems Ieee Access : Practical Innovations, Open Solutions 2021 9 118379 118405 10.1109/ACCESS.2021.3101039 Sarwar, S., Qayyum, Z. U., García-Castro, R., Safyan, M., Munir, R. F., J., M. T., & Applications (2019). Ontology based E-learning framework: A personalized, adaptive and context aware model. 78(24), 34745–34771. Segal A Gal K Shani G Shapira B A difficulty ranking approach to personalization in E-learning International Journal of Human-Computer Studies 2019 130 261 272 10.1016/j.ijhcs.2019.07.002 Su, Y., & Zhang, Y. (2020). Automatic Construction of Subject Knowledge Graph based on Educational Big Data. Paper presented at the Proceedings of the 2020 The 3rd International Conference on Big Data and Education. Su CY Guo Y Factors impacting university students’ online learning experiences during the COVID-19 epidemic Journal of Computer Assisted Learning 2021 37 6 1578 1590 10.1111/jcal.12555 34908643 Supriyatno T Susilawati S Ahdi H E-learning development in improving students’ critical thinking ability Cypriot Journal of Educational Sciences 2020 15 5 1099 1106 10.18844/cjes.v15i5.5154 Tadeo, C. P. S. (2021). ONLINE DISTANCE LEARNING: A TEACHING STRATEGY ON IMPROVING STUDENTS’PERFORMANCE IN SOCIAL STUDIES IN THE NEW NORMAL. Tüysüz M Yıldıran D Demirci N What is the motivation difference between university students and high school students? Procedia-Social and Behavioral Sciences 2010 2 2 1543 1548 10.1016/j.sbspro.2010.03.232 Troussas, C., Krouska, A., & Virvou, M. (2019). Adaptive e-learning interactions using dynamic clustering of learners’ characteristics. Paper presented at the 10th International Conference on Information, Intelligence, Systems and Applications, IISA 2019. Troussas, C., Krouska, A., & Virvou, M. (2019a). Adaptive e-learning interactions using dynamic clustering of learners’ characteristics. Paper presented at the 2019 10th International Conference on Information, Intelligence, Systems and Applications (IISA). Troussas C Krouska A Virvou M MACE: mobile artificial conversational entity for adapting domain knowledge and generating personalized advice International Journal on Artificial Intelligence Tools 2019 28 04 1940005 10.1142/S0218213019400050 Venkatesh M Sathyalakshmi SJEG Smart learning using personalized recommendations in web-based learning systems using artificial bee colony algorithm to improve learning performance an International Journal 2020 16 1–2 101 117 Wongwatkit C Panjaburee P Srisawasdi N A proposal to develop a guided-inquiry mobile learning with a mastery learning mechanism for improving students’ learning performance and attitudes in physics International Journal of Mobile Learning and Organisation 2017 11 1 63 86 10.1504/IJMLO.2017.080898 Wu, S., Liu, C., Shi, H., & Cai, S. (2019). Using Augmented Reality Technology to Learn Cube Expansion Diagram in Spatial Geometry of Elementary Mathematics. Paper presented at the 2019 IEEE International Conference on Engineering, Technology and Education, TALE 2019. Yaniawati P Kariadinata R Sari N Pramiarsih E Mariani M Integration of e-learning for mathematics on resource-based learning: increasing mathematical creative thinking and self-confidence International Journal of Emerging Technologies in Learning (iJET) 2020 15 6 60 78 10.3991/ijet.v15i06.11915 Zhang, Y. (2021). Personalized recommendation method of multimedia network assisted English teaching resources based on particle swarm optimization. Paper presented at the 13th International Conference on Measuring Technology and Mechatronics Automation, ICMTMA 2021. Zhang, Y., & Goh, W. B. (2021). Personalized task difficulty adaptation based on reinforcement learning (pp. 1–32). User Modeling and User-Adapted Interaction. Zhu X Chen Z Dual-modality spatiotemporal feature learning for spontaneous facial expression recognition in e-learning using hybrid deep neural network The Visual Computer 2020 36 4 743 755 10.1007/s00371-019-01660-3	0	PMC9734490	NO-CC CODE	2022-12-14 23:28:29	no	Educ Inf Technol (Dordr). 2022 Dec 5;:1-22	utf-8	Educ Inf Technol (Dordr)	2,022	10.1007/s10639-022-11489-4	oa_other
==== Front Anal Bioanal Chem Anal Bioanal Chem Analytical and Bioanalytical Chemistry 1618-2642 1618-2650 Springer Berlin Heidelberg Berlin/Heidelberg 36477496 4444 10.1007/s00216-022-04444-2 Trends From research to market: correlation between publications, patent filings, and investments in development and production of technological innovations in biosensors Cagnani Giovana Rosso [email protected] da Costa Oliveira Thiago Mattioli Isabela A. Sedenho Graziela C. Castro Karla P. R. Crespilho Frank N. [email protected] grid.11899.38 0000 0004 1937 0722 São Carlos Institute of Chemistry, University of São Paulo, São Carlos, 13560-970 Brazil 7 12 2022 19 12 9 2022 28 10 2022 15 11 2022 © 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 global population grows and science and technology development evolve, fulfilling basic human needs has been even more linked to technological solutions. In this review, we present an overview of the biosensor market and discuss the factors that make certain countries more competitive than others in terms of technology and innovation and how this is reflected in the trends in publication and patent filling. Additionally, we expose briefly how the COVID-19 pandemic acts as a catalyst for the integration of research and development, business, and innovation sectors to bring solutions and ideas that have been predicted as tendencies for the future. Graphical abstract Keywords Biosensors Technological innovation Biosensor research Biosensor market Biosensor development ==== Body pmcIntroduction Biosensing and bioelectronics devices act as one of the most important technological solutions already developed. Nowadays, it is possible to find biosensors with multiple applications in a wide range of areas such as point-of-care (POC) monitoring of treatment and disease progression, environmental monitoring, food and biohazard control, drug discovery, forensics, and biomedical research [1]. Among all of biosensing areas, biosensors dedicated to healthcare and medical analyses are one of the most promising fields for delivering marketable and accessible biosensors, with a commercial purpose and high expectation to fulfill human needs and demands [2]. The increasing wellness market over the years is also responsible for many of the emerging biosensing technologies and innovation. This market aims to fulfill the needs of a population (with or without a diagnosed pathology) who wants to prevent a variety of consequences of aging and improve its self-esteem through body-care [3]. Recently, a McKinsey and Company report showed that consumers define the “wellness market” across six different dimensions: better health, appearance, fitness and nutrition, mindfulness, and high-quality sleep [4]. For monitoring these well-being facets, global biosensing market have invested on wearable and self-powered biosensors [5] capable of connecting with telemedicine apps working through data-driven care, just-in-time diagnosis, and quick symptoms monitoring [4]. In this context, a study reported that total global wearable device revenues are expected to reach USD 73 billion by 2022, especially based on the Asia Pacific fast-growing rate [6]. These statistics include not only biosensors devices, but also hybrid watches, smartwatches, and other electronic technologies that can be integrated to the biosensing assay itself. The global biosensing devices market, by its turn, is projected to reach USD 27.1 billion by 2022, and by 2028, it is anticipated to reach a range of USD 31.5 billion [7–9]. In 2019, the global biosensors market was valued at USD 19.6 billion, expanding at a compound annual growth rate (CAGR) of 7.9% during the forecast period. In comparison, in the same period, the global pharmaceutical manufacturing market was valued at USD 324.42 billion in 2019 and is expected to grow at a CAGR of 13.7% [10]. Furthermore, the global healthcare market reached nearly USD 8.45 trillion in 2018, with a CAGR of 7.3% since 2014. This is expected to grow to nearly USD 11.9 trillion with a CAGR of 8.9% by 2022 [11]. Although the biosensors field is correlated to the pharmaceutical market and healthcare market, why does not it share the same growth as the latter markets? Perhaps the biosensor industry has difficulty in converting knowledge and applying this to technology, that is, marketable products [12, 13]. Recently, COVID-19 pandemics outbreak seemed to worse innovation purposes. A McKinsey and Company report on the impact of COVID-19 in innovation, surveyed across more than 200 companies around the world, presented that innovation is not seem anymore as a high-level priority in different industry areas during COVID-19 crisis [14]. In Medical and Health industry, which presents the majority of biosensors production companies, only approximately 30% of executives considered innovation as priority after COVID-19 outbreak, in comparison to 60% of executives considering it in precrisis moments [14]. These metrics indicate that the conversion of academic or industrial knowledge into future marketable biosensing products may be affected in the next years. In this context, this review aims to explore the academic research and applied technology by comparing the number of scientific publications versus patent filings. Additionally, we present a briefly overview of the integration of R&D, business, and innovation sectors to bring solutions for the COVID-19 pandemic and discuss the impact of science and technology investments that make certain countries more competitive than others in this field. Technological innovations: publications biosensors versus biosensor patents Since the 1970s, biosensors have attracted the interest of researchers from numerous areas, which is supported by the constant growth of scientific literature in this field, with over 150,000 articles, and numerous reviews, books, and chapters published, as can be seen in Fig. 1A. Between the 1990s and 2010, the USA was the country that published the most articles on the subject. However, because of the establishment of the National Medium to Long-term Plan for the Development of Science and Technology (2005–2020) [15] which represents an important milestone in China’s scientific modernization, over the last decade, the country has been leading the publication of articles on biosensors, widely surpassing countries such as the USA and Germany [16], as shown in Fig. 1B. In addition to the countries already mentioned, Korea and India are the remaining of the five main countries that publish on the subject (Web of Science database).Fig. 1 A Types of publications found in the Web of Science for the term "biosensors" for the period from 1960 to 2022. B Progression of publications over the years involving biosensors to five countries that publish the most on this topic In turn, the number of patent applications has also been growing worldwide every year since 2004, with the sole exception of 2009 when growth decreased by 3.8% due to the financial crisis [17]. In 2018, the patent applications in all fields worldwide grew by 5.2%, representing 3.3 million patent applications. In the performed patent search in October 2022, we found around 132,056 patents in the field of biosensors had been registered at World Intellectual Property Organization (WIPO) (Fig. 2A). Of these, contrary to the trend observed in the search for scientific production, when it comes to intellectual property protection, the USA, represented by US Patent and Trademark Office (USPTO), continues to be the largest patent depositary in the biosensor field annually (55,023 patents). In second place is the Patent Cooperation Pact (PCT), which has 27,764 patents. The PCT, an agreement signed on June 19, 1970, in Washington, was created with the purpose of developing a system of patents and technology transfer to promote cooperation between industrialized and developing countries [18]. Nowadays, filing an international patent application under the PCT allows applicants to protect an invention in 153 countries simultaneously, including the largest biosensor markets in the world. Next is the European Patent Office (EPO), which protects 14,057 of patents on biosensors, followed by Intellectual Property Australia (IPAustralia, 9192) and the Canadian Intellectual Property Office (CIPO, 10,229). The China National Intellectual Property Administration (CNIPA, 4,927) showed the same trend observed in article publications, i.e., achieving some protagonism after the “National Medium to Long-term Plan for the Development of Science and Technology.” The Indian Patent Office (IPI, 3,592), Japan Patent Office (JPO, 2,322), Korean Intellectual Property Office (KIPO, 1,739) (WIPO, 2022), and Intellectual Property Office of New Zealand (IPONZ, 594) finalized the ranking of the ten offices that most filed patents at WIPO up to October 2022. Germany, on the other hand, was one of the first countries to protect its intellectual property (along the USA). However, currently, it does not play a leading role in the biosensor patent scenario. Figure 2A shows the evolution of patents filed to biosensor field from 1970 to 2022 and Fig. 2B shows the patent percentages of the offices that most have registered documents at WIPO.Fig. 2 A Evolution of patents filed to biosensor field all long the years in biosensor field and B patent percentages of the offices that most have registered documents at WIPO (search term: biosensor, from 1970 to 2022) Concerning the registration of patents, the WIPO database makes it possible to evaluate the patent classification which depends on the type of innovation that is being protected, as shown in Table 1. With regard to the published articles, the advanced search of the Web of Science was used, and the words highlighted in Table 1 were inputted in the topic field together with the word “biosensor.”Table 1 International class of patent filed related to biosensors, from 1976 to October 2022 (WIPO database) International class code Patent applications (%) Articles published in the topic (%) Description G01N 24.7 19.7 Investigating* or analyzing* materials by determining their chemical or physical properties C07K 18.8 3.9 Peptides* A61K 17.2 3.4 Preparations for medical, dental, or toilet purposes C12N 11.4 21.8 Microorganisms* or enzymes; compositions thereof; propagating, preserving, or maintaining microorganisms; mutation or genetic engineering; culture media C12Q 10.8 18.5 Measuring or testing processes involving enzymes, nucleic acids, or microorganisms A61P 7.4 18.3 Specific therapeutic* activity of chemical compounds or medicinal* preparations A61B 6.5 8.5 Diagnosis; surgery; identification C12P 4.1 5.7 Fermentation* or enzyme-using processes to synthesize a desired chemical compound or composition or to separate* optical isomers from a racemic* mixture *Keywords used in the advanced search field of Web of Science to compare patents and published papers By analyzing the numbers of Table 1 in terms of subjects contemplated in the academic research and intellectual property protection, it can be seen that there is a relationship between the percentage of patents and published articles on biosensors in relation to fermentation processes (4.1% and 5.7%), diagnosis and identification (6.5% and 8.5%), and investigation and analysis (24.7% and 19.7%) of physical and chemical properties. Alternatively, articles involving the application of biosensors in the medical and dental fields comprised the least number of articles among the search results (3.4%), while the corresponding patents add up to 17.2% of the results found. Based on the biosensor scenario presented, it is possible to analyze the relationship between research generated in academia and number of patents and technological transfer. The success of the innovation industry can be intricately linked to the effective technology transfer between the research and development (R&D) sectors, either in universities or industries, and the manufacturing sector. The ability of the industry to assimilate and apply the new knowledge originating from R&D allows the conception and/or maintenance of a competitive advantage. In addition, investing in research has a double effect: it develops new processes and product innovations and develops and expands the company’s ability to identify, assimilate, and exploit the information available in the market [19]. In this sense, the application of the absorptive capacity (ACAP) concept is fundamental. Succinctly, ACAP is defined as the ability to identify and to amass knowledge, by assimilating, internalizing, transforming, and applying it, resulting in the creation of valuable products and services for commercial purposes [20, 21]. To verify how the industry concentrates the knowledge generated in R&D, patents have become the standard measure for innovation in most fields [22]. However, there are those who argue that this is not the most appropriate method, especially because some patents are never commercialized, or some companies use patenting to prevent others from entering their field [23]. Nevertheless, the filing of patents is still a widely used indicator to assess the capacity for technological innovation [24]. Based on that, this indicator was chosen to discuss the innovations coming from university and private initiative in the present work. In WIPO database, we perform a search using the terms “biosensor” and patent applicant (PA). As a result, it obtained a total of the 17% of patents with universities or institutes as applicants. The other results (83% of the total patents) were considered to be originated from private initiative. Analyzing individually each patent office was found the following percentages of patents considered academic: 17.2% from USPTO, 7.2% from JPO and IPI, 15.8% from CIPO, 14.6% from IPAustralia, 22.3% from PCT, and 0.3% from EPO. CNIPA and KIPO, patent offices from China and Republic of Korea, presented almost the same percentage of patents filed among academy (46.8% CNIPA and 44.2% KIPO) and private initiative (53.2% CNIPA and 55.8% KIPO). Figure 3 presents the percentages related to the search for patent academic and private initiative.Fig. 3 Percentages related to patent academic and private initiative in WIPO database (search term: biosensor and PA (university or institute); October 2022) In terms of technologic transfer, the numbers presented through search in WIPO database indicates that there is still a gap between the knowledge production carried out by the academy and its conversion into technological innovation (evaluated here in terms of the filing of patents involving biosensors). This discrepancy can be associated with the fact that not all research developed at universities will be converted into technological innovation, often because the knowledge generated is more focused on unknown phenomena or fundamental research. Another issue is the difficulty in the development of research involving these areas, such as aspects of the structure necessary for this type of research, as well as the stringent measures imposed by the ethics committees. Furthermore, it may be difficult for researchers in academic centers to contemplate and meet the needs of the market, with the appropriate speed to face the problem in question, partially due to its volatility, or even due to not being in direct contact with such demands, or because some of these research centers think that this is not the primary role of academia. Thus, it is imperative to establish effective communication between representatives of the private sector (aware of market demands) and leading researchers of groups in the area of biosensors (holders of know-how), so that the transfer of technology is carried out successfully, and the needs of the general population are met. In contrast, the private initiative holds most of the patents filed at WIPO, because it is important for a company to have patented products in its portfolio to maintain its market share, its current net worth, its sale value when going public, and many other factors. Furthermore, companies use intellectual protection not only to protect the original idea, but also to protect newer versions of the same product, which ends up creating a family of patents. Therefore, this could compromise the idea of using patent numbers as an indicator of technology transfer. However, when performing a new search in WIPO using the term “biosensor” only with the “patent families” field checked, we found a total of 44,713 filed patents. The same search was performed using the term “biosensor and PA (university or institute),” obtaining the value of 9501 filed patents. From these numbers, we obtain the percentage of 21.2% of academic patents and 79.8% of patents from the private sector, values that are not very far from the percentage found for “single patents” (17.2% and 82.8%, respectively). Therefore, based on the concept that technology transfer is a set of steps that describe the formal transfer of inventions resulting from scientific research conducted by the R&D sector (which include universities) to the productive sector [25], the indicators (quantity of patents) presented above demonstrate that, for the biosensors field, the technological flow is better consolidated in the private sector. In an ideal scenario, if technology acquisition came through transfer between academia and the private sector, it would allow companies to acquire new products, processes, or technology without the need to participate in the initial, expensive, and volatile stages of research and development [26], enabling the sharing of risks and costs with other institutions. Science and applied technology as result of innovation investments From 2013 to 2019, UNESCO (United Nations Educational, Scientific and Cultural Organization) has gathered statistical data regarding science, technology, and innovation investment patterns for more than 200 countries [27]. Data were organized either according to the percentage of gross domestic product (GDP) of the country that is allotted to R&D (research & development) or the total invested amount. The obtained trends found for the first 15 countries, in both scenarios, are shown in Table 2.Table 2 Data on R&D funding of the 15 countries that have the highest investments in R&D according to UNESCO’s statistical study conducted from 2013 to 2019 [27] Country Percentage of GDP allotted for R&D Country Total investment in R&D (USD, billions) Republic of Korea 4.1% USA 476 Japan 3.4% China 346 Switzerland 3.2% Japan 170 Austria 3.1% Germany 110 Finland 3.1% Republic of Korea 73 Sweden 3.1% France 61 Denmark 2.9% India 48 Germany 2.9% UK 44 USA 2.7% Brazil 41 Slovenia 2.4% Russian Federation 40 Belgium 2.4% Italy 29 France 2.3% Canada 28 Australia 2.2% Australia 23 Singapore 2.1% Spain 19 Czechia 2.0% Netherlands 16 Generally, countries attempt to allot a fixed percentage of the total GDP to fund science, technology, and innovation, considering R&D as an important business sector of its national economy [27]. However, the GDP percentage allotted to this business sector (Table 2) is still low and dependent on a series of factors, such as working population size, the economic performance of each country in the global market, transparency and political issues, and unemployment rates [28]. Despite being dependent on these factors, there is a consensus in literature that R&D expenditure leads to long-term productivity growth for the country [29]. This is due to the direct correlation between science, technology, and innovation investments and the technological development and independence of countries. These relationships are supported by two main arguments: (i) the higher the R&D expenditure, the greater the technology transferred through science and innovation; and (ii) these expenditures directly affect the industrial innovation potential [29]. The investments of a country in science, technology, and innovation are reflected by the trends in its publication. According to a statistical study carried out by NSF (National Scientific Foundation) in 2018, the five countries that contributed the most to publications in science- and engineering-related academic articles and conference proceedings were 1st China; 2nd USA, 3rd India, 4th Germany, and 5th Japan [30, 31]. Interestingly, except for India, all these countries are part of the top five nations that have the biggest total investment in R&D reported in Table 2. Concerning research on biosensors, the top five in terms of activity in publishing articles under this field in scientific databases (Web of Science and PubMed) until 2020 were 1st China, 2nd USA, 3rd Germany, 4th South Korea, and 5th India. These countries were listed according to the first author’s affiliation, considering the countries with 500 or more publications based on author’s affiliation. This ranking is correlated to the total investment of the first five countries presented in Table 2 (under the R&D column), and this relationship is presented in Fig. 4. The Republic of Korea and UK were also added as they were ranked 5th in terms of total investment and biosensing publication rankings. Four out of the top five most active countries in biosensing literature are among the five countries with the highest investments in science and technology in general. Among them, China is one of the most prominent nations due to its rapid and remarkable economic and scientific growth [16], whereas the other countries have kept similar positions in the past years. The contributions made by the Republic of Korea are also significant and may be attributed to its massive investments in general education over the years [32]. These data are also consistent with the percentage of GDP allotted to the R&D business sector. Moreover, Republic of Korea has the largest ratio of researches per inhabitants, resulting from their recent policies on education and innovation [27]. For a more critical review of publication trends in biosensing according to other kinds of classifications, as well as publication counts for other countries, please see the Olson and Bae report [33].Fig. 4 R&D and biosensor publication trends for the top five most active countries in these fields. Data from the Republic of Korea were added as it is ranked 5th in terms of total R&D expenditure in billions of dollars. Circle sizes are proportional to the total R&D expenditures presented in Table 2 The data shown above confirm what is already known in the research and innovation community: the higher the R&D expenditure, the greater its projection in the number of academic publications in a certain nation. Regarding biosensing technology, these trends are still valid. This is one of the most important factors that influence the innovation stats of a nation. Furthermore, the information from a research article shared with the scientific community, for example, can also be converted into an innovative commercial biosensor, depending on the stage of this research. In this context, a country with several academic biosensing publications has the potential to lead in terms of biosensor innovation globally, if other social and technological barriers are successfully surpassed with a short research-innovation conversion time. COVID-19 pandemics: how has it influenced business, R&D, and innovation for biosensing market? The COVID-19 pandemic outbreak in February of 2020 led to a worldwide public health crisis which evidenced different social and technological limitations and inequalities of society in this scope [34]. For example, we have seen the facility of access that developed communities had in relation to vaccines and high-quality hospitals in contrast to the precarious conditions of public health systems and disorganized pandemic control strategies of emerging and non-developed countries. Despite these social aspects, COVID-19 pandemics also brought challenges for R&D, business, and innovation, which directly or indirectly affected biosensing market (and others) and led to the need of further integration between these areas. In 2020, they were rapidly mobilized to provide solutions for pandemic control in order to fulfill a main human need with social, political, and economic implications: reducing social distancing. For this, science, technology, and innovation gained much attention from governments and worldwide organizations, and, as a result, billions of dollars were mobilized for researching solutions [35]. In this sense, innovation was considered one of the pillars for pandemic control and overcoming, as the demand for it remarkably increased. Despite the executives’ pessimism toward innovation during pandemics previously discussed in the “Introduction” section [14], a variety of policies was adopted to ease some barriers frequently faced during innovation process, such as regulatory flexibilities, stimulation of collaborations between startups, industries and academic institutions, hackathons and competitions, and fast-track support [35]. According to an OECD report, the result of these efforts was reflected on the expressive rapidness of vaccine development. As of November 2020, approximately 10 months after pandemic outbreak, it was reported more than 200 vaccine candidates under development, according to WHO data [35]. Moreover, alternative COVID-19 treatments and diagnosis methodologies were also rapidly studied and proposed. As of April 2020, the WHO had more than 200 reports of it [35]. However, the impressive rapidness of innovation seen, especially during 2020, is also a reflection from integration with R&D institutions, which also faced some facilities that improved its performance, as knowledge diffusion through research made publicly available, extensive adoption of preprint publication, access to critical research infrastructure in some institutions, and others [35]. The success of R&D, business, and innovation integration was also reflected on the generation of new technologies and implementation of past knowledge into emerging solutions. Besides, in some cases, COVID-19 pandemic outbreak acts as a catalyst to solutions and ideas that have been forecasted as tendencies for the future. As an example, the rise of digital healthcare and telemedicine [36, 37] for patient self-care in a social isolation context has been predicted by several authors in literature since, at least, the 2010s, as a future trend for medical area [38]. On the other hand, viral screening in a population was adopted as one of the most important strategies for pandemic control, relying on mass-testing by employing rapid, easy-to-handle and accessible biosensors and POC assays [39], as lateral flow immunochromatographic devices (LFIDs). This sort of device has already been largely studied over the years [40], since its first conceptualizations in the 1960s [41], and a large number of patents have been deposited since then. Therefore, in this context, the successful integration of R&D efforts over the past years with LFID market knowledge was of great usefulness for guiding the rapid development of COVID-19 rapid lateral flow tests. Moreover, the swiftness of the conversion of research into marketable products seen in COVID-19 LFIDs can be attributed to the previous market and industry experience, as, for example, with pregnancy immunochromatographic rapid tests and other technologies based on the same working principles. Concluding remarks After analysis of the databases, it is possible to conclude that a country’s investment in science, technology, and innovation is reflected by the trends in its publication. This was evidently seen when the top five countries in terms of activity in a publication related to biosensors were the same five countries which invested the most in science and technology. However, there is still some gap between scientific research and technological innovation which hinders the production of a commercially viable biosensor and its introduction into the market. This highlights the need for researchers to better understand consumer behavior and the importance of interactions between researchers from different fields (chemistry, biology, medicine, and engineering, for example) as well as between the academy and companies. Government and/or private financial investments remain essential to the development of translational research. An alliance of experts with different backgrounds and significant R&D investments will provide high-impact scientific production, which consequently leads to the filing of patents for new high-impact products in the market. However, the entire process of technological innovation is not simple and involves several steps. Furthermore, in light of the present review, there is a perception that a lack of well-established methodology in conjunction with ineffective communication between the involved parts, hinders, even more, the innovative technological transfer. Thus, the transfer of technological innovation between universities and the biosensors market is a field of research to be explored with the possibility of carrying out future studies. Outlook Studies have been conducted with the goal of developing biosensors in different fields of application. Many of these projects are deposited in important patent offices. In the other hand, only few patent projects became commercial products. This scenery tends to change with integralization of the sectors. In other words, by the technology transfer between academic areas (as engineering, chemistry, materials science, and computation) and industry. Countries that invest in multidisciplinary teams will be a step ahead when it comes to resolving technological barriers that often prevent the launch of new products. This makes them emerge as leaders in the main biosensor market segments. In addition, in the emergence of unexpected demands, as in the case of the COVID-19 pandemic, these research centers will be able to develop solutions to control the spread of disease through rapid diagnose using specific biosensors. Furthermore, there are other areas that need strategic attention, such as the segment of cancer biomarkers. Many researches are carried out and patents are filed; however, there are few commercial products. In such cases, there are problems that need to be solved, whether regulatory or technological, and once solved; the demands of a vast portion of the market can be met. Author contribution Giovana Cagnani: Conceptualization; visualization; writing, original draft; writing, review and editing; formal analysis; investigation Thiago Oliveira: Conceptualization; visualization; writing, original draft; writing, review and editing; formal analysis; investigation Isabela Mattioli: Conceptualization; visualization; writing, original draft; writing, review and editing Graziela Sedenho: Visualization; writing, original draft; formal analysis; investigation Karla Castro: Writing, original draft; investigation Frank Crespilho: Conceptualization; resources, supervision Funding Authors gratefully acknowledge the financial support provided by the São Paulo Research Foundation (FAPESP) (Projects 2020/04796–8; 2018/22214–6; 2019/15333–1; 2019/12053–8; 2022/09164-5), Coordinating Agency for Advanced Training of Graduate Personnel (CAPES) (Project 88881.504532/2020–01), and MeDiCo Network CAPES-Brazil grant number: 23038.003012/2020–16. Declarations Conflict of interest The authors declare no competing interests. Published in the topical collection Young Investigators in (Bio-)Analytical Chemistry 2023 with guest editors Zhi-Yuan Gu, Beatriz Jurado-Sánchez, Thomas H. Linz, Leandro Wang Hantao, Nongnoot Wongkaew, and Peng Wu. Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Giovana Rosso Cagnani and Thiago da Costa Oliveira contributed equally to this work. ==== Refs References 1. Alegret S, Merkoçi A. Comprehensive analytical chemistry: electrochemical sensor analysis. 1 St. Alegret S, Merkoçi A, editors. Vol. 49, Elsevier. Elsevier Science; 2007. 1308 p. 2. Sadana A. Market size and economics for biosensors. In: Fractal binding and dissociation kinetics for different biosensor applications. 1st ed. Elsevier Science; 2005. p. 265–299. 3. Sacavém A Correia A The industry of wellness: the improvement of well-being Int J Sport Manag Mark 2009 5 4 417 425 4. Callaghan S, Lösch M, Pione A, Teichner W. The future of the $1.5 trillion wellness market \| McKinsey. McKinsey and Company. 2021. Available from: https://www.mckinsey.com/industries/consumer-packaged-goods/our-insights/feeling-good-the-future-of-the-1-5-trillion-wellness-market. Accessed July 2021. 5. Zeng X, Peng R, Fan Z, Lin Y. Self-powered and wearable biosensors for healthcare. Mater Today Energy. 2021;100900. 6. Mück JE Ünal B Butt H Yetisen AK Market and patent analyses of wearables in medicine Trends Biotechnol 2019 37 6 563 566 10.1016/j.tibtech.2019.02.001 30851983 7. ReportLinker. Biosensors market size, share & trends analysis report by application by technology, by end use, by region and segment forecasts, 2020 - 2027. 2020. 8. Ahmadi A, Moradi S. Recent Advances in Biosensor Technology. Open Acc J Bio Sci. 2021;3(6):1376. 9. Grand View Research. Biosensors market worth $41.8 billion by 2028 \| CAGR: 7.9%. Grand View Research. 2021. Available from: https://www.grandviewresearch.com/industry-analysis/biosensors-market. Accessed July 2021. 10. Manufacturing, Pharmaceutical size, market analysis, trends by, report development, drug formulation, by of, by route area, by therapy prescription, by group, by age forecasts, segment [Internet]. Grand View Research. 2020;2027:2020–7. Available from: https://www.grandviewresearch.com/industry-analysis/pharmaceutical-manufacturing-market#:~:text=The global pharmaceutical manufacturing market,13.74%25 from 2020 to 2027. Accessed July 2021. 11. Wood L. The $11.9 Trillion global healthcare market: key opportunities & strategies (2014–2022) [Internet]. BusinessWire. 2019 [cited 2021 Oct 25]. p. 11–3. Available from: %22Healthcare Global Market Opportunities And Strategies To 20%22. 12. Bathelt H Kogler DF Munro AK Social foundations of regional innovation and the role of university spin-offs: the case of Canada’s Technology Triangle Ind Innov 2011 18 5 461 486 10.1080/13662716.2011.583462 13. Fontes M The process of transformation of scientific and technological knowledge into economic value conducted by biotechnology spin-offs Technovation 2005 25 4 339 347 10.1016/j.technovation.2003.08.004 14. Bar-Am J, Furstenthal L, Jorge F, Roth E. Innovation in a crisis: why it is more critical than ever. Prioritizing innovation today is the key to unlocking postcrisis growth. McKinsey & Company. 2020. 15. Huang C Wu Y State-led technological development: a case of China’s nanotechnology development World Dev 2012 40 5 970 982 10.1016/j.worlddev.2011.11.013 16. Xie, Qingnan and Freeman RB. Bigger than you thought: China’s contribution to scientific publications. Natl Bur Econ Res [Internet]. 2018;24829:27. Available from: https://ssrn.com/abstract=3214375. 17. WIPO. WIPO statistics database, Patents [Internet]. World Intellectual Property Indicators 2019. Geneva; 2019. Available from: https://www.wipo.int/publications/en/details.jsp?id=4464. Accessed July 2021. 18. WIPO. Patent Cooperation Treaty (PCT). 2001. 19. Cohen WM Levinthal DA Innovation and learning: the two faces of R & D Econ J 1989 99 397 569 10.2307/2233763 20. Cappellari G, Welter CVN, Hermes LCR, Sausen JO. Absorptive capacity: components and organizational mechanisms for its development. Rev Adm Mackenzie. 2019;20(6). 10.1590/1678-6971/eRAMD190028. 21. Cohen WM Levinthal DA Absorptive capacity: a new perspective on learning and innovation Adm Sci Q 1990 35 1 128 10.2307/2393553 22. Fernandez DJ A simple index of innovation with complexity J Informetr 2017 11 1 1 17 10.1016/j.joi.2016.10.009 23. Sheikh NJ, Sheikh O. Bibliometrics and patents: case of forecasting biosensor technologies for emerging point-of-care and medical iot applications. Innov Discov Netw Anal Res Invent Act Technol Manag. 2018;25–43. 24. Dutta S, Lanvin B, Wunsch-Vincent S. Global innovation index 2020: who will finance innovations? 2021. Available from: https://www.wipo.int/edocs/pubdocs/en/wipo_pub_gii_2020.pdf. Accessed July 2021. 25. Van Gils M, Vissers G, De Wit J. Selecting the right channel for knowledge transfer between industry and science: consider the R&D‐activity. Eur J Innov Manag. 2009. 26. Hung S-W Tang R-H Factors affecting the choice of technology acquisition mode: an empirical analysis of the electronic firms of Japan, Korea and Taiwan Technovation 2008 28 9 551 563 10.1016/j.technovation.2007.10.005 27. UNESCO. How much does your country infest in R&D? [Internet]. 2019 [cited 2021 Sep 13]. p. 6. Available from: http://uis.unesco.org/apps/visualisations/research-and-development-spending/. 28. Ilter C What economic and social factors affect GDP per capita a study on 40 countries J Glob Strateg Manag 2017 11 2 51 62 10.20460/JGSM.2018.252 29. Coccia M. What is the optimal rate of R&D investment to maximize productivity growth? Technol Forecast Soc Change [Internet]. 2009;76(3):433–46. Available from: 10.1016/j.techfore.2008.02.008. 30. McCarthy N. The countries leading the world in scientific publications. Forbes [Internet]. 2019;1–2. Available from: https://www.statista.com/chart/20347/science-and-engineering-articles-published/. 31. NSB. Publications output: U.S. Trends and International Comparisons. 2019. Available from: https://ncses.nsf.gov/pubs/nsb20214. Accessed August 2021. 32. Hultberg P, Calonge DS, Kim S, Hultberg P, Calonge DS, Kim S. Education policy in South Korea: a contemporary model of human capital accumulation? Education policy in South Korea: a contemporary model of human capital accumulation? Cogent Econ Financ. 2017;5(1):1389804. 33. Olson N, Bae J. Biosensors—publication trends and knowledge domain visualization. Sensors (Switzerland). 2019;19(11):2615. 34. Gorzelany-Dziadkowiec M. COVID-19: business innovation challenges. Sustainability. 2021;13(20):11439. 35. Paunov C, Planes-Satorra S. Science, technology and innovation in the time of COVID-19. OECD Sci Technol Ind Policy Pap. 2021;99(99). 10.1787/234a00e5-en. 36. Agarwal R Gao GG DesRoches C Jha AK The digital transformation of healthcare: current status and the road ahead Inf Syst Res 2010 21 4 796 809 10.1287/isre.1100.0327 37. Kichloo A Albosta M Dettloff K Wani F El-Amir Z Singh J Telemedicine, the current COVID-19 pandemic and the future: a narrative review and perspectives moving forward in the USA Fam Med Community Health 2020 8 3 1 9 38. Lupton D The digitally engaged patient: self-monitoring and self-care in the digital health era Soc Theory Heal 2013 11 3 256 270 10.1057/sth.2013.10 39. Mattioli IA, Hassan A, Oliveira ON, Crespilho FN. On the challenges for the diagnosis of SARS-CoV-2 based on a review of current methodologies. Vol. 5, ACS Sensors. 2020. p. 3655–77. 40. Paek SH Lee SH Cho JH Kim YS Development of rapid one-step immunochromatographic assay Methods 2000 22 1 53 60 10.1006/meth.2000.1036 11020318 41. Pristoupil TI Fricová V Hrubá A Microchromatography, immunochromatography and detection of proteolytic activity on nitrocellulose membranes impregnated with proteins J Chromatogr 1997 26 127 131 10.1016/S0021-9673(01)98846-6	36477496	PMC9734492	NO-CC CODE	2022-12-14 23:28:29	no	Anal Bioanal Chem. 2022 Dec 7;:1-9	utf-8	Anal Bioanal Chem	2,022	10.1007/s00216-022-04444-2	oa_other
==== Front Front Med Front Med Frontiers of Medicine 2095-0217 2095-0225 Higher Education Press Beijing 36469233 917 10.1007/s11684-021-0917-7 Research Article Achievements of the national malaria control and elimination program in the People’s Republic of China: the Atlas of Malaria Transmission in China Feng Jun 12 Zhang Li 1 Xia Zhigui 1 Zhou Shuisen 1 Xiao Ning 12 Zhou Xiao-Nong [email protected] 12 1 grid.508378.1 National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention (Chinese Center for Tropical Diseases Research); NHC Key Laboratory of Parasite and Vector Biology; WHO Collaborating Centre for Tropical Diseases, National Center for International Research on Tropical Diseases, Shanghai, 200025 China 2 grid.16821.3c 0000 0004 0368 8293 School of Global Health, Chinese Center for Tropical Diseases Research, Shanghai Jiao Tong University School of Medicine, Shanghai, 200025 China 5 12 2022 18 28 7 2021 21 12 2021 © Higher Education 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. In 2017, China achieved the target of zero indigenous malaria case for the first time, and has been certified as malaria free by World Health Organization in 2021. To further summarize the historical achievements and technical experiences of the elimination program, a project on the Roadmap Analysis and Verification for Malaria Elimination in China was carried out. Results of the project were compiled and published as the Atlas of Malaria Transmission in China (The Atlas). The Atlas using modern digital information technologies, has been supported by various data from 24 malaria endemic provinces of China since 1950, to assess the changes in malaria epidemic patterns from 1950 to 2019 at national and provincial levels. The Atlas is designed as two volumes, including a total of 1850 thematic maps and more than 130 charts, consisting of introductory maps, thematic maps of malaria epidemic and control at national and provincial levels. It objectively and directly shows the epidemic history, evolution process, and great achievements of the national malaria control and elimination program in China. The Atlas has important reference value for summing up historical experience in the national malaria elimination program of China, and malaria control and elimination in other endemic countries in the world. Keywords malaria transmission control elimination China atlas ==== Body pmcAcknowledgements The work was supported by the key techniques in collaborative prevention and control of major infectious diseases in the Belt and Road (No. 2018ZX10101002-004). We thank all the staffs from the provincial CDCs in China and Chinese Center for Disease Control and Prevention who provided invaluable support in the study. Compliance with ethics guidelines Jun Feng, Li Zhang, Zhigui Xia, Shuisen Zhou, Ning Xiao, and Xiao-Nong Zhou declare that they have no competing interests. The data in the study were obtained from paper-source and web-source, therefore the ethics and participatory consent was not needed, and this study was approved by the National Institute of Parasitic Diseases, Chinese Center for Diseases Control and Prevention. ==== Refs References 1. Zhou ZJ The malaria situation in the People’s Republic of China Bull World Health Organ 1981 59 6 931 936 6978199 2. Tang LH Chinese achievements in malaria control and research Chin J Parasitol Parasit Dis (Zhongguo Ji Sheng Chong Xue Yu Ji Sheng Chong Bing Za Zhi) 1999 17 257 259 3. Feng J Zhang L Huang F Yin JH Tu H Xia ZG Zhou SS Xiao N Zhou XN Ready for malaria elimination: zero indigenous case reported in the People’s Republic of China Malar J 2018 17 1 315 10.1186/s12936-018-2444-9 30157876 4. Feng J Zhou SS From control to elimination: the historical retrospect of malaria control and prevention in China Chin J Parasitol Parasit Dis (Zhongguo Ji Sheng Chong Xue Yu Ji Sheng Chong Bing Za Zhi) 2019 37 5 505 513 5. Tang LH Qian HL Xu SH Malaria and its control in the People’s Republic of China Southeast Asian J Trop Med Public Health 1991 22 4 467 476 1820632 6. Ministry of Health of China National Malaria Elimination Action Plan (2010-2020) 2010 Beijing Ministry of Health of China 7. Chen JH Fen J Zhou XN From 30 million to zero malaria cases in China: lessons learned for China—Africa collaboration in malaria elimination Infect Dis Poverty 2021 10 1 51 10.1186/s40249-021-00839-y 33875017 8. Feng J Xiao H Xia Z Zhang L Xiao N Analysis of malaria epidemiological characteristics in the People’s Republic of China, 2004–2013 Am J Trop Med Hyg 2015 93 2 293 299 10.4269/ajtmh.14-0733 26078326 9. Feng J Xia ZG Vong S Yang WZ Zhou SS Xiao N Preparedness for malaria resurgence in China: case study on imported cases in 2000–2012 Adv Parasitol 2014 86 231 265 10.1016/B978-0-12-800869-0.00009-3 25476887 10. Feng J Yan H Feng XY Zhang L Li M Xia ZG Xiao N Imported malaria in China, 2012 Emerg Infect Dis 2014 20 10 1778 1780 10.3201/eid2010.140595 25279813 11. Feng J Tu H Zhang L Xia Z Zhou S Imported malaria cases—China, 2012–2018 China CDC Wkly 2020 2 17 277 284 10.46234/ccdcw2020.072 34594639 12. Feng J Zhang L Tu H Zhou SS Xia ZG From elimination to post-elimination: characteristics, challenges and re-transmission preventing strategy of imported malaria in China China Trop Med (Zhongguo Re Dai Yi Xue) 2021 21 1 5 10 13. National Development and Reform CommissionMinistry of Foreign Affairs, Ministry of Commerce of the People’s Republic of China Vision and Actions on Jointly Building Silk Road Economic Belt and 21st-Century Maritime Silk Road 2015 Beijing The State Council of the People’s Republic of China 14. Zhou XN Xia ZG Song YJ Feng J Yin JH Feng XY Liu XW Yang Y Zhang L Zhou HJ The Atlas of Malaria Transmission in China 2020 Shanghai Sinomap Press 15. Yang GJ Tanner M Utzinger J Malone JB Bergquist R Chan EY Gao Q Zhou XN Malaria surveillance-response strategies in different transmission zones of the People’s Republic of China: preparing for climate change Malar J 2012 11 1 426 10.1186/1475-2875-11-426 23256579 16. Feng J Liu J Feng X Zhang L Xiao H Xia Z Towards malaria elimination: monitoring and evaluation of the “1-3-7” approach at the China—Myanmar border Am J Trop Med Hyg 2016 95 4 806 810 10.4269/ajtmh.15-0888 27527626 17. Chen T Zhang S Zhou SS Wang X Luo C Zeng X Guo X Lin Z Tu H Sun X Zhou H Receptivity to malaria in the China—Myanmar border in Yingjiang County, Yunnan Province, China Malar J 2017 16 1 478 10.1186/s12936-017-2126-z 29162093 18. Tambo E Ai L Zhou X Chen JH Hu W Bergquist R Guo JG Utzinger J Tanner M Zhou XN Surveillance-response systems: the key to elimination of tropical diseases Infect Dis Poverty 2014 3 1 17 10.1186/2049-9957-3-17 24971165 19. Liu Y Hsiang MS Zhou H Wang W Cao Y Gosling RD Cao J Gao Q Malaria in overseas labourers returning to China: an analysis of imported malaria in Jiangsu Province, 2001–2011 Malar J 2014 13 1 29 10.1186/1475-2875-13-29 24460982 20. Wong J Achieving universal health coverage Bull World Health Organ 2015 93 9 663 664 10.2471/BLT.14.149070 26478632 21. Wang D Chaki P Mlacha Y Gavana T Michael MG Khatibu R Feng J Zhou ZB Lin KM Xia S Yan H Ishengoma D Rumisha S Mkude S Mandike R Chacky F Dismasi C Abdulla S Masanja H Xiao N Zhou XN Application of community-based and integrated strategy to reduce malaria disease burden in southern Tanzania: the study protocol of China-UK-Tanzania pilot project on malaria control Infect Dis Poverty 2019 8 1 4 10.1186/s40249-018-0507-3 30646954 22. Mlacha YP Wang D Chaki PP Gavana T Zhou Z Michael MG Khatib R Chila G Msuya HM Chaki E Makungu C Lin K Tambo E Rumisha SF Mkude S Mahende MK Chacky F Vounatsou P Tanner M Masanja H Aregawi M Hertzmark E Xiao N Abdulla S Zhou XN Effectiveness of the innovative 1,7-malaria reactive community-based testing and response (1, 7-mRCTR) approach on malaria burden reduction in Southeastern Tanzania Malar J 2020 19 1 292 10.1186/s12936-020-03363-w 32799857 23. Zhou XN China declared malaria-free: a milestone in the world malaria eradication and Chinese public health Infect Dis Poverty 2021 10 1 98 10.1186/s40249-021-00882-9 34253259 24. Feng J Zhang L Zhang SS Xia ZG Zhou SS The progress of national malaria elimination and epidemiological characteristics of malaria in China in 2017 Chin J Parasitol Parasit Dis (Zhongguo Ji Sheng Chong Xue Yu Ji Sheng Chong Bing Za Zhi) 2018 36 3 201 209 25. Galappaththy GN Fernando SD Abeyasinghe RR Imported malaria: a possible threat to the elimination of malaria from Sri Lanka? Trop Med Int Health 2013 18 6 761 768 10.1111/tmi.12097 23506152 26. Feng X Levens J Zhou XN Protecting the gains of malaria elimination in China Infect Dis Poverty 2020 9 1 43 10.1186/s40249-020-00661-y 32326979 27. Lin CY Chen Z Wang SQ Luo PZ Wu DL Zheng AJ Wei JJ Investigation of a rare local epidemic of Plasmodium malariae infection in Sanya City, Hainan Province China Trop Med (Zhongguo Re Dai Yi Xue) 2016 16 5 481 484 28. Sun YW Yu DM Chen J Li X Wang B Wang ZJ Mao LL Yao WQ Two individual incidences of vivax malaria in Dandong municipality of Liaoning Province Chin J Publ Heal (Zhongguo Gong Gong Wei Sheng) 2017 33 2 314 316 29. Feng J Zhang L Zhang SS Xia ZG Zhou SS Epidemiological characteristics of malaria and the progress towards its elimination in China in 2018 Chin J Parasitol Parasit Dis (Zhongguo Ji Sheng Chong Xue Yu Ji Sheng Chong Bing Za Zhi) 2019 37 3 241 247 30. Liu H Zhou Y Deng Y Lin Z Zhang C Chen Q Wei C Duan K Tian P Zhou H Xu J Malaria from hyperendemicity to elimination along international borders in Yunnan, China during 2003–2020: a case study Infect Dis Poverty 2022 11 1 51 10.1186/s40249-022-00972-2 35538510 31. Yang HL Baloch Z Xu JW Sun XD Lin ZR Zhou YW Zhao XT Lv Q Xu SY Ding CL Chen QY Tian P Dung KX Xia XS Zhou HN Malaria: elimination tale from Yunnan Province of China and new challenges for reintroduction Infect Dis Poverty 2021 10 1 101 10.1186/s40249-021-00866-9 34289905 32. Huang F Zhang L Xue JB Zhou HN Thi A Zhang J Zhou SS Xia ZG Zhou XN From control to elimination: a spatial-temporal analysis of malaria along the China—Myanmar border Infect Dis Poverty 2020 9 1 158 10.1186/s40249-020-00777-1 33213516 33. Wang D Lv S Ding W Lu S Zhang H Kassegne K Xia S Duan L Ma X Huang L Gosling R Levens J Abdulla S Mudenda M Okpeku M Matengu KK Serge P D Xiao N Zhou XN Could China’s journey of malaria elimination extend to Africa? Infect Dis Poverty 2022 11 1 55 10.1186/s40249-022-00978-w 35578325 34. Qian MB Li SZ Zhou XN After malaria: which parasitic disease will China eliminate next? Nature 2021 596 7871 189 10.1038/d41586-021-02188-0	36469233	PMC9734496	NO-CC CODE	2022-12-14 23:28:29	no	Front Med. 2022 Dec 5;:1-8	utf-8	Front Med	2,022	10.1007/s11684-021-0917-7	oa_other
==== Front Educ Inf Technol (Dordr) Educ Inf Technol (Dordr) Education and Information Technologies 1360-2357 1573-7608 Springer US New York 11471 10.1007/s10639-022-11471-0 Article Instructional design with ADDIE and rapid prototyping for blended learning: validation and its acceptance in the context of TVET Bangladesh Shakeel Shariful Islam [email protected] http://orcid.org/0000-0001-8490-5891 Al Mamun Md Abdullah [email protected] Haolader Md Faruque Ahmed [email protected] grid.443073.7 0000 0001 0582 2044 Department of Technical and Vocational Education, Islamic University of Technology, 1704 Gazipur, Bangladesh 7 12 2022 130 4 8 2022 22 10 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. Following COVID-19, the global educational landscape shifted dramatically. Almost every educational institute in Bangladesh undertook a strategic move to begin offering online or blended learning courses to mitigate the challenges created by the pandemic. The TVET sector, particularly the polytechnic institute of Bangladesh, endeavored to explore the blended learning approach as an immediate and long-term solution to address the educational dislocation caused by the pandemic. This study attempts to conceptualize a pedagogical design based on the ADDIE and rapid prototyping model to make a reliable and robust instructional design to be used in the blended learning context. A content validity index (CVI) was used to validate the proposed model; a technology acceptance model (TAM) was employed to examine its acceptability to students; and finally, students’ academic performances were analysed to evaluate the overall performance of the proposed instructional design. The findings reveal that the proposed instructional design can be a reliable and valid pedagogical approach to be implemented in the blended learning context for polytechnic students. The proposed instructional design may help TVET educators and course designers to create a robust blended learning environment in the TVET sector and in other similar disciplines, such as science and engineering education. Keywords Instructional design Blended learning Polytechnic institutes ADDIE and RAPID prototyping TVET Bangladesh Islamic University of TechnologyIslamic University of Technology Shakeel Shariful Islam ==== Body pmcIntroduction The delivery of learning is rapidly evolving with the advent of modern technologies. Researchers are continually exploring different ways and methods to create effective online environments for students (Al Mamun et al., 2020; Al Mamun, Hossain, et al., 2022; Lawrie et al., 2016). The recent outburst of COVID-19 further accelerated the adaptation of online learning among educators to address the immediate educational challenges caused by the pandemic. In response, the technical and vocational education and training (TVET) sector in Bangladesh is trying to meet these educational challenges by shifting the course delivery from face-to-face to distance learning mode. However, the report suggests that, in general, the educational institutions of Bangladesh and, in particular, the TVET course designers are facing multifaceted problems in shifting traditional learning to the online environment (Uzzaman et al., 2020). For example, weaknesses are prevalent in the following areas: infrastructure, a lack of modern technologies and low internet speed (Al-Amin et al., 2021), poor preparation of teacher training in pedagogical knowledge and technology use (Rony & Awal, 2019), of teachers awareness of the potential use of technology and poor technical competency (Saidu & Al Mamun, 2022), and student readiness and motivation towards online learning (Al Mamun, Hossain, et al., 2022; Jahan et al., 2021), etc. are key factors obstructing the effective implementation of online education in Bangladesh. Educators and course designers of the polytechnic institutes of Bangladesh are facing even more difficulties in implementing the online learning environment as research has shown that it was far more difficult for technical subjects to be delivered online as some of the topics require hands-on activities and special training to master the skills (Kamal, 2020). In addition, students studying in the polytechnic institutes of Bangladesh usually have low socioeconomic status (Khan, 2019) and, thus, lack modern digital devices and high-speed internet connections in their households (Das, 2021). Also, effective and immediate personal feedback, which is readily available in face-to-face classroom situations, cannot be offered in the online mode effectively. Despite decades of research, online learning lacks the development of such a pedagogical method that integrates a quick and effective feedback mechanism system within the online learning environment (Li et al., 2019). Nonetheless, several research studies have attempted to provide synchronous feedback to students in online and self-regulated learning environments (Al Mamun, 2018; Al Mamun et al., 2020; Al Mamun et al., 2022; Timonen & Ruokamo 2021). However, other studies show that fully online learning delivery may have an unsatisfactory impact on academic achievement due to the absence of direct teacher supervision (Adedoyin & Soykan, 2020). Thus, standalone online learning may not fully meet students’ learning needs. Previous research argues that blended learning (BL) can improve educational approaches to support effective student learning (Kang & Seomun, 2018). Blended learning is a pedagogical approach that offers students experience in both face-to-face and online educational learning modes. This approach allows students to learn anytime, anywhere, and can potentially minimize all the drawbacks of online and traditional learning systems coherently (Carman, 2002). The blended classrooms encourage more active classroom learning by promoting self-confidence and academic success by increasing students’ behavioral, emotional, and intellectual participation in the learning process (Wang et al., 2009). For example, the flipped classroom, which is a popular form of BL, engages students more actively in solving complicated tasks and helps students to develop higher-order thinking skills in different disciplines such as engineering and medical science (Al Mamun, Azad, et al., 2022; M. K. Lee, 2018; Tang et al., 2022). Thus, the BL approach can offer a practical solution to the polytechnic institutes of Bangladesh in minimizing the educational challenges caused by the pandemic. Although several studies used the ADDIE model and rapid prototyping individually to create online instructional designs (Dong, 2021) and blended learning modules (Chen, 2016; Islam et al., 2022; Stapa & Mohammad, 2019), the combination of the two to develop instructional designs for the blended learning environments is non-existent. Particularly in the Bangladesh TEVT context, there is a dearth of research that may give educators a thorough guideline for creating a BL environment that is both rigorous and time-efficient. Thus, this study endeavors to address this gap and develops a BL framework for the TVET educators and course designers to implement in the polytechnic institutes of Bangladesh. The following research questions were posed to meet this end:RQ1: Is the proposed instructional design reliable and valid for the BL environment of the polytechnic institutes of Bangladesh? RQ2: To what extent do the polytechnic students of Bangladesh accept the proposed instructional design? RQ3: Is the proposed instructional design effective in improving students’ academic performances? Blended learning: the context of polytechnic institutes in Bangladesh Blended learning is defined as the coexistence of face-to-face and online learning, where a significant portion of the content is delivered online (Means et al., 2013; Marsh & Drexler, 2001) explained BL as a self-paced but teacher-led, online, and face-to-face classroom delivery system to achieve a flexible and cost-effective education to cater to individual learning approaches. BL combines several modes of instruction, i.e., live online classrooms, face-to-face classrooms, and self-paced learning (Singh, 2003). A designer can create a course from scratch or add extra online activities to a traditional learning setup in this method of learning (Alammary et al., 2014). The effectiveness of BL depends on its design approach, such as how the course is structurally organized and how every aspect of the learning objectives is aligned with the delivery method. The instructors can integrate various instructional components, e.g., lectures, discussions, and different synchronous and asynchronous learning activities, to deliver an effective learning environment. Though online learning is not a new concept in Bangladesh, there is little or no research available that offers a comprehensive BL environment in the context of technical and vocational education and training (TVET), especially for polytechnic students. In fact, designing a suitable BL environment is found to be a complicated task in the TVET context of Bangladesh (Raihan & Han, 2013). Several studies identified key factors that hinder the effective implementation of educational technology in the polytechnic institutes of Bangladesh (Al Mamun, 2012; Al Mamun & Tapan, 2009; M. S. H. Khan et al., 2012). For example, Hossain & Ahmed (2013) reported that due to a shortage of trained resource personnel and the absence of a positive mindset toward technology, the intensity of technology use in educational institutions is fragile and backdated. According to research, the use of technology in the polytechnic institutions of Bangladesh has been limited due to a lack of pedagogical knowledge, technological incompetency, shortage of training, and the absence of modern technology (Al Mamun, 2012). Often instructors fail to integrate technology into their lessons as a result of a huge teaching load (Chowdhury, 2018). Thus, teachers have been unable to find the necessary time to redesign the course components in light of technological advancements (Mndzebele, 2013). Moreover, polytechnic institutes of Bangladesh often struggle with inadequate financial support for developing the ICT infrastructure, which results in insufficient computers, labs, libraries, and modern classroom facilities (Raihan & Shamim, 2008). Thus, these issues have become the key impediments for policymakers and educators in developing and implementing an effective BL environment for the polytechnic institutes of Bangladesh. Technical and vocational education and training (TVET) curricula, specifically at the polytechnic level, are often meant to educate learners for direct entrance into a specific career or trade. Curriculum and instructional designers for online learning need to ensure that polytechnic students will have the same opportunity to achieve those sets of skills and training as they might in the traditional learning environment. However, creating an online environment as per the requirements of the TVET qualification framework while offering carefully designed tasks and activities to develop the desired skill sets in that mode is a formidable task. Though the concept of BL is exciting for some, it is still a daunting task to implement. Nonetheless, the pandemic has compelled polytechnic institutes of Bangladesh to search for a quick, immediate solution to continue the education provisions without any interruption. This study attempts to customize the popular analysis, design, development, implementation, and evaluation (ADDIE) model to build an instructional prototype in the BL environment. As ADDIE has its limitations, e.g., it requires adequate funding and a longer timeframe to implement, this study integrates the concept of rapid prototyping (RP) with the ADDIE approach to design an instructional design for the BL environment (Dong, 2021). However, RP has its drawbacks too. It has a propensity to promote informal design that is not fully formed. Thus, a potential design might be adopted uncritically in the hands of irresponsible or harried designers. Therefore, to achieve a plausible, functional instructional design, this study integrates the concept of ADDIE with rapid prototyping (RP). This approach potentially cancels out the shortcomings of each model. Review of ADDIE and RAPID prototyping model The ADDIE model has become influential since its inception at Florida State University in 1975 (Branson et al., 1975). It is an iterative instructional design method in which the instructional designer may return to any previous step based on the formative evaluation undertaken during the process (Aldoobie, 2015; Kulvietiene & Sileikiene, 2006). This process-based model enables instructional designers, content developers, and even teachers to produce efficient and effective teaching practices. The ADDIE model consists of five phases: analysis, design, development, implementation, and evaluation (see Fig. 1a). In this approach, the result of one phase serves as the starting point for the next one (Aldoobie, 2015). Fig. 1 Instructional design of (a) ADDIE model adapted from Kulvietiene & Sileikiene (2006), and (b) Rapid Prototyping (RP) adapted from Meier and Miller (2016) The analysis phase in the ADDIE model defines, identifies, and determines the feasible solutions to a problem (Kulvietiene & Sileikiene, 2006). The design phase is primarily concerned with implementing the directives that the designer received from the analysis phase. Additionally, throughout the design phase, the instructional designer concentrates on choosing a course format and designing an appropriate instructional approach and assessment technique for the subject (Aldoobie, 2015). Based on the design phase, the development phase aggregates all the separate pieces to create a complete working prototype ready for implementation. Often, instructional designers argue that the components envisioned in the design phase must “come to life” in the development phase (Onguko et al., 2013). The implementation phase is concerned with transacting a plan; it entails three primary steps: training, preparing learners, and structuring the learning environment. Evaluation is the final step of the ADDIE model. The evaluation phase checks each step of the instructional design to ensure that they are aligned with the program’s intended goals. Two types of evaluation need to be undertaken in the ADDIE model. The first one is a formative evaluation, which can be undertaken after completing each phase. The second one is the summative evaluation which assesses the actual value of the whole instructional design at the end of the program. However, the ADDIE approach has a few downsides to creating a quick and effective instructional design. An inherent problem with the ADDIE methodology is that each step is a resource-intensive procedure, requires a longer time for content design and development, and is sometimes expensive (Dong, 2021). Attempting to adhere to all phases of ADDIE during a pandemic is a difficult task for teachers who want a quick instructional design for the effective delivery of the courses. In contrast, rapid prototyping (RP) requires less design time, faster implementation, less cost, and the benefits of more frequent evaluations over the ADDIE approach (Jones & Richey, 2000). Researching is the first stage of the RP model, which is similar to the ADDIE model’s analysis phase (Dong, 2021). In the second stage, designers experiment with the system, identify potential problems, and contribute to selecting a suitable interface for designing the online environment (Miller, 2008) through regular assessments and evidence-based design adjustments (Meier & Miller, 2018). Additionally, Dick et al., (2009) suggest that RP can be implemented through concurrent design and development, which means that most of the analysis work is done concurrently with the production of the initial draft of the design materials. Thus, RP techniques usually minimize production time because they (a) reduce the implementation time by utilizing the working prototypes as the final product and (b) offer continuous modification of the working prototypes. Finally, the summative and formative evaluation of the prototype is undertaken to ensure the viability of the model. Instructional design with ADDIE model and RAPID prototyping In the middle of a pandemic, a course designer needs the quick development of a trustworthy and robust model. Thus, combining ADDIE with rapid prototyping may be used to develop a personalized instructional design that can be communicated, implemented, measured, evaluated, and modified to fine-tune the model for the BL context. In this study, to reduce the design time and make a robust prototype, some features of the RP approach have been integrated with the ADDIE approach. In this process, the review phase has been introduced between the design and development phases of the prototype (Fig. 2). It offers a continuous review process that will accentuate the pace of development for the prototype design and development. Thus, the design and development of this prototype occur concurrently (Dong, 2021). The following is the modified version of an instructional design prototype based on the ADDIE and RP approaches. Fig. 2 Proposed instructional design and development process for the blended learning context In the first phase, designers are engaged with the analysis process to determine the course objectives, contents, instructional strategy, and learners’ background to formulate the conceptual structure of the instructional strategy. The designer must be aware of the types of virtual and social settings that are necessary to assist students in achieving their learning objectives. There is widespread agreement that the most effective techniques for course design begin with a precise definition of course objectives before developing course activities, assignments, and evaluations (McGee & Reis, 2012). Course objectives are especially crucial for blended courses since they may guide material development, mode of delivery (face-to-face class or online), and developing related instructional strategies. In the second phase, designing the instructional prototype and its development takes place. In designing the prototype, three key design components are considered, i.e., the learning activities, assessment procedures, and instructional design. The learning activities and elements of instructional strategies are identified, and the criteria for assessing and evaluating students’ performance are formulated to begin the prototype design. In designing the prototype, the designers and students review the design elements continually. This is a concurrent process where the designer is allowed to work parallel with many design segments while the end-user (students) reviews the prototype that has been forwarded to them. The next step is to develop the instructional prototype in the second phase. The development of the prototype consists of creating learning materials, tutorials, and student activities. Class activities, assessments, assignments, lectures, discussion forums, etc., must be already prepared at this stage. In this stage, the designer must deal with the steps to deliver instructions to the user. The concept of rapid prototyping has been employed in the design and development phase. That means students will have the opportunity to review the prototype even before it is completed. After developing each segment of the prototype, it was forwarded to the reviewer and the students for their views and suggestions. In fact, the opportunity to engage the students in the RP approach significantly improves the evaluation of prototypes alongside the reviewers’ feedback and can be readily used to update the instructional design (Jones & Richey, 2000). The final phase is the implementation stage, where the plan is converted to action. This phase involves three significant tasks, e.g., training the tutor, making students ready, and structuring and deploying the learning environment. In this phase, the course designer must incorporate training sessions to use all the tools integrated into the prototype. The teacher needs to make sure the learners are well-trained to use the prototype and that the learning environment is well-structured and user-friendly. Poon (2013) suggests integrating weekly discussions, teacher feedback, practice sessions, face-to-face meetings, etc., into the course delivery to ensure the quality of learning. Students are found to be motivated to engage with the activities and be responsible for the learning when such activities are integrated within the learning environment (McGee & Reis, 2012). It is to be noted that the students, reviewers, and even the other fellow designers can continually evaluate the prototype and give feedback on the course contents. Implementation of the proposed model in the BL context The proposed model was used to prepare a BL environment for a diploma engineering course titled 67,911: Internal Combustion (IC) Engine Principal, which is offered in the first semester of the Diploma in Marine Technology program. The details of the course curriculum and the required learning materials are available on the website of the Bangladesh Technical Education Board (BTEB) (BTEB, 2016). The BTEB designed the curriculum for face-to-face delivery. The current study restructured the course curriculum in such a way that forty-five per cent (45%) of the content was delivered online and 55% through the face-to-face classroom. For the online part of the course delivery, Google classroom was used. Google Classroom is a freely available learning management system for anyone with a Google account. Google classroom is found to be a simple student-friendly platform for delivering and managing online classrooms (Saidu & Al Mamun, 2022). However, Google classroom does not have any group discussion feature; thus, a separate online discussion forum, i.e., a Facebook group, was created for this class. Table 1 shows the implementation of the proposed model in the BL context. Table 1 Implementation of the proposed model in the BL context Phases Tasks Design Input Design Output Online (45% course) Face-to-face (55% course) Analysis Goal Analysis What are the learning objectives? What are the expected learning outcomes? The learning objectives and the expected learning outcomes are defined in the BTEB curriculum used for designing the BL course materials (BTEB, 2016). The learning objectives and the expected learning outcomes defined in the BTEB curriculum are used for designing the face-to-face course materials (BTEB, 2016). Content Analysis Does the course content align with educational objectives? Does the instructional material consistent with learning outcomes? Learning objectives focus on more factual and conceptual knowledge, Less challenging content, unequivocal and unidirectional learning contents Learning objectives focus on more procedural and meta-cognitive knowledge, More challenging content, equivocal and multidirectional learning contents Instructional Strategy Analysis What types of virtual and social settings are necessary to assist students in achieving these objectives? The online classroom (Google meet/ Zoom) Online chat room Online discussion group Traditional classroom Group discussion Labs/ hands-on Learner Analysis What attitudes, abilities, and prior knowledge are required of students? Assessing students’ attitudes through the technology acceptance model Early assessment of prior knowledge Analysis of Assessment Techniques What abilities should the learner gain after the completion of instruction? How can we know if pupils have met these objectives? Test/ Quizzes Feedback Summative/ formative assessments Design Learning Activity Design What should be the learning activities? Classroom activities/ exercises/ group discussions Assessment Design How to assess the student’s performance? What should be the evaluating criteria? Quiz/ Assignments Test/Activities Projects/ Presentations Instructional Strategy Design What strategies to follow for creating the instructional design? Lectures, discussions, reading activities, online chat room, online discussion group Lectures, discussions, hands-on practices, labs Development Material Development Developing tutorials/ activities How to give overall direction to the classroom? How to align learning objectives with learning outcomes? Online announcements, Worksheets, video clips, images, PowerPoint slides and recorded lectures Lecture notes, activity development, workbooks Implementation Training the instructor Preparing the learner Organizing the learning environment How to train the students and teachers? How to coordinate the learner’s space? Video tutorials, workshops, and seminars Workshops and seminars Evaluation Feedback How to survey the data? How to interpret test results? Review and feedback through opinion forms and survey instrument Methods Context and participants The proposed instructional design was implemented with the IC Engine Principle course offered in the first year at the two TVET institutions of Bangladesh, i.e., Faridpur and Munshiganj institutes of marine technology. There were three groups of participants (Table 2) whose reviews and opinions were elicited at different levels to validate the instructional design and examine its acceptance and usability. Table 2 Participants and types of validation Validation types Validation time Validation Instrument N Role Expertise domain Relevant experiences Content Validation Continuous Reflective journal opinion form 4 Instructors = 2 Students = 2 Mechanical engineering 2 years Content Validation End of each lesson Content Validation Index (CVI) 6 Faculty Engineering/ Educational technology 5 Years Acceptance and usability End of the course Technology Acceptance Model (TAM) 81 Students Throughout the design and development phase, one instructor and one student from each of the polytechnic institutes were involved in reviewing and giving their feedback on the contents employed by the proposed instructional design prototype (Table 2). A reflective journal form was used to obtain feedback from both the instructors and students. Their reviews were considered highly important for further rectification of the course designs. Further, to validate the content, six experts were asked to provide their opinion on the course contents. Their specialization was mechanical, electrical, and computer science engineering, with the research, focus on educational technology. They all have about 5 years of experience doing research related to teaching-learning in an online context. Finally, 81 students who participated in the IC Engine Principle course voluntarily gave their opinions through a survey questionnaire. The survey was designed based on the technology acceptance model (TAM) to examine the acceptance and usability of the proposed instructional design. Model validation, acceptance, and effectiveness Effective instructional design has three key phases –development, validation, and usability (J. Lee et al., 2017). The details of the instructional design and development phases have been discussed in the above sections with supporting theoretical and background information (see Sects. 2, 3, and 4). The following sections discuss the methodology of model validation, acceptance, and usability, i.e., model effectiveness. Validation of the contents To answer RQ1, two sources of data have been utilized in validating the proposed instructional model and its content. First, a reflective journal opinion form (Chitpin, 2006; Cooper & Stevens, 2006) was devised to record the review generated by the reviewers (i.e., two instructors and two students) (Appendix, Table A). It is a potent technique that can help to develop a better and more organized instructional strategy. This reflective journal approach appears to be beneficial in allowing students to reflect on their learning as well as the effectiveness of the instructional method (Cooper & Stevens, 2006). Thorpe (2004) found a wide range of reflections from the reviewers while using this approach. These reviews lead to a more thorough understanding of the instructional technique and help to fine-tune it. Second, the content validation index (CVI) has been used to validate the content of the proposed instructional design (Appendix, Table B). The content Validity Index (CVI) is the most frequently used quantitative measure for determining the validity of the contents of any given course material (Rodrigues et al., 2017). Content validity is a critical early step in enhancing the construct validity of an instrument (Yusoff, 2019). Content validation determines whether the items included in the course design accurately represent all the domains of learning. As a result, content validity works as the primary indicator of how well content is developed (Waltz et al., 2016). Research shows that good content validity implies that the materials are well-developed following current evidence and best practices (Yusoff, 2019). This study formulated the CVI scale based on the methods from several published articles (J. Lee et al., 2017; Polit et al., 2007; Waltz et al., 2016). The data derived from the content validity index (CVI) scale has been analysed to measure whether the experts agree on the content’s suitability for the blended learning context. In the CVI scale, the TVET experts analysed the appropriateness of the content on four different dimensions, i.e., content reliability, comprehensibility, user-friendliness, and the generality of the instructional design (Banyen et al., 2016; Hoffman, 2013; J. Lee et al., 2017). A 4-point rating scale on a continuum from not clear (1) to very clear (4) has been utilized to collect the expert opinion on 14 different items about the contents (Appendix, Table B). Acceptance and usability of the instructional design To answer the RQ2, this study used the technology acceptance model (TAM) to examine its acceptance and usability by the end-users, i.e., students. TAM has been widely utilized in studies on the acceptance of new technologies by users (Davis, 1985). The primary goal of the TAM model is to give insight into users’ views about new technology adoption. The original TAM proposed four key areas to evaluate, i.e., perceived usefulness, perceived ease of use, behavioral intention to use, and actual use (Davis, 1989). Perceived usefulness (PU) is defined as the extent to which an individual feels that utilizing a certain technology would improve his or her work performance (Alshurideh et al., 2019). Several empirical investigations have shown that PU is the most important factor in deciding whether or not to use a particular technology (Tan et al., 2012; Tarhini et al., 2017). Students usually adopt a new technology system when they believe that its use will improve their learning performance (Davis, 1985). Perceived ease of use (PEU) of a system refers to the degree to which an individual believes that utilizing a certain technology is simple and has no difficulties involved (Davis, 1989). In an online learning context, Lin et al. (2011) defined PEOU as the degree to which users perceive that utilizing an e-learning system will be effortless. Previous studies have shown that the perceived ease of use has a major impact on the perceived usefulness of a product (Binyamin et al., 2019; Zogheib et al., 2015). Hence, the authors hypothesize that the students’ perceived ease of use of the proposed instructional model influences the perceived usefulness of the proposed instructional model in the blended learning context.H1: Perceived ease of use (PEU) has a positive influence on perceived usefulness (PU) PU has been shown to have a substantial effect on behavioral intentions (BI) toward e-learning adoption (Ritter, 2017; Teo, 2012; G. K. W. Wong, 2015; Zogheib et al., 2015). There is a substantial positive link between perceived usefulness (PU) and behavioral intention to utilize the e-learning system (Mahmodi, 2017). Thus, the following hypothesis has been formulated:H2. Perceived usefulness (PU) has a positive influence on the Behavioural intention (BI) to use the proposed instructional design Behavioral intention is a cognitive process of a person’s readiness to undertake a particular activity and is a direct precursor of usage behavior. Behavioral intention (BI) refers to the intention of learners to utilize e-learning systems, which often includes a long-term commitment (Liao & Lu, 2008). Research shows that PEU is positively associated with behavioral intention to employ it, both directly and indirectly (Sandjojo & Wahyuningrum, 2016). Therefore, it is hypothesized that:H3. Perceived ease of use (PEU) has a positive influence on the Behavioural intention (BI) to use the proposed instructional design Research has also demonstrated that the behavioral intention of an e-learning system directly and considerably determines the actual usage of the new technology system (Mou et al., 2016). We, therefore, hypothesized that students’ behavioral intention to use the proposed instructional design prototype impacts their actual use of it in the blended learning context.H4: Behavioural intention of use (BI) has a positive influence on the actual use (AU) of the proposed instructional design Based on the TAM literature, we have formulated a 12 items survey to examine the acceptance and usability of the proposed instructional model (Fig. 3). Altogether, 7 items have been adopted from Al-Maroof and Al-Emran (2018) and Mailizar et al., (2021). Finally, five items were newly created to develop the 12 items survey instrument (Appendix, Table C). Fig. 3 Technology Acceptance Model (TAM) to examine students’ acceptance of the proposed instructional model Partial least squares (PLS) based structural equation modelling (SEM) was utilized to analyze the data gathered from the TAM model for evaluating the students’ acceptance of the proposed instructional model. Data from the TAM model helped to re-examine the underlying variables that contribute to students’ adoption of the new instructional design. Confirmatory factor analysis (CFA) was used to validate the construct of each dimension of the TAM model. CFA is a theory-driven approach to determine whether or not the number of factors and their loadings with measured variables adhere to the pre-established theory (Hung et al., 2010). Effectiveness of the instructional design Finally, to evaluate the effectiveness of the proposed instructional design (RQ3), we compared students’ performance in the IC Engine Principle course in the BL context with other engineering courses conducted in the traditional learning context. A paired-sample t-test has been used to compare the students’ performances. Results Model validation Reviewers’ reflections from the opinion form were analysed first to examine the content validity. Reviewers have carefully gone through each of the learning items and online lessons and mentioned whether the items and lessons are suitable for the proposed instructional design for the BL context. The following table shows the reviewers’ (students and faculty members) reflections on the suitability of the contents. Four reviewers were engaged in this section. They were given a model validation form in which they were required to indicate the suitability of the learning contents. Table 3 reveals that most of the reviewers agreed with the suitability of the online course contents for the BL approach. However, some class tests and assignments were extensively modified as per reviewers’ suggestions. Reviewers were also advised to provide suggestions about the learning materials and course contents. Their suggestions were then analyzed qualitatively to rectify the items and improve the fidelity of each lesson. The reviewers offered several recommendations for making the online classroom more engaging and participatory. For example, one of the reviewers wrote,The rubrics of the assignment are necessary for online classrooms to ensure clarity. Table 3 The reviewers’ reflection on the suitability of the contents Classes Items Suitability (in %) Online Class 1 Online lecture 75 Video contents 100 Assignment 75 Class test 50 Online Class 2 Online lecture 100 Video contents 100 Assignment 75 Class test 75 Online Class 3 Online lecture 100 Video contents 75 Assignment 100 Class test 75 Online Class 4 Online lecture 100 Video contents 75 Assignment 75 Class test 100 Online Class 5 Online lecture 100 Video contents 75 Assignment 100 Class test 100 Based on the reviewer’s feedback, a rubric has been developed, which is designed to assist students in reflecting upon their progress in completing activities in online courses. Some reviewers commented on the need to restructure the sequence of the contents; this suggestion was accepted and acted upon. For example, one reviewer said-It will be helpful for the students if the class materials are organized properly in the course stream. It will help the students to follow the classroom properly. . Thus, online learning has been reorganized lesson by lesson in a single interface of the Google classroom so that students can easily follow the learning materials. Also, some incremental changes have been made for each of the lessons as per the reviewers’ feedback, which ensured the prototype’s fidelity. Further, the content validation index (CVI) has been used to examine the experts’ feedback to validate the contents of the proposed instructional design. CVI is the most generally reported technique to validate the content of an instrument or intervention (Zamanzadeh et al., 2015). Typically, the value of item CVI (I-CVI) and Scale-level-CVI (S-CVI) have been used for content validation. The CVI with a value of 1.00 or near 1.00 indicates very good content validity, whereas a value of 0.50 or less indicates an inadequate degree of content validity (Martuza, 1977). The S-CVI is determined based on the number of components in a tool that have received a very positive rating, such as ‘very suitable. Finally, S-CVI using the Universal Agreement (UA) (S-CVI/UA) and (S-CVI/AVE) has been calculated to check the content validity of the course designed for the BL context (Zamanzadeh et al., 2015). As revealed in Table 4, the content validity was found to be high as the I-CVI was greater than 0.80 for each of the constructs. Also, the mean S-CVI/UA was greater than 0.700, and the mean S-CVI/AVE was found to be greater than 0.90, which was deemed satisfactory (Polit et al., 2007). For items’ reliability, the Cronbach alpha coefficient for each construct was more than 0.60, which is considered reliable and satisfactory (Hair et al., 2009). Table 4 Content Validity Index (CVI) of the proposed instructional design Construct Total Items S-CVI/UA S-CVI/AVE Cronbach Alpha Content Reliability 5 0.667 0.889 0.769 Comprehensibility 5 0.667 0.925 0.618 User-friendliness 5 0.833 0.914 0.804 Generality 5 0.667 0.880 0.625 Avg. S-CVI/UA 0.709 Avg. S-CVI/AVE 0.902 Model acceptance Confirmatory factor analysis (CFA) Based on the TAM constructs, this study conducted structural equation modelling (SEM) using smart PLS to evaluate the student’s acceptance of the proposed model (Ramayah et al., 2018). PLS-SEM produces more accurate estimates when the sample size is limited and is recommended for predicting the relationships of the theoretical construct (Hair et al., 2020). PLS analysis employs two distinct models- the measurement and the structural model. The measurement model, also known as the outer model, depicts the underlying relationships of the latent constructs, whereas the structural model, also known as the inner model, defines the relationships between the exogenous and endogenous variables of the model. Gefen et al., (2000) and Hair et al., (2017) presented various recommendations regarding the validation of the measurement and structural models. Based on the recommendations, this study examined the outer loadings of the survey items and the average variance extracted (AVE) to determine the measurement model’s convergent validity. Discriminant validity was determined using cross-loading and the Fornell-Larcker criteria. Additionally, this study used the bootstrapping technique for determining the statistical significance of the path coefficients of the relationships, the Heterotrait-Monotrait (HTMT) ratio, and the coefficient of determination (R2) values (Hair et al., 2017). Henseler et al. (2015) proposed that the HTMT needs to be examined to develop a more stringent discriminant validity of the constructs. The R2 in the structural model was investigated to predict the proportion of the variation of the dependent variable, i.e., students’ acceptance of other independent variables in the model. Table 5 shows the convergent validity and the reliability of the TAM constructs. The composite reliability (CR) and Cronbach’s alpha (α) value are larger than 0.7, and the AVE value larger than 0.50 provide excellent convergent validity and reliability of the model (Fornell & Larcker, 1981). However, in determining the measurement model, one item (AU1) was dropped as the factor loading was found to be below 0.40 for the item (Hulland, 1999). Further, as shown in Table 5, the square root of the AVE on the diagonal (in bold numbers) is greater than the correlations of the constructs, confirming the validity of the measurement model (Hair et al., 2017). Table 5 Reliability, convergent, and discriminant validity Measures Items CR AVE Reliability (α) AU BI PEU PU Actual Use (AU) 2 0.878 0.782 0.722 0.885 Behavioral Intention of Use (BI) 3 0.837 0.632 0.711 -0.242 0.795 Perceived Ease of Use (PEU) 3 0.845 0.646 0.730 -0.113 0.447 0.804 Perceived Usefulness (PU) 3 0.909 0.769 0.856 0.141 0.303 0.144 0.877 Table 6 shows good convergent and discriminant validity as all items had larger loadings (> 0.700) on their respective constructs and lower loadings on other constructs. These data show that the psychometric characteristics of the TAM constructs were excellent for the proposed instructional design (Hair et al., 2017). The HTMT values shown in Table 7 indicate that all model construct values fall below the threshold value of 0.85, which satisfies the condition of strict discriminant validity (Henseler et al., 2015). Table 6 Multicollinearity assessment and factor structure matrix of the model Constructs Items AU BI PEU PU VIF Actual Use (AU) AU2 0.897 -0.225 -0.045 0.018 1.470 AU3 0.872 -0.203 -0.16 0.244 1.470 Behavioral Intention of Use (BI) BI1 -0.145 0.823 0.393 0.367 1.366 BI2 -0.176 0.803 0.355 0.216 1.453 BI3 -0.275 0.758 0.309 0.107 1.366 Perceived Ease of Use (PEU) PEU1 -0.08 0.447 0.853 0.103 1.489 PEU2 -0.089 0.31 0.746 0.146 1.332 PEU3 -0.109 0.286 0.809 0.101 1.631 Perceived Usefulness (PU) PU1 0.132 0.306 0.126 0.911 2.340 PU2 0.197 0.185 0.002 0.818 2.070 PU3 0.083 0.275 0.193 0.899 2.054 Table 7 Results of Heterotrait-Monotrait (HTMT) ratio for discriminant validity Construct AU BI PEU PU Actual Use (AU) Behavioral Intention of Use (BI) 0.347 Perceived Ease of Use (PEU) 0.181 0.594 Perceived Usefulness (PU) 0.231 0.353 0.163 This study also checks the collinearity issue because it affects weight estimates and the statistical significance of the relationships of the items. Table 6 shows that the variance inflation factor (VIF) of all the items is below 5.0, indicating that the model is free from multicollinearity issues (Hair et al., 2017). Finally, the structural model using the TAM constructs has been examined to check the predictive explanatory power (R2), and the cross-validated redundancy (Q2) of the model (Fig. 4). The predictive explanatory power (R2) indicates the degree to which the independent variables adequately explain the dependent variables. Cohen (1988) recommended that predictive explanatory power can be classified as substantial, moderate, or weak when R2 values are above 0.26, 0.13, or 0.02, respectively. Our model shows in Table 8 a moderate explanatory power (R2 = 0.258) for behavioral intention to use (BI) and weak explanatory power for both actual use and perceived usefulness for the proposed instructional design (R2 = 0.059, 0.021). Fig. 4 Results of the structural model using TAM The cross-validation redundancy (Q2) is used to assess the model’s predictive relevance for the latent dependent variables. If Q2 > 0, the model is considered predictively relevant (Geisser, 1975; Stone, 1976). According to the results in Table 8, the structural model is acceptable since the exogenous constructions have predictive relevance for the model’s endogenous components. Table 8 Results of Structural Model Constructs R 2 Q 2 Actual Use (AU) 0.059 0.033 Behavioral Intention of Use (BI) 0.258 0.133 Perceived Usefulness (PU) 0.021 0.005 Table 9 explores the hypotheses test results between different TAM constructs. It reveals that both perceived usefulness (H2) (β = 0.244, t = 2.229, p < 0.05) and perceived ease of use (H3) (β = 0.411, t = 4.025, p < 0.05) had a positive influence on behavioural intention to use the proposed instructional design. Similarly, hypothesis H4 revealed the behavioral intention of use had a positive influence on actual usage of the proposed instructional design (β = -0.242, t = 2.231, p < 0.05), But hypothesis H1 suggests that perceived ease of use did not have a positive influence on perceived usefulness (β = 0.144, t = 0.919, p > 0.05). Table 9 Results of hypotheses testing using path analysis Hypothesis Relationship Std. beta (β) SD t - value p-value decision f 2 H 1 PEU → PU 0.144 0.156 0.919 0.359 Not supported 0.021 H 2 PU → BI 0.244 0.110 2.229 0.026 Supported 0.079 H 3 PEU → BI 0.411 0.102 4.025 0.000 Supported 0.223 H 4 BI → AU -0.242 0.109 2.231 0.026 Supported 0.062 Sullivan and Feinn (2012) stressed that a p-value indicates the presence of a statistically significant impact but does not provide insights into the strength of these relationships. Thus, it is important to present and evaluate the impact size (f2) to understand the strength of these relationships. Cohen (1988) reported that the f2 values of 0.02, 0.1, and 0.35 indicate small, medium, and large impact sizes, respectively. The results in Table 9 show that the impact of perceived usefulness (PU) on behavioral intention (BI) and the impact of behavioral intention (BI) on actual use (AU) are small. In contrast, perceived ease of use (PEU) has a medium impact size on behavioral intention (BI) to use the proposed instructional design. Model effectiveness After running a full semester in the BL context, students’ final exam score in the IC Engine Principle course was compared with the scores of other non-BL courses. A paired-samples t-test was used to examine student performances in both contexts. As revealed in Table 10, the results showed an improved performance in the IC Engine Principle course (M = 3.228, SD = 0.60417) compared to the overall CGPA of other non-BL courses (M = 3.087, SD = 0.78382). The result is statistically significant at t (66) = -2.410, p < 0.05. The effect size (0.081) further shows that this is a moderate improvement in students’ performances due to the intervention, e.g., the implementation of the proposed instructional design in the BL context (Cohen, 1988). Table 10 Paired sample t-test results for students’ improvement of performance in the BL context Mean Std. Dev. Paired Differences t df Sig. Mean SD St. error Lower bound Upper bound Overall CGPA (other courses) 3.087 0.784 − 0.140 0.477 0.058 − 0.257 − 0.024 -2.410 66 0.019 GPA (IC Engine course at BL context) 3.228 0.604 Also, the graduate progression chart (Table 11) indicates the passing rate of the students in the IC Engine Principle course. The passing rate is found to be higher for the academic year 2021 when the “IC Engine Principle” course has been delivered in BL mode with the proposed instructional design. Table 11 Students’ progression chart for the IC Engine Principle course Academic Year 2019 2020 2021 Total number of students 84 80 81 Total number of passing students 61 61 67 Graduate progression rate 73% 76% 83% Discussion This study conceptualized a pedagogical framework combining the ADDIE and rapid prototyping model for the blended learning context to be used by the course designers in the TVET context of Bangladesh. The proposed model has been validated by the reviewers and experts, and its effectiveness and acceptance by the students were examined. The findings revealed that the proposed pedagogical design is reliable and valid and thus might be appropriately implemented in the blended learning context of the polytechnic institutes of Bangladesh. The polytechnic students demonstrated a positive attitude towards the model and performed better in the achievement test compared to the students without the blended learning session. Decades of research show the importance of blending online and face-to-face classrooms to offer effective learning experiences for students (J. Lee et al., 2017; Mason et al., 2013). In the context of polytechnic institutes of Bangladesh, a customized ADDIE-RP instructional design confirmed the same for a marine engineering course in the BL environment through content validation (Polit et al., 2007). As revealed, the course implemented with the customized ADDIE-RP model received positive reviews from experts. Research shows that due to its adaptability, ADDIE can contribute to and satisfies most instructional needs (Campbell, 2014). This might be the key contributing factor to receiving such positive reviews from experts and reviewers. However, as ADDIE requires more time to adapt, the RP Model complements this deficiency by providing formative feedback and quick adoption of necessary technologies at an early stage (Dong, 2021). This unique feature of RP offers effective communication among the instructors and facilitates their focus on the teaching and learning activities through trialability in the quickest possible time (Botturi et al., 2007). In summary, RP could address the limitations of ADDIE by integrating formative feedback elements and early adaptations. Thus, the proposed model provides a unique, open, and flexible pedagogical framework to be implemented in the BL environment of the polytechnic institutes of Bangladesh. The findings of this study are consistent with other recent studies that suggested that rapid prototyping and ADDIE should be used together to create instructional design since they both have the potential to enhance blended learning environments (J. Lee et al., 2017). The CVI index of the proposed model also indicates content interpretability, comprehensibility, usability, and generality of the instructional design (J. Lee et al., 2017). Interpretability is described as the capacity to explain or convey the meaning to a person in a way that can be easily understood (Barredo Arrieta et al., 2020). This is the major strength of this model. In a similar vein, the easy comprehensibility of the model allowing the users to grasp what the proposed instructional design can deliver, is a key strength of this model. The usability of the model also received a very positive rating from the experts, which confirms it is easy to use, and effective for interaction (de Oliveira et al., 2021). This study utilized the Technical Acceptance Model (TAM) to recognize students’ acceptance of the proposed instructional design and examine its actual use by the students. The findings revealed that perceived usefulness (PU) has a significant positive influence on the behavioral intention (BI) to use this prototype. This result is in line with other studies where there is a strong relationship between perceived usefulness and behavioral intention to use a new instructional prototype (Salloum et al., 2019; K. T. Wong et al., 2013). It can be argued that when students believe that modern technology could enhance their performance, it inherently influences their behavioral intentions to use the technology. This model also revealed that perceived ease of use (PEU) had positively influenced the behavioral intention (BI) to use it. This finding is consistent with the studies conducted by Davis (1989), Motaghian et al. (2013), and Park (2009). It is evident that when students found the proposed model comfortable and easy to use, it positively affected their behavioral intention to use it. It is to be noted that perceived ease of use (PEU) had no positive impact on perceived usefulness (PU), which is in contrast with some other studies that reported a direct positive relationship between them (Akmal, 2017; Cigdem & Topcu, 2015). Liu et al. (2010) also argued that course design and user interface are the most important factors that directly affect PEU and encourage students to opt for new technology. However, Motaghian et al. (2013) also argued that significant positive relations between perceived ease of use (PEU) and perceived usefulness (PU) may not always be established. It is argued that some other variables, e.g., age, gender, subject, instructor preparation and support, and years of teaching experience, could be responsible for this deviation (Dai et al., 2020). Future research can explore further the relationships among these variables with the perceived ease of use and perceived usefulness. Finally, the effectiveness of this model has been examined by implementing the proposed instructional design in a blended learning context. This study designed a marine engineering course with the proposed model and offered it formally to the students for a full semester. The results showed moderate improvement in students’ academic performance in the BL course compared to the non-BL courses. The findings of this study are consistent with earlier studies that found significant improvements in students’ attitudes and satisfaction while taking BL courses and these improvements are directly related to the students’ academic performances (Bland, 2006; Kellogg, 2009; Kintu et al., 2017) concluded that students’ academic performance might be improved by implementing the proper web technology for assignments and exams. Tian & Suppasetseree (2013) also found that an online task-based instructional model significantly improves students’ performance. In fact, educators continually look for innovative approaches to keep classes exciting and engaging while utilizing technology in the blended learning context (Arghode et al., 2018). Implications of the research Though ADDIE is popular among instructional designers (DeBell, 2020), it comes with its limitations. Educators, course designers, and academic libraries should take measures to overcome these limitations while maintaining the quality of the ADDIE model during the instructional design process. The current study offers a modified framework that includes some aspects of RP into ADDIE to improve the overall efficiency of this theoretical approach. Creating such a unique theoretical framework can lay down the foundation around which educators could construct compelling instructional materials for a blended learning environment. Also, this study has several practical implications for TVET educators in the context of higher education in Bangladesh. TVET educators can organize training for the instructional designers to solve certain difficulties related to the instructional design in the blended learning context by employing the proposed model. Specifically, the collaboration between the course designers of different institutions can take place to develop a universal instructional design for higher institutions (Linh & Suppasetseree, 2016). Given that instructional design has an impact on the quality of instruction, this proposed model would assist Bangladeshi TVET course designers in gaining the skill set necessary to create blended learning sessions that will increase the learning effectiveness and efficiency of the students. Since there is no instructional design for establishing a BL environment in Bangladesh, the proposed model, intended for TVET education as well as other higher education programs, can fill this gap. To give educators a better grasp of the underlying design processes and to enable them to make better, more informed decisions, this study explicitly explained the steps of the key design component of the proposed model. Additionally, the outcome of this study might encourage individual TVET teachers to design and implement their courses for blended learning environments at their respective institutes. In this research, Google Classroom and Facebook were shown to be more readily available to be used as educational tools that are suitable for polytechnic students’ learning styles and preferences. The TVET educators could also consider the potential use of social media as viable tools to utilize in instructional design development. Also, this proposed ID can be used as a framework for developing similar courses in other similar domains of learning, i.e., medical, nursing, engineering, and science education. Limitations and future research direction This study was limited to only two polytechnic institutes, and only eighty-one students were engaged in validating the acceptance of the proposed instructional design. All the students came from a single discipline. Recruiting students from different disciplines and institutes could help to scale up the potential effectiveness of the model. Keeping these limitations into consideration, the findings of this study may apply and be generalizable only to the disciplines taught in the polytechnic institutes of Bangladesh. Methodically, this paper focuses mostly on the quantitative data to validate and measure the acceptance and effectiveness of the customized prototype. Though in the reflective opinion form, a limited amount of qualitative data was used for designing and developing the course contents, more qualitative data could be incorporated for subsequent studies to strengthen the validation of the proposed model. This study only focused on micro-level course design (J. Lee et al., 2017), overlooking the macro-level design aspects of the curriculum. In fact, this instructional framework lends itself to an individualistic approach to designing a BL environment for the courses. To secure a comprehensive understanding of the course design, the TVET educators and course designers need to consider aspects of instructional design for both macro and micro levels. This study did not compare the implementation time of the current project with other similar projects. Future studies can investigate how much time could be saved using this modified ADDIE-RP framework compared to other ADDIE approaches. Another methodological limitation of RP is due to the fast-paced approach, which often prevails in the instructional design at the expense of quality. This quick, fast-paced approach can have a detrimental impact on subsequent advances, impeding comprehension, teamwork, and commitment. Though the ADDIE model has elements of quality control, this study did not explicitly examine the drawbacks of the fast-paced RP approach. Future research might consider all the drawbacks of the RP and ADDIE models and can control them during the prototype design process. Conclusion Amid COVID-19, course instructors of the polytechnic institutes of Bangladesh were under high pressure to deliver their teaching. Thus, the TVET course designers urgently looked for options to create a rapid BL session for their students within a short space of time. Nonetheless, it was a daunting task for the researcher to create a compelling and quick prototype convenient for TVET educators, as the concept of online learning is new for many TVET institutions in Bangladesh. This study endeavored to make an effective and quick instructional design for the TVET educators to support the course designers during the pandemic. The proposed instructional design has the potential to solve the immediate educational challenges and can offer a long-term solution even beyond the post-pandemic situations for this group. The core strength of this model was to offer the resource-constrained TVET course designers a framework to accomplish a reliable and robust instructional design within a brief time frame. In this regard, this ADDIE-RP instructional design could bring a major break-through in the BL environment for the TVET institutions of Bangladesh. Appendix Table A Reviewers Opinion Journal. Class No. Item Suitable / Not suitable Suggestions Online Class 1 Online lecture Video contents Assignment Class test Online Class 2 Online lecture Video contents Assignment Class test Online Class 3 Online lecture Video contents Assignment Class test Online Class 4 Online lecture Video contents Assignment Class test Online Class 5 Online lecture Video contents Assignment Class test Table B Content Validation of The Prototype. Constructs Features Items/Dimensions 1 2 3 4 Contents Reliability i. Clearly defined learning goals ii. Alignment of content with educational objectives iii. Exclusion of Unnecessary Information iv. Constructive feedback (i.e., quizzes, self-check questions, exercises, activities, tests, and other practice exercises or testing activities) v. Accuracy of information vi. Consistency of instructional materials with the intended learning outcome Item 1/Online Class 1 Item 2/ Online Class 2 Item 3/Online Class 3 Item 4/ Online Class 4 Item 5/ Online Class 5 Comprehensibility of the prototype i. Prototype actions and understanding ii. Auditory and visual compatibility Item 1/Online Class 1 Item 2/ Online Class 2 Item 3/Online Class 3 Item 4/ Online Class 4 Item 5/ Online Class 5 User-friendliness of the prototype i. A user-friendly interface ii. The ability to divert from the course flow. iii. Distinctive navigation technique iv. The authority of students to evaluate their abilities and practice Item 1/Online Class 1 Item 2/ Online Class 2 Item 3/Online Class 3 Item 4/ Online Class 4 Item 5/ Online Class 5 The generality of the prototype i. Positive Interaction with other instructional designers ii. Reliability of Prototype in designing a comparable course Item 1/Online Class 1 Item 2/ Online Class 2 Item 3/Online Class 3 Item 4/ Online Class 4 Item 5/ Online Class 5 Table C TAM for assessing of proposed instructional design. Constructs Items 1 2 3 4 Perceived Usefulness PU1 PU2 PU3 I believe technology improves my quality of learning. I believe web platforms should be used regularly in teaching. I believe a blended learning environment improved my learning capacity. Perceived Ease of Use PEU1 PEU2 PEU3 The prototype has a user-friendly interface. All the contents in the prototype are easily accessible. I find it easy to navigate through the classroom Behavioral Intention of Use BI1 BI2 BI3 I would like to keep myself updated with new educational technology. I am more comfortable with blended learning than only face-to-face learning. I want to attend more courses that offer blended learning. Actual Use AU1 AU2 AU3 I use all the features of this prototype alone regularly I take part in test activities frequently (e.g., quizzes and assignments). I like to access all digital learning materials daily and can download/ upload files. Funding This research is funded by the Islamic University of Technology (IUT). Data Availability All data are available from the corresponding author upon reasonable request. Declarations Ethical declaration. This study is approved by Committee for Advanced Studies and Research (CASR), Islamic university of Technology (IUT), Bangladesh. 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 Adedoyin, O. B., & Soykan, E. (2020). Covid-19 pandemic and online learning: the challenges and opportunities. Interactive Learning Environments, 1–13. 10.1080/10494820.2020.1813180. Akmal, A. (2017). Influence of Perceived Usefulness and Ease of Use on Students’ Continuous Intention in Learning on-Line English Lessons: an Extended Tam. UAD TEFL International Conference, 1, 28. 10.12928/utic.v1.146.2017 Al-Amin M Zubayer A Al, Deb B Hasan M Status of tertiary level online class in Bangladesh: students’ response on preparedness, participation and classroom activities Heliyon 2021 7 1 e05943 10.1016/j.heliyon.2021.e05943 33506126 Al-Maroof RAS Al-Emran M Students acceptance of google classroom: an exploratory study using PLS-SEM approach International Journal of Emerging Technologies in Learning 2018 13 6 112 123 10.3991/ijet.v13i06.8275 Al Mamun MA Technology in Educational Settings in the Government Polytechnic Institutes of Bangladesh: a critical analysis International Journal of Computer Applications 2012 54 13 32 40 10.5120/8628-2502 Al Mamun, M. A. (2018). The role of scaffolding in the instructional design of online, self-directed, inquiry-based learning environments: student engagement and learning approaches [The University of Queensland]. 10.14264/uql.2018.607 Al Mamun MA Azad MAK Mamun A Boyle M Review of flipped learning in engineering education: scientific mapping and research horizon Education and Information Technologies 2022 27 1 1261 1286 10.1007/s10639-021-10630-z 34257512 Al Mamun MA Hossain MA Salehin S Khan MSH Hasan M Engineering students’ readiness for online learning amidst the COVID-19 pandemic: scale validation and lessons learned from a developing country Educational Technology & Society 2022 25 3 30 45 Al Mamun MA Lawrie G Wright T Instructional design of scaffolded online learning modules for self-directed and inquiry-based learning environments Computers & Education 2020 144 103695 10.1016/j.compedu.2019.103695 Al Mamun MA Lawrie G Wright T Exploration of learner-content interactions and learning approaches: the role of guided inquiry in the self-directed online environments Computers & Education 2022 178 104398 10.1016/j.compedu.2021.104398 Al Mamun MA Tapan SM Using ICT in teaching-learning at the polytechnic institutes of Bangladesh: constraints and limitations Teacher’s World-Journal of Education and Research 2009 33 1 207 217 Alammary A Sheard J Carbone A Blended learning in higher education: three different design approaches Australasian Journal of Educational Technology 2014 30 4 30 10.14742/ajet.693 Aldoobie, N. (2015). ADDIE Model.American International Journal of Contemporary Research, 5(6). Alshurideh, M., Salloum, S. A., Kurdi, A., B., & Al-Emran, M. (2019). Factors affecting the social networks acceptance: An empirical study using PLS-SEM approach. PervasiveHealth: Pervasive Computing Technologies for Healthcare, Part F1479, 414–418. 10.1145/3316615.3316720 Arghode V Brieger E Wang J Engaging instructional design and instructor role in online learning environment European Journal of Training and Development 2018 42 7–8 366 380 10.1108/EJTD-12-2017-0110 Banyen W Viriyavejakul C Ratanaolarn T A blended learning model for learning achievement enhancement of thai undergraduate students International Journal of Emerging Technologies in Learning 2016 11 4 48 55 10.3991/ijet.v11i04.5325 Barredo Arrieta A Díaz-Rodríguez N Del Ser J Bennetot A Tabik S Barbado A Garcia S Gil-Lopez S Molina D Benjamins R Chatila R Herrera F Explainable Artificial Intelligence (XAI): concepts, taxonomies, opportunities and challenges toward responsible AI Information Fusion 2020 58 82 115 10.1016/j.inffus.2019.12.012 Binyamin SS Rutter MJ Smith S Extending the technology acceptance model to understand students’ use of learning management systems in Saudi higher education International Journal of Emerging Technologies in Learning 2019 14 3 4 21 10.3991/ijet.v14i03.9732 Bland, L. (2006). Applying flip/inverted classroom model in electrical engineering to establish life-long learning. ASEE Annual Conference and Exposition, Conference Proceedings. 10.18260/1-2--491 Botturi, L., Cantoni, L., Lepori, B., & Tardini, S. (2007). Fast Prototyping as a Communication Catalyst for E-Learing Design. In J. Willis (Ed.), Making the Transition to E-Learning (pp. 266–283). IGI Global. 10.4018/978-1-59140-950-2.ch016 Branson, R. K., Rayner, G. T., Cox, J. L., Furman, J. P., King, F. J., & Hannum, W. H. (1975). Interservice procedures for instructional systems development.Center for Educational Technology, 4(AD-A019 468). BTEB (2016). Bangladesh Technical Education Board Course Structure & Syllabus (2016 Probidhan). https://bteb.portal.gov.bd/sites/default/files/files/bteb.portal.gov.bd/page/84dc01b3_d072_48f9_8aaa_aca776301210/2099.1.pdf Campbell PC Modifying ADDIE: incorporating New Technologies in Library instruction Public Services Quarterly 2014 10 2 138 149 10.1080/15228959.2014.904214 Carman, J. M. (2002). Blended learning design: Five key ingredients. Learning, 11. 10.1109/CSSE.2008.198 Chen LL A model for effective online Instructional Design Literacy Information and Computer Education Journal 2016 7 2 2303 2308 10.20533/licej.2040.2589.2016.0304 Chitpin S The use of reflective journal keeping in a teacher education program: a popperian analysis Reflective Practice 2006 7 1 73 86 10.1080/14623940500489757 Chowdhury SA Impacts of ICT integration in the Higher Education Classrooms: Bangladesh Impacts of ICT Integration in the Higher Education Classrooms: Bangladesh Perspective Journal of Education and Practice 2018 9 15 35 Cigdem H Topcu A Predictors of instructors’ behavioral intention to use learning management system: a turkish vocational college example Computers in Human Behavior 2015 52 22 28 10.1016/j.chb.2015.05.049 Cohen, J. (1988). Statistical power analysis for the behavioral Sciences (2nd ed.). Routledge. Cooper JE Stevens DD Journal-keeping and academic work: four cases of higher education professionals Reflective Practice 2006 7 3 349 366 10.1080/14623940600837566 Dai Z Wang M Liu S Tang L Design and the technology acceptance model analysis of instructional mapping Computer Applications in Engineering Education 2020 28 4 892 907 10.1002/cae.22261 Das TK Online Education during COVID-19: prospects and Challenges in Bangladesh Space and Culture India 2021 9 2 65 70 10.20896/saci.v9i2.1220 Davis, F. D. (1985). A technology acceptance model for empirically testing new end-user information systems: theory and results [Massachusetts Institute of Technology]. https://dspace.mit.edu/handle/1721.1/15192 Davis FD Perceived usefulness, perceived ease of use, and user acceptance of information technology MIS Quarterly: Management Information Systems 1989 13 3 319 339 10.2307/249008 de Oliveira, M., Mattedi, A. P., & Seabra, R. D. (2021). Usability evaluation model of an application with emphasis on collaborative security: an approach from social dimensions. Journal of the Brazilian Computer Society, 27(1), 10.1186/s13173-021-00108-8. DeBell, A. (2020). What is the ADDIE Model of Instructional Design?Water Bear Learning. https://waterbearlearning.com/addie-model-instructional-design/ Dong H Adapting during the pandemic: a case study of using the rapid prototyping instructional system design model to create online instructional content Journal of Academic Librarianship 2021 47 3 102356 10.1016/j.acalib.2021.102356 Fornell C Larcker DF Evaluating structural equation models with unobservable variables and measurement error Journal of Marketing Research 1981 18 1 39 50 10.1177/002224378101800104 Gefen, D., Straub, D., & Boudreau, M. C. (2000). Structural equation modeling and regression: guidelines for Research Practice. Communications of the Association for Information Systems, 4(Oc), 10.17705/1cais.00407. Geisser S The predictive sample reuse method with applications Journal of the American Statistical Association 1975 70 350 320 328 10.1080/01621459.1975.10479865 Hair, J., Black, Black, B., & Anderson. (2009). &. Multivariate Data Analysis (7th ed.). Pearson. Hair, J., Black, W. C., Babin, B. J., & Anderson, R. E. (2009). Multivariate Data Analysis (7th edition). Hair, J., Howard, M. C., & Nitzl, C. (2020). Assessing measurement model quality in PLS-SEM using confirmatory composite analysis. Journal of Business Research, 109(August 2019), 101–110. 10.1016/j.jbusres.2019.11.069 Hair, J., Hult, G. T. T., Ringle, C., & Sarstedt, M. (2017). A primer on partial least squares structural equation modeling (PLS-SEM). Sage (2nd ed.). Sage Publications. Henseler J Ringle CM Sarstedt M A new criterion for assessing discriminant validity in variance-based structural equation modeling Journal of the Academy of Marketing Science 2015 43 1 115 135 10.1007/s11747-014-0403-8 Hoffman, J. S. (2013). Instructional Design—Step by Step: Nine Easy Steps for Designing Lean, Effective, and Motivational Instruction. iUniverse. https://books.google.com/books?id=xCvPUzd0DmcC&pgis=1 Hossain, A., & Ahmed, A. (2013). Prospect of ICT-based Distance (online) Learning for Pedagogical Development in Bangladesh.National Academy for Educational Management (NAEM) Journal, 9(7). Hulland J Use of partial least squares (PLS) in strategic management research: a review of four recent studies Strategic Management Journal 1999 20 2 195 204 10.1002/(sici)1097-0266(199902)20:2<195::aid-smj13>3.0.co;2-7 Hung MLL Chou C Chen CHH Own ZYY Learner readiness for online learning: scale development and student perceptions Computers and Education 2010 55 3 1080 1090 10.1016/j.compedu.2010.05.004 Islam, M. K., Sarker, M. F. H., & Islam, M. S. (2022). Promoting student-centred blended learning in higher education: a model. E-Learning and Digital Media, 204275302110277. 10.1177/20427530211027721. Jahan, I., Rabbi, F., & Islam, U. N. (2021). Can Blended Learning be the New-Normal in Higher Education of Bangladesh ?Internasional Jurnal of Educational Reserch Review, June,306–317. Jones TS Richey RC Rapid prototyping methodology in action: a developmental study Educational Technology Research and Development 2000 48 2 63 80 10.1007/BF02313401 Kamal, A., & UNICEF Education COVID-19 Case Study. (2020). UNICEF Education COVID-19 (Case Study Malaysia – Empowering Teachers to Deliver Blended Learning after School Reopening 8 July 2020), March, 1–3. Kang J Seomun GA Evaluating web-based nursing education’s Effects: a systematic review and Meta-analysis Western Journal of Nursing Research 2018 40 11 1677 1697 10.1177/0193945917729160 28875770 Kellogg, S. (2009). evelop ping On nline Materi ials to Facilit tate an n In nverte d Clas ssroom m Appr roach. 39th ASEE/IEEE Frontiers in Education Conference, 1–6. Khan, M. A. (2019). Situation Analysis of Bangladesh TVET Sector: A background work for a TVET SWAp. August. Khan MSH Hasan M Clement CK Barriers to the introduction of ICT into education in developing countries: the example of Bangladesh International Journal of Instruction 2012 5 2 61 80 Kintu, M. J., Zhu, C., & Kagambe, E. (2017). Blended learning effectiveness: the relationship between student characteristics, design features and outcomes. International Journal of Educational Technology in Higher Education, 14(1), 10.1186/s41239-017-0043-4. Kulvietiene, R., & Sileikiene, I. (2006). the Blended Learning Delivery Design Model. Conference on Distance Learning and Web Engineering, 1–5. Lawrie, G. A., Schultz, M., Bailey, C. H., Mamun, A., Micallef, M. A., Williams, A. S., M., & Wright, A. H. (2016). Development of Scaffolded Online Modules to Support Self-Regulated Learning in Chemistry Concepts. In M. Schultz, S. Schmid, & T. Holme (Eds.), ACS Symposium Series (Vol. 1235, Issue 1235, pp. 1–21). American Chemical Society. 10.1021/bk-2016-1235.ch001 Lee J Lim C Kim H Development of an instructional design model for flipped learning in higher education Educational Technology Research and Development 2017 65 2 427 453 10.1007/s11423-016-9502-1 Lee MK Flipped classroom as an alternative future class model?: implications of South Korea’s social experiment Educational Technology Research and Development 2018 66 3 837 857 10.1007/s11423-018-9587-9 Li C He J Yuan C Chen B Sun Z The effects of blended learning on knowledge, skills, and satisfaction in nursing students: a meta-analysis Nurse Education Today 2019 82 February 51 57 10.1016/j.nedt.2019.08.004 31437783 Liao HL Lu HP The role of experience and innovation characteristics in the adoption and continued use of e-learning websites Computers and Education 2008 51 4 1405 1416 10.1016/j.compedu.2007.11.006 Lin KM Chen NS Fang K Understanding e-learning continuance intention: a negative critical incidents perspective Behaviour & Information Technology 2011 30 1 77 89 10.1080/01449291003752948 Linh, N. D., & Suppasetseree, S. (2016). The development of an Instructional Design Model on Facebook Based Collaborative Learning to Enhance EFL Students’ writing skills. IAFOR Journal of Language Learning, 2(1), 10.22492/ijll.2.1.04. Liu IF Chen MC Sun YS Wible D Kuo CH Extending the TAM model to explore the factors that affect intention to use an online Learning Community Computers and Education 2010 54 2 600 610 10.1016/j.compedu.2009.09.009 Mahmodi M The analysis of the factors affecting the Acceptance of E-learning in Higher Education Interdisciplinary Journal of Virtual Learning in Medical Sciences 2017 8 1 1 9 10.5812/ijvlms.11158 Mailizar M Burg D Maulina S Examining university students’ behavioural intention to use e-learning during the COVID-19 pandemic: an extended TAM model Education and Information Technologies 2021 26 6 7057 7077 10.1007/s10639-021-10557-5 33935579 Marsh, J., & Drexler, P. (2001). How to design effective blended learning. Sunnyvale: Brandon-Hall. Martuza, V. R. (1977). Applying norm-referenced and criterion-referenced measurement in education. Allyn and Bacon. Mason GS Shuman TR Cook KE Comparing the effectiveness of an Inverted Classroom to a Traditional Classroom in an Upper-Division Engineering Course IEEE Transactions on Education 2013 56 4 430 435 10.1109/te.2013.2249066 McGee, P., & Reis, A. (2012). Blended Course Design: a synthesis of Best Practices. Online Learning, 16(4), 10.24059/olj.v16i4.239. Means, B., Toyama, Y., Murphy, R., & Baki, M. (2013). The effectiveness of online and blended learning: A meta-analysis of the empirical literature.Teachers College Record, 115(3). Meier JJ Miller RK Turning the revolution into an evolution: the case for design thinking and rapid prototyping in libraries College and Research Libraries News 2016 77 6 283 286 10.5860/crln.77.6.9506 Meier JJ Miller RK Fail early and often to succeed: a case for rapid prototyping in libraries Library Leadership and Management 2018 33 1 1 19 10.5860/llm.v33i1.7290 Miller, B. G. (2008). Metabolic expansion via designed enzyme recruitment. Abstracts of Papers, 235th ACS National Meeting, New Orleans, LA, United States, April 6–10, 2008 FIELD Full Journal Title:Abstracts of Papers, 235th ACS National Meeting, New Orleans, LA, United States, April 6–10, 2008, 38, BIOL-012. Mndzebele N Teachers readiness in using ICT in the Classroom: the case of a developing country International Journal of Information and Education Technology 2013 3 4 409 412 10.7763/ijiet.2013.v3.309 Motaghian H Hassanzadeh A Moghadam DK Factors affecting university instructors’ adoption of web-based learning systems: Case study of Iran Computers and Education 2013 61 1 158 167 10.1016/j.compedu.2012.09.016 Mou, J., Shin, D. H., & Cohen, J. (2016). Understanding trust and perceived usefulness in the consumer acceptance of an e-service: a longitudinal investigation. 10.1080/0144929X.2016.1203024 Onguko B Jepchumba L Gaceri P For us it was a learning experience”: design, development and implementation of blended learning European Journal of Training and Development 2013 37 7 615 634 10.1108/EJTD-10-2012-0052 Park SY An analysis of the technology acceptance model in understanding University students’ behavioral intention to use e-Learning Educational Technology and Society 2009 12 3 150 162 10.2307/jeductechsoci.12.3.150 Polit, D. F., Beck, C. T., Owen, S. V., Polit, D. F., & Beck, C. T. (2007). O. S. Focus on Research Methods Handling Missing Data in Self-Report Measures. 488–495. 10.1002/nur Poon, J. (2013). Blended Learning: An Institutional Approach for Enhancing Students’ Learning Experiences. InJournal of Online Learning and Teaching(Vol. 9, Issue 2). Raihan MA Han SL Blended-learning in TVET for Empowering Engineering Students’ with Technological Innovations: a case of Bangladesh International Journal of Educational Science and Research (IJESR) 2013 3 1 61 74 Raihan MA Shamim MRH A study to explore the practice of ICTs in TVET in Bangladesh and South Korea Certified International Journal of Engineering Science and Innovative Technology 2008 9 4 2319 5967 Ramayah T Cheah J Ting FCH Memon MA Partial least squares structural equation modeling (PLS-SEM) using SmartPLS 3.0: an updated and practical guide to statistical analysis Practical Assessment Research and Evaluation 2018 4 October 291 Ritter NL Technology Acceptance Model of Online Learning Management Systems in Higher Education: a Meta-Analytic Structural equation Mode International Journal of Learning Management Systems 2017 5 1 1 15 10.18576/ijlms/050101 Rodrigues IB Adachi JD Beattie KA MacDermid JC Development and validation of a new tool to measure the facilitators, barriers and preferences to exercise in people with osteoporosis BMC Musculoskeletal Disorders 2017 18 1 1 9 10.1186/s12891-017-1914-5 28052768 Rony HAZ Awal ST University teachers’ training on online teaching-learning using online platform during COVID-19: a case study Bangladesh Education Journal 2019 18 2 57 64 Saidu MK Al Mamun MA Exploring the factors affecting behavioural intention to Use Google Classroom: University Teachers’ perspectives in Bangladesh and Nigeria TechTrends 2022 66 4 681 696 10.1007/s11528-022-00704-1 35128544 Salloum SA Mohammad Alhamad Q Al-Emran A Monem MA Shaalan K Exploring students’ acceptance of e-learning through the development of a comprehensive technology acceptance model Ieee Access : Practical Innovations, Open Solutions 2019 7 128445 128462 10.1109/ACCESS.2019.2939467 Sandjojo, N., & Wahyuningrum, T. (2016). Measuring e-learning systems success: Implementing D and M is success model. Proceedings of the 2015 4th International Conference on Interactive Digital Media, ICIDM 2015, Icidm, 4–9. 10.1109/IDM.2015.7516343 Singh, H. (2003). Building Effective Blended Learning Programs. In Issue of Educational Technology (Vol. 43, Issue 6). Stapa MA Mohammad N The Use of Addie Model for Designing blended Learning Application at Vocational Colleges in Malaysia Asia-Pacific Journal of Information Technology & Multimedia 2019 08 01 49 62 10.17576/apjitm-2019-0801-05 Stone M Cross-Validatory Choice and Assessment of statistical predictions (with discussion) Journal of the Royal Statistical Society: Series B (Methodological) 1976 38 1 102 102 10.1111/j.2517-6161.1976.tb01573.x Sullivan GM Feinn R Using effect size—or why the P value is not enough Journal of Graduate Medical Education 2012 4 3 279 282 10.4300/jgme-d-12-00156.1 23997866 Tan GWH Ooi KB Sim JJ Phusavat K Determinants of mobile learning adoption: an empirical analysis Journal of Computer Information Systems 2012 52 3 82 91 10.1080/08874417.2012.11645561 Tang H Mao L Wang F Zhang H A validation study for a short-version scale to assess 21st century skills in flipped EFL classrooms Oxford Review of Education 2022 48 2 148 165 10.1080/03054985.2021.1935226 Tarhini A Hone K Liu X Tarhini T Examining the moderating effect of individual-level cultural values on users’ acceptance of E-learning in developing countries: a structural equation modeling of an extended technology acceptance model Interactive Learning Environments 2017 25 3 306 328 10.1080/10494820.2015.1122635 Teo T Examining the intention to use technology among pre-service teachers: an integration of the Technology Acceptance Model and Theory of Planned Behavior Interactive Learning Environments 2012 20 1 3 18 10.1080/10494821003714632 Thorpe K Reflective learning journals: from concept to practice Reflective Practice 2004 5 3 327 343 10.1080/1462394042000270655 Tian X Suppasetseree S Development of an instructional model for online task-based interactive listening for EFL learners English Language Teaching 2013 6 3 30 41 10.5539/elt.v6n3p30 Timonen, P., & Ruokamo, H. (2021). Designing a preliminary Model of Coaching Pedagogy for Synchronous Collaborative Online Learning. Journal of Pacific Rim Psychology, 15, 10.1177/1834490921991430. Uzzaman MN Jackson T Uddin A Rowa-Dewar N Chisti MJ Habib GMM Pinnock H Continuing professional education for general practitioners on chronic obstructive pulmonary disease: feasibility of a blended learning approach in Bangladesh BMC Family Practice 2020 21 1 1 10 10.1186/s12875-020-01270-2 31901226 Dick, W., Carey, L., & Carey, J. O. (2009). The Systematic Design of Instruction. In Merrill/Pearson (7th ed.). Waltz, C. F., Strickland, O. L., & Lenz, E. R. (2016). Measurement in Nursing and Health Research. In Measurement in Nursing and Health Research (4th ed.). Springer. 10.1891/9780826170620 Wang M Shen R Novak D Pan X The impact of mobile learning on students’ learning behaviours and performance: report from a large blended classroom British Journal of Educational Technology 2009 40 4 673 695 10.1111/j.1467-8535.2008.00846.x Wong GKW Understanding technology acceptance in pre-service teachers of primary mathematics in Hong Kong Australasian Journal of Educational Technology 2015 31 6 713 735 10.14742/ajet.1890 Wong KT Osman Goh R Rahmat MK Understanding student teachers’ behavioural intention to use technology: technology acceptance model (TAM) validation and testing International Journal of Instruction 2013 6 1 89 104 Yusoff MSB ABC of Content Validation and Content Validity Index calculation Education in Medicine Journal 2019 11 2 49 54 10.21315/eimj2019.11.2.6 Zamanzadeh V Ghahramanian A Rassouli M Abbaszadeh A Alavi-Majd H Nikanfar AR Design and implementation content validity study: development of an instrument for measuring patient-centered communication Journal of Caring Sciences 2015 4 2 165 178 10.15171/jcs.2015.017 26161370 Zogheib, B., Rabaa’i, A., Zogheib, S., & Elsaheli, A. (2015). University student perceptions of technology use in mathematics learning. Journal of Information Technology Education: Research, 14(2015), 417–438. 10.28945/2315	0	PMC9734497	NO-CC CODE	2022-12-14 23:28:29	no	Educ Inf Technol (Dordr). 2022 Dec 7;:1-30	utf-8	Educ Inf Technol (Dordr)	2,022	10.1007/s10639-022-11471-0	oa_other
==== Front J Clin Monit Comput J Clin Monit Comput Journal of Clinical Monitoring and Computing 1387-1307 1573-2614 Springer Netherlands Dordrecht 36464761 948 10.1007/s10877-022-00948-5 Original Research Automated electrocardiogram signal quality assessment based on Fourier analysis and template matching Menon Kartikeya M. [email protected] 12 Das Subrat 123 Shervey Mark 12 Johnson Matthew 12 Glicksberg Benjamin S. 1234 Levin Matthew A. 1235 1 grid.59734.3c 0000 0001 0670 2351 Mount Sinai Clinical Intelligence Center, Icahn School of Medicine at Mount Sinai, New York, NY USA 2 grid.59734.3c 0000 0001 0670 2351 The Hasso Plattner Institute for Digital Health at Mount Sinai, Icahn School of Medicine at Mount Sinai, New York, NY USA 3 grid.59734.3c 0000 0001 0670 2351 Department of Medicine, Icahn School of Medicine at Mount Sinai, New York, NY USA 4 grid.59734.3c 0000 0001 0670 2351 Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai, New York, NY USA 5 grid.59734.3c 0000 0001 0670 2351 Department of Anesthesiology, Perioperative and Pain Medicine, Icahn School of Medicine at Mount Sinai, New York, NY USA 5 12 2022 19 7 7 2022 10 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. We developed and tested a novel template matching approach for signal quality assessment on electrocardiogram (ECG) data. A computational method was developed that uses a sinusoidal approximation to the QRS complex to generate a correlation value at every point of an ECG. The strength of this correlation can be numerically adapted into a ‘score’ for each segment of an ECG, which can be used to stratify signal quality. The algorithm was tested on lead II ECGs of intensive care unit (ICU) patients admitted to the Mount Sinai Hospital (MSH) from January to July 2020 and on records from the MIT BIH arrhythmia database. The algorithm was found to be 98.9% specific and 99% sensitive on test data from the MSH ICU patients. The routine performs in linear O(n) time and occupies O(1) heap space in runtime. This approach can be used to lower the burden of pre-processing in ECG signal analysis. Given its runtime (O(n)) and memory (O(1)) complexity, there are potential applications for signal quality stratification and arrhythmia detection in wearable devices or smartphones. Keywords Fourier ECG Electrocardiogram AI ML ==== Body pmcIntroduction We present a method for rapid computational validation of high-volume, high frequency electrocardiography (ECG) data. As a cheap, universally available diagnostic tool, ECG data are increasingly used as an input to a variety of processing algorithms that range from simple heart rate detection to cardiac disease risk stratification vis-a-vis sophisticated predictive models [1–3]. Since the power of any downstream computational analysis depends on the quality of input data, much work in electrical engineering and signal processing has been done to try and distinguish or distill clear ECG voltage data from noisy interference [4, 5]. The automated morphological and clinical analyses of ECGs are well documented [6–9]. High-quality signal processing of the timing, magnitude, and axis of voltage propagation within the heart can all help predictive algorithms classify ECG data into diagnostically useful categories. However, ECGs that have high levels of noise can saturate the corrective capacity of Butterworth, Kalman, or Savitzky-Golay filters that are canonically used in ECG analysis routines, including the Pan-Tompkins algorithm [5]. Therefore, these unusable regions must be discarded to proceed with any sort of clinical analysis. Examples of these unusable regions are seen in Fig. 1.Fig. 1 Example of noisy signals (left) and high-quality signals (right). Regions of interference, demonstrated on the left, are not reparable through typical signal processing techniques, like Butterworth or Kalman filtering In ECG data sets that collectively span several months, (for example, cohorts of COVID-19 patients admitted to intensive care units), manually searching for and excising regions of high interference can become onerous, if not outright impossible. As such, there is an opportunity to quickly and programmatically identify regions of data that are sufficiently noisy as to warrant exclusion. Since the late 1980s, there have been many techniques proposed to detect and remove artifacts in ECGs, which were recently comprehensibly documented and reviewed [10]. The authors concluded that these artifact detection (AD) algorithms were domain-specific (i.e., only interpretable in the specific context of one type of patient or clinical setting), not yet organized for integration into clinical workflows, not evaluated in real-time, and often needed to be coupled with clinical event detection software. Most of the AD algorithms had hard-coded parameters for the recording frequency, which would require modification and validation for use in other contexts. Lastly, many AD algorithms use pre-processed voltage data or rely on public databases (such as the MIT BIH arrhythmia collection [11]) for training and testing [4–6, 9, 12]. With voltage data from commercial monitors, the pre-processing routines are not disclosed and may result in bias or lack of generalizability for their downstream AD algorithms. The MIT BIH database contains several 30-min records from 1975 to 1979 that annotate different types of arrhythmias for patients on a variety of cardiovascular medications or pacemakers [11]. However, given their length and scope, these records do not showcase the same variance in signal quality or noise as compared to ECG data collected for many hours or days from hospitalized patients. These challenges are tackled in [13, 14], where the authors test a combination of physiological constraints, template matching, and machine learning to stratify the signal quality of ambulatory electrocardiography and photoplethysmography (PPG). In [13], a QRS-complex and PPG-pulse peak estimator is used in the first step of the proposed routine, which adds variability based on which morphological detection algorithm is used. In [14], the results are generalizable and validated on data from different sensors, but the routine introduces algorithmic complexity in its analytical methods, which include wavelet transforms. In our work, the proposed algorithm is developed by approximating the shape of a QRS complex through Fourier analysis and then scoring ECGs on a sample-by-sample basis using template matching. The routine was designed to be computationally (both time and memory) inexpensive, easy to implement and replicate, and high-fidelity in regard to filtering clinically unusable regions of ECG voltage data. Methods The steps involved in the scoring and classification of ECG epochs are described below. The data used in this paper were collected under approval of the Program for the Protection of Human Subjects Institutional Review Board the Icahn School of Medicine at Mount Sinai/Mount Sinai Health System (STUDY-20-00338, MSCIC Predictive Modeling and Consultation Tool, approved 09-17-2020). Overview of method The proposed algorithm is developed by first approximating a Fourier series to the QRS complex. This function is the kernel that is then piecewise cross-correlated (“template matched”) with the patient's entire ECG, divided into subsections of at least two seconds duration. These correlations are then squared and numerically integrated. The resulting scores can be interpreted as a metric for the quality of an ECG, marking values outside of a threshold range eligible for exclusion. The routine runs in linear O(n) time and requires O(1) constant heap space during its execution. Database and testing The input data set was comprised of raw 240 Hz lead II ECG waveform data collected from patients admitted to intensive care units at The Mount Sinai Hospital (MSH) between January 2020 and July 2020. Ten adult patients with at least one hour of recorded ECG data were randomly selected from this ECG data repository. The ECG data were subdivided into 244,903 2-s intervals (a cumulative 136 h of data). For each patient, the algorithm described below was allowed to process each interval and generate a score for it. To test performance, 100 consecutive epochs from each patient (1000 total), not labeled or processed by the algorithm, were annotated as ‘usable’ or ‘unusable’ by author SD. The assessment of ‘usable’ or ‘unusable’ was made on the following grounds: if a human expert can derive a heart rate from it by finding typical morphological features of an ECG (for example, the QRS complex), the epoch was labeled ‘usable.’ Using the distribution generated over the course of training, the algorithm then classified the test data, and performance statistics were generated. The proposed algorithm was also evaluated against records from the MIT BIH arrhythmia database. The database records were chosen based on signal-to-noise ratio (number of non-beat annotations relative to the number of beat annotations). The algorithm produces a metric si for each i-th subset Mi of an ECG and characterizes Mi as noise or signal based on the distribution of si. Algorithm development Fourier approximation We begin with the Fourier series approximation of representative lead II QRS complexes. Fourier series are sums of sine and cosine waves that approximate periodic functions. An ECG recording represents a propagation of voltage in time with periodicity (i.e., a series of heartbeats), so Fourier series are well-suited to approximate the sinusoidal morphologies of cardiac activities like P waves, QRS complexes, and repolarization (T) waves. If vN(t) represents the N-th order Fourier approximation (different orders shown in Fig. 2) of the evolution of ECG voltage in time t, where L is the time interval of the QRS complex and ψn is the frequency phase:Fig. 2 Example of kernel. The original heartbeat (shown in green, the peak is the QRS complex) can be approximated by a Fourier series. Two approximations (N = 3—blue and N = 10—orange) are shown. The higher the N, the better the approximation to the actual heartbeat vNt=A02+∑n=1NAncos2πLnt-ψn=a02+∑n=1Nancos2πLnt+bnsin2πLnt The larger the N, the more sine and cosine terms are included in the approximation, and the closer the Fourier series looks to the actual heartbeat, which is demonstrated in Fig. 2. The Fourier coefficients an and bn are defined as follows and can be calculated via Simpson’s rule for numerical approximation of definite integrals:an=2L∫LvNt×cos2πLntdt bn=2L∫LvNt×sin2πLntdt The Fourier approximation of the QRS complex is done on a semi-automatic basis by first manually selecting the desired order of approximation (N = 5 was used). The kernel is then calculated on an automatic basis. The algorithm streams ECG data, computing a Fourier approximation to each processed QRS complex, until all N sequentially computed coefficients are within 5% from each other. Cross-correlation Let F(t) represent the voltage propagation in time t of the lead II ECG recording to be analyzed, and vN(t) be the N-th order Fourier approximation to the QRS complex. The next step in the algorithm computes a 1-dimensional cross-correlation between this approximation and the ECG recording of interest. This method, called template matching, is commonly applied in ECG analysis and has previously been used to identify ventricular ectopy [15], detect heart beats [16], and in earlier work on ECG signal quality assessment [13]. The “template” or “kernel” serves as a computational guideline for the shape of a heartbeat represented on ECG. When voltage data correlate strongly with the pre-calculated kernel, it indicates high probability of quality signal, and vice versa. This is demonstrated in Fig. 3, where low-quality signal has poor overlap with the kernel. The degree of overlap, or correlation, can be quantified with the 1-dimensional cross-correlation (also known as a convolution with a symmetric kernel) between the template and the ECG recording.Fig. 3 Visualization of template matching. On the left, an ECG segment with baseline wander, and on the right, the template (approximation to QRS complex) is shown in green with the noisy ECG peaks overlaid in grey. There is significant variation in the morphology of these QRS complexes versus the template To perform the cross-correlation, the ECG is broken down into intervals Mi=ti,ti+τ, where i is the cardinality of the interval (i.e., M5 is the 5th such interval in an ECG recording), ti refers to the starting time of the interval, and τ refers to the length of the interval in seconds. The smaller the interval, the more specific the algorithm can be in excising regions of noisy interference from the ECG. Given a physiologically feasible lower limit on bradycardia (30 bpm was used), the ECG can be subdivided into intervals of two seconds (i.e., each Mi is [ti,ti+2]). This ensures each interval should have at least one QRS complex. The cross correlation is visualized in Fig. 4. For k=0,1,⋯,F(t)+vNt-2, C=maxF(t),vNt, and vNt∗ the complex conjugate of the Fourier series, the cross-correlation z for two discrete 1-dimensional arrays at all points in series k is defined as:Fig. 4 Cross-correlation plots with the pre-computed kernel. Cross-correlation of ECG data with a pre-computed sinusoidal kernel specific to the QRS complex produces peak values at regions of consistent signal, and irregular values at regions of inconsistent signal. On the left, a clean ECG signal and the squared value of its cross-correlation with the Fourier approximation. On the right, motion artifact and baseline wander add noise to the signal. The cross-correlation peaks are visibly irregular compared with those of the clean signal zk=F(t)∗vNtk-C+1=∑j=0Ft-1Fj(t)∗vNtj-k+C-1∗ Because the template approximates the shape of the QRS complex, its correlation with the ECG recording is very high (seen in the four highest peaks in the clean signal cross-correlation in Fig. 4) when it passes over a heartbeat. When this similarity between QRS approximation and ECG recording is not as profound, the correlation is lower, which produces lesser and more irregular peaks (seen in the motion noise cross-correlation in Fig. 4). Score calculation The cross-correlation between the template and ECG can be used to generate a metric whose distributional qualities help distinguish noisy interference from valid ECG data. As seen in Fig. 4, as the ECG voltage data propagates, the template matching generates a correlation as a function of time. When the match between kernel and ECG is better, the peak of the correlation is higher, so the integral of the correlation values FMit∗vN(t) can be used to assign an ECG recording a score. The equation for the score si is as follows, where vN(t) is the Fourier kernel, FMi is the voltage propagation for ECG in segment Mi, and nQRS is the number of QRS complexes within segment Mi:si=1nQRS∫MiFMit∗vNt2dt=1nQRS∫titi+τFMit∗vNt2dt The scores si, calculated for each interval Mi of an ECG, are roughly mesokurtic, which means that a multiple of the standard deviation of the score can be used to identify noise. Figure 5 shows the prior example of the score of a usable versus unusable ECG signal within the distribution of scores.Fig. 5 Score distributions. The scores si are normally distributed so the central tendency can be used to distinguish noise from signal. In the clean signal, the integral of the cross-correlation peaks gives a score of 9.8. In the noisy signal, the integral of the peaks yields a score of 15.5 In Fig. 5, the same ECG recording and cross-correlation as in Fig. 4 are used, but the integral (area under curve) of the convolution is highlighted in green (clean signal) and red (motion noise). The bell curve underneath shows the distribution of scores produced by the ECG recording of a given ICU patient. Given the Gaussian property of the distribution of scores, the integral value of the motion noise cross-correlation can be used to identify it as an outlier, and therefore as low-quality signal. Figure 6 illustrates this distributional character of the scores that separates high-quality from low-quality signals. As the number of intervals tested increases, the histogram will better approximate the bell curve shown in Fig. 5.Fig. 6 Example histogram of scores. The high-quality signal scores cluster around the distribution mean, where the noisy signal scores lie further away Results Across ten patients, the algorithm was allowed to form a distribution of scores on n = 244,903 epochs and tested on n = 1000 and found to perform consistently well. In patients 1, 3, and 8, no regions of noise were identified in the entire test samples, so no specificity is calculated. Per Table 1, the overall performance of the algorithm was strong, producing an average specificity of 98.8%, sensitivity of 99.1%, and accuracy of 98.9%. As shown in Table 2, the algorithm was also evaluated against records from the MIT BIH arrhythmia database, where the sensitivity of the template matching was high enough to identify the handful of morphologically different heartbeats out of several thousand heartbeats. Table 1 Performance data on MSH patients Patient ID Number of Noisy Segments Specificity Sensitivity Accuracy 1 0 100 N/A 100 2 2 100 100 100 3 0 100 N/A 100 4 3 100 100 100 5 8 98.9 100 98.9 6 17 90.2 94.1 90.9 7 9 98.9 100 98.9 8 0 100 N/A 100 9 2 100 100 100 10 25 100 100 100 Algorithm performance on each of the MSH patients sampled for raw 240 Hz ECG data Table 2 Performance data on MIT BIH data Record ID Number of annotations Atypical annotations Atypical annotations identified si kurtosis 100 2,273 1 1 2.96 115 1,961 0 0 2.59 122 2,478 0 0 4.09 123 1,518 3 3 2.79 Records 100 and 123 combined have 1,300,000 data points and almost 4,000 heart beats annotated. The algorithm was sensitive enough to capture the only 4 instances of ventricular ectopy across all these annotations, and specific enough to not misclassify any unannotated heartbeat Since the proposed algorithm evaluates ECGs with a pre-computed kernel, it achieves an O(n) time complexity, making it much faster than methods requiring autocorrelation or machine learning (Fig. 7). In any given iteration, only the data in Mi=ti,ti+τ are held in memory for evaluation, so the routine is also memory efficient. It uses O(1) heap space in runtime, which means that a 32-bit architecture yields only one float (the score) for each interval Mi, requiring 4×dim(M) bytes of storage to process an ECG with this algorithm.Fig. 7 Runtime analysis. As expected, the algorithm performs in O(n) time under evaluation with varying data size. The size of the ECG file in megabytes is on the right axis, and the corresponding length of the ECG recording (in hours) is on the left Discussion In this report, we present a Fourier series-based template matching algorithm that is a highly accurate and scalable means of quickly identifying regions of untenable signal interference in ECG data. Across 1,000 test ECG data segments from 10 ICU patients, the average sensitivity for identifying unusable ECG segments was 99%, the average specificity was 98.9%, and the average accuracy was 98.9%. As ECGs are increasingly used in studies employing big data analytical techniques, our approach provides an easy-to-implement routine to lower the burden of data pre-processing. Given the low memory overhead in both disk space and run-time heap space, this method can also help with signal processing for wearable devices. Calculating a day's worth of signal scoring data would cost less than 700 kB, well within the storage capacity of modern wearable devices. The algorithm's distribution-based filtering suggests flexibility in different use cases. If a researcher wants only the very highest quality signal, a more stringent score cutoff may be used. If the data in question exhibit high morphological variability (for example, inverted T waves, multi-morphic ventricular ectopy, atrial fibrillation, etc.) then the threshold may be appropriately loosened. This aspect also means that the specificity of the algorithm may decline if a looser threshold is selected. The threshold between signal and noise is made based on distributional qualities of the recorded data (the standard deviation), and as the threshold for noise classification is lowered, valid signal may be incorrectly classified, and vice versa. Since the algorithm is dependent on the distribution of scores (and hence, distribution of quality signal) within an ECG, if the input data is particularly corrupted, the distribution that is formed may not lend itself to accurate stratification of signal quality. Put another way, the algorithm detects noise based on correlation to itself, as opposed to an absolute standard of high-quality ECG. This template matching approach fills a useful niche within the research oeuvre around ECG signal assessment. Many ECG analyses are platform specific [10], developed and validated specifically for public databases [4, 5, 9, 12], or were trained on device data potentially biased by proprietary signal processing. Our algorithm is platform agnostic and doesn’t require hard-coded parameters to describe the ECG, like frequency. The accuracy of this algorithm is demonstrated in 240 Hz (MSH ICU ECGs) and 360 Hz (MIT BIH ECGs) without prior knowledge of the recording frequency. Further, the voltage data from the MSH ICU patients had no a priori signal processing or associated clinical event detection annotations, which are data augmentations that can create bias or reproducibility issues. Orphanidou et al. [13, 14] presented a template matching method that resolved many these issues but noted that the performance of the signal assessment strongly relies on accurate and reliable QRS complex detection under ambulatory and exercise ECG recording conditions. The performance testing results (sensitivity = 94%, specificity = 97%) in Orphanidou et al. are comparable to those produced by our algorithm (sensitivity = 99%, specificity = 98.9%), but they do not report runtime analysis or computational memory efficiency. Using a new scoring method, the algorithm presented here validates the template matching approach as a highly accurate modality for signal quality assessment while not requiring specific ECG recording conditions, a particular platform, or a particular frequency and format of input data. In comparison to prior ECG signal quality assessment algorithms described in comprehensive literature reviews [10, 12] and AD algorithms using similar computational techniques [13, 15, 16], this is the first ECG AD algorithm to specifically report the results of runtime testing and memory analysis. As the convolutional kernel is based on the Fourier approximation of a heartbeat, this algorithm can be adapted to scan ECGs for specific morphological subtypes. For example, a kernel may be calculated that is highly sensitive to premature ventricular contractions or bundle branch blocks. This kernel will then register high correlation, and therefore high integrated scores, with the desired morphology, as opposed to a typical heartbeat. This may have utility in wearable devices where continuous monitoring may alert a user's cardiologist or emergency services if dangerous electrical patterns are recognized. In summary, we anticipate our algorithm will help accelerate the productivity of research on raw ECG waveforms. Given its computational efficiency, the algorithm has practical use as a starting point for signal filtration and pattern identification in real-time continuous ECG monitoring, including in wearable devices. Acknowledgements This work was supported in part through the computational resources and staff expertise provided by Scientific Computing at the Icahn School of Medicine at Mount Sinai. Author contributions KMM wrote the main manuscript text and SD provided labeled ECGs to compute performance metrics. MS and MJ sourced and assembled the data and BSG and MAL are senior authors. All authors reviewed the manuscript. Funding This work was supported by 1R01LM013766-01A1. Data availability The public data set is available at the MIT-BIH arrhythmia database. The corresponding author is available for questions and clarification at [email protected]. The MSH data is PHI-sensitive and available at appropriate request. A worked example of signal quality stratification and sample code for the algorithm are available at https://github.com/kartikmenon/ECG_JCMC. Declarations Conflict of interest None of the authors have financial or non-financial interests related to this work. Ethical approval Program for the Protection of Human Subjects Institutional Review Board approved on 09/17/2020 for human subjects data under STUDY-20-00338 (title: MSCIC Predictive Modeling & Consultation Tool). Informed consent N/A. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References: 1. Cohen-Shelly M Electrocardiogram screening for aortic valve stenosis using artificial intelligence Eur Heart J 2021 42 30 2885 2896 10.1093/eurheartj/ehab153 33748852 2. Adedinsewo DA Detecting cardiomyopathies in pregnancy and the postpartum period with an electrocardiogram-based deep learning model Eur Heart J Digit Health 2021 2 4 586 596 10.1093/ehjdh/ztab078 34993486 3. Akbilgic O ECG-AI: electrocardiographic artificial intelligence model for prediction of heart failure Eur Heart J Digit Health 2021 2 4 626 634 10.1093/ehjdh/ztab080 34993487 4. D’Aloia M Longo A Rizzi M Noisy ECG signal analysis for automatic peak detection Information 2019 10 2 35 10.3390/info10020035 5. Pan J Tompkins WJ A real-time QRS detection algorithm IEEE Trans Biomed Eng 1985 32 3 230 236 10.1109/TBME.1985.325532 3997178 6. Merino M Gómez IM Molina AJ Envelopment filter and K-means for the detection of QRS waveforms in electrocardiogram Med Eng Phys 2015 37 6 605 609 10.1016/j.medengphy.2015.03.019 25922210 7. Lee WK Yoon H Park KS Smart ECG monitoring patch with built-in R-peak detection for long-term HRV analysis Ann Biomed Eng 2016 44 7 2292 2301 10.1007/s10439-015-1502-5 26558395 8. Moody GB Mark RG The impact of the MIT-BIH arrhythmia database IEEE Eng Med Biol Mag 2001 20 3 45 50 10.1109/51.932724 11446209 9. Elgendi M Fast QRS detection with an optimized knowledge-based method: evaluation on 11 standard ECG databases PLoS ONE 2013 8 9 e73557 10.1371/journal.pone.0073557 24066054 10. Satija U Ramkumar B Manikandan MS A review of signal processing techniques for electrocardiogram signal quality assessment IEEE Rev Biomed Eng 2018 11 36 52 10.1109/RBME.2018.2810957 29994590 11. Goldberger A PhysioBank, PhysioToolkit, and PhysioNet: components of a new research resource for complex physiologic signals Circulation 2000 101 23 e215 e220 10.1161/01.CIR.101.23.e215 10851218 12. Mir HY Singh O ECG denoising and feature extraction techniques: a review J Med Eng Technol 2021 45 8 672 684 10.1080/03091902.2021.1955032 34463593 13. Orphanidou C Bonnici T Charlton P Clifton D Vallance D Tarassenko L Signal-quality indices for the electrocardiogram and photoplethysmogram: derivation and applications to wireless monitoring IEEE J Biomed Health Inform 2015 19 3 832 838 25069129 14. Orphanidou C Drobnjak I Quality assessment of ambulatory ECG using wavelet entropy of the HRV signal IEEE J Biomed Health Inform 2017 21 5 1216 1223 10.1109/JBHI.2016.2615316 28113529 15. Krasteva V Jekova I QRS template matching for recognition of ventricular ectopic beats Ann Biomed Eng 2007 35 12 2065 2076 10.1007/s10439-007-9368-9 17805974 16. Chan HL, Chen GU, Lin MA, Fang SC. Heartbeat detection using energy thresholding and template match. In: 2005 IEEE Engineering in Medicine and Biology 27th Annual Conference; 2005, pp. 6668–6670.	36464761	PMC9734499	NO-CC CODE	2022-12-14 23:28:29	no	J Clin Monit Comput. 2022 Dec 5;:1-9	utf-8	J Clin Monit Comput	2,022	10.1007/s10877-022-00948-5	oa_other
==== Front Occup Health Sci Occup Health Sci Occupational Health Science 2367-0134 2367-0142 Springer International Publishing Cham 130 10.1007/s41542-022-00130-y Major Empirical Contribution Worrying About Finances During COVID-19: Resiliency Enhances the Effect of Worrying on Both Proactive Behavior and Stress http://orcid.org/0000-0002-9488-7110 Fa-Kaji Naomi M. 1 http://orcid.org/0000-0002-4779-0949 Silver Elisabeth R. 1 http://orcid.org/0000-0002-8319-7851 Hebl Mikki R. 1 http://orcid.org/0000-0002-1277-5669 King Danielle D. 1 http://orcid.org/0000-0002-7069-5130 King Eden B. [email protected] 1 http://orcid.org/0000-0002-7066-5089 Corrington Abby 2 http://orcid.org/0000-0002-0244-0232 Bilotta Isabel 1 1 grid.21940.3e 0000 0004 1936 8278 Department of Psychological Sciences, Rice University, 6100 Main St., Houston, TX 77005 USA 2 grid.418778.5 0000 0000 9812 3543 Department of Management, Providence College School of Business, Providence, RI USA 3 12 2022 132 29 1 2022 25 10 2022 9 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. Amidst the COVID-19 pandemic and resulting economic instability, many people are contending with financial insecurity. Guided by Conservation of Resources Theory (Hobfoll, American Psychologist 44:513–524, 1989; Hobfoll et al., Annual Review of Organizational Psychology and Organizational Behavior 5:103–128, 2018), the current research explores the consequences of experiencing financial insecurity during a pandemic, with a focus on individuals who report relatively higher rates of financial insecurity, performance challenges, and stress during such experiences: working parents (American Psychological Association, 2022). This research also examines the role that personal resources, in the form of trait resiliency, play in the relationships between financial insecurity and behavioral and psychological outcomes including worrying, proactive behaviors, and stress. In a study of 636 working parents and their children, we find that financial insecurity heightens worrying, underscoring the threatening nature of the loss or anticipated loss of material resources. Worrying, in turn, promotes proactive behaviors at work—an effect that is more pronounced among high-resiliency individuals. However, worrying is also associated with elevated stress among high-resiliency individuals, providing support for a trait activation perspective (rather than buffering hypotheses) on ongoing, uncontrollable adversities. Taken together, our results help to (1) illuminate the impact of financial insecurity on work and well-being, (2) reveal a mechanism (i.e., worrying) that helps explain the links between financial insecurity and work and personal outcomes, and (3) expand our knowledge of the implications trait resiliency has for both psychological and behavioral reactions to ongoing crises. Keywords Resiliency Financial insecurity Well-being Worrying http://dx.doi.org/10.13039/100007863 Rice University ==== Body pmc “I finally know what distinguishes man from the other beasts: financial worries.”- Jules Renard Financial worries are a ubiquitous part of modern life. The overwhelming majority of Americans (65%) surveyed by the American Psychological Association in 2022 indicated that money is a significant source of stress (compared to 58% in 2021; APA, 2022). Economic concerns were exacerbated among parents, of whom 80% reported finances as a significant stressor (compared to 58% of non-parents). This increase in financial concerns can be traced in part to the unfolding ramifications of the COVID-19 pandemic, such as rising consumer prices and inflation, uncertain labor market conditions, stagnant wages, and supply chain disruptions (see Phetmisy & King, 2021; Wanberg et al., 2020). In this paper, we build on an emerging body of research to explore psychological and behavioral responses to financial insecurity from the lens of Conservation of Resources (COR) theory (Halbesleben et al., 2014; Hobfoll, 1989). We further expand COR theory by empirically examining the role that trait resiliency—one’s ability to weather stressful or traumatic events (Bonanno, 2004; Britt et al., 2016)—plays in determining employee responses to resource loss. This research is important for three primary reasons. First, this research directs needed attention to the impact of financial insecurity on both well-being and work-related outcomes. Past work has demonstrated that concerns about money can be related to critical indicators of well-being such as mental and physical health (Angel et al., 2003; Richardson et al., 2017). Yet, aside from a handful of studies examining the role financial insecurity plays in work-family conflict (Lawrence et al., 2013) and workplace safety outcomes (Pettita et al., 2020), the impact of financial instability or uncertainty on work-related, behavioral outcomes has largely been ignored. In a commentary, Phetmisy and King (2021) speculated about potentially competing predictions regarding such behavioral effects, suggesting that financial concerns may have the effect of “influencing employees to overexert themselves to be in good standing with supervisors and subsequently reduce the risk of being laid off” or could instead “make it more difficult for employees to engage or produce at work” (p. 91). That is, financial insecurity could either lead employees to overwork themselves to maintain employment or cause employees to disengage at work due to stress. We build on this and other previous scholarship by empirically testing not only the implications of financial insecurity for stress, but also for proactive behaviors at work. Second, this work provides both theoretical rationale and empirical evidence for a mechanism (i.e., worrying) that helps explain the link between financial insecurity and the outcomes of stress and proactive work behaviors. We reason that anticipated or experienced resource loss, in the form of financial insecurity, will give rise to worrying, which in turn can have strong implications for work behavior and psychological well-being. Thus, this work sheds light on a previously unexplored psychological process that underlies the experience of insecurity. We reason that it is not just the possible loss of resources, but also the resulting thought processes (i.e., worrying), that are associated with attempts to preserve remaining resources. Given that COR research does not typically focus on people’s cognitive responses to anticipated resource loss (see Halbesleben et al., 2014), integrating the concept of worrying as a proximal outcome of financial insecurity offers potential new directions for COR theory. Third, the results of this research expand our knowledge of how traits influence reactions to ongoing crises. Studying resiliency is particularly important because it, by definition, concerns whether, when, and how people persist and overcome adversity (Kossek & Perrigino, 2016). Prior work theoretically positions resiliency as a potential source of protection against resource loss effects (see Bardoel et al., 2014; Chen et al., 2015; Hobfoll et al., 2015; Hobfoll & Shirom, 1993). However, research on resiliency in the workplace is sparse, and has not sufficiently addressed how significant adversities impact employees. Although a small number of studies have examined the moderating role of resiliency in the stress process, most of this work has looked at everyday work-related stress as the context (e.g., García-Izquierdo et al., 2018; Hao et al., 2015; Lanz & Bruk-Lee, 2017); it has not methodologically included an adverse experience (see Britt et al., 2016; King et al., 2015). The COVID-19 pandemic and ensuing economic instability represent a widespread, collective trauma impacting all members of society, and thus offers a unique opportunity to address this phenomenon. Scholars have recently asserted that it is important to differentiate between psychological and behavioral resiliency (see King et al., 2022; Robertson & Cooper, 2013), and Hobfoll (2011) uses this distinction to link COR to potential resource loss outcomes. This work contributes one of the first empirical assessments connecting trait resiliency to both psychological and behavioral outcomes that are relevant to the workplace. In line with this rationale, we explore the potential for trait resiliency to strengthen the impact of financial insecurity and pandemic-related worries on proactive responses, such as increased work efforts, while also potentially exacerbating or buffering the experience of stress. By investigating behavioral and psychological responses to insecurity and worry as a function of resiliency, our findings shape understanding of the complex correlates of resiliency. In sum, this work will build new understanding of how the implications of financial insecurity on well-being and work-related behaviors vary for people who score lower or higher on a relevant individual-level characteristic (i.e., resiliency). The key role of financial concerns in the contemporary human experience—highlighted in the opening quote and polling statistics presented—makes this work timely and important. Psychological Impact of Financial Insecurity Financial insecurity has been defined and operationalized in a number of ways (see Sinclair & Cheung, 2016) ranging from objective financial measures (e.g., direct income) to more subjective measures (e.g., needs or wants). In this paper, we define financial insecurity as a global, cognitive assessment that people report about their deprivation in objective and/or subjective financial resources. Conservation of Resources (COR) theory is a useful framework for understanding people’s responses to fluctuations in resources. A basic COR theory tenet is that people are motivated to acquire and maintain valued resources (Hobfoll, 1989; Hobfoll et al., 2018). These resources can be both material (e.g., financial reserves) and personal (e.g., skills and personality traits; Halbesleben et al., 2014), and a foundational principle of COR is that resource loss or threat produces stress reactions (Halbesleben et al., 2014; Hobfoll et al., 2018). Importantly, COR theory’s emphasis on valued resources speaks to the importance of subjective appraisals in this process (Hobfoll, 1989; Hobfoll et al., 2018). In other words, it is ultimately individuals’ psychological experience or expectation of loss, rather than objective loss, that determines how one feels and responds. Scholars have emphasized the importance of subjective threat perceptions in understanding individuals’ reactions to stressful events (Brief & Atieh, 1987; Folkman et al., 1986). Supporting the idea that financial insecurity represents a threat, previous research found that financial insecurity often gives rise to worrying (Hacker et al., 2013; Wilson et al., 2020), which is characterized by a cascade of negative thoughts stemming from, and contributing to, fear or threat perception (Borkovec et al., 1983). In the current study, we define worrying as generalized cognitions stemming from concerns about finances, health, relationships, and/or other matters related to one’s circumstances (see Meyer et al., 1990). Thus, financial insecurity is the cognitive assessment of one’s financial situation, and worrying represents generalized fear or threat-related cognitions that arise from such assessments. We predict that increased perceptions of financial insecurity will be appraised as more threatening—which will be reflected in greater reported cognitions of worrying. More formally, we expect the following: H1: Financial insecurity will be positively related to worrying. Worry as a Mediating Mechanism We anticipate that individuals will experience the depletion of financial resources as a threat—giving rise to the psychological experience of worrying—and that this appraisal (i.e., worry) will ultimately relate to psychological and behavioral consequences of financial insecurity. In other words, we examine a mechanism, worry, that may help to explain the COR-posited connection between resource inadequacy and both stress and behavior (Hobfoll et al., 2018; Westman et al., 2005). COR theory explains that current resource availability relates to future resource possibilities (Hobfoll et al., 2018). Resource loss hampers efforts to obtain and conserve resources, yet resource investment is necessary to recover from past losses and guard against future losses (Bardoel et al., 2014). The nature of the COVID-19 pandemic and resulting economic instability make these difficult problems to directly alleviate, but this does not preclude efforts to limit the negative consequences of these threats. Here, we uncover the impact of financial insecurity on both proximal and more distal work-related cognitions and behaviors. We anticipate that psychological and behavioral consequences of financial insecurity will be explained by worrying, because worrying represents cognitive appraisals linking potentially threatening situations to problem-focused affect and behavior (Borkovec et al., 1983; Brosschot et al., 2006; Mathews, 1990). Indeed, worrying is often associated with the investment of cognitive resources in the form of problem-solving aimed at mitigating or preempting threats (Borkovec et al., 1983; Mathews, 1990). Due to this, worrying may lead to greater stress, yet it may also prompt workers to engage in proactive behaviors that help them address the problem, such as demonstrating their value to their organization. Importantly, we do not anticipate that financial insecurity's relationships with stress and proactive work behaviors will be uniform across all individuals. People vary in their responses to resource loss, and central to understanding this variation is the trait of resiliency—one’s tendency to “bounce back” from and overcome adversity (Bonanno & Mancini, 2008; Britt et al., 2016). We expect resiliency to moderate the relationship between worrying and downstream behavioral and psychological consequences (see Fig. 1 for full theoretical model). In the following section, we outline how resiliency may affect responses to financial insecurity.Fig. 1 Theoretical Model for the Effect of Financial Insecurity on Behavioral and Psychological Outcomes The Moderating Role of Resiliency In addition to varying in their material resources, people also vary in their personal and psychological resources (e.g., personality and skills; Hobfoll, 1989, 2011). A personal resource that drives effects during times of crisis is resiliency. Hobfoll et al. (2018) asserted that resiliency (versus vulnerability) is particularly relevant to conversations about resource possession, gain, and protection (or lack thereof), and Hobfoll and Shirom (1993) detail the benefit of COR theory for understanding stress and resiliency effects in work settings. Resiliency can act as a buffer, blunting the impact of stressors on psychological and behavioral outcomes and, thus, enabling people to recover from setbacks (Boudrias et al., 2011; Ong et al., 2006; Shoss et al., 2018). Related concepts, like psychological capital (of which resiliency is a component), are positively related to job satisfaction (Youssef & Luthans, 2007) and job performance (Luthans et al., 2005; 2007a), as well as negatively related to stress (Avey et al., 2009; Boudrias et al., 2011; see Avey et al., 2011 for a meta-analysis) and burnout (Harker et al., 2016). At the same time, resiliency may have negative effects on employees when it encourages continued goal-striving even after facing barriers to goal attainment that severely deplete psychological and physical resources, or when it encourages prioritizing subordinate goals (e.g., “maintain employment”) over superordinate ones (e.g., “maintain well-being”; King & Burrows, 2021). This potentially depleting, maladaptive effect of continued persistence despite adversity was termed its “shadow side” by Adler (2013). Related concepts—like hardiness, an individual difference that characterizes how effectively people handle difficult situations (Kobasa, 1979; Kobasa et al., 1982)—have been shown to exacerbate the negative effect of workplace stressors (e.g., poor safety climate) on stress-related outcomes (e.g., frustration; Golubivach et al., 2014). Our work suggests that the impact of financial insecurity on psychological and behavioral outcomes via worrying may differ depending on trait resiliency, particularly in contexts of ongoing and uncontrollable adversity (e.g., a global pandemic and economic crisis). We expand on the limited empirical resiliency research in working adult populations (see Britt et al., 2016) and address the call from King et al. (2022) to begin to consider both psychological and behavioral outcomes of resiliency. Specifically, we directly examine whether trait resiliency buffers or exacerbates the impact of worrying on stress and proactivity at work. Resiliency and Behavioral Outcomes One potential outcome of worrying after appraising financial insecurity is engagement in proactive work behavior—defined as “anticipatory action that employees take to impact themselves and/or their environments” (Grant & Ashford, 2008, p. 8). This is an important outcome because when workers adopt proactive behaviors, their efforts can benefit their organizations and coworkers. In line with COR theory, such actions may serve a functional resource recovery and/or protection purpose by allowing workers to demonstrate their value to the organization. Existing research suggests that work-related resources (e.g., job autonomy and control, coworker trust, supervisor support) and individual resources (e.g., proactive personality, learning goal orientation) relate to proactive behavior at work (Ohly & Fritz, 2010; Parker et al., 2006; Shin & Kim, 2015). Further, Parker et al. (2006) suggest that cognitive and motivational factors, such as appraisals of situational control, explain the effects that individual and work characteristics have on proactive workplace behavior. We extend this research by examining worrying as a cognitive factor that helps explain the link between financial insecurity—a novel antecedent of proactive behavior that reflects an employee’s appraisal of their available resources—and proactive behavior as a potential behavioral resource-seeking response to a cognitive focus on the need to address financial threat (i.e., worrying). We examine whether resiliency, a personal resource that supports goal-directed behavior in the face of adversity (Britt et al., 2016; King et al., 2022), enhances attempts to maintain employment (i.e., by being more proactive at work), particularly when experiencing worry-related cognition (i.e., higher threat appraisal). Resiliency is associated with greater positive adaptation in the face of stressful events (Boudrias et al., 2011; Shoss et al., 2018; Tugade & Fredrickson, 2004; Utsey et al., 2008) and positively relates to proactive work behavior (Caniëls & Baaten, 2019), but what is not understood is whether and how resiliency interacts with cognitive responses to stressors (e.g., worry) in the prediction of resource-related outcomes. In the face of experienced or anticipated hardships, worrying motivates people to problem-solve, which can lead to increased engagement in proactive behaviors (Borkovec, et al., 1983; Brosschot et al., 2006). This leads us to expect, overall, that worrying will be positively related to proactivity. However, some scholars have found that financial worries can diminish workers’ cognitive capacity, limiting their job performance (Meuris & Leana, 2015). Thus, whether worrying over financial insecurity promotes high or low levels of proactive work behaviors likely hinges on workers’ access to other resources like resiliency. Extending prior work, we test whether resiliency is a resource that workers are able to draw upon to alleviate negative effects of financial insecurity and worry, and to capitalize on their positive aspects, such as enacting proactive work behavior. People who are high in trait resiliency are more averse to lacking control than those low in trait resiliency (Parker et al., 2015), and resiliency allows employees to better maintain goal-directed behavior in the face of adversity (King & Burrows, 2021). As a result, resiliency may enhance the effect of worrying about uncontrollable aspects of one’s situation on proactive work behavior, with the ultimate goal of maintaining employment amid financial adversity. We predict:H2: Financial insecurity will have a positive indirect relationship with proactive behavior via worrying (H2a), and we anticipate that this indirect relationship will be stronger for individuals who are high (versus low) in resiliency (H2b). Resiliency and Stress Although scholars have examined how resiliency influences health outcomes following stressful events (e.g., Black et al., 2017; Zautra et al., 2005), as well as its general relationship with mental well-being (e.g., Boudrias et al., 2011; Kinman & Grant, 2011), there is not enough research on how resiliency may influence the propensity to experience stress in response to prolonged hardships (see Ong et al., 2006 for an exception). Britt et al. (2016) remark that the overreliance of resilience research on acute, one-time stressors limits our understanding of resiliency and its effects. Thus, we seek to empirically address the important question of whether and how resiliency shapes one’s outcomes when managing ongoing stressors that are not under the individual’s control. This question is of theoretical interest; the answer is important to our understanding of how trait resiliency relates to psychological well-being during a crisis. The present research extends prior work by examining the effect of resiliency on reactions to ongoing stressors and resulting mental health-related outcomes. Even though previous work has found that resilient people tend to suffer fewer negative outcomes following hardships (e.g., Ong et al., 2006; Shoss et al., 2018), we might not expect to find the same pattern of results during ongoing hardships (e.g., the COVID-19 pandemic and continuing financial instability). Thus, we outline two competing predictions for how resiliency may relate to stress. First, in line with its expected consequences for more favorable behavioral outcomes, resiliency may have a buffering effect on the relationship between financial insecurity, worrying cognitions, and stress. The Buffering Hypothesis of resilience (e.g., Johnson et al., 2011; Rutter et al., 2008) has received support in clinical research, illustrating that resiliency can buffer the negative effect of stressors on suicide ideation. Thus, the relationship between worrying and experienced stress might be weaker for high-resiliency individuals. This prediction aligns with COR theory, which posits that people must invest, and can draw from, existing resources in order to protect against and recover from future resource loss (Hobfoll, 2011; Hobfoll, et al., 2018). This prediction is also in line with empirical research on nurses in the U.S. showing that resiliency buffers the negative effect of interpersonal work conflict on burnout (Lanz and Bruk-Lee, 2017). Resiliency is an important personal resource, and those who have access to more of this resource may be better equipped to recover from the cognitively depleting experience of worrying. Overall, this buffering logic is consistent with evidence that resiliency tends to be negatively related to outcomes like distress and burnout (e.g., Boudrias et al., 2011; Hao et al., 2015; Harker et al., 2016; Kinman & Grant, 2011; West et al., 2020). It follows that:H3alt1: Financial insecurity will have a positive indirect relationship with stress via worrying (H3alt1a), and the positive relationship between worrying and stress will be weaker for individuals who are high (versus low) in resiliency (H3alt1b). H3alt1b predicts a buffering (i.e., beneficial) effect of resiliency consistent with our predictions about resiliency’s positive effect on proactive work behaviors. However, there is also reason to anticipate that the beneficial effect of resiliency in protecting against poor psychological outcomes would be different from its beneficial effect on positive behavioral outcomes. At times, worrying may be relatively benign—or potentially even beneficial, given its association with problem-solving—but worrying is also associated with a host of negative mental and physical health outcomes and may itself become a source of stress (Brosschot et al., 2006; Segerstrom et al., 2000). Threats that are perceived as uncontrollable, such as the COVID-19 pandemic and economic uncertainty, are particularly likely to result in prolonged and/or severe worrying (Brosschot et al., 2006), heightening the likelihood of negative mental health outcomes. Further, there is evidence that high-resiliency people find a lack of control to be more aversive than low-resiliency people (Parker et al., 2015), suggesting that trait resiliency may not only fail to act as a buffer in the context of a continuing, uncontrollable set of stressors (e.g., COVID-19 and financial insecurity), but may actually represent a liability. Further, research suggests that those high on the resiliency-relevant trait of hardiness are more likely to use coping strategies that focus on solving problems, rather than emotion-focused coping strategies that involve trying to understand, express, and accept emotional reactions to threats (Eschleman et al, 2010)—the latter (i.e., emotion-focused coping) being more adaptive when threats are uncontrollable (e.g., Folkman & Tedlie, 2004; Terry & Hynes, 1998). Although resiliency sometimes mitigates the negative impact of hardships, high resiliency may heighten the stress resulting from worrying amid uncertain, precarious, and individually uncontrollable socioeconomic hardships. This expectation is concordant with Trait Activation Theory (Tett & Burnett, 2003; Tett & Guterman, 2000), which details the relevance of certain traits to particular situations, and the resulting accentuation of thoughts and reactions associated with those traits. In the current study, more worry (an indicator of threat appraisal) likely heightens the relevance of resiliency to one’s experiences and outcomes. That is, this trait may produce more pronounced threat reactions (i.e., stress) for high-resiliency individuals—who are typically able to overcome adversity with ease and problem-focused effort—because experiencing a situation that produces worry and that they can not easily “fix” (i.e., remove) or control, such as a pandemic, may be more stressful for them than it is for their low-resiliency counterparts. Thus, unlike the potential beneficial moderating effects of resiliency on positive work behavior (H2), the relationship between worry and stress may be stronger for high-resiliency individuals—meaning that rather than protecting against negative mental health outcomes (H3alt1), resiliency will actually exacerbate these. More formally, this line of reasoning leads to the prediction:H3alt2: Financial insecurity will have a positive indirect relationship with stress via worrying (H3alt2a), and the positive relationship between worrying and stress will be stronger for individuals who are high (versus low) in resiliency (H3alt2b). Conservation of Resources (COR) theory is a useful framework for uncovering whether resiliency's moderating effect on the relationship between worrying and stress is consistent with either the buffering hypothesis or trait activation theory. Possessing higher trait resiliency may facilitate a resource gain cycle (Chen et al., 2015; Hobfoll, 1989), where resiliency is leveraged to recover the lost resources reflected in worrying; alternately, high resiliency could facilitate a loss cycle (Hobfoll et al., 2018), where worrying reflects a fundamental misalignment between individuals’ current adversity experience and their typical (and desired) experience of overcoming threats, leading to more stress. Given these competing predictions, we posit that financial insecurity will impact stress via worrying and we will offer new evidence regarding the ways that these processes influence individuals who are high and low in trait resiliency. Below, we detail the results of a study examining financial insecurity during the COVID-19 pandemic and its impact on behavioral and psychological outcomes. Method Participants and Procedure We used online Qualtrics surveys to collect multi-source data from working parents and their college-aged children. We relied on a snowball sampling approach (Biernacki & Waldorf, 1981), which involved asking undergraduate research assistants to share the survey with their contacts and inviting college-aged students and/or working parents of college-aged students to participate in the survey via posts on social media platforms (e.g., Facebook, Instagram, LinkedIn). In each case, the college-aged student or working parent (whomever was recruited first) was asked to recruit one of their working parents or college-aged children, respectively, for an accompanying survey. To ensure that parent–child responses could be matched, the student generated a special six digit code in their survey and then shared that code with their parents to enter in the survey that they later took. For 76% of participants, if both the parent and college-aged child completed the survey, the child was compensated with a $20 Amazon gift card.1 For others (24%), survey completion was voluntary and neither the child nor the parent was compensated. All participants were 18 years of age or older (Mparent = 50.9, SDparent = 7.0; Mchild = 20.7, SDchild = 3.0), and a total of 636 parent–child pairs completed the survey. Both the parent (72% women, 28% men, < 1% non-binary) and child (71% women, 28% men, 1% non-binary) samples were largely comprised of women. We also observed a relatively high level of racial diversity for both the parents (28% Asian, 24% White, 22% Hispanic/Latino, 17% Black, 8% Other) and children (28% Asian, 22% Hispanic/Latino, 20% White, 17% Black, 14% Other). Of the parent–child pairs that participated in the study, 78% indicated that they were living together during the pandemic (i.e., at the time that they completed the survey in April 2020). Parents responded to items about their financial insecurity both prior to and during the COVID-19 pandemic, their behaviors at work (i.e., their engagement in proactive behaviors), worrying, resiliency, and demographics. College-aged children responded to a question about their parent’s stress. Measures Financial Insecurity Parents’ financial insecurity was measured using two items developed for the purposes of this study. The instructions stating, “The following items pertain to your financial security before the COVID-19 crisis and at present,” were presented to ensure that the items were framed in light of the COVID-19 pandemic. Participants were asked, (1) “How financially secure do you CURRENTLY feel?” and (2) “How financially secure did you feel FIVE MONTHS BEFORE the COVID-19 crisis began?” They responded on a 7-point Likert-type scale (1 = Extremely insecure; 7 = Extremely secure). We reverse coded these variables such that higher scores indicate greater financial insecurity. Worrying Parents’ worrying was measured using four items developed for this study. As with the financial insecurity measure, the instructions, “Since the COVID-19 crisis began, to what extent have you engaged in or experienced the following,” were used to ensure that the items were framed in light of the COVID-19 pandemic. Participants used a 7-point Likert-type scale (1 = Never; 7 = Almost always; Cronbach’s alpha = .81) to indicate how frequently they experienced concerns like “Worrying about access to groceries” (see Appendix A for full scale). Resiliency Parents’ trait resiliency was assessed using six items adapted from Luthans et al. (2007b) measure. Participants used a 7-point Likert-type scale (1 = Strongly disagree; 7 = Strongly agree; Cronbach’s alpha = .76) to indicate their agreement with items including, “When I have a setback, I have trouble recovering from it, moving on” and “I usually take stressful things in stride” (see Appendix A for full scale). Children’s Reports of Parent Stress College-aged children’s perceptions of their parents’ stress were measured with one item developed for this study.2 They were asked to respond to the following item using a 7-point Likert-type scale (1 = Not stressed at all; 7 = Extremely stressed): “How would you categorize this parent’s (i.e., the parent who is taking this survey) overall level of stress with respect to COVID-19?”. Proactive Work Behaviors Work-related behaviors were measured using items from a social identity-based impression management scale (Little et al., 2015) that we adapted to fit the pandemic context. Participants were asked “Since the COVID-19 crisis began, to what extent do you feel like you've been engaging in each of the following behaviors?” Participants responded to four items comprising one of the subscales, which assessed whether respondents were engaging in more proactive behaviors since the onset of the pandemic (e.g., “I try to work harder in my job than I did before COVID-19”; Cronbach’s alpha = .84). They responded using a 7-point Likert-type scale (1 = Strongly disagree; 7 = Strongly agree). See Appendix A for full scale. Parent Socioeconomic Status To achieve an objective measure of parents’ approximate socioeconomic status (SES), we asked about their household income. Parents were asked “What is your total yearly household income?” and chose from one of eight income brackets (see Appendix A). Results Descriptive statistics and correlations can be found in Table 1. We used a linear regression analysis in R to test H1, followed by analyses using the lavaan package (Rosseel, 2012) to test for the indirect effects predicted in H2 and H3 with bootstrap samples (see Preacher et al., 2007). In all analyses, we included all 636 parent–child pairs and controlled for participants’ reported pre-pandemic level of financial insecurity.Table 1 Descriptive Statistics and Correlations for Study Variables Variable M SD 1 2 3 4 5 6 1. Financial Insecurity During COVID-19 3.85 1.75 — 2. Financial Insecurity Pre-COVID-19 3.08 1.65 .69* — 3. Worry 3.73 1.39 .28* .17* — 4. Resiliency 5.27 0.97 -.16* -.18* -.01 — 5. Proactive Work Behaviors 4.41 1.29 .02 .00 .17* .11 — 6. Children’s Report of Parent Stress 4.18 1.37 .25* .18* .40* -.03 .13 — 7. Parent SES 4.64 2.21 -.52* -.44* -.16* .16*** .09* -.07† n = 636 † p < .10 * p < .05 p < .01 * p < .001 Validation of the Measures To better assess the convergent and discriminant validity of the measures relative to other existing measures of similar constructs, we invited 250 members of a paid, online Qualtrics research panel to complete the original scales and other existing measures. Eligibility to participate was set to align with the primary study; all participants had to be employed parents over the age of 18. A total of 249 participants completed the survey after excluding one participant who indicated they were unemployed (64.7% women; 0.4% American Indian or Alaska Native, 2.0% Asian, 11.6% Black or African American, 0.4% German, 4.0% Hispanic or Latinx, 2.8% Multiracial, 78.7% White/Caucasian; average age = 42.2 years, SD = 12.3). The measures that were assessed in this validation study include: Financial Stress (Ullah, 1990; Warr & Jackson, 1987), InCharge Financial Distress/Financial Well-Being Scale (Prawitz et al., 2006), Perceived Stress Scale (Cohen et al., 1983), Penn State Worry Questionnaire (Stober & Bittencourt, 1998), Personal Initiative (Frese et al., 1997), and the shortened version of Bateman and Crant’s (1993) Proactive Personality Scale found in Seibert et al. (1999). Full scales are listed in Appendix B. Evidence for convergent validity of the original measures can be inferred from the direction, magnitude, and statistical significance of correlations between the original items and existing scales of similar constructs Table 2. Evidence of discriminant validity can be inferred by these features of correlations between the original scales and measures of dissimilar constructs.Table 2 Correlational Evidence of Measure Convergent and Discriminant Validity M (SD) 1 2 3 4 5 6 7 8 9 10 11 1. Original: Financial insecurity now 3.74 (1.93) – 2. Original: Financial insecurity pre-COVID 3.05 (1.67) .50 – 3. Validation: InCharge Financial Distress Scale 5.69 (2.47) .64 .36 .93 4. Validation: Financial Stress 2.81 (1.18) .33 .16* .67 .89 5. Original: Worrying 3.57 (1.56) .09 .11 .41 .44 .88 6. Validation: Penn State Worry Questionnaire 3.98 (1.08) .27 .23 .59 .48 .52 .90 7. Original: Stress 3.67 (1.73) .10 .08 .29 .30 .56 .46 – 8. Validation: Perceived Stress Scale 2.32 (0.61) -.03 -.04 .20 .33 .33 .34 .30 .82 9. Original: Proactive Work Behavior 5.09 (0.84) -.18 -.18 .06 .09 .18 .10 -.03 .35 .81 10. Validation: Personal Initiative 5.30 (1.03) -.16 -.27** -.06 .07 .15* -.02 .02 .32 .62 .88 11. Validation: Proactive Personality 5.33 (1.02) -.21 -.27 -.13* .02 .18 -.07 .03 .34 .45 .72 .92 12. Original: Resiliency 5.27 (0.89) -.25 -.30 -.33 -.20 -.08 -.36** -.16* .10 .29 .36 .40 .69 p < .01 p < .05 Values for Cronbach’s alpha are presented in italics on the diagonal The pattern of correlations between scales using this additional sample provides general support for convergent and discriminant validity of these original measures. For example, in the case of the measures of financial insecurity, positive and significant relationships between existing measures of financial stress and the item measuring current financial insecurity (rs = .64 and .33, ps < .01) were greater in magnitude than correlations with other variables. The same is true in the case of the positive and significant relationships between the original measure of proactive work behavior and the related scales of personal initiative (r = .65, p < .01) and proactive personality (r = .45, p < .01), where all other correlations tended to be smaller in magnitude (rs ranging from .06 (financial distress) to .35 (perceived stress)). In addition, the measure of worrying used in the primary study was positively and significantly correlated with another existing measure of worrying (r = .52, p < .01). In sum, these supplemental analyses provide evidence to support the validity of the measures used and robustness of the primary patterns of correlations in the current study. Main Effects We tested H1 by regressing the hypothesized mediator, worrying, on financial insecurity during the pandemic, controlling for pre-pandemic financial insecurity. Financial insecurity during the pandemic was significantly positively related to worrying, ß = 0.31, SE = 0.05, t(633) = 5.75, p < .001. The effect of pre-pandemic financial insecurity on worrying was not significant, p = .49. Thus, the results fully support H1. Next, we examined the relationship between financial insecurity during the pandemic and our first outcome variable, proactive work behaviors. Financial insecurity during the pandemic was not significantly related to engaging in proactive behaviors at work, p = .52. Pre-pandemic financial insecurity was also not significantly related to proactivity, p = .61. Testing the relationship between financial insecurity during the pandemic and stress revealed a significant effect of financial insecurity during the pandemic on stress, ß = 0.25, SE = 0.05, t(633) = 4.69, p < .001. The effect of pre-pandemic financial insecurity on stress was not significant, p = 0.91. Despite not finding a significant main effect of financial insecurity on proactive work behaviors, scholars have argued that it is still appropriate to test for indirect effects because such effects might be masked by other mediators working in opposition (i.e., suppressor variables; Hayes, 2009; Rucker et al., 2011; Zhao et al., 2010). For this reason, we conducted indirect effects and moderated indirect effects analyses using this outcome variable. Indirect Effects Analyses We tested for indirect effects using the lavaan package in R (Rosseel, 2012) with 5,000 bootstrap samples (Preacher et al., 2007), controlling for pre-pandemic financial insecurity in all analyses. Coefficients for these analyses are reported in Table 3. We first tested H2a, that there would be a positive indirect effect of financial insecurity on proactive work behaviors via worrying. Despite the absence of a significant total effect, the indirect path was significant, as hypothesized, ß = 0.05, SE = 0.02, z = 3.01, p = .003, 95% CI = [0.03, 0.10]. Thus, H2a was supported.Table 3 Coefficients for Indirect Effects Analyses DV Financial Insecurity → Worry (a) Worry → DV (b) Total effect (c) Direct effect (c’) Indirect Effect (ab) 95% CI Proactive Work Behaviors 0.31 0.18* 0.04 -0.02 0.05 [0.03, 0.10] Stress 0.31* 0.35* 0.25* 0.14 0.11* [0.07, 0.16] n = 636. The predictor variable in these analyses is financial insecurity during the pandemic and the mediator is worry. These analyses control for pre-COVID-19 financial insecurity. Coefficients presented are standardized linear regression coefficients. Analyses were conducted using 5,000 bootstrap samples * p < .05 p < .01 * p < .001 We then tested, consistent with H3alt1a and H3alt2a, whether there was a positive indirect effect of financial insecurity on stress via worrying and found support for this: the indirect path from financial insecurity to stress via worrying was significant, ß = 0.11, SE = 0.02, z = 4.82, p < .001, 95% CI = [0.07, 0.16]. Moderated Indirect Effects Analyses We tested for moderated indirect effects using the lavaan package in R (Rosseel, 2012) with 5,000 bootstrap samples (Preacher et al., 2007). We controlled for pre-pandemic financial insecurity in all analyses. Results are summarized in Table 4.Table 4 Coefficients for Moderated Indirect Effects Analyses Coefficient No Demographic Controls Controlling for SES Controlling for SES & Demographic Variables a Proactivity Stress Proactivity Stress Proactivity Stress Financial Insecurity → Worry (a) 0.31* 0.31* 0.30* 0.30* 0.28* 0.27* Worry → DV (b) Interaction 0.14*** 0.09* 0.15*** 0.09* 0.13** 0.09* Low Resiliency 0.03 0.26* 0.02 0.26 0.03 0.26* High Resiliency 0.32* 0.45* 0.32* 0.45* 0.28* 0.43* Total effect (c) 0.04 0.25* 0.09 0.28* 0.07 0.25* Direct effect (c’) -0.02 0.14 0.04 0.17* 0.03 0.15* Indirect Effect (ab) Low Resiliency 0.01 0.08* 0.01 0.08 0.01 0.07 High Resiliency 0.10* 0.14* 0.10 0.13* 0.01 0.12* Index of Moderated Mediation 0.04** 0.03* 0.04** 0.03* 0.08** 0.02* 95% CI b [0.02, 0.08] [0.01, 0.06] [0.02, 0.08] [0.01, 0.06] [0.01, 0.07] [0.002, 0.05] n = 636. The predictor variable in these analyses is financial insecurity during the pandemic and the mediator is worry. All analyses in this table control for pre-COVID-19 financial insecurity. Coefficients presented are standardized linear regression coefficients. Analyses were conducted using 5,000 bootstrap samples a Demographic variables included in the analysis: parents’ age, race, and gender in addition to parents’ SES level b We report 95% confidence intervals for the index of moderated mediation * p < .05 p < .01 * p < .001 In H2b we predicted that the indirect effect of financial insecurity on proactive work behaviors would be moderated by trait resiliency such that the positive relationship between worry and proactive behaviors would be stronger for high-resiliency individuals. To test this, we performed a moderated indirect effects analysis (see Fig. 2). The first leg of this model indicated that financial insecurity predicted greater worry, ß = 0.31, SE = 0.06, z = 5.38, p < .001, 95% CI = [0.20, 0.42]. The second leg of this model, which regressed proactive behaviors on worry, resiliency, and their interaction, yielded a significant interaction between worry and resilience, ß = 0.14, SE = 0.04, z = 3.64, p < .001, 95% CI = [0.07, 0.22]. There was a significant positive relationship between worry and proactive work behaviors for high-resiliency individuals, ß = 0.32, SE = 0.05, z = 6.67, p < .001, 95% CI = [0.22, 0.40], but not for low-resiliency individuals, p = .51. For high-resiliency individuals (one standard deviation above the mean), the indirect effect through worry was significant, ß = 0.10, SE = 0.03, z = 3.52, p < .001, 95% CI = [0.05, 0.16]. For low-resiliency individuals (one standard deviation below the mean), the indirect effect through worry was not significant, p = .65. The index of moderated mediation was significant, ß = 0.04, SE = 0.02, z = 3.03, p = .002, 95% CI = [0.02, 0.08]. Thus, the results partially support H2b: the relationship between worry and resilience was indeed stronger for high- versus low-resiliency individuals, but there was only a significant positive relationship between worry and proactive work behaviors for high-resiliency individuals, not for low-resiliency individuals.Fig. 2 Moderated Indirect Effects Model for the Effect of Financial Insecurity on Proactive Work Behaviors. Note. n = 636. This analysis controls for pre-COVID-19 financial insecurity. The c notation indicates the total effect of financial insecurity on stress; c’ indicates the direct effect after controlling for the path through the mediator. Coefficients presented are standardized linear regression coefficients. Analysis was conducted using 5,000 bootstrap samples. Results remain effectively the same when controlling for parents’ socioeconomic status. * p < .05. p < .01. * p < .001 We expected trait resiliency to moderate the relationship between worrying and stress, but had two competing hypotheses for the nature of this interaction. H3alt1b predicted that the positive relationship between worrying and stress would be weaker for high-resiliency individuals whereas H3alt2b predicted that this relationship would be stronger for high-resiliency individuals. To test these hypotheses, we performed a moderated indirect effects analysis (see Fig. 3). As in the simple indirect effects analysis we conducted, the first leg of this model indicated that financial insecurity predicted greater worry, ß = 0.31, SE = 0.06, z = 5.34,3p < .001, 95% CI = [0.20, 0.42]. The second leg of this pathway, which regressed stress on worry, resiliency, and their interaction, yielded a significant interaction between worry and resiliency, ß = 0.09, SE = 0.04, z = 2.44, p = .015, 95% CI = [0.02, 0.17]. The relationship between worry and stress was stronger for high-resiliency individuals, ß = 0.45, SE = 0.04, z = 11.65, p < .001, 95% CI = [0.37, 0.52], compared to low-resiliency individuals, ß = 0.26, SE = 0.04, z = 6.88, p < .001, 95% CI = [0.18, 0.33]. For high-resiliency individuals, the indirect effect through worry was significant, ß = 0.14, SE = 0.03, z = 4.63, p < .001, 95% CI = [0.09, 0.20]. For low-resiliency individuals, the indirect effect through worry was still significant, albeit weaker, ß = 0.08, SE = 0.02, z = 3.62, p < .001, 95% CI = [0.04, 0.13]. The index of moderated mediation was significant, ß = 0.03, SE = 0.01, z = 2.25, p = .028, 95% CI = [0.01, 0.06]. Thus, the results support H3alt2: the positive relationship between worrying and stress is more pronounced for individuals high in resiliency.Fig. 3 Moderated Indirect Effects Model for the Effect of Financial Insecurity on Stress. Note. n = 636. This analysis controls for pre-COVID-19 financial insecurity. The c notation indicates the total effect of financial insecurity on stress; c’ indicates the direct effect after controlling for the path through the mediator. Coefficients presented are standardized linear regression coefficients. Analysis was conducted using 5,000 bootstrap samples. Results remain effectively the same when controlling for parents’ socioeconomic status. * p < .05. p < .01. * p < .001 Supplemental Exploratory Analyses We further conducted two sets of exploratory analyses: potential first-stage moderation and subjective versus objective judgments. Potential First-Stage Moderation We examined the possibility that resiliency buffers the positive relationship between financial insecurity and worry, or that resilient employees are more likely to perceive financial insecurity as less threatening, thereby mitigating its unfavorable consequences. Although we conducted this analysis, it is important to recognize that we hypothesized second-stage, rather than first-stage, moderation for three reasons. First, conceptual work (Britt et al., 2016; Fisher et al., 2018) asserts that an experience must be felt as a “significant adversity” to the person for the concept of resiliency to be relevant. Specifically, Britt et al. (2016) state that exposure to traditional stressors alone (e.g., financial insecurity) “may not constitute adversity in the context of resilience assessment.” Thus, we assessed resiliency as a resource that may buffer the effects of stress responses that illustrate an experienced adversity is present (i.e., worrying), rather than assuming financial insecurity is a significant adversity to all and that resiliency necessarily is, therefore, relevant. Second, conceptual work by Adler (2013) on the “shadow side” of resilience asserts that scholars and practitioners should be careful not to associate resilience with an absence of stress responses following a stressful experience, as that may stigmatize typical human reactions to stress and limit necessary help-seeking behavior. Thus, Adler writes that resiliency is being “misapplied” (Adler, 2013, p. 227) when it is used to “inadvertently convey the message that faltering in the face of profound stress signals a lack of resilience” (Adler, 2013, p. 227). We believe that resiliency is relevant to how we recover, respond to, and “bounce back” from experienced stress reactions (i.e., worrying; Block & Block, 1980), and is not the absence of stress reactions in difficult times (first-stage moderation). Finally, we drew on empirical work by Tugade and colleagues (Tugade & Fredrikson, 2004; Tugade et al., 2004) on resiliency and emotions, which has demonstrated that anxiety and worrying about a stressful event or experience is not statistically predicted by trait resiliency, and trait resiliency is not associated with less negative emotions in response to stressful experiences (Tugade et al.,2004). In line with these findings, our study’s hypotheses focused on the prediction that, despite (understandably) worrying, resiliency buffers the negative effects of worrying when experiencing financial insecurity during the pandemic. An examination of the alternative hypothesis of resilience as a first-stage moderator did not yield strong support. Specifically, for the first leg of the model, which regressed worry on financial insecurity, resiliency, and their interaction, there was a marginally significant interaction between financial insecurity and resiliency, ß = 0.08, SE = 0.04, z = 1.72, p = .09, 95% CI = [-0.01, 0.16]. The relationship between financial insecurity and worry was significant for both high-resiliency individuals, ß = 0.38, SE = 0.07, z = 5.33, p < .001, 95% CI = [0.23, 0.51], and for low-resiliency individuals, ß = 0.22, SE = 0.07, z = 3.03, p = .002, 95% CI = [0.08, 0.37]. Looking at proactive work behaviors as the dependent variable for the second leg of the model, worry was significantly positively related to proactive behaviors, ß = 0.18, SE = 0.05, z = 3.76, p < .001, 95% CI = [0.08, 0.27]. The indirect effect through worry was significant for both high-resiliency individuals, ß = 0.07, SE = 0.02, z = 2.99, p = .003, 95% CI = [0.03, 0.12], and for low-resiliency individuals, ß = 0.04, SE = 0.02, z = 2.26, p = .02, 95% CI = [0.01, 0.08]. The index of moderated mediation was not significant, ß = 0.01, SE = 0.01, z = 1.54, p = .12, 95% CI = [0.00, 0.03]. Looking at stress as the dependent variable for the second leg of the model, worry was significantly positively related to stress, ß = 0.35, SE = 0.04, z = 9.54, p < .001, 95% CI = [0.28, 0.43]. The indirect effect through worry was significant for both high-resiliency individuals, ß = 0.13, SE = 0.03, z = 4.69, p < .001, 95% CI = [0.08, 0.20], and for low-resiliency individuals, ß = 0.08, SE = 0.03, z = 2.89, p = .004, 95% CI = [0.03, 0.14]. The index of moderated mediation was only marginally significant, ß = 0.03, SE = 0.02, z = 1.71, p = .09, 95% CI = [-0.003, 0.06]. Testing Reverse Relationship Between Mediator and Dependent Variable We also tested models in which we switched the order of the mediator (worrying) and the dependent variable (proactive work behaviors and stress), with moderation by resiliency on the second leg. There was not strong evidence for these alternate models. Proactive Work Behaviors. The first leg of the model regressed proactive behaviors on financial insecurity; the relationship between these two variables was not significant, ß = 0.04, SE = 0.05, z = 0.68, p = .50, 95% CI = [-0.07, 0.14]. The second leg of the model, which regressed worrying on proactive work behaviors, and their interaction, yielded a significant interaction between proactive behaviors and resiliency, ß = 0.09, SE = 0.04, z = 2.33, p = .02, 95% CI = [0.01, 0.17]. The relationship between financial insecurity and worry was significant for high-resiliency individuals, ß = 0.24, SE = 0.04, z = 5.48, p < .001, 95% CI = [0.15, 0.32], but was not significant for low-resiliency individuals, ß = 0.05, SE = 0.04, z = 1.19, p = .24, 95% CI = [-0.03, 0.13]. However, the indirect effect of financial insecurity on worry through proactive work behaviors was neither significant for high-resiliency, ß = 0.01, SE = 0.01, z = 0.65, p = .51, 95% CI = [-0.02, 0.04], nor for low-resiliency, ß = 0.002, SE = 0.01, z = 0.38, p = .71, 95% CI = [-0.003, 0.02], individuals. The index of moderated mediation was not significant, ß = 0.003, SE = 0.01, z = 0.62, p = .54, 95% CI = [-0.01, 0.02]. Stress. Looking at the first leg of the model, financial insecurity was significantly related to stress, ß = 0.25, SE = 0.05, z = 4.62, p < .001, 95% CI = [0.14, 0.35]. The second leg of the model regressed worrying on stress, resilience and their interaction. The main effect of stress on worrying was significant, ß = 0.35, SE = 0.04, z = 9.11, p < .001, 95% CI = [0.28, 0.43], but the interaction between stress and resilience was not significant, ß = 0.07, SE = 0.04, z = 1.63, p = .10, 95% CI = [-0.01, 0.14]. The indirect effect of financial insecurity on worrying via stress was significant, ß = 0.09, SE = 0.02, z = 4.11, p < .001, 95% CI = [0.05, 0.13], but the index of moderated mediation was not significant, ß = 0.02, SE = 0.01, z = 1.53, p = .13, 95% CI = [-0.002, 0.04]. Subjective Versus Objective Judgments Our theorizing draws upon the idea that subjective judgements of resources are more predictive of outcomes than objective resources. To test this, we conducted similar analyses with parents’ SES as a covariate, and additional analyses controlling for parents’ gender, race, and age. Results are summarized in Table 4. The contrast codes used in these analyses are listed in Appendix C. There was no substantive change in our results when including these control variables. Discussion The current study demonstrates that financial insecurity impacts employees’ stress via the explanatory mechanism of pandemic-related worrying, and that this relationship is stronger for high-resiliency individuals. We also find that financial insecurity is associated with greater proactive work behaviors, via pandemic-related worrying, but only among high-resiliency individuals. We observe that trait resiliency, widely regarded as protective for its buffering role in helping people cope with traumatic experiences, may not be universally beneficial. Despite appearing to “bounce back” by remaining proactive at work, resilient individuals simultaneously experience greater stress. Theoretical and Practical Implications Our findings contribute to both theory and practice in several ways. First, consistent with predictions from COR theory, we find that perceived resource loss (indexed by self-reported financial insecurity amid the COVID-19 pandemic) is associated with stress. COR theory suggests that resource loss has a tendency to “spiral”—that is, the loss of resources engenders stress, which in turn depletes resources in an ongoing feedback loop (Hobfoll, 2011; Hobfoll et al., 2018). Our finding of a positive relationship between resource loss (financial insecurity) and stress via worrying highlights a potential cognitive mechanism through which loss spirals operate, thereby extending COR theory’s predictions. COR theory suggests that resources are not isolated but rather interconnected, and that social and environmental contexts can either pave or obstruct the way for accumulating or protecting resources (Halbesleben et al., 2014; Hobfoll, 2011). The COVID-19 pandemic is a context that has generated or exacerbated precarious socioeconomic conditions for many. Our data demonstrate the importance of considering people's subjective experiences of this precariousness, given its relation to both well-being and workplace productivity. Practitioners can communicate these findings to organizational leaders to impart the importance of providing sufficient resources to employees to offset the negative implications of financial insecurity on employee well-being. Our study additionally contributes to the relatively nascent literature on the role of trait resiliency, both in general and in the workplace, specifically by exploring, in tandem, its impact on stress and work behaviors (Chen et al., 2015; Hobfoll & Shirom, 1993) in alignment with the recommendation of Robertson and Cooper (2013) to differentiate between psychological and behavioral resiliency effects. COR theory’s “desperation principle” posits that dwindling resources can lead to maladaptive responses (Hobfoll et al., 2018). Our findings demonstrate that perceived resource loss (i.e., financial insecurity) is indirectly related to proactivity at work via maladaptive cognitions (i.e., worry), for those high in resiliency. Indeed, the indirect relationship between financial insecurity and proactive behavior via worrying was not significant for individuals low in resiliency. A greater tendency to “go the extra mile” at work when facing financial insecurity could be interpreted as consistent with COR theory’s prediction that responses to stressors become increasingly acute as one’s resources become progressively depleted or threatened, but the precise patterns of results in this study are supportive of a trait activation interpretation, rather than a buffering interpretation. That is, our finding that only people who are high in resiliency responded to worrying by increasing proactivity at work underscores the importance of considering individual differences to build understanding of how people manage the sequelae of financial insecurity. Although increasing one’s effort at work could result in positive workplace outcomes, diverting attention and efforts away from the self and toward one’s job during a time of unprecedented hardship and uncertainty may not be adaptive. Supporting this concern, the development of burnout has been conceptualized as analogous to COR theory’s loss spirals (William et al., 2020). Some have suggested that hyperactivity is the first step in developing burnout (Weber & Jaekel-Reinhard, 2000), and that the most dedicated employees are most susceptible to burnout as they ramp up activity in hopes of garnering favorable outcomes, only to be disappointed when their efforts do not materialize (Maslach et al., 2001). Together, our findings could be interpreted as evidence that highly resilient people draw on resiliency as a resource to cope with the negative cognitive processes (i.e., worrying) associated with resource loss (i.e., financial insecurity), which in turn further depletes resources needed to manage stress. From a practical standpoint, the current findings suggest that financial insecurity comes with unfavorable sequelae of worrying and stress that are not buffered by resiliency. This means that programs or policies focusing on creating or bolstering resiliency may not have the intended positive impacts on employees who are managing financial stress. In addition, these results suggest that managers interested in the well-being of employees need to look beneath the surface of productive behaviors. In other words, just because a person is performing well in times of crisis does not mean that they are coping well. Shifting focus from only encouraging task-related resilience to also emphasizing psychological recovery and well-being, particularly in response to persistent stressors, may be a helpful approach for maintaining productivity and well-being (see King & McSpedon, 2022; Sonnentag et al., 2022). For example, leaders can create and model norms for maintaining reasonable boundaries between work and non-work, and can implement interventions that allow and encourage clearer boundary management (Karabinski et al., 2021), thereby replenishing resources needed for stress management. Limitations and Future Research Directions This study is not without limitations. First, the application of a cross-sectional design limits claims of causality. However, in line with Spector’s (2019) recommendations regarding the optimization of cross-sectional designs, we included multi-source data (i.e., third-party ratings of the dependent variable, stress) to reduce same-source bias. Incorporating longitudinal designs could further elucidate the downstream effects of financial insecurity. Second, because we used a convenience sample of college students attending a private university as our second source of data, the generalizability of these results to those with lower objective and/or subjective SES may be limited. As such, additional scholarship is needed to explore these relationships among lower-SES groups. Despite this limitation, when we controlled for SES in our exploratory analyses, we found that perceived financial insecurity had predictive power above and beyond objective SES (as well as other demographic variables, see Table 4), suggesting the unique predictive and theoretical value of this more specific and mutable measure of financial well-being. Future research might also address the limitation of the measures used in the current study, such as clearly articulating the cause of the stress (e.g., stress with respect to COVID-19). Finally, further research could explore how “going the extra mile” at work amid extreme stress relates to individual differences such as proactive personality and how it impacts long-term outcomes for those high and low in resiliency. The long-term impacts of increasing proactivity at work amid a widespread crisis are unknown; although highly resilient individuals’ tendency to do so in the face of financial insecurity and COVID-related worry may seem to be an asset for organizations at face value, there could very well be long-term losses associated with, for example, burnout associated with this increased proactivity. Conclusion The extraordinary conditions of the pandemic have provoked unforeseen and ongoing stressors, including the relatively understudied experience of financial insecurity. Using COR theory as a backdrop, the current research provides crucial information about the consequences of experiencing financial insecurity amid the COVID-19 pandemic. We evaluate worrying as an explanatory mechanism linking financial insecurity to both psychological and behavioral outcomes. Furthermore, we shed light on the role of resiliency, a personal resource that is particularly relevant in the wake of adversity, and its relationship with increased employee proactive work behaviors and stress during a crisis. The results of this multi-source study suggest that resiliency may encourage proactive responses to worrying, but also that resiliency in the face of continuing stressors may come at the cost of stress itself. Appendix A Study Measures Worrying Measure of worrying, developed for this study. Respondents used a 7-point Likert-type scale ranging from Never (1) to Almost always (7) to answer how frequently they worried about various concerns. Since the COVID-19 crisis began, to what extent have you engaged in or experienced the following?Worrying about access to groceries, Worrying about access to medicines Worrying about older friends’ or family members’ health Worrying about personal health Trait Resiliency Measure of resiliency, adapted from Luthans et al. (2007b). Respondents used a 7-point Likert-type scale ranging from Strongly disagree (1) to Strongly agree (7) to indicate the extent to which they agreed with each statement. Please indicate the extent to which you agree with each of the following statements:When I have a setback, I have trouble recovering from it, moving on. I usually manage difficulties one way or another. I can be “on my own,” so to speak, if I have to I usually take stressful things in stride. I can get through difficult times because I’ve experienced difficulty before. I feel I can handle many things at a time. Proactive Work Behaviors Measure of proactive work behaviors, adapted from a subscale in the Social Identity-based Impression Management During Pregnancy scale (SIMp; Little et al., 2015). Respondents used a 7-point Likert-type scale ranging from Strongly disagree (1) to Strongly agree (7) to indicate the extent to which they agreed with each statement. Since the COVID-19 crisis began, to what extent do you feel like you've been engaging in each of the following behaviors?:I try to work harder in my job than I did before COVID-19 I have taken on more responsibility at work I volunteer for more duties than are required I try to get more done and be more productive at work Parents’ SES What is your total yearly household income?Less than $20,000 $20,000 to $34,999 $35,000 to $49,999 $50,000 to $74,999 $75,000 to $99,999 $100,000 to $149,999 $150,000 to $199,999 $200,000 or more Appendix B Measures Included in Validation Study Proactive Personality I am constantly on the lookout for new ways to improve my life. Wherever I have been, I have been a powerful force for constructive change. Nothing is more exciting than seeing my ideas turn into reality. If I see something I don’t like, I fix it. No matter what the odds, if I believe in something I will make it happen. I love being a champion for my ideas, even against others’ opposition. I excel at identifying opportunities. I am always looking for better ways to do things. If I believe in an idea, no obstacle will prevent me from making it happen. I can spot a good opportunity long before others can. Personal Initiative: I actively attack problems. Whenever something goes wrong, I search for a solution immediately. Whenever there is a chance to get actively involved, I take it. I take initiative immediately even when others don’t. I use opportunities quickly in order to attain my. Goals. Usually I do more than I am asked to do. I am particularly good at realizing ideas. InCharge Financial Distress Scale: What did you feel was the level of your financial stress this past month? Using the stair steps above, indicate how satisfied you were with your financial situation this past month. The “1” at the bottom of the steps represents complete dissatisfaction. The “10” at the top of the stair steps represents complete satisfaction. The more dissatisfied you are, the lower the number you should circle. The more satisfied you are, the higher the number you should circle. How did you feel about your financial situation this past month? How often did you worry about being able to meet normal monthly living expenses this past month? In the past month, how confident were you that you could find the money to pay for a financial emergency that would cost about $1,000? How often did this happen to you in the past month? You wanted to go out to eat, go to a movie, or do something else and didn’t go because you couldn’t afford to? How frequently did you find yourself just getting by financially and living paycheck to paycheck this past month? How stressed did you feel about your personal finances this past month? Financial Stress: Have you had serious financial worries? Were you unable to do the things you like to do because of shortages of money? Were you unable to do the things you need to do because of shortages of money? Were you unable to manage with the money you have? Perceived Stress Scale: been upset because of something that happened unexpectedly? felt that you were unable to control the important things in your life? felt nervous and “stressed”? dealt successfully with irritating life hassles? felt that you were effectively coping with important changes that were occurring in your life? felt confident about your ability to handle your personal problems? felt that things were going your way? found that you could not cope with all the things that you had to do? been able to control irritations in your life? felt that you were on top of things? been angered because of things that happened that were outside of your control? found yourself thinking about things that you have to accomplish? been able to control the way you spend your time? felt difficulties were piling up so high that you could not overcome them? Penn State Worry Questionnaire: If I didn’t have enough time to do everything, I didn’t worry about it My worries overwhelmed me I didn’t tend to worry about things Many situations made me worry I knew I shouldn’t have worried about things, but I just couldn’t help it When I was under pressure, I worried a lot I was always worrying about something I found it easy to dismiss worrisome thoughts As soon as I finished one task, I started to worry about everything else that I had to do I did not worry about anything When there was nothing more I could do about a concern, I didn’t worry about it anymore I noticed that I had been worrying about things Once I started worrying, I couldn’t stop I worried all the time I worried about projects until they were all done Appendix C Contrast Variables for Categorical Demographic Controls Tables Table 5 Contrast Variables for Parents’ Gender Gender V1: Female vs. Mean V2: Non-Binary vs. Mean Female 1 0 Non-Binary 0 1 Male -1 -1 In this table, “mean” refers to the grand mean across all levels of gender 5, Table 6 Contrast Variables for Parents’ Race Race V1: Asian vs. Mean V2: Black vs. Mean V3: Latinx vs. Mean V4: Other vs. Mean Asian 1 0 0 0 Black 0 1 0 0 Latinx 0 0 1 0 Other 0 0 0 1 White -1 -1 -1 -1 In this table, “mean” refers to the grand mean across all levels of gender 6 Data Availability The data sets generated during and/or analyzed during the current study are not publicly available but are available from the corresponding author on reasonable request. Declarations Conflicts of Interest There are no known conflicts of interest associated with the research, authorship, or production of this article. 1 Compensation was available to some participants because after we began conducting the study, we were awarded a modest grant to help us recruit a more diverse sample with respect to race and socioeconomic background. 2 We specifically chose to examine children’s reports of stress because we wanted to use multi-source data to overcome some of the challenges inherent in single-source analyses (e.g., Podsakoff et al., 2003; Williams et al., 1989) and to address limitations of self-report measures of stress (e.g., Razavi, 2001; Semmer et al., 2003). We chose to focus on stress because it is a potentially observable outcome that can manifest in daily interpersonal interactions, and research has shown that children (even when parents attempt to suppress their experiences and emotions) are able to sense stress in their parents (e.g., Waters et al., 2020). We believe this contributes to implications of our work (e.g., potential relevance to children’s pandemic experiences and outcomes), while also addressing common method bias concerns inherent in single-source data. 3 Note that, despite the fact that the first leg of the model was the same for both outcome variables, the z-statistics vary slightly across the different analyses due to the bootstrapping procedure. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Naomi M. Fa-Kaji, Elisabeth R. Silver, Mikki R. Hebl, Danielle D. King, and Eden B. King contributed equally. ==== Refs References Adler, A. B. (2013). Resilience in a military occupational health context: Directions for future research. In R. R. Sinclair & T. W. Britt (Eds.), Building psychological resilience in military personnel: Theory and practice (pp. 223–235). American Psychological Association. 10.1037/14190-010. Angel RJ Frisco M Angel JL Chiriboga DA Financial strain and health among elderly Mexican-origin individuals Journal of Health and Social Behavior 2003 44 4 536 551 10.2307/1519798 15038148 American Psychological Association. (2022). Stress in America: Money, inflation, war pile on to nation stuck in COVID-19 survival mode. Retrieved September 17, 2022, from https://www.apa.org/news/press/releases/stress/2022/march-2022-survival-mode#ukraine Avey JB Luthans F Jensen SM Psychological capital: A positive resource for combating employee stress and turnover Human Resource Management 2009 48 5 677 693 10.1002/hrm.20294 Avey JB Reichard RJ Luthans F Mhatre KH Meta-analysis of the impact of positive psychological capital on employee attitudes, behaviors, and performance Human Resource Development Quarterly 2011 22 2 127 152 10.1002/hrdq.20070 Bardoel EA Pettit TM De Cieri H McMillan L Employee : An emerging challenge for HRM Asia Pacific Journal of Human Resources 2014 52 3 279 297 10.1111/1744-7941.12033 Biernacki P Waldorf D Snowball sampling—problems and techniques of chain referral sampling Sociological Methods & Research 1981 10 141 163 10.1177/004912418101000205 Black JK Balanos GM Whittaker AC Resilience, work engagement and stress reactivity in a middle-aged manual worker population International Journal of Psychophysiology 2017 116 9 15 10.1016/j.ijpsycho.2017.02.013 28238816 Block JH Block J Collins WA The role of ego-control and ego-resiliency in the organization of behavior Development of cognition, affect and social relations: The Minnesota symposia on child psychology 1980 Erlbaum 39 101 Bonanno GA Mancini AD The human capacity to thrive in the face of potential trauma Pediatrics 2008 121 2 369 375 10.1542/peds.2007-1648 18245429 Bonanno GA Loss, trauma, and human resilience: Have we underestimated the human capacity to thrive after extremely aversive events? American Psychologist 2004 59 1 20 28 10.1037/0003-066X.59.1.20 14736317 Borkovec TD Robinson E Pruzinsky T DePree JA Preliminary exploration of worry: Some characteristics and processes Behaviour Research and Therapy 1983 21 1 9 16 10.1016/0005-7967(83)90121-3 6830571 Boudrias JS Desrumaux P Gaudreau P Nelson K Brunet L Savoie A Modeling the experience of psychological health at work: The role of personal resources, social-organizational resources, and job demands International Journal of Stress Management 2011 18 4 372 395 10.1037/a0025353 Brief AP Atieh JM Studying job stress: Are we making mountains out of molehills? Journal of Organizational Behavior 1987 8 2 115 126 10.1002/job.4030080203 Britt TW Shen W Sinclair RR Grossman MR Klieger DM How much do we really know about employee resilience? Industrial and Organizational Psychology: Perspectives on Science and Practice 2016 9 2 378 404 10.1017/iop.2015.107 Brosschot JF Gerin W Thayer JF The perseverative cognition hypothesis: A review of worry, prolonged stress-related physiological activation, and health Journal of Psychosomatic Research 2006 60 2 113 124 10.1016/j.jpsychores.2005.06.074 16439263 Caniëls MC Baaten SM How a learning-oriented organizational climate is linked to different proactive behaviors: The role of employee resilience Social Indicators Research 2019 143 2 561 577 10.1007/s11205-018-1996-y Chen S Westman M Hobfoll SE The commerce and crossover of resources: Resource conservation in the service of Stress and Health: Journal of the International Society for the Investigation of Stress 2015 31 2 95 105 10.1002/smi.2574 25873421 Eschleman KJ Bowling NA Alarcon GM A meta-analytic examination of hardiness International Journal of Stress Management 2010 17 4 277 307 10.1037/a0020476 Fisher DM Ragsdale JM Fisher EC The importance of definitional and temporal issues in the study of resilience Applied Psychology 2018 68 4 583 620 10.1111/apps.12162 Folkman S Tedlie J Coping: Pitfalls and promises Annual Review of Psychology 2004 55 745 10.1146/annurev.psych.55.090902.141456 14744233 Folkman S Lazarus RS Dunkel-Schetter C DeLongis A Gruen RJ Dynamics of a stressful encounter: Cognitive appraisal, coping, and encounter outcomes Journal of Personality and Social Psychology 1986 50 5 992 1003 10.1037/0022-3514.50.5.992 3712234 García-Izquierdo, de Pedro, M. M., Ríos-Risquez, I., & Soler Sánchez, I. (2018). Resilience as a moderator of psychological health in situations of chronic stress (burnout) in a sample of hospital nurses. Journal of Nursing Scholarship, 50(2), 228-236.10.1111/jnu.12367. Grant AM Ashford SJ The dynamics of proactivity at work Research in Organizational Behavior 2008 28 3 34 10.1016/j.riob.2008.04.002 Hacker, J. S., Rehm, P., & Schlesinger, M. (2013). The insecure American: Economic experiences, financial worries, and policy attitudes. Perspectives on Politics, 23-49.10.1017/S1537592712003647. Halbesleben JR Neveu JP Paustian-Underdahl SC Westman M Getting to the “COR” understanding the role of resources in conservation of resources theory Journal of Management 2014 40 5 1334 1364 10.1177/01492063145271 Hao S Hong W Xu H Zhou L Xie Z Relationship between resilience, stress, and burnout among civil servants in Beijing, China: Mediating and moderating effect analysis Personality and Individual Differences 2015 83 65 71 10.1016/j.paid.2015.03.048 Harker R Pidgeon AM Klaassen F King S Exploring resilience and mindfulness as preventative factors for psychological distress burnout and secondary traumatic stress among human service professionals Work 2016 54 3 631 637 10.3233/WOR-162311 27286075 Hayes AF Beyond Baron and Kenny: Statistical mediation analysis in the new millennium Communication Monographs 2009 76 4 408 420 10.1080/03637750903310360 Hobfoll SE Conservation of resources: A new attempt at conceptualizing stress American Psychologist 1989 44 3 513 524 10.1037/0003-066X.44.3.513 2648906 Hobfoll, S. E. (2011). Conservation of resources theory: Its implication for stress, health, and . In S. Folkman (Ed.), Oxford library of psychology. The Oxford handbook of stress, health, and coping (p. 127–147). Oxford University Press. Hobfoll SE Halbesleben J Neveu JP Westman M Conservation of resources in the organizational context: The reality of resources and their consequences Annual Review of Organizational Psychology and Organizational Behavior 2018 5 103 128 10.1146/annurev-orgpsych-032117-104640 Hobfoll, S. E., & Shirom, A. (1993). Stress and burnout in the workplace: Conservation of resources. Handbook of Organizational Behavior, 1, 41–60. M. Dekker. Hobfoll SE Stevens NR Zalta AK Expanding the science of resilience: Conserving resources in the aid of adaptation Psychological Inquiry 2015 26 2 174 180 10.1080/1047840X.2015.1002377 27069335 Johnson J Wood AM Gooding P Taylor PJ Tarrier N Resilience to suicidality: The buffering hypothesis Clinical Psychology Review 2011 31 4 563 591 10.1016/j.cpr.2010.12.007 21276646 Karabinski T Haun VC Nübold A Wendsche J Wegge J Interventions for improving psychological detachment from work: A meta-analysis Journal of Occupational Health Psychology 2021 26 3 224 10.1037/ocp0000280 34096763 King DD Newman A Luthans F Not if, but when we need resilience in the workplace Journal of Organizational Behavior 2015 37 5 782 786 10.1002/job.2063 King, D. D., DeShon, R., Phetmisy, C., & Burrows, D. (2022). What is resilience? Offering construct clarity to address “quicksand” and “shadow side” resilience concerns. In P. Perrewé, P. Harms, & C. Chang (Eds.), Research in occupational stress and well being (Vol. 20, pp. 25–50). Emerald Publishing Limited. King, D. D., & Burrows, D. (2021). Resilience in the goal hierarchy: Strategy change as a form of perseverance. In G. Todt, J. Backmann, & M. Weiss (Eds.), Work life after failure?: How employees bounce back, learn, and recover from work-related setbacks (pp. 99–108). Emerald Publishing Limited. 10.1108/978-1-83867-519-620211007. King, D. D. & McSpedon, M. R. (2022). What leaders get wrong about resilience. Harvard Business Review. Retrieved October 14, 2022, from https://hbr.org/2022/06/what-leaders-get-wrong-about-resilience. Kinman G Grant L Exploring stress resilience in trainee social workers: The role of emotional and social competencies The British Journal of Social Work 2011 41 2 261 275 10.1093/bjsw/bcq088 Kobasa SC Stressful life events, personality, and health: An inquiry into hardiness Journal of Personality and Social Psychology 1979 37 1 11 10.1037/0022-3514.37.1.1 458548 Kobasa SC Maddi SR Kahn S Hardiness and health: A prospective study Journal of Personality and Social Psychology 1982 42 168 177 10.1037/0022-3514.42.1.168 7057354 Kossek EE Perrigino MB Resilience: A review using a grounded integrated occupational approach The Academy of Management Annals 2016 10 1 729 797 10.1080/19416520.2016.1159878 Lanz JJ Bruk-Lee V Resilience as a moderator of the indirect effects of conflict and workload on job outcomes among nurses Journal of Advanced Nursing 2017 73 2973 2986 10.1111/jan.13383 28714151 Lawrence ER Halbesleben JRB Paustian-Underdahl SC The influence of workplace injuries on work-family conflict: Job and financial insecurity as mechanisms Journal of Occupational Health Psychology 2013 18 371 383 10.1037/a00339 24099157 Little LM Major VS Hinojosa AS Nelson DL Professional image maintenance: How women navigate pregnancy in the workplace Academy of Management Journal 2015 58 1 8 37 10.5465/amj.2013.0599 Luthans F Avolio BJ Avey JB Norman SM Positive psychological capital: Measurement and relationship with performance and satisfaction Personnel Psychology 2007 60 3 541 572 10.1111/j.1744-6570.2007.00083.x Luthans F Avolio BJ Walumbwa FO Li W The psychological capital of Chinese workers: Exploring the relationship with performance Management and Organization Review 2005 1 2 249 271 10.1111/j.1740-8784.2005.00011.x Luthans, F., Youssef, C. M., & Avolio, B. J. (2007b). Psychological capital: Developing the human competitive edge. Oxford University Press. Maslach C Schaufeli WB Leiter MP Job burnout Annual Review of Psychology 2001 52 1 397 422 10.1146/annurev.psych.52.1.397 11148311 Mathews A Why worry? The cognitive function of anxiety Behaviour Research and Therapy 1990 28 6 455 468 10.1016/0005-7967(90)90132-3 2076083 Meuris J Leana CR The high cost of low wages: Economic scarcity effects in organizations Research in Organizational Behavior 2015 35 143 158 10.1016/J.RIOB.2015.07.001 Meyer TJ Miller ML Metzger RL Borkovec TD Development and validation of the Penn State Worry Questionnaire Behaviour Research and Therapy 1990 28 6 487 495 10.1016/0005-7967(90)90135-6 2076086 Ohly S Fritz C Work characteristics, challenge appraisal, creativity, and proactive behavior: A multi-level study Journal of Organizational Behavior 2010 31 4 543 565 10.1002/job.633 Ong AD Bergeman CS Bisconti TL Wallace KA Psychological, positive emotions, and successful adaptation to stress in later life Journal of Personality and Social Psychology 2006 91 4 730 749 10.1037/0022-3514.91.4.730 17014296 Parker SK Williams HM Turner N Modeling the antecedents of proactive behavior at work Journal of Applied Psychology 2006 91 3 636 652 10.1037/0021-9010.91.3.63691 16737360 Parker SL Jimmieson NL Walsh AJ Loakes JL Trait fosters adaptive coping when control opportunities are high: Implications for the motivating potential of active work Journal of Business and Psychology 2015 30 3 583 604 10.1007/s10869-014-9383-4 Pettita, L., Probst, T.M., Ghezzi, V., & Barbanelli, C. (2020). Economic stress, emotional contagion and safety outcomes: A cross-country study. Work, 66(2), 421–435. 10.3233/WOR-203182. Phetmisy CN King DD The ubiquitous effects of financial stress during pandemics and beyond: Opportunities for industrial and organizational psychology Industrial and Organizational Psychology: Perspectives on Science and Practice 2021 14 1–2 90 93 10.1017/iop.2021.16 Podsakoff P MacKenzie S Lee J Podsakoff N Common method biases in behavioral research: A critical review of the literature and recommended remedies Journal of Applied Psychology 2003 88 879 890 10.1037/0021-9010.88.5.879 14516251 Preacher KJ Rucker DD Hayes AF Addressing moderated mediation hypotheses: Theory, methods, and prescriptions Multivariate Behavioral Research 2007 42 1 185 227 10.1080/00273170701341316 26821081 Razavi, T. (2001). Self-report measures: An overview of concerns and limitations of questionnaire use in occupational stress research. Discussion Paper: University of Southampton. http://eprints.epwp.eprints-hosting.org/id/eprint/93/. Richardson T Elliott P Roberts R Jansen M A longitudinal study of financial difficulties and mental health in a national sample of British undergraduate students Community Mental Health Journal 2017 53 3 344 352 10.1007/s10597-016-0052-0 27473685 Robertson I Cooper CL Resilience [Editorial] Stress and Health: Journal of the International Society for the Investigation of Stress 2013 29 3 175 176 10.1002/smi.2512 23913839 Rosseel, Y. (2012). Lavaan: An R package for structural equation modeling and more. Version 0.5–12 (BETA). Journal of Statistical Software, 48(2), 1–36. 10.18637/jss.v048.i02 Rucker DD Preacher KJ Tormala ZL Petty RE Mediation analysis in social psychology Social and Personality Psychology Compass 2011 5 6 359 371 10.1111/j.1751-9004.2011.00355.x Rutter PA Freedenthal S Osman A Assessing protection from suicidal risk: Psychometric properties of the suicide resilience inventory Death Studies 2008 32 142 153 10.1080/07481180701801295 18693384 Segerstrom SC Tsao JCI Alden LE Craske MG Worry and rumination: Repetitive thought as a concomitant and predictor of negative mood Cognitive Therapy and Research 2000 24 6 671 688 10.1023/A:1005587311498 Semmer NK Grebner S Elfering A Perrewe PL Ganster DC Beyond self-report: Using observational, physiological, and situation-based measures in research on occupational stress Emotional and physiological processes and positive intervention strategies 2003 Emerald Group Publishing Limited 205 263 Shin Y Kim M-J Antecedents and mediating mechanisms of proactive behavior: Application of the theory of planned behavior Asia Pacific Journal of Management 2015 32 1 289 310 10.1007/s10490-014-9393-9 Shoss MK Jiang L Probst TM Bending without breaking: A two-study examination of employee resilience in the face of job insecurity Journal of Occupational Health Psychology 2018 23 1 112 126 10.1037/ocp0000060 27786505 Sinclair RR Cheung JH Money matters: Recommendations for financial stress research in occupational health psychology Stress and Health 2016 32 3 181 193 10.1002/smi.2688 27400815 Sonnentag S Cheng BH Parker SL Recovery from work: Advancing the field toward the future Annual Review of Organizational Psychology and Organizational Behavior 2022 9 33 60 10.1146/annurev-orgpsych-012420-091355 Spector PE Do not cross me: Optimizing the use of cross-sectional designs Journal of Business and Psychology 2019 34 2 125 137 10.1007/s10869-018-09613-8 Terry DJ Hynes GJ Adjustment to a low-control situation: Reexamining the role of coping responses Journal of Personality and Social Psychology 1998 74 4 1078 1092 10.1037/0022-3514.74.4.1078 Tett RP Burnett DD A personality trait-based interactionist model of job performance Journal of Applied Psychology 2003 88 3 500 517 10.1037/0021-9010.88.3.500 12814298 Tett RP Guterman HA Situation trait relevance, trait expression, and cross-situational consistency: Testing a principle of trait activation Journal of Research in Personality 2000 34 4 397 423 10.1006/jrpe.2000.2292 Tugade MM Fredrickson BL Resilient individuals use positive emotions to bounce back from negative emotional experiences Journal of Personality and Social Psychology 2004 86 2 320 333 10.1037/0022-3514.86.2.320 14769087 Tugade MM Fredrickson BL Feldman Barrett L Psychological resilience and positive emotional granularity: Examining the benefits of positive emotions on coping and health Journal of Personality 2004 72 6 1161 1190 10.1111/j.1467-6494.2004.00294.x 15509280 Utsey SO Giesbrecht N Hook J Stanard PM Cultural, sociofamilial, and psychological resources that inhibit psychological distress in African Americans exposed to stressful life events and race-related stress Journal of Counseling Psychology 2008 55 1 49 62 10.1037/0022-0167.55.1.49 Wanberg CR Csillag B Douglass RP Zhou L Pollard MS Socioeconomic status and well-being during COVID-19: A resource-based examination Journal of Applied Psychology 2020 105 12 1382 1396 10.1037/apl0000831 33090858 Waters SF Karnilowicz HR West TV Mendes WB Keep it to yourself? Parent emotion suppression influences physiological linkage and interaction behavior Journal of Family Psychology 2020 34 7 784 793 10.1037/fam0000664 32324017 Weber A Jaekel-Reinhard A Burnout syndrome: A disease of modern societies? Occupational Medicine (oxford, England) 2000 50 7 512 517 10.1093/occmed/50.7.512 11198677 West, C. P., Dyrbye, L. N., Sinsky, C., Trockel, M., Tutty, M., Nedelec, L., ... & Shanafelt, T. D. (2020). Resilience and burnout among physicians and the general US working population. JAMA Network Open, 3(7), e209385-e209385.10.1001/jamanetworkopen.2020.9385 Westman M Hobfoll SE Chen S Davidson OB Laski S Perrewé PL Ganster DC Organizational stress through the lens of Conservation of Resources (COR) theory Exploring interpersonal dynamics 2005 Elsevier Science/JAI Press 167 220 Williams LJ Cote JA Buckley MR Lack of method variance in self-reported affect and perceptions at work: Reality or artifact? Journal of Applied Psychology 1989 74 3 462 10.1037/0021-9010.74.3.462 Wilson JM Lee J Fitzgerald HN Oosterhoff B Sevi B Shook NJ Job insecurity and financial concern during the COVID-19 pandemic are associated with worse mental health Journal of Occupational and Environmental Medicine 2020 62 9 686 691 10.1097/JOM.0000000000001962 32890205 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 Zautra AJ Johnson LM Davis MC Positive affect as a source of resilience for women in chronic pain Journal of Consulting and Clinical Psychology 2005 73 2 212 220 10.1037/0022-006X.73.2.212 15796628 Zhao X Lynch JG Chen Q Reconsidering Baron and Kenny: Myths and truths about mediation analysis Journal of Consumer Research 2010 37 2 197 206 10.1086/651257	0	PMC9734500	NO-CC CODE	2022-12-14 23:28:29	no	Occup Health Sci. 2022 Dec 3;:1-32	utf-8	Occup Health Sci	2,022	10.1007/s41542-022-00130-y	oa_other
==== Front Br Dent J Br Dent J British Dental Journal 0007-0610 1476-5373 Nature Publishing Group UK London 5266 10.1038/s41415-022-5266-7 General Dental calculus - oral health, forensic studies and archaeology: a review Forshaw Roger [email protected] grid.5379.8 0000000121662407 KNH Centre for Biomedical Egyptology, Faculty of Biology, Medicine and Health, Manchester University, Stopford Building, Oxford Road, Manchester, M13 9PL, UK 9 12 2022 2022 233 11 961967 28 1 2022 9 5 2022 © The Author(s), under exclusive licence to the British Dental Association 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. Dental calculus is recognised as a secondary aetiological factor in periodontal disease, and being a prominent plaque retentive factor, it is routinely removed by the dental team to maintain oral health. Conversely, dental calculus can potentially be useful in forensic studies by supplying data that may be helpful in the identification of human remains and assist in determining the cause of death. During the last few decades, dental calculus has been increasingly recognised as an informative tool to understand ancient diet and health. As an archaeological deposit, it may contain non-dietary debris which permits the exploration of human behaviour and activities. While optical and scanning electron microscopy were the original analytical methods utilised to study microparticles entrapped within the calcified matrix, more recently, molecular approaches, including ancient DNA (aDNA) and protein analyses, have been applied. Oral bacteria, a major component of calculus, is the primary target of these aDNA studies. Such analyses can detect changes in the oral microbiota, including those that have reflected the shift from agriculture to industrialisation, as well as identifying markers for various systemic diseases. Key points Provides an overview of dental calculus from the viewpoint of the dental surgeon, the forensic specialist and the archaeologist. Dental calculus, being among the richest bimolecular source in the archaeological record, is able to provide significant new insights and answers to questions, relevant to both archaeology and anthropology. Considers how dental calculus analyses can be useful in shedding light on ancient diet, disease and lifestyle patterns. issue-copyright-statement© The Author(s), under exclusive licence to the British Dental Association 2022 ==== Body pmcIntroduction Dental calculus can be defined as a complex mineralised plaque biofilm which is sequentially generated and entraps microbial, dietary, host and ancient debris during spontaneous calcification events.1,2 It is composed primarily of calcium phosphate mineral salts deposited between and within remnants of formerly viable microorganisms and is covered by an unmineralised bacterial layer.3 Dental calculus forms throughout an individual's life on the subgingival and/or supragingival tooth surfaces. There is a cohesive bond between crystals in the calculus and the enamel, dentine or cementum apatite crystals at the calculus-tooth interface.4 Calculus is found in all known populations, past and present, but the extent varies widely among individuals and populations. Factors such as oral hygiene routine, frequency of dental care, age, systemic health, diet and ethnicity all affect its formation. Dental calculus has long been recognised as a significant factor in the aetiology of periodontal disease. It is a rich source of human DNA and as such, has the potential to serve as an investigative tool in forensic studies. As calculus is mineralised, it often survives well in archaeological contexts and is useful when studying the dental pathology of our ancestors. It is only comparatively recently that the vast research potential of dental calculus for archaeological study has been appreciated, providing as it does insights into ancient diet, health, disease and evolutionary history. This present study aims to review dental calculus in relation to these three topics of oral health, forensic investigations and archaeological research. Dental significance Epidemiological studies have long demonstrated a strong association between calculus and periodontal disease.1,5,6 Dental calculus does not contribute directly to gingival irritation but it provides a nidus for the continued accumulation of plaque. It retains plaque in close proximity to the gingivae and creates areas where plaque removal is difficult or impossible, all contributing to gingival inflammation and periodontal disease.7,8 Calculus provides an ideal porous vehicle for bacterial plaque retention and growth and is therefore regarded as a secondary aetiological factor in periodontal disease. Once formed, the presence of calculus may compromise oral hygiene procedures and promote the growth of pathogenic plaque. As a prominent plaque retentive factor, the deposits must therefore be removed for adequate periodontal therapy and the maintenance of oral health.3 The removal of plaque and calculus is the basis of periodontal treatment and the dental team expends considerable effort and resources on this aspect of modern-day dental care. There are wider health implications for the build-up of calculus deposits. Individuals with a relatively high disease susceptibility and who have poor oral hygiene and calculus deposits that have resulted in marked destruction of the periodontal tissues are more at risk of systemic disease.9 Periodontal disease may predispose the patient to an increased incidence of bacteraemia which can cause infective endocarditis, a condition that ultimately may be fatal.10 There is a link between chronic inflammation of the periodontal tissues and the risk of the affected cells of the oral epithelium becoming malignant.11 Also, there is a two-way relationship between diabetes and periodontal disease. Diabetes increases the risk of developing periodontal disease and periodontal inflammation adversely affects glycaemic control.12 After removal from the teeth of patients, calculus debris is routinely disposed of, as being a clinical waste product, it is of little use. However, in recent years, dental calculus has been suggested as having a possible role in forensic investigations and accepted as a tool in helping to understand ancient biographical and dietary information. Forensic studies Analysing genetic material is the most accurate method for establishing human identity in forensic examinations. Where human remains are degraded or fragmented, teeth and bones may be the only sources of DNA available for identification. The unique composition of teeth and their locations in the mandible and maxilla provide additional protection to DNA compared to bones, making them a preferred source of DNA in many cases.13 The dental pulp is the usual source of DNA but where there is poor preservation of the human remains or where permission for destructive sampling is denied, then accessing this source can be problematic. However, previous studies have reported that DNA is present in dental calculus (see the human DNA analysis section) and so the biofilm can provide an alternative source. A pilot study has been able to demonstrate the potential of dental calculus to serve as an investigative tool in forensic studies.14 Charlier and his co-workers15 investigated archaeological samples of calculus of individuals from a number of historic sites, utilising microscopy coupled with elemental surface analysis. Their results provided information about food, environmental habits and work-related exposure to pollutants. Such data can be helpful in determining individual habits and pathologies, potentially useful in identification and determining cause of death. Recently, a new, sensitive, ultra-high-performance, liquid chromatography-tandem mass spectrometry technique has been developed, a method that is capable of demonstrating a large variety of pharmaceutical and psychoactive drugs entrapped and preserved within calculus. Such inclusions would have been derived from direct contact with sources in the oral cavity, from inhalation of smoke or vapour and from the release of serum into the saliva and gingival crevicular fluids. Many drugs, including heroin, cocaine and opium, are able to be identified by this method - drugs that may not have been originally detected in the blood at the time of autopsy but which may have been a factor in the cause of death. This technique is not only applicable to modern forensic investigations but can also provide an overview of drugs and stimulants used in ancient time periods.16 Drug entrapment in calculus Archaeological evidence has demonstrated that ancient peoples used plant-based remedies for the treatment of medical conditions. Originally, most of the information about these preparations was derived from the material culture, osteological evidence of pathologies and surgical interventions, archaeobotanical analyses and textual sources. These data are now supported by the detection, in calculus, of plants now known for their medicinal properties.17,18,19 The wide and varied range of plants that have been identified in the dental calculus of Neanderthals would suggest that they had a sophisticated understanding of their environment. Certain bitter-tasting plants, with no nutritional value, containing compounds such as yarrow and camomile, suggest that they had the ability to select and use particular flora for medicinal purposes.17 Dental calculus analyses of an early medieval Italian population of Colonna in central Italy, dating back to the eighth to the tenth century AD, also demonstrated similar results. Not only was a detailed qualitative reconstruction of the food habits of this community able to be obtained, but the identification of specific chemical indicators suggested the pharmaceutical application of a number of medicinal plants. Among these were: Digitalis sp., used in treatment for heart conditions; Hyssopus officinalis, recognised for its antiseptic and expectorant properties; Artemisia sp., having digestive, antiseptic, antimalarial and expectorant capabilities; and Ephedra sp., a bronchodilator and vasoconstrictor.19 Again, dietary intake and plant-based treatments were studied in a woman buried in the Late Preceramic site of Huaca El Paraíso, Peru (2100-1500 BC). In this case, an analysis of the calculus was supplemented by examining the sediment enclosed inside her grossly carious mandibular molars (Figures 1 and 2).20Fig. 1 a, b, c) Selected phytoliths and starch grains identified in dental calculus samples from a woman from the Late Preceramic period, Peru. Micrographs taken at x400 magnification. Reprinted from Annals of Anatomy, Vol 240, Allende et al., 'Dental anthropological report: Exploring plant-based treatments through the analysis of dental calculus and sediment of dental caries in a woman from the Late Preceramic period, Peru', 2022, with permission from Elsevier20 Fig. 2 a, b, c, d) Selected phytoliths identified in dental caries in a mandibular molar of a woman from the Late Preceramic period, Peru. Micrographs taken at x400 magnification. Reprinted from Annals of Anatomy, Vol 240, Allende et al., 'Dental anthropological report: Exploring plant-based treatments through the analysis of dental calculus and sediment of dental caries in a woman from the Late Preceramic period, Peru', 2022, with permission from Elsevier20 Heavy metal poisoning Heavy metal exposure has become a serious health concern in recent decades, particularly with the ubiquity of these elements in our daily environment. Various analytical methods, such as transmission electron microscopy, have been used in the assays of the major and minor elements found in supragingival dental calculus.21 Cadmium is a widespread-trace toxic heavy metal with a long biological half-life and is considered to induce a higher risk of cancer in multiple organs of the human body. A recent study has confirmed the relationship between cadmium levels in dental calculus, due to betel-quid chewing and smoking, with the subsequent risk of oral cancer.22 Tobacco smoke is a complex and reactive mixture of numerous chemicals that include a number of heavy metals. Again, as calculus is a biological material that can be collected non-invasively, it can be useful in monitoring oral heavy metal exposure.23 Analysis of the calculus from the mandibular teeth of Agnès Sorel, mistress of the French King Charles VII, who was buried in Loches, France in 1450 AD, revealed a very high level of mercury, sufficient to have caused death by acute mercury poisoning. Evidence suggests that Agnes suffered from roundworms for which mercury was a common treatment at that time. Mercury was also used to treat women in labour for difficult deliveries and Agnes was known to have borne four daughters by the king. However, the evidence of a massive dose of this heavy metal hints at foul play and while it is not known if her death was deliberate or accidental, it has been suggested that her undue influence over the king would have created enemies for her at court.15,24 COVID-19 The SARS-CoV-2 virus is the causal agent for COVID-19 and the high yield of the virus in salivary secretion is a common finding in the infection.25 As calculus contains biomolecules, such as nucleic acids, it has been hypothesised that following infection, SARS-CoV-2 ribonucleic acid could be preserved in calculus. A recent study in previously infected individuals has been able to confirm this proposition.23 Samples of dental calculus were analysed by performing reverse transcription polymerase chain reaction assays following nucleic acid extraction and amplification. As well as in infected cases, this technique may also be able to detect traces in asymptomatic or mild asymptomatic patients. However, further studies are needed to define the method's reliability, cost-effectiveness and suitability for large scale epidemiological studies or post-mortem analyses.26 Human DNA analysis The bulk of DNA in dental calculus is microbial and originates from the oral microbiota (community of microorganisms within the oral cavity); however, a small, consistent and genetically rich proportion is endogenous human DNA. The mechanisms by which human DNA is incorporated into dental calculus are not fully understood. They are presumed to include passive adsorption of human DNA from oral fluids and shed mucosal cells, with more active incorporation through host inflammatory responses, such as an immune response mediated by neutrophils.2,27,28,29 Today, DNA technology has many applications, including studies into ancient DNA (aDNA), which can provide snapshots of the genetics of ancient populations. aDNA has revolutionised how we study many aspects of the biological past, including population origins and movements, natural selection, evolutionary relationships and identifying the presence of disease pathogens. As dental calculus is less porous and thus less susceptible to degradation by environmental microbes than dentine or bone, it offers an alternative source of ancient human DNA that may persist when other skeletal tissues fail to yield aDNA. Archaeological research into dental calculus The study of dental calculus goes back to the 1980s when calculus deposits were partially decalcified and then viewed under an optical microscope. Supragingival calculus deposits are typically chosen as subgingival calculus is more tightly adherent to the tooth surface, more heavily mineralised and is affected by haemorrhagic components from the gingival crevicular fluid. The deposits need to be carefully sampled, prepared and degraded with dilute hydrochloric following a standard protocol before microscopic examination.30,31,32,33 The technique is able to identify microparticles preserved within the matrix, which includes fragments of cereals, vegetable fibres, phytoliths, pollens, seeds, animal hairs, parasites and even insects that accidentally become entrapped (Figures 3 and 4).34,35 Subsequent work examined the effects of ancient diets on calculus formation36 and utilised calculus to study the diet and palaeoenvironment of Neanderthals and ancient humans.30,37,38Fig. 3 Medieval mandible showing calculus build-up. Site A24, 2003, Vine Street, Leicester, UK. Reproduced with permission from Anita Radini Fig. 4 A small invertebrate entrapped in dental calculus of a wild chimpanzee revealed in a scanning electron microscope image. Reproduced with permission from Robert Power and Heiko Temming Far greater magnification can be achieved by the use of scanning electron microscopy (SEM), a technique often used as complementary with optical microscopy (OM). Calculus samples are mounted on stubs and sputter coated with gold to prevent surface charging by the electron beam.31 Elemental analysis of particles can be studied by using SEM with energy-dispersive x-ray spectroscopy.39 With the application of newer scientific methods in archaeology, bacterial DNA and host mitochondrial DNA preserved within the calculus are now successfully able to be extracted and analysed.40,41 These continuing technological developments have radically improved the ability not only to extract DNA, but also proteins and metabolites. The research is reshaping our understanding of past diet, behaviour, ancestry, occupational activities and the health of past individuals and populations. In bimolecular studies, aDNA and proteins are extracted from decalcified dental calculus, often utilising a unified protocol, which decreases the material needed for analyses while maximising the information yield, as described for example by Mackie et al.42 and Fagernäs et al.43 Dietary information The study of plant microremains (phytoliths and starch grains) retained in dental calculus is a technique that is being increasingly utilised to determine ancient dietary information, as well as evidence about past environments and human cultures (Fig. 5). The analysis has the potential for revealing the genera and species of dietary plants, patterns of cultivation and methods of food preparation.37,44,45,46,47,48 Additionally, it is a useful method for studying population-level dietary trends.49Fig. 5 Images of starch granules and other microremains entrapped within dental calculus utilising light microscopy. a, b, c, d, e, f, g, h, i) Archaeological samples. j, k, l) Modern starch references. The scale bar indicates 20 μm. Reprinted with permission from Gismondi et al., 'Back to the roots: dental calculus analysis of the first documented case of coeliac disease', Archaeological and Anthropological Sciences, 2020, Springer Nature74 Today, starch-based foods constitute 50-70% of the energy intake of most humans and they also had this important role in the pre-agricultural human diet.44 The use of starches trapped inside dental calculus provides a direct link to food consumption and as the particles are confined inside a calcified matrix, they are less likely to alter over time. Calculus analysed from the Shanidar (Iraq) and Spy (Belgium) Neanderthals indicates that they consumed a considerable amount of plant materials. These included date palms, legumes and grass seeds, suggesting that their diet was not primarily based on meat as had previously been proposed. The range of local plant foods that they consumed was diverse, some of which they cooked, implying an overall sophistication in Neanderthal dietary regimes.38 A study of teeth from a middle Holocene (circa 5500-4500 BC) archaeological site in Syria was able to determine that the individuals were consuming a variety of plant foods. Domesticated cereals, such as wheat and barley, which the archaeological record had previously hypothesised as supplying the major sources of starches, was found to make up a surprisingly small portion of the diet.38,50 Researchers investigating a forager of the Central Mediterranean region, dating to the end of the eighth millennium BC (the Mesolithic Period), utilised a combination of dental calculus analysis and stable isotope techniques to reveal inclusions of fish scales, fish and bird flesh, starch granules and other plant and animal microdebris. The results indicated that marine resources, together with a variety of plant foods, were regularly consumed by the individual, whereas previous stable isotope data alone had indicated that it was mainly terrestrial-based resources that were the main contribution to Mesolithic diets.51 Similarly, data obtained from individuals of the older Magdalenian culture (17000-12000 BP) in Northern Iberia, showed that plant and plant-like foods were also parts of their diet. The results indicated a mixed subsistence economy, rather than the previously held supposition of a diet largely based on protein.52 Microfossil analysis has more recently been combined with proteomic (large scale study of proteins) and aDNA-based approaches to provide more accurate interpretations of entrapped dietary information within calculus. Microfossil analysis utilising OM and SEM allows morphological matches of entrapped calculus particles, which, when associated with bimolecular studies, provides complementary information that is able to characterise many dietary components at a higher taxonomic resolution.17,28 Migrations and dietary shifts Previous studies have indicated that aDNA derived from human skeletal remains can be used to determine historical human migrations around the globe.53,54,55 More recently, it has been established that an analysis of aDNA from microorganisms preserved within calculus also has the potential to reconstruct human migration and interaction networks. Tracking human migrations by this method involves the identification of genetic mutations within bacterial species. The composition of the human oral microbiota is relatively distinct to each culture and geographic region. This characteristic highlights the potential ability of microbiota DNA within dental calculus to provide an ancient genetic signal of cultural affinity, therefore advancing our understanding of human prehistory.56,57,58 These types of analyses can also identify changes in human oral microbiota communities that correlates with major dietary alterations over time. The transition from hunter-gatherer to farming shifted the oral microbial community to a disease-associated configuration, with a marked increase in the prevalence of dental calculus and oral pathology.59 The composition of oral microbiota remained constant between Neolithic and medieval times, after which cariogenic bacteria became dominant during the Industrial Revolution, with the advent of industrially processed flour and sugar.57 Prehistory Reconstructing detailed aspects of the lives of our distant ancestors has always proved challenging due to the restricted nature of the surviving evidence. Poor preservation of plant remains and the lack of systematic recovery techniques for organic residues have prevented an understanding of dietary habits in pre-agrarian societies. However, the recovery of botanical microfossils and other debris from archaeological dental calculus has the potential to provide not only plant-related evidence, but also to contribute to the reconstruction of non-dietary related practices in everyday life. Such information is not always available in the archaeological record. Dental calculus can be retrieved from skeletal material dating to most archaeological periods. Despite their extreme age, in some cases, the results from the analysis of the material can shed light on hitherto unknown aspects of ancient life. Excavations that were undertaken in Qesem cave in Israel has produced information relating to Lower Palaeolithic hominids who lived some 420000-200000 BP. The population were found to have consumed a very broad range of plant species, suggesting an ecological knowledge sufficiently developed to permit the selection of physiologically essential plant foods. In addition, the presence of micro-charcoal fragments of up to 70 μm indicates a smoky atmosphere inside the cave and perhaps points to the ingestion of cooked or smoked food.32 Calculus samples from the multi-period (pre-Mesolithic to Meriotic) Central Sudanese site of Al Khiday has also produced evidence for cooking and smoke inhalation. Among the plant evidence was found samples of Cyperus rotundus (purple nut sedge), a perennial found to have been consumed during all periods. The non-nutritional qualities of the plant suggest that its use may have been directed at its aromatic or medicinal properties. C. rotundus is known to inhibit Streptococcus mutans, which is associated with the initiation of dental caries.60,61 Chewing of C. rotundus tubers may have contributed to the unexpectedly low prevalence of caries in the Meriotic samples from Al Khiday.62 Calculus obtained from individuals buried at Nemrik 9, a Neolithic site in Northern Iraq, dating to 9100-8600 BP, was not only able to provide insights into diet at this early farming community, but was also able to provide some evidence that plants may have being used as tools. The shape and pattern of entrapped wooden splinters and phytoliths indicated that either wood and reeds may have been processed using teeth as an implement, or that these materials were used as tooth picks.63 Stable isotope analysis Stable isotope analysis of biomaterials, such as bone, teeth, fingernails and hair, has become a recognised technique for palaeodietary analysis in bioarchaeology.64,65 The technique is destructive, as the sample material is destroyed during the analytical process and so curatorial concerns may prohibit such analyses. A method in which calculus can be analysed for stable carbon and nitrogen concentrations has been described by Scott and Poulson.66 The advantage of using calculus is that it is not an inherent part of the skeleton but a secondary material and thus may overcome curatorial concerns regarding preservation of the specimen.67 However, more recent research indicates that as the formation processes and composition of dental calculus can be highly variable among and within individuals, results from carbon and nitrogen isotopic analysis compared to those from bone can be inconsistent. Consequently, caution has to be exercised in using these results in dietary interpretations and subsequent conclusions may be invalid.68,69 Trade and craft activities During the process of biomineral maturation of dental plaque, not only are dietary microfossils entrapped within the calcifying matrix, but also a wide variety of airborne and waterborne debris. Among them can be waste products associated with craft and trade activities, which includes materials such as ground stone grit and plant and animal fibres.41 Lapis lazuli crystals have been found embedded within the calculus of a 9-14th century AD woman buried in a church-monastery complex at Dalheim, Germany. The blue particles were dispersed across many dental calculus fragments from different teeth, suggesting the particles entered the calculus in separate episodes rather than as a single localised event. Lapis lazuli is mined from a single region in Afghanistan and was a long-distance luxury trade good in the premodern era. Members of religious orders were the prime producers of books in the Middle Ages and lapis lazuli was used by the scribes and painters to illuminate high-quality texts.39 Although it is not certain how the mineral came to be embedded in the calculus, it is known that craftspeople occasionally licked their brushes to make a fine point when embellishing manuscripts.70 Elemental analysis of crystals found in the dental calculus of an individual from an Etruscan-Celtic necropolis at Monterenzio Vecchia in Northern Italy revealed a very high level of manganese. This element is frequently found in ancient ceramic technology, such as pottery and paintings, and in this instance, could reflect work exposure to the pollutant.15 Evidence of preserved cellulose fibres, consistent with the characteristics of cotton, were found embedded within the matrix of dental calculus from four Late Woodland individuals (6-12th century AD) from the Danbury site, Ohio. The particular cotton fibres, Gossypium spp., are not indigenous to that region and would have to have been introduced from a south-western source. Such an interaction between Northern Ohio and southern coastal regions is supported by archaeological evidence of both cotton and marine shell exchange. How the cotton fibres became entrapped in the calculus is unclear but it is possible that the teeth were used as tools to separate the fibres when engaged in spinning cotton. Archaeological information of this nature is able to help shed new light on the craft industry and long-distance trade in prehistoric North America.71 Ancient diseases Not only is ancient calculus a rich reservoir of oral microbiota, food particles and other debris, but as mentioned, acquired pathogens are also trapped within its matrix. From a study of these ancient pathogens, it is possible to identify the origins, causes and evolution of specific infectious diseases. These opportunistic pathogens include those that are involved in periodontal, respiratory, cardiovascular and various other systemic diseases.28 One such condition is leprosy and evidence of the Mycobacterium leprae genome, the causative agent of leprosy, has been recovered via shotgun sequencing from the calculus of a sixteenth-century individual from Trondheim, Norway. The presence of M. leprae DNA and peptides in the calculus suggest an oral manifestation of the disease, considered perhaps to be the mucosa or soft palate. Calculus represents an alternative sample source to bones and teeth, especially in the absence of definite osteological markers and where human remains are poorly preserved or too valuable to warrant destructive bone sampling.72 Next-generation sequencing of calculus samples from an older woman at a prehistoric site in San Francisco Bay (CA-SCL-919) has revealed high levels of Neisseria meningitidis, one of the most common causes of bacterial meningitis. This, combined with the presence of incipient endocranial lesions and pronounced meningeal grooves, suggests an ancient case of meningococcal disease.73 Investigations into the calculus of a Roman woman, housed at the National Archaeological Museum of Cosa, Tuscany, has revealed what is considered to be the first historical evidence of coeliac disease. Molecular analyses demonstrated the HLA-DQ 2.5 haplotype, typically associated with a high predisposition to coeliac disease, while the results of the stable isotope analysis were suggestive of chronic malnutrition. Optical microscopic analysis revealed a gluten-rich diet and the skeletal remains displayed enamel hypoplasia and cribra orbitalia. In addition, there were specific molecular markers supporting the use of several medicinal herbal products, possibly aimed at treating this condition. One of these was for metabolites typical of exotic rhizomes, which are recognised for their anti-inflammatory and immunomodulatory properties. These were not native to Italy, instead perhaps coming from Eastern Asia, which would sustain historical information about the existence of trade routes at that time.74 Conclusions The build-up of dental calculus on teeth is an important oral health issue. In forensic studies, calculus has the potential to serve as a useful investigative resource. But, it is in the fields of anthropology and particularly archaeology that the analysis of ancient calculus has been revealed as having significant applications. Techniques have progressed considerably since its ability to inform on past human lives was first recognised. The quality and value of information that can be obtained from the identification of embedded microfossil remains and other debris is improving and expanding all the time. Dental calculus is among the richest bimolecular source in the archaeological record and in recent years, bimolecular investigation of dental calculus has increasingly been utilised as an important tool in archaeological investigations. It is a hardy, long-term, biomolecular reservoir of ancient disease and dietary information and has important applications in the fields of medicine, archaeology and human evolutionary studies. Dental calculus is providing significant new insights and answers to questions, relevant to both archaeology and anthropology, and further advancements in this field can be expected. Ethics declaration The author declares no conflicts of interest. ==== Refs References 1. White D J. Dental calculus: recent insights into occurrence, formation, prevention, removal and oral health effects of supragingival and subgingival deposits. Eur J Oral Sci 1997; 105: 508-522. 2. Mann A E, Sabin S, Ziesemer K et al. Differential preservation of endogenous human and microbial DNA in dental calculus and dentin. Sci Rep 2018; 8: 9822. 3. Jepsen S, Deschner J, Braun A, Schwarz F, Eberhard J. Calculus removal and the prevention of its formation. Periodontol 2000 2011; 55: 167-188. 4. Rohanizadeh R, Legeros R Z. Ultrastructural study of calculus-enamel and calculus-root interfaces. Arch Oral Biol 2005; 50: 89-96. 5. Akali A, Lang N P. Dental calculus: the calcified biofilm and its role in disease development. Periodontol 2000 2018; 76: 109-115. 6. Dumitrescu A L, Kawamura M. Etiology of periodontal disease: Dental plaque and calculus. In Dumitrescu A L (ed) Etiology and Pathogenesis of Periodontal Disease. pp 1-38. Berlin: Springer, 2010. 7. Balaji V R, Niazi T M, Dhanasekaran M. An unusual presentation of dental calculus. J Indian Soc Periodontol 2019; 23: 484-486. 8. Hinrichs J E, Thumbigere-Math V. The role of dental calculus and other local predisposing factors. In Newman M G, Takei H G, Klokkevold P R, Carranza F A (eds) Newman and Carranza's Clinical Periodontology. 13th ed. pp 190-207. Philadelphia: Elsevier, 2019. 9. Mealey B L, Klokkevold P R. Impact of periodontal infection on systemic health. In Newman M G, Takei H G, Klokkevold P R, Carranza F A (eds) Newman and Carranza's Clinical Periodontology. 13th ed. pp 225-236. Philadelphia: Elsevier, 2019. 10. Söder B, Meurman J H, Söder P-Ö. Dental calculus is associated with death from heart infarction. Biomed Res Int 2014; 2014: 569675. 11. Söder B, Yakob M, Meurman J H, Andersson L C, Söder P-Ö. The association of dental plaque with cancer mortality in Sweden. A longitudinal study. BMJ Open 2012; DOI: 10.1136/bmjopen-2012-001083. 12. Preshaw P M, Alba A L, Herrera D et al. Periodontitis and diabetes: a two-way relationship. Diabetologia 2012; 55: 21-31. 13. Higgins D, Austin J J. Teeth as a source of DNA for forensic identification of human remains: a review. Sci Justice 2013; 53: 433-441. 14. Singh U, Goel S. Estimation and quantification of human DNA in dental calculus: A pilot study. J Forensic Dent Sci 2017; 9: 149-152. 15. Charlier P, Huynh-Charlier I, Munoz O, Billard M, Brun L, de la Grandmaison G L. The microscopic (optical and SEM) examination of dental calculus deposits (DCD). Potential interest in forensic anthropology of a bio-archaeological method. Leg Med (Tokyo) 2010; 12: 163-171. 16. Sørensen L K, Hasselstrøm J B, Larsen L S, Bindslev D A. Entrapment of drugs in dental calculus - Detection validation based on test results from post-mortem investigations. Forensic Sci Int 2021; 319: 110647. 17. Hardy K, Buckley S, Collins M et al. Neanderthal medics? Evidence for food, cooking, and medicinal plants entrapped in dental calculus. Naturwissenschaften 2012; 99: 617-626. 18. Fiorin E, Sáez L, Malgosa A. Ferns as healing plants in medieval Mallorca, Spain? Evidence from human dental calculus. Int J Osteoarchaeol 2018; 29: 82-90. 19. Gismondi A, D'Agostino A, Canuti L et al. Dental calculus reveals diet habits and medicinal plant use in the Early Medieval Italian population of Colonna. J Archaeol Sci Rep 2018; 20: 556-564. 20. Allende M K-G, Samplonius A. Dental anthropological report: Exploring plant-based treatments through the analysis of dental calculus and sediment of dental caries in a woman from the Late Preceramic period, Peru. Ann Anat 2022; 240: 151849. 21. McDougall W A. Analytical transmission electron microscopy of the distribution of elements in human supragingival dental calculus. Arch Oral Biol 1985; 30: 603-608. 22. Zhang B, Tan X, He X, Yang H, Wang Y, Zhang K. Evaluation of cadmium levels in dental calculus of male oral SCC patients with Betel-Quid chewing in Hunan Province of China. Biol Trace Elem Res 2019; 191: 348-353. 23. Yaprak E, Yolcubal I, Sinanoğlu A, Doğrul-Demiray A, Guzeldemir-Akcakanat E, Marakoğlu I. High levels of heavy metal accumulation in dental calculus of smokers: a pilot inductively coupled plasma mass spectrometry study. J Periodontal Res 2017; 52: 83-88. 24. Charlier P. Les dents d'Agnès Sorel. L'Information dentaire 2005; 25: 1512-1513. 25. Lee R A, Herigon J C, Benedetti A, Pollock N R, Denkinger C M. Performance of saliva, oropharyngeal swabs, and nasal swabs for SARS-Cov-2 molecular detection: A systematic review and meta-analysis. J Clin Microbiol 2021; DOI: 10.1128/JCM.02881-20. 26. Berton F, Rupel K, Florian F, Biasotto M, Pallavicini A, Di Lenarda R. Dental calculus - a reservoir for detection of past SARS-CoV-2 infection. Clin Oral Investig 2021; 25: 5113-5114. 27. Fuchs T A, Abed U, Goosmann C et al. Novel cell death programme leads to neutrophil extracellular traps. J Cell Biol 2007; 176: 231-241. 28. Warinner C, Rodrigues J F, Vyas R et al. Pathogens and host immunity in the ancient human oral cavity. Nat Genet 2014; 46: 336-344. 29. Ozga A T, Nieves-Colón M A, Honap T P et al. Successful enrichment and recovery of whole mitochondrial genomes from ancient human dental calculus. Am J Phys Anthropol 2016; 160: 220-228. 30. Henry A G, Piperno D R. Using plant microfossils from dental calculus to recover human diet: a case study from Tell al-Raqā'i, Syria. J Archaeol Sci 2008; 35: 1943-1950. 31. Power R C, Salazar-García D C, Wittig R M, Henry A G. Assessing use and suitability of scanning electron microscopy in the analysis of micro remains in dental calculus. J Archaeol Sci 2014; 49: 160-169. 32. Hardy K, Radini A, Buckley S et al. Dental calculus reveals potential respiratory irritants and ingestion of essential plant-based nutrients at Lower Palaeolithic Qesem Cave Israel. Quatern Int 2016; 398: 126-135. 33. D'Agostino A, Di Marco G, Rubini M et al. Environmental implications and evidence of natural products from dental calculi of a Neolithic-Chalcolithic community (central Italy). Sci Rep 2021; 11: 10665. 34. Dobney K, Brothwell D. Dental calculus: its relevance to ancient diet and oral ecology. In Cruwys G, Foley R (eds) Teeth and Anthropology. pp 55-81. Oxford: British Archaeological Reports International Series, 1986. 35. Dobney K, Brothwell D. A scanning electron microscope study of archaeological dental calculus. In Olsen S L (ed) Scanning Electron Microscopy in Archaeology. pp 372-385. Oxford: British Archaeological Reports International Series, 1988. 36. Lieverse A R. Diet and the aetiology of dental calculus. Int J Osteoarchaeol 1999; 9: 219-232. 37. Wesolowski V, de Souza S M F M, Reinhard K J, Ceccantini G. Evaluating microfossil content of dental calculus from Brazilian sambaquis. J Archaeol Sci 2010; 37: 1326-1338. 38. Henry A G, Brooks A S, Piperno D R. Microfossils in calculus demonstrate consumption of plants and cooked foods in Neanderthal diets (Shanidar III, Iraq; Spy I and II, Belgium). Proct Natl Acad Sci U S A 2011; 108: 486-491. 39. Radini A, Tromp M, Beach A et al. Medieval women's early involvement in manuscript production suggested by lapis lazuli identification in dental calculus. Sci Adv 2019; DOI: 10.1126/sciadv.aau7126. 40. De la Fuente C P, Flores S V, Moraga M L. Human bacterial DNA from dental calculus: a new source of genetic material. Am J Phys Anthropol 2012; 147: 127. 41. Warinner C, Speller C, Collins M J. A new era in palaeomicrobiology: prospects for ancient dental calculus as a long-term record of the human oral microbiome. Philos Trans R Soc Lond B Biol Sci 2015; 370: 20130376. 42. Mackie M, Radini A, Speller C. The sustainability of dental calculus for archaeological research. In Favreau J, Patalano R (eds) Shallow Pasts, Endless Horiozons: Sustainability & Archaeology: Proceedings of the 48th Annual Chacmool Archaeology Conference. pp 74-81. Calgary: The Chacmool Archaeological Association of the University of Calgary, 2017. 43. Fagernäs Z, Garcia Collado M I, Hendy J et al. A unified protocol for simultaneous extraction of DNA and proteins from archaeological dental calculus. J Archaeological Sci 2020; 118: 105135. 44. Hardy K, Blakeney T, Copeland L, Kirkham J, Wrangham R, Collins M. Starch granules, dental calculus and new perspectives on ancient diet. J Archaeol Sci 2009; 36: 248-255. 45. Weyrich L S, Dobney K, Cooper A. Ancient DNA analysis of dental calculus. J Hum Evol 2015; 79: 119-124. 46. Lippi M M, Pisaneschi L, Sarti L, Lari M, Moggi-Cecchic J. Insights into the Copper-Bronze Age diet in Central Italy: Plant microremains in dental calculus from Grotta dello Scoglietto (Southern Tuscany, Italy). J Archaeol Sci Rep 2017; 15: 30-39. 47. D'Agostino A, Canini A, Di Marco G, Nigro L, Spagnoli F, Gismondi A. Investigating plant micro-remains embedded in dental calculus of the Phoenician inhabitants of Motya (Sicily, Italy). Plants (Basel) 2020; 9: 1395. 48. Jovanović J, Power R C, de Becdeliévre C, Goude G, Stefanović S. Microbotanical evidence for the spread of cereal use during the Mesolithic-Neolithic transition in the Southeastern Europe (Danube Gorges): Data from dental calculus analysis. J Archaeol Sci 2021; 125: 105288. 49. Kinaston R, Willis A, Miszkiewicz J J, Tromp M, Oxenham M F. The dentition: Development, disturbances, disease, diet and chemistry. In Buikstra J E (ed) Ortner's Identification of Pathological Conditions in Human Skeletal Remains. 3rd ed. pp 749-797. London: Academic Press, 2019. 50. Curvers H H, Schwartz G M. Excavations at Tell al-Raqā'i: a small rural site of early urban northern Mesopotamia. Am J Archaeol 1990; 94: 3-23. 51. Cristiani E, Radini A, Borić D et al. Dental calculus and isotopes provide direct evidence of fish and plant consumption in Mesolithic Mediterranean. Sci Rep 2018; 8: 8147. 52. Power R C, Salazar-García D C, Straus L G, González Morales M R. Henry A G. Microremains from El Mirón Cave human dental calculus suggest a mixed plant-animal subsistence economy during the Magdalenian in Northern Iberia. J Archaeol Sci 2015; 60: 39-46. 53. Lazaridis I, Patterson N, Mittnik A et al. Ancient human genomes suggest three ancestral populations for present-day Europeans. Nature 2014; 513: 409-413. 54. Allentoft M E, Sikora M, Sjögren K-G et al. Population genomics of Bronze Age Eurasia. Nature 2015; 522: 167-172. 55. Haak W, Lazaridis I, Patterson N et al. Massive migration from the steppe was a source for Indo-European languages in Europe. Nature 2015; 522: 207-211. 56. Dominguez-Bello M G, Blaser M J. The human microbiota as a marker for migrations of individuals and populations. Annu Rev Anthropol 2011; 40: 451-474. 57. Adler C J, Dobney K, Weyrich L S et al. Sequencing ancient calcified dental plaque shows changes in oral microbiota with dietary shifts of the Neolithic and Industrial revolutions. Nat Genet 2013; 45: 450-455. 58. Eisenhofer R, Anderson A, Dobney K, Cooper A, Weyrich L S. Ancient microbial DNA in dental calculus: A new method for studying rapid human migration events. J Isl Coast Archaeol 2019; 14: 149-162. 59. Aufderheide A C, Rodriguez-Martin C, Langsjoen O. The Cambridge Encyclopaedia of Human Paleopathology. Cambridge: Cambridge University Press, 1988. 60. Yu H-H, Lee D-H, Seo S-J, You Y-O. Anticariogenic properties of the extract of Cyperus rotundus. Am J Chin Med 2007; 35: 497-505. 61. Najah A M. In vitro Inhibitory effect of Cyperus rotundus L crude extracts on mouth isolates of Streptococcus mutans and Candida albicans. Al-Mustansiriyah J Pharma Sci 2012; 11: 85-91. 62. Buckley S, Usai D, Jakob T, Radini A, Hardy K. Dental calculus reveals unique insights into food items, cooking and plant processing in Prehistoric Central Sudan. PLoS One 2014; DOI: 10.1371/journal.pone.0100808. 63. Cummings L S, Yost C, Sołtysiak A. Plant microfossils in human dental calculus from Nemrik 9, a Pre-Pottery Neolithic site in Northern Iraq. Archaeol Anthropol Sci 2018; 10: 883-891. 64. Schoeninger M J. Stable isotope evidence for the adoption of maize agriculture. Curr Anthropol 2009; 50: 633-640. 65. Schoeninger M J. Diet reconstruction and ecology using stable isotope ratios. In Larsen C S (ed) A Companion to Biological Anthropology. pp 445-464. Chichester: Wiley-Blackwell, 2010. 66. Scott G R, Poulson S R. Stable carbon and nitrogen isotopes of human dental calculus: a potentially new non-destructive proxy for paleodietary analysis. J Archaeol Sci 2012; 39: 1388-1393. 67. Forshaw R J. Dental indicators of ancient dietary patterns: dental analysis in archaeology. Br Dent J 2014; 216: 529-535. 68. Salazar-García D C, Richards M P, Nehlich O, Henry A G. Dental calculus is not equivalent to bone collagen for isotope analysis: a comparison between carbon and nitrogen stable isotope analysis of bulk dental calculus, bone and dentine collagen from same individuals from the Medieval site of El Raval (Alicante, Spain). J Archaeol Sci 2014; 47: 70-77. 69. Price S D R, Keenleyside A, Schwarcz H P. Testing the validity of stable isotope analyses of dental calculus as a proxy in paleodietary studies. J Archaeol Sci 2018; 91: 92-103. 70. Coccato A, Moens L, Vandenabeele P. On the stability of medieval inorganic pigments: A literature review of the effect of climate, material selection, biological activity, analysis and conservation treatments. Herit Sci 2017; 5: 12. 71. Blatt S H, Redmond B G, Cassman V, Sciulli P W. Dirty teeth and ancient trade: Evidence of cotton fibres in human dental calculus from Late Woodland, Ohio. Int J Osteoarchaeol 2011; 21: 669-678. 72. Fotakis A K, Denham S D, Mackie M et al. Multi-omic detection of Mycobacterium leprae in archaeological human dental calculus. Philos Trans R Soc Lond B Biol Sci 2020; 375: 20190584. 73. Eerkens J W, Nichols R W, Murray G G R et al. A probable prehistoric case of meningococcal disease from San Francisco Bay: Next generation sequencing of Neisseria meningitidis from dental calculus and osteological evidence. Int J Paleopathol 2018; 22: 173-180. 74. Gismondi A, D'Agostino A, Di Marco G et al. Back to the roots: dental calculus analysis of the first documented case of coeliac disease. Archaeol Anthropol Sci 2020; 12: 6.	36494546	PMC9734501	NO-CC CODE	2022-12-14 23:28:29	no	Br Dent J. 2022 Dec 9; 233(11):961-967	utf-8	Br Dent J	2,022	10.1038/s41415-022-5266-7	oa_other
==== Front Food Anal Methods Food Anal Methods Food Analytical Methods 1936-9751 1936-976X Springer US New York 2429 10.1007/s12161-022-02429-6 Article A Fast and Simple DNA Mini-barcoding and RPA Assay Coupled with Lateral Flow Assay for Fresh and Canned Mackerel Authentication Frigerio Jessica [email protected] 12 Gorini Tommaso 1 Palumbo Cassandra 2 De Mattia Fabrizio 1 Labra Massimo 2 Mezzasalma Valerio 1 1 FEM2-Ambiente, Piazza Della Scienza 2, I-20126 Milan, Italy 2 grid.7563.7 0000 0001 2174 1754 Department of Biotechnology and Biosciences, University of Milano-Bicocca, FEM2-Ambiente, Piazza Della Scienza 2, I-20126 Milan, Italy 5 12 2022 110 6 7 2022 28 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. Nowadays, food authentication is more and more required given its relevance in terms of quality and safety. The seafood market is heavily affected by mislabelling and fraudulent substitutions/adulterations, especially for processed food products such as canned food items, due to the loss of morphological features. This study aims to develop new assays based on DNA to identify fresh mackerel (Scomber spp.) and commercial products. A new primer pair was de novo designed on the 5S rRNA gene and non-transcribed spacer (NTS), identifying a DNA mini-barcoding region suitable for species identification of processed commercial products. Moreover, to offer a fast and low-cost analysis, a new assay based on recombinase polymerase amplification (RPA) was developed for the identification of fresh ‘Sgombro’ (Scomber scombrus) and ‘Lanzardo o Occhione’ (Scomber japonicus and Scomber colias), coupled with the lateral flow visualisation for the most expensive species (Scomber scombrus) identification. This innovative portable assay has great potential for supply chain traceability in the seafood market. Graphical Abstract Supplementary Information The online version contains supplementary material available at 10.1007/s12161-022-02429-6. Keywords DNA mini-barcoding Food fraud Recombinase polymerase amplification (RPA) Food traceability Food authentication Mislabelling http://dx.doi.org/10.13039/501100002954 Università degli Studi di Milano-Bicocca ND ND Frigerio Jessica Labra Massimo FEM2-AmbienteND ND ND ND Frigerio Jessica Gorini Tommaso De Mattia Fabrizio Mezzasalma Valerio ==== Body pmcIntroduction Nowadays, food authentication has become a major concern for addressing quality and safety issues (Soon 2022). The seafood market is heavily affected by inadvertent or deliberate events of adulteration or species substitution (Silva & Hellberg 2021), especially for processed food products, such as canned food, due to the loss of morphological features (Xing et al. 2020). In Europe, seafood labelling is regulated by two legislations: regulations (EU) No.1169/2011 and (EU) No 1379/2013, according to which the label ought not to mislead consumers. The Regulation (EU) 1169/2011 includes general assessments regarding mandatory food information that has to be available and easily accessible (Art. 12, comma 1,3), but does not include the commercial denomination or scientific name of the fish species sold as processed products (Paolacci et al. 2021). This lack of information could be misleading for consumers and encourage misidentification processes and fraud. Canned seafood products such as tuna, salmon, anchovies and mackerel are the most widely consumed fish products across the world due to their practicality of storage and consumption (Mottola et al. 2022). The COVID-19 pandemic has increased the European consumption of canned fish, and the sales of these food items boomed during the lockdown period in Europe, particularly in Southern European nations such as Spain, France and Italy. The European Centre for the Promotion of Imports from developing countries (CBI) reported that canned tuna consumption went up by 12% from January to May 2020 compared to the same period in 2019, reaching a 38.6% increase in countries like Italy (https://www.cbi.eu/market-information/fish-seafood/canned-fish/market-potential). Although tuna is the bestselling canned product, mackerel (Scomber spp.) is an emerging alternative (conserved in brine, olive or vegetable oil) due to its (reduced o) lower costs and its increasing use in the seafood market (Mottola et al. 2022). From a taxonomic point of view, the genus of mackerel is composed offour different species, namely S. scombrus (Linnaeus, 1758), S. japonicus (Houttuyn, 1782), S. colias (Gmelin, 1789) and S. australasicus (Cuvier & Valenciennes, 1832). The Atlantic mackerel S. scombrus is more expensive than the other species due to the excellent properties of the meat (Infante et al. 2007). The average price for S. japonicus/colias in 2016 was 0.58 €/Kg while for S. scombrus was 1.24 €/Kg (Working Document to WGWIDE, 2017). Because the morphological features are removed during processing, identifying the species is difficult and fraudulent substitution with cheaper species increases. To fight fraud, DNA-based techniques are widely presented in the scientific literature (Böhme et al. 2019; Nehal et al. 2021; Barbuto et al. 2010; Shokralla et al. 2015; Frigerio et al. 2021a, b). DNA barcoding is the most used approach, especially for the seafood market, and the U.S. Food and Drug Administration (FDA) uses it to control and identify fish species for regulatory compliance (Yancy et al. 2008). The standard DNA barcode presented by Hebert in 2003 for animal identification was the 5’ end portion of the mitochondrial cytochrome c oxidase subunit I (COI). Nevertheless, due to the length of this region (650 bp), it cannot be used in highly processed products, where industrial treatment such as high temperature and pressure can lead to DNA fragmentation. For this reason, a DNA mini-barcoding approach (about 100–200 bp) is more suitable for this typology of products (Filonzi et al. 2021). This study aims to develop new DNA-based assays to identify mackerel commercial products. To identify the species of canned mackerel products, a new primer pair was de novo designed on the 5S rRNA gene and non-transcribed spacer (NTS), identifying a DNA mini-barcoding region suitable for transformed and processed commercial products. 5S rRNA gene and non-transcribed spacer (NTS) werechosen because for mackerel identification the COI fragment is not able to provide an unambiguous identification at species level (Mottola et al. 2022). Despite the fact that DNA mini-barcoding is efficient in analysing processed products such as canned mackerel, this technique requires a few days, higher costs and specialised laboratory personnel to obtain the results. To overcome this obstacle in this study, a new assay for fresh specimens was developed, based on recombinase polymerase amplification (RPA), for the identification of the most common mackerel species sold in Italy, ‘Sgombro’ (Scomber scombrus) and ‘Lanzardo o Occhione’ (Scomber japonicus and Scomber colias). RPA in literature has been reported mainly in food pathogen detection tests such as Escherichia coli O157 (Zhao et al., 2022), Campylobacter jejuni (Geng et al. 2019), Salmonella (Hu et al. 2019), Listeria monocytogenes and Staphylococcus aureus (Guo et al. 2019) but thanks to the easiness of the assay it is a promising method also for species detection. RPA assay coupled with lateral flow provides a fast (less than an hour) and cheap (< 5 €) test for companies and consumers, reducing the time required to get the results and allowing it to be used also by non-specialised laboratory personnel. Material and Methods Specimen Collection In this study, a total of 27 samples were collected (Table 1). Reference specimens (SCFEM_01-04) were sampled at the Milan fish market (Milan, Italy) and morphologically identified by the quality manager and veterinary surgeon, Dr Valerio Ranghieri, while the commercially canned specimens (SCFEM_05-27) were collected from Italian supermarkets coming from four different companies. Commercial samples were chosen from the most important Italian brands and considering three different conservation liquids(olive oil, vegetable oil andbrine).Table 1 In the table are indicated the lab specimen, the declared species, the typology of the sample processing stage, the company and finally the verified species through the DNA mini-barcoding analysis Lab specimen Declared species Typology of sample Company DNA quantification (ng/μL) Verified species SCFEM_01 S. colias Fresh tissue Milan fish market 18.2 S. colias SCFEM_02 S. japonicus Fresh tissue Milan fish market 21.3 S. japonicus SCFEM_03 S. scombrus Fresh tissue Milan fish market 16.6 S scombrus SCFEM_04 S. australasicus Fresh tissue Milan fish market 24.2 S. australasicus SCFEM_05 Scomber spp. Vegetable oil Company 1 6.9 S. colias SCFEM_06 Scomber spp. Vegetable oil Company 1 8.6 S. japonicus SCFEM_07 Scomber spp. Vegetable oil Company 1 8.7 S. japonicus SCFEM_08 Scomber spp. Vegetable oil Company 1 10.2 S. japonicus SCFEM_09 Scomber spp. Vegetable oil Company 1 9.6 S. japonicus SCFEM_10 Scomber spp. Vegetable oil Company 1 7.6 S. japonicus SCFEM_11 Scomber spp. Vegetable oil Company 1 11.2 S. japonicus SCFEM_12 Scomber spp. Vegetable oil Company 1 9.7 S. colias SCFEM_13 Scomber spp. Vegetable oil Company 1 6.3 S. colias SCFEM_14 Scomber spp. Vegetable oil Company 1 7.2 S. colias SCFEM_15 Scomber spp. Vegetable oil Company 1 6.8 S. colias SCFEM_16 Scomber spp. Vegetable oil Company 1 8.9 S. colias SCFEM_17 Scomber spp. Vegetable oil Company 1 5.7 S. colias SCFEM_18 Scomber spp. Vegetable oil Company 1 9.2 S. colias SCFEM_19 S. japonicus – S. colias Olive oil Company 2 11.4 S. colias SCFEM_20 S. japonicus – S. colias Olive oil Company 2 10.2 S. colias SCFEM_21 S. japonicus – S. colias Vegetable oil Company 2 9.8 S. colias SCFEM_22 S. japonicus – S. colias Brine Company 2 7.2 S. japonicus SCFEM_23 S. japonicus – S. colias Olive oil Company 2 9.3 S. colias SCFEM_24 Scomber colias Brine Company 3 11.2 S. colias SCFEM_25 S. japonicus – S. colias Olive oil Company 3 9.2 S. colias SCFEM_26 S. colias Olive oil Company 4 7.1 S. colias SCFEM_27 S. colias Olive oil Company 4 6.3 S. colias Primer Design for DNA Mini-barcoding Primer pairs for DNA mini-barcoding were newly designed in silico. The region 5S RNA gene and NTS was identified as one of the most variable markers to distinguish between all the Scomber species (Aranishi, 2005). All nucleotide sequences of the 5S rRNA gene and NTS (104 sequences) for Scomber spp. were obtained from NCBI Nucleotide and were aligned using ClustalW2 software (www.ebi.ac.uk/Tools/msa/clustalw2/). The most conserved regions were identified using Bioedit software and a primer pair specific for the genus Scomber spp. was de novo designed. 5S rRNA region was tested with Primer–Blast tool available from NCBI (www.ncbi.nlm.nih.gov/tools/primer-blast/) to verify the specificity to the Scomber genus. Primer Design for RPA Analysis Primer pairs for RPA were identified in silico and were designed on the 5S rRNA gene and NTS. Differently from PCR primers, RPA primers require a length of 30–35 nucleotides. In order to identify species-specific couples of primers, all nucleotide sequences of the 5S rRNA region for Scomber spp. were obtained from NCBI Nucleotide and primers were designed as shown in the previous paragraph but focusing on variable regions. DNA Extraction For all samples listed in Table 1, gDNA was obtained starting from 20 mg of tissue by using DNeasy Blood & Tissue Kit (QIAGEN, Germany). Canned specimens (SCFEM_05-27) were pre-treated in order to clean the tissue from the conservation liquid such as oil (vegetable and olive) or brine. The products conserved under brine were washed three times with a physiological solution (NaCl 0.7%) mixing overnight at 4 °C. Oil and lipids were removed by soaking in chloroform/methanol/water (1:2:0.8) mixing overnight at room temperature (Chapela et al. 2007). Purified gDNA was checked for concentration and purity by using a Qubit 4 Fluorometer and Qubit dsDNA HS Assay Kit (Invitrogen, Carlsbad, California, United States). DNA Mini-barcoding A standard PCR amplification was performed using PCR Mix Plus (A&A Biotechnology, Poland) following the manufacturer’s instructions in a 25-μL reaction containing 1 μL 10 mM of each primer and 3 μL of gDNA. PCR cycles consisted of an initial denaturation step for 5 min at 95 °C, followed by 35 cycles of denaturation (45 s at 95 °C), annealing (45 s at 50 °C) and extension (1 min at 72 °C), and, hence, a final extension at 72 °C for 7 min. The amplicon was visualised by electrophoresis on agarose gel using 1.5% agarose Tris–acetate-EDTA (TAE) gel. Purified amplicons were bidirectionally sequenced at Eurofins Genomics (Germany). After manual editing, primer removal and pairwise alignment, all the tested samples (Table 1) identities were assessed by adopting a standard comparison approach against the GenBank database with BLASTn (Altschul et al. 1990). Each barcode sequence was taxonomically assigned to the animal species with the nearest matches (maximum identity > 99% and query coverage of 100%). RPA Assay An RPA reaction mix (TwistAmp® Basic, England and Wales) was prepared in a total volume of 50 μL containing 2.5 μL of Magnesium Acetate (MgOAc, added at the end of mix preparation), 2.4 μL of 10 mM for each primer, 29.5 μL of rehydration buffer, 10.2 μL of sterile water and 3 μL of gDNA and tested on reference specimens (SCFEM_01-04). Before proceeding with the amplification, the mixture was shaken vigorously to start the reaction. The amplification reaction consisted of 4 min at 39 °C, subsequently, it was further stirred and then put back at 39 °C for 20 min. Unlike DNA barcoding, RPA requires the purification of the amplicons prior to gel electrophoresis to guarantee better performance. The QIAquick Gel Extraction Kit (QIAGEN) was used for the purification of amplicons. Amplicons occurrence was assessed by electrophoresis on agarose gel using 1.5% agarose Tris–acetate-EDTA (TAE) gel. Lateral Flow Assay HybriDetect—Universal Lateral Flow Assay Kit (Milenia Biotec GmbH, Germany) with gold particles was used for lateral flow assay. The lateral flow strip is designed to develop qualitative or semi-quantitative rapid test systems. Primer forward was labelled with fluorescein-5,6-isothiocyanate (FICT) and primer reverse with biotin (BIO) by Eurofins Genomics (Germany). The samples were mixed with the solution supplied in the kit, and then the strip was placed into the solution. The DNA of interest, labeled with FITC and biotin, binded first to the gold-labeled FITC-specific antibodies in the sample application area of the strip. The gold complexes travelled through the membrane, driven by capillary forces. Only the DNA with the gold particles binded the test line with the immobilized biotin-ligand molecules, generating a grey-blue band. Unbound gold particles migrate over the control band and will be captued by species-specific antibodies. This assay was only tested with S. scombrus primer pairs on reference specimens (SCFEM_01-04). The assay was performed with a modification of the company protocol, consisting in using only 3 μL of DNA for the visualisation on the lateral flow. For limit of detection (LOD) evaluation, the dilution of 1:10–1, 1:10–2, 1:10–3, 1:10–4 and 1:10–5 starting from the reference of S. scombrus was tested. Results DNA Mini-barcoding The primer pair for DNA mini-barcoding was designed to be specific to the Scomber genus. To analyse highly processed products with expected degraded DNA, the primer pair amplifies a region of 151–160 base pairs. In Fig. 1 is described the sequence alignment for primer pair design and the primers’ sequences are described in Table 2. To evaluate a specific approach for Scomber genus, the most conserved region (indicated in green in Fig. 1) was identified for primer pair design.Fig. 1 Sequence alignment of the 5S rRNA gene and NTS from NCBI Nucleotide. The sequences displayed are representative of all the haplotypes generated by using FaBox (1.61) (https://users-birc.au.dk/palle/php/fabox/). The coloured nucleotide bases indicated a mismatch. On the top of the consensus sequence, at position 259 and 418 are indicated the designed primer. This alignment was obtained by Geneious Prime Table 2 In the table are shown the sequence 5’-3’ of primer de novo designed in this study for DNA mini-barcoding and RPA, annealing temperature and amplicon expected dimensions Specificity Primer name Sequence 5’-3’ Annealing temperature (°C) Amplicon dimension (bp) Scomber spp. Sco5S_F CTCACTGTTACAGCCTG 50 °C 151–160 Scomber spp. Sco5S_R CAAACACATGCTATCCTT 50 °C 151–160 S. scombrus RPA_S.sco F: ACACACAGGGCGTTGAGAAACAAAGCTGCAATCA R: TCAGGCTATTTGTGTACATGCGCTTATAAGATG 39 °C 178 S. japonicus/S. colias RPA_S.jap_col F: GTCTGAATGCACGCCAGAGAGGTGGCACTGAGACG R: TTTCTGCGGAGAAACACACAGCTGGAAGGACTGC 39 °C 173 DNA extraction was successful for all the samples with high DNA quality and good yield (i.e. 5.7–24.2 ng/μL). Reference samples (SCFEM_01-04) were analysed with the primer pair designed in this study with success. All the electropherograms obtained were high quality and allowed to uniquely identify the species. Therefore, for all the reference specimens, it was possible to identify the species, proving the ability of this marker to correctly identify all the species belonging to the Scomber genus. All commercial samples (SCFEM_05-27) were successfully identified at the species level by Sanger sequencing, despite their processing stage (cooked at high temperature and under pressure, conserved in vegetable oil, olive oil and brine). Also for commercial samples, high-quality electropherograms were obtained. Amplicons obtained for all samples (SCFEM_01-27) are represented in Fig. 2.Fig. 2 In the figure, the electropherogram agarose gel of the reference specimens (SCFEM_01-04) and canned specimens (SCFEM_05-27) is shown. 151–160 base pairs amplicons were obtained as expected. A ladder 3000–100 was used RPA Assay For RPA analysis, two primer pairs were designed on variable regions to be species-specific to S. scombrus (“Sgombro”) and S. japonicus/colias (“Lanzardo o Sgombro Occhione”), in order to distinguish these two groups (see Figure S.1 in supplementary information). The specificity was confirmed by developing in silico PCR through the software primer BLAST. The primer pairs amplify a product of 173–178 bp and are shown in Table 2. Both the primer pairs were tested on the reference samples S. scombrus, S. colias, S. japonicus and S. australasicus (SCFEM_01-04). After the amplification process and amplicon purification, amplicons were visualised on an electrophoresis agarose gel. Results are shown in Fig. 3. Both primer pairs were specific only to target species.Fig. 3 In the figure are shown the electrophoresis results of the RPA assay. In (a) are shown the results for the primer pair specific for S. scombrus (178 bp), in (b) are shown the results for the primer pair specific for S. japonicus/S. colias (173 bp). A ladder 3000–100 was used These results allow us to distinguish S. japonicus and S. colias from S. scombrus which is the most expensive mackerel species, confirming the feasibility of the assay. Lateral Flow Assay A lateral flow assay was developed only for the primer pair specific for S. scombrus, the most expensive species belonging to the Scomber genus. All the reference specimens (S. scombrus, S. colias, S. japonicus and S. australasicus) (SCFEM_01-04) and negative control were tested. The assay showed a band only for S. scombrus (Fig. 4a). In order to investigate the limit of detection (LOD), dilution series were created. The initial concentration of the S. scombrus sample was 16.6 ng/μL. Three microlitres of DNA dilution from 10–1 up to 10–5 was tested. Results of RPA reaction showed a high sensitivity of this assay, with a detection up to 0.0048 ng of total DNA (Fig. 4b).Fig. 4 a shows the results obtained from the lateral flow assay starting from RPA amplification. The upper band is the control band, the lower band shows the success of amplification. b shows the results of the dilution of the reference specimen of S. scombrus. Dilutions from 10–1 (4.98 ng of DNA) up to 10–5 (0.000498 ng of DNA) were tested Discussion The Misleading Labelling for Mackerel Since it is not mandatory to provide the fish scientific name on labels for processed products, most of the commercial items purchased (i.e. 14) reported only the commercial name ‘Sgombro’ on their packaging. Based on Annex 1 of the Italian MiPAAF Decree dated September 22, 2017 (MiPAAF, DM 19,105, 2017), this term corresponds only to the S. scombrus species but it is commonly used in the market for all the Scomber species. In the other nine products analysed (see Table 1), only three provided the species name (S. colias) and the remaining six were labelled both as S. colias and S. japonicus which correspond to the commercial name ‘Lanzardo o Sgombro Occhione’. If we wanted to calculate a misidentification rate only considering the commercial name (“Sgombro”, corresponding to S.scombrus), 60% of canned products analysed in this study would result as mislabelled, a value higher than those reported in the scientific literature (Neo et al. 2022; Xing et al. 2020; Hu et al. 2018; Panprommin & Manosri 2022). Nevertheless, the product “Sgombro” sold canned, due to European Regulation not requiring the scientific name on the label for processed products, can be misinterpreted because it can be related to all mackerel species. Moreover, another labelling issue is related to the interchange of the two species S. colias and S. japonicus. Despite being two different species, as one inhabits the Atlantic/Mediterranean Sea (S. colias) and the other the Pacific Ocean (S. japonicus), both of these are accepted and sold under the same vernacular name ‘Lanzardo o Sgombro Occhione’ (Mottola et al. 2022). The old scientific name indeed was S. japonicus-colias, but only recent studies recognised the taxa as two allopatric species. Canned and Fresh Mackerel Molecular Routine Analysis and Future Perspectives Highly processed products such as canned fish can undergo transformation processes like high temperature and high pressure. These industrial processes could damage DNA, which can be fragmented and degraded. Literature shows that the DNA of canned fish products, such as tuna, sardines and mackerel is usually fragmented (Chapela et al. 2007; Pecoraro et al. 2020; Servusova & Piskata 2021). For this reason, a DNA mini-barcoding approach is required (Frigerio et al. 2021a, b; Roungchun et al. 2022). In this study, a couple of primers on the region 5S rRNA and NTS were de novo designed with success. The 5S rDNA consists of a 120 bp conserved region, but the length and sequence of the NTS may vary among species. Among nuclear markers, the 5S rRNA is the most interesting in taxonomic identification because of its unique structure making it a species specific gene in higher eukaryotes, including teleost fishes such as mackerel (Aranishi & Okimoto 2004). For this reason, the 5S rRNA region was chosen for the discrimination of Scomber species. Moreover, due to its lower length compared with the standard DNA barcoding region (about 650 bp), it can be useful to overcome the problem of processed and fragmented DNA. Using the primer pair designed in this study, it was possible to analyse and identify all commercial store-bought products, despite their processing stage (high temperature, under brine or oil conservation). Even if this technique can be successfully used for transformed products such as canned items, sometimes the fish market allows for short timeframes to complete the analysis. In fact, often it is not possible to store the fish for long periods (differently from other food sectors), especially when the product is to be sold fresh and not processed. Therefore, a standard DNA mini-barcoding approach is not suitable to meet the needs of this market. For this reason, in this study, we also wanted to develop an assay based on the RPA methodology for fresh mackerel detection, which allowed us to analyse a product in less than two hours and without the expensive instrumentation of a molecular biology laboratory. The lateral flows assay provides a visible result in less than 15 min with the visualisation of a band (in addition to the control band) in presence of the target species. This test is a ready-to-use, test strip based on lateral flow technology using gold particles. In contrast to DNA mini-barcoding (see Table 3), the RPA assay is a specific analysis thataims to identify a specific species. In this study, we focused on differentiating the fresh products sold under the name of “Sgombro” (S. scombrus) and “Lanzardo” or “Sgombro occhione” (S. japonicus and S. colias) which are the most common in the seafood market. The former is a more expensive, higher quality fish compared to the latter. We had successfully developed and tested a specific couple of primers for S. scombrus and S. japonicus-S. colias for RPA assay with gel electrophoresis visualisation. In addition, in order to develop a rapid and cheap assay, the lateral flow assay for the couple of primers for S. scombrus, which is the most expensive species, was successfully developed and can detect very low quantity of DNA (0.0048 ng of total DNA). For companies, it could be a revolution in terms of supply chain traceability using a rapid and cheap kit that gives a result in less than two hours.Table 3 In the table are shown the comparison between the DNA barcoding and mini-barcoding and the RPA (with and without lateral flow) approaches Analysis typology Result visualisation Pro Cons DNA barcoding Electropherogram after Sanger sequencing Universal analysis Complete gene analysis Longer time for results Expensive Needs a specialised laboratory DNA mini-barcoding Electropherogram after Sanger sequencing Universal analysis Shorter gene besides DNA barcoding Longer time for results Expensive Needs a specialised laboratory RPA Agarose Gel Cheap analysis Results in 2 h for fresh samples Non universal analysis Agarose gel visualisation needs a specialised laboratory RPA + Lateral Flow assay Lateral flow Cheap analysis Results in less than 2 h for fresh samples Non universal analysis Excluding DNA barcoding and mini-barcoding techniques which allow universal analyses, other techniques besides RPA are presented in the literature for fish and mackerel authentication. A fast and easy tecnique similar to RPA is Loop-Mediated Isothermal Amplification (LAMP). An assay for fish detection has been reported to be fast (about 3 h) and with a limit of detection of 0.2 ng/μL (But et al. 2020). Prado et al. presented a real-time PCR method for mackerel detection with high sensitivity (up to 0.005 ng of DNA) (Prado et al. 2013). Although LAMP assay is faster and cheaper than real-time PCR, its sensitivity is low. RPA combine both the short timing and the high sensitivity (up to 0.0048 ng of total DNA), revealing to be the most promising methodology.For this reason, future studies will be focused on the development of the RPA assay with the lateral flow for the most common mackerel species, on further analysis on different specimen’s typology (such as canned products and multispecies products) and on assay validation before industrial scale-up and commercialization of a mock-up for companies A kit with this assay would be cheaper (< 5 €) and faster (less than an hour) than a laboratory test because it would not require expensive laboratory instrumentation and skilled technicians. Conclusion In this study, a DNA mini-barcoding and an RPA assay for the Scomber species identification were developed. The DNA mini-barcoding analysis enables the recognition of each species of Scomber (S. scombrus, S. colias, S. japonicus and S. australasicus). However, European legislation is still too permissive, as the determination of species on processed food labels is not yet mandatory. This can worsen fraud and mislabelling issues, which are already very common in the seafood sector. Furthermore, the development of a quick and cheap test, such as RPA and lateral flow assay, can be a huge change for companies, allowing for an economical control of the entire supply chain. Considering that this promising test is simple and easy to use, it would be used directly by consumers in the future, making them more aware of the products they buy and eat and protecting them from food fraud and mislabelling. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 147 KB) Acknowledgements We thank Dr Valerio Ranghieri, veterinarian surgeon and quality manager of the Milan Fish Market, who provided and morphologically recognised the SCFEM_01-04 samples. We thank Dr Paola Re for graphical support and Dr Filippo Bargero for English revision. Author Contribution Conceptualization J.F, V.M., Methodology J.F., T.G., C.P., Validation J.F., T.G., V.M., Data Curation J.F., V.M., Writing—Original Draft J.F., V.M, T.G., Project administration V.M., J.F., M.L., Supervision Funding acquisition F.D.M. Funding FEM2-Ambiente s.r.l., provided support in the form of a salary for authors J.F., V.M., T.M. and F.D.M.. The company only provided financial support in the form of research materials. No funding was received for this study. Data Availability The data that support the findings of this study are available from the corresponding author J.F. Declarations Competing interests The authors declare no competing interests. Conflict of Interest Jessica Frigerio declares that she has no conflict of interest. Tommaso Gorini declares that he has no conflict of interest. Cassandra Palumbo declares that she has no conflict of interest. Fabrizio De Mattia declares that he has no conflict of interest. Massimo Labra declares that he has no conflict of interest. Valerio Mezzasalma declares that he has no conflict of interest. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Altschul SF Gish W Miller W Basic local alignment search tool J Mol Biol 1990 215 3 403 410 10.1016/S0022-2836(05)80360-2 2231712 Aranishi F Rapid PCR-RFLP method for discrimination of imported and domestic mackerel Mar Biotechnol 2005 7 6 571 575 10.1007/s10126-004-4102-1 Aranishi F Okimoto T PCR-based detection of allergenic mackerel ingredients in seafood J Genet 2004 83 2 193 195 10.1007/BF02715826 15536259 Barbuto M, Galimberti, A, Ferri E, et al (2010) DNA barcoding reveals fraudulent substitutions in shark seafood products: the Italian case of “palombo”(Mustelus spp.). Food Res Int 43(1):376–381. 10.1016/j.foodres.2009.10.009 Böhme K Calo-Mata P Barros-Velázquez J Ortea I Review of recent DNA-based methods for main food-authentication topics J Agric Food Chem 2019 67 14 3854 3864 10.1021/acs.jafc.8b07016 30901215 But GWC Wu HY Rapid detection of CITES-listed shark fin species by loop-mediated isothermal amplification assay with potential for field use Sci Rep 2020 10 1 1 14 10.1038/s41598-020-61150-8 31913322 Chapela MJ Sotelo CG Pérez-Martín RI Comparison of DNA extraction methods from muscle of canned tuna for species identification Food Control 2007 18 10 1211 1215 10.1016/j.foodcont.2006.07.016 Cuvier G, Valenciennes A (1832) Histoire naturelle des poissons. Tome huitième. Livre neuvième. Des Scombéroïdes. v. 8: i-xix + 5 pp. + 1-509, Pls. 209-245. [Cuvier authored pp. 1-470; Valenciennes 471-509. Date of 1831 on title page. i-xv + 1-375 in Strasbourg edition] Filonzi L Vaghi M Ardenghi A Efficiency of DNA mini-barcoding to assess mislabeling in commercial fish products in Italy: an overview of the last decade Foods 2021 10 7 1449 10.3390/foods10071449 34206502 Frigerio J Marchesi C Magoni C Application of DNA mini-barcoding and infrared spectroscopy for the authentication of the Italian product “bottarga” LWT 2021 139 110603 10.1016/j.lwt.2020.110603 Frigerio J Agostinetto G Mezzasalma V DNA-based herbal teas’ authentication: an ITS2 and psbA-trnH multi-marker DNA metabarcoding approach Plants 2021 10 10 2120 10.3390/plants10102120 34685929 Geng Y Liu G Liu L Real-time recombinase polymerase amplification assay for the rapid and sensitive detection of Campylobacter jejuni in food samples J Microbiol Methods 2019 157 31 36 10.1016/j.mimet.2018.12.017 30576752 Gmelin JF (1789) Aves Anseres. Tome I. Pars II. In: Gmelin J.F. (Ed.) Caroli a Linnaei Systema Naturae per Regna Tria Naturae 1(2):501–1032 Guo Z Chen Z Liu X Detection of Listeria monocytogenes and Staphylococcus aureus via duplex recombinase polymerase amplification J Food Saf 2019 39 4 e12628 10.1111/jfs.12628 Houttuyn M (1782) Beschryving van eenige Japanese visschen, en andere zee-schepzelen. Verh Holl Maatsch Wet Haarlem 20(2):311–350 Hu Y Huang SY Hanner Study of fish products in Metro Vancouver using DNA barcoding methods reveals fraudulent labeling Food Control 2018 94 38 47 10.1016/j.foodcont.2018.06.023 Hu J Huang R Sun Y Sensitive and rapid visual detection of Salmonella Typhimurium in milk based on recombinase polymerase amplification with lateral flow dipsticks J Microbiol Methods 2019 158 25 32 10.1016/j.mimet.2019.01.018 30703446 Infante C Blanco E Zuasti E Phylogenetic differentiation between Atlantic Scomber colias and Pacific Scomber japonicus based on nuclear DNA sequences Genetica 2007 130 1 1 8 10.1007/s10709-006-0014-5 16897460 Linnaeus C (1758) Systema Naturae per regna tria naturae, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis. Editio decima, reformata [10th revised edition], vol. 1: 824 pp. Laurentius Salvius: Holmiae Mipaaf-Decreto Ministeriale n° (1910) 5 del 22 settembre 2017 - Denominazioni in lingua italiana delle specie ittiche di interesse commerciale. Available at: https://www.politicheagricole.it/flex/cm/pages/ServeBLOB.php/L/IT/IDPagina/11953#:~:text=di%20interesse%20commerciale-,Decreto%20Ministeriale%20n%C2%B019105%20del%2022%20settembre%202017%20%2D%20Denominazioni,tutte%20le%20Capitanerie%20di%20Porto. Accessed 29 April 2022 Mottola A Piredda R Catanese G Species authentication of canned mackerel: challenges in molecular identification and potential drivers of mislabelling Food Control 2022 137 108880 10.1016/j.foodcont.2022.108880 Nehal N Choudhary B Nagpure A DNA barcoding: a modern age tool for detection of adulteration in food Crit Rev Biotechnol 2021 41 5 767 791 10.1080/07388551.2021.1874279 33530758 Neo S, Kibat C, Wainwright BJ (2022) Seafood mislabelling in Singapore. Food Control 108821. 10.1016/j.foodcont.2022.108821 Panprommin D Manosri R DNA barcoding as an approach for species traceability and labeling accuracy of fish fillet products in Thailand Food Control 2022 136 108895 10.1016/j.foodcont.2022.108895 Paolacci S Mendes R Klapper R Labels on seafood products in different European countries and their compliance to EU legislation Mar Policy 2021 134 104810 10.1016/j.marpol.2021.104810 Pecoraro C Crobe V Ferrari A Canning processes reduce the DNA-based traceability of commercial tropical tunas Foods 2020 9 10 1372 10.3390/foods9101372 32992465 Prado M Boix A von Holst C Development of a real-time PCR method for the simultaneous detection of mackerel and horse mackerel Food Control 2013 34 1 19 23 10.1016/j.foodcont.2013.04.007 Regulation (EU) No (1169), 2011 Of The European Parliament And Of The Council 2011, Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32011R1169&from=IT. Accessed 29April 2022 Regulation (EU) No (1379), 2013 Of The European Parliament And Of The Council 2013, Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:32013R1379&from=IT. Accessed 29th April 2022 Roungchun JB Tabb AM Hellberg RS Identification of tuna species in raw and processed products using DNA mini-barcoding of the mitochondrial control region Food Control 2022 134 108752 10.1016/j.foodcont.2021.108752 Servusova E Piskata Z Identification of selected tuna species in commercial products Molecules 2021 26 4 1137 10.3390/molecules26041137 33672711 Shokralla S Hellberg RS Handy SM A DNA mini-barcoding system for authentication of processed fish products Sci Rep 2015 5 1 1 11 10.1038/srep15894 Silva AJ, Hellberg RS (2021) DNA-based techniques for seafood species authentication. Adv Food Nutr Res 95:207–255. 10.1016/bs.afnr.2020.09.001 Soon JM (2022) Food fraud countermeasures and consumers: a future agenda. In Future Foods 597–611. 10.1016/B978-0-323-91001-9.00027-X The European and market potential for canned fish (2022). Available at: https://www.cbi.eu/market-information/fish-seafood/canned-fish/market-potential. Accessed 29 April 2022 Working Document to WGWIDE (2017) Available at: http://www.repositorio.ieo.es/e-ieo/bitstream/handle/10508/11071/Villamor_et_al.2017_WD%20WGWIDE_SColias.pdf?sequence=1&isAllowed=y. Accessed 29th April 2022 Xing RR Hu RR Han JX DNA barcoding and mini-barcoding in authenticating processed animal-derived food: a case study involving the Chinese market Food Chem 2020 309 125653 10.1016/j.foodchem.2019.125653 31670116 Yancy HF, Fry SF, Randolph SC et al (2008) A protocol for validation of DNA-barcoding for the species identification of fish for FDA regulatory compliance. FDA Laboratory Information Bulletin Zhao L, Wang J, Chen M, et al (2022) Development and application of recombinase polymerase amplification assays for rapid detection of Escherichia coli O157 in food. Food Analytical Methods 1-8 10.1007/s12161-022-02250-1	0	PMC9734502	NO-CC CODE	2022-12-14 23:28:29	no	Food Anal Methods. 2022 Dec 5;:1-10	utf-8	Food Anal Methods	2,022	10.1007/s12161-022-02429-6	oa_other
==== Front Drugs Ther Perspect Drugs Ther Perspect Drugs & Therapy Perspectives 1172-0360 1179-1977 Springer International Publishing Cham 971 10.1007/s40267-022-00971-1 Adis Drug Q&a Nirmatrelvir plus ritonavir in COVID-19: a profile of its use Blair Hannah A. [email protected] grid.420067.7 0000 0004 0372 1209 Springer Nature, Private Bag 65901, Mairangi Bay, Auckland, 0754 New Zealand 6 12 2022 17 14 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. Oral nirmatrelvir plus ritonavir (Paxlovid™) is an effective treatment option for coronavirus disease 2019 (COVID-19), the illness caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Nirmatrelvir inhibits the main protease of SARS-CoV-2, with ritonavir acting as a pharmacokinetic booster. In the phase II/III EPIC-HR trial, nirmatrelvir plus ritonavir reduced the risk of progression to severe COVID-19 in symptomatic, unvaccinated, non-hospitalized adults with mild-to-moderate COVID-19 at high risk for progression to severe disease. The incidence of COVID-19-related hospitalization or death through day 28 was significantly lower with nirmatrelvir plus ritonavir than with placebo. The efficacy of nirmatrelvir plus ritonavir has also been demonstrated in the real-world setting. Nirmatrelvir plus ritonavir is generally well tolerated, with most adverse events being of mild or moderate severity. Supplementary Information The online version contains supplementary material available at 10.1007/s40267-022-00971-1. Plain Language Summary Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the cause of the coronavirus disease 2019 (COVID-19) global pandemic. Nirmatrelvir plus ritonavir (Paxlovid™) is an oral antiviral treatment for COVID-19 that reduces the ability of SARS-CoV-2 to multiply in the body. The active substance nirmatrelvir blocks the activity of an enzyme needed by the virus to multiply, while ritonavir slows the breakdown of nirmatrelvir to increase its therapeutic benefit. Nirmatrelvir plus ritonavir reduces the risk of hospitalization or death in patients who are at increased risk for progression to severe COVID-19. The drug is generally well tolerated; most adverse events were mild or moderate in severity. Drug–drug interactions are possible. Nirmatrelvir plus ritonavir is an important treatment option for COVID-19 in patients whose age or underlying health puts them at high risk for becoming severely ill. Supplementary Information The online version contains supplementary material available at 10.1007/s40267-022-00971-1. ==== Body pmc Digital Features for this Adis Drug Q&A can be found at 10.6084/m9.figshare.21554598 Adis evaluation of nirmatrelvir plus ritonavir in the treatment of COVID-19 Potent inhibitor of the main protease of SARS-CoV-2, co-packaged with a pharmacokinetic booster Administered orally twice daily for 5 days Reduces the risk of hospitalization or death in patients at high risk of progressing to severe COVID-19 Generally well tolerated May be associated with significant drug–drug interactions What is the rationale for developing nirmatrelvir plus ritonavir to treat COVID-19? Coronavirus disease 2019 (COVID-19), the illness caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), first appeared in late 2019 and was quickly declared a global pandemic [1]. Despite the rapid development and authorisation of numerous COVID-19 vaccines, increasing rates of breakthrough infection due to waning vaccine efficacy and the emergence of new variants of concern have necessitated the use of further measures to control the pandemic, including booster doses and the development of several therapeutic agents [2]. Some treatments are only recommended for use in patients who are at high risk of progression to severe COVID-19 (e.g. older age, immunodeficiencies, comorbidities, lack of vaccination) [3]. Nirmatrelvir plus ritonavir (Paxlovid™) was one of the first orally available antiviral treatments for COVID-19 [4, 5]. Nirmatrelvir is a peptidomimetic inhibitor of the main protease (Mpro) of SARS-CoV-2 [5]. The HIV-1 protease inhibitor ritonavir has no activity against SARS-CoV-2 Mpro, but acts as a pharmacokinetic boosting agent [6, 7]. Nirmatrelvir plus ritonavir is available in several countries, including the USA [7] and those of the EU [6], for the treatment of COVID-19 in patients who are at increased risk for progression to severe disease. Table 1 provides a summary of the prescribing information for nirmatrelvir plus ritonavir in the aforementioned regions.Table 1 Summary of the prescribing information of nirmatrelvir plus ritonavir (Paxlovid™) for the treatment of COVID-19 in the USA [7] and the EU [6] What is the authorized indication for nirmatrelvir plus ritonavir? USA Treatment of mild-to-moderate COVID-19 in adults and paediatric pts (aged ≥ 12 years weighing ≥ 40 kg) with positive results of direct SARS-CoV-2 viral testing, and who are at high risk for progression to severe COVID-19, including hospitalization or death (EUA) EU Treatment of COVID-19 in adults who do not require supplemental oxygen and who are at increased risk for progression to severe COVID-19 What is the recommended dosage of nirmatrelvir plus ritonavir? Nirmatrelvir 300 mg (two 150 mg tablets) with ritonavir 100 mg (one 100 mg tablet) all taken together orally twice daily for 5 days How should nirmatrelvir plus ritonavir be administered? Take with or without food; swallow tablets whole (do not chew, break or crush) Initiate treatment as soon as possible after COVID-19 diagnosis and within 5 days of symptom onset Complete full 5-day treatment course even if pt requires hospitalization due to severe or critical COVID-19 after starting treatment What are the contraindications to the use of nirmatrelvir plus ritonavir? Pts with a history of clinically significant hypersensitivity reactions to nirmatrelvir, ritonavir, or any other components of the product Pts receiving drugs that are highly dependent on CYP3A for clearance and for which elevated concentrations are associated with serious and/or life-threatening reactions Pts receiving drugs that are potent CYP3A inducers where significantly reduced nirmatrelvir or ritonavir plasma concentrations may be associated with the potential for loss of virological response and possible resistance How should nirmatrelvir plus ritonavir be used in special populations? Kidney impairment Mild: no dosage adjustment required Moderate (eGFR ≥ 30 to < 60 mL/min): reduce dosage of nirmatrelvir to 150 mg twice daily Severe (eGFR < 30 mL/min): not recommended for use Hepatic impairment Mild or moderate (Child-Pugh Class A or B): no dosage adjustment required Severe (Child-Pugh Class C): not recommended for use Pts receiving other products containing ritonavir or cobicistat No dosage adjustment required What other special warnings/precautions pertain to the use of nirmatrelvir plus ritonavir? Hepatotoxicity May cause hepatic transaminase elevations, clinical hepatitis and jaundice; exercise caution in pts with pre-existing liver disease, liver enzyme abnormalities or hepatitis HIV-1 resistance May increase risk of HIV-1 developing resistance to HIV protease inhibitors in pts with uncontrolled or undiagnosed HIV-1 infection Allergic reactions/hypersensitivity May cause hypersensitivity reactions; discontinue immediately if signs/symptoms of clinically significant hypersensitivity or anaphylaxis occur and initiate appropriate medications and/or supportive care (USA) Unless otherwise stated, information applies to both the USA and the EU. Consult local prescribing information for further details COVID-19 coronavirus disease 2019; eGFR estimated glomerular filtration rate, EUA Emergency Use Authorization, pt(s) patient(s), SARS-CoV-2 severe acute respiratory syndrome coronavirus 2 How does nirmatrelvir plus ritonavir work? Nirmatrelvir is a potent inhibitor of the SARS-CoV-2 Mpro, also known as 3C-like protease or nsp5 protease [7]. SARS-CoV-2 Mpro is responsible for processing two large polyprotein precursors (pp1a and pp1ab) into the smaller functional units required for viral replication [8]. Nirmatrelvir binds directly to the SARS-CoV-2 Mpro active site, inhibiting the activity of SARS-CoV-2 Mpro and thus preventing viral replication [6, 7]. In a biochemical assay, nirmatrelvir inhibited the activity of SARS-CoV-2 Mpro with an inhibition constant of 3.1 nM and a half-maximal inhibitory concentration of 19.2 nM [7]. Ritonavir lacks activity against SARS-CoV-2 Mpro, but increases plasma concentrations of nirmatrelvir by inhibiting the CYP3A-mediated metabolism of nirmatrelvir [6, 7]. Nirmatrelvir demonstrated antiviral activity against SARS-CoV-2 infection of differentiated normal human bronchial epithelial cells, with 50 and 90% maximal effective concentrations of 62 and 181 nM, respectively, after 3 days of drug exposure [6, 7]. The potency of nirmatrelvir against current and emerging SARS-CoV-2 variants is similar to that against wild-type SARS-CoV-2 [9, 10]. In vitro, nirmatrelvir had potent cell culture antiviral activity against all isolates belonging to the Alpha (B.1.1.7), Beta (B.1.351), Gamma (P.1), Delta (B.1.617.2) and Omicron (B.1.1.529) SARS-CoV-2 variants of concern, and the Lambda (B.1.1.1.37) and Mu (B.1.621) SARS-CoV-2 variants of interest [10, 11]. Nirmatrelvir also demonstrated antiviral activity in a murine model of SARS-CoV-2, reducing lung viral titres and improving disease indicators relative to placebo [7]. The impact of naturally occurring SARS-CoV-2 Mpro polymorphisms (of unknown clinical significance) on nirmatrelvir activity was assessed in a biochemical assay using recombinant Mpro enzyme [6]. G15S, T135I, S144A, H164N, H172Y, Q189K and D248E amino acid substitutions were associated with 3.5- to 233-fold reductions in the activity of nirmatrelvir. Of note, G15S is present in the Lambda variant of SARS-CoV-2, which did not have reduced susceptibility to nirmatrelvir in cell culture. In the phase II/III EPIC-HR trial, a number of treatment-emergent substitutions in SARS-CoV-2 Mpro gene or cleavage regions were detected. However, none of these substitutions occurred in hospitalized patients receiving nirmatrelvir plus ritonavir. The P132H/L/S, A260V and A266V substutitions did not reduce the activity of nirmatrelvir in a biochemical assay. Due to their different mechanisms of action, no cross-resistance is expected between nirmatrelvir and remdesivir or anti-SARS-CoV-2 monoclonal antibodies (mAbs) [6]. What are the pharmacokinetic properties of nirmatrelvir plus ritonavir? Ritonavir is coadministered with nirmatrelvir and acts as a pharmacokinetic enhancer [6, 7]. In healthy volunteers, ritonavir increased systemic nirmatrelvir exposure and prolonged its half-life [6, 7, 12]. This supported the selection of a regimen of nirmatrelvir 300 mg plus ritonavir 100 mg twice daily for 5 days in subsequent phase II/III trials [12]. The increase in systemic exposure of nirmatrelvir at steady state appears to be less than dose proportional following repeated twice-daily oral administration of nirmatrelvir plus ritonavir (75 mg + 100 mg, 250 mg + 100 mg, and 500 mg + 100 mg) [6]. Steady state is achieved on day 2, with ≈ 2-fold accumulation [6, 7]. Relative to fasting conditions, administration of nirmatrelvir plus ritonavir with a high-fat meal modestly increased nirmatrelvir exposure; therefore, nirmatrelvir plus ritonavir can be administered with or without food (Table 1) [6, 7]. Following a single oral dose of nirmatrelvir 300 mg plus ritonavir 100 mg in healthy volunteers, maximum plasma concentrations of nirmatrelvir and ritonavir were reached in a median of 3.00 and 3.98 h, respectively [6, 7]. Nirmatrelvir is ≈ 69% bound to plasma proteins [6, 7], with a blood-to-plasma ratio of 0.60 [7]. Ritonavir is highly protein-bound (≈ 98–99%) [6, 7], with a blood-to-plasma ratio of 0.14 [7]. The mean volume of distribution is 10.5 L for nirmatrelvir and 112.4 L for ritonavir [7]. Metabolism of nirmatrelvir is by CYP3A4 primarily, but this metabolism is inhibited by the coadministration of ritonavir [6, 7]. Ritonavir is metabolized predominantly by CYP3A4, with a minor contribution from CYP2D6 to form the oxidation metabolite M-2. Nirmatrelvir is excreted via the urine (≈ 50% of a 300 mg dose) and faeces (≈ 35%), mostly as unchanged drug. Ritonavir is eliminated primarily via the hepatobiliary system, with ≈ 86% of a radiolabeled dose excreted via the faeces [6, 7]. The mean oral clearance is 8.99 L/h for nirmatrelvir and 13.92 L/h for ritonavir [7]. The mean half-life of both nirmatrelvir and ritonavir is 6.1 h [6, 7]. Significant and clinically relevant pharmacokinetic drug–drug interactions may occur when nirmatrelvir plus ritonavir is coadministered with various other drugs (e.g. CYP3A substrates, inducers or inhibitors) [6, 7]. Some contraindications apply (Table 1) and dosage adjustments or additional monitoring may be required [6, 7]. Consult local prescribing information for further detailed information. What is the efficacy of nirmatrelvir plus ritonavir in COVID-19? Nirmatrelvir plus ritonavir reduces the risk of progression to severe COVID-19 in symptomatic, unvaccinated, non-hospitalized adults with mild-to-moderate COVID-19 at high risk for progression to severe disease, based on the results of a randomized, double-blind, multicentre, placebo-controlled, phase II/III trial (EPIC-HR) [13]. Participants in EPIC-HR were aged ≥ 18 years with confirmed SARS-CoV-2 infection, symptom onset ≤ 5 days prior to randomization, at least one sign or symptom of COVID-19 on the day of randomization, and at least one characteristic or co-existing condition associated with high risk of progression to severe COVID-19 [13]. Exclusion criteria included patients with previous SARS-CoV-2 infection or vaccination. Eligible patients were randomized to receive either nirmatrelvir 300 mg plus ritonavir 100 mg or placebo, administered orally every 12 h for 5 days. Randomization was stratified by geographic region and by receipt or expected receipt of COVID-19 mAbs. At baseline, the median age of patients was 46 years and the median time since first symptom was 3 days (range 0–9 days). Most (71.5%) patients were white and 51.1% of patients were male. The most common risk factors for progression to severe COVID-19 were body mass index (BMI) ≥ 25 kg/m2 (80.5%), current smoking (39.0%) and hypertension (32.9%). The primary endpoint was the proportion of patients with COVID-19-related hospitalization or death from any cause through day 28, assessed in the modified intention-to-treat (mITT) population (i.e. patients whose treatment began within 3 days of symptom onset, who did not receive nor were expected to receive COVID-19 mAbs) [13]. In a planned interim analysis of data from 774 patients in the mITT population, the incidence of COVID-19-related hospitalization or death through day 28 was significantly lower with nirmatrelvir plus ritonavir than with placebo, corresponding to a relative risk reduction (RRR) of 89.1% (Table 2) [13]. Consistent results were seen in the final analysis of data from 1379 patients in the mITT population (Table 2). The efficacy of nirmatrelvir plus ritonavir was also consistent across subgroups based on age (< 65 and ≥ 65 years), sex, BMI (< 25, 25 to > 30 and ≥ 30 kg/m2), diabetes, time since symptom onset (≤ 3 and > 3 days), baseline SARS-CoV-2 serology status (negative and positive) and receipt or expected receipt of COVID-19 mAbs [13].Table 2 Efficacy of nirmatrelvir plus ritonavir in the phase II/III EPIC-HR trial [13] Endpoint [no. of pts (%)]a Nirmatrelvir + ritonavir Placebo Difference [% (95% CI)] RRR (%) Interim analysis (n = 389) (n = 385) COVID-19-related hospitalization or death through day 28b 3 (0.77)* 27 (7.01) − 6.32 (− 9.04, − 3.59) 89.1 Final analysis (n = 697) (n = 682) COVID-19-related hospitalization or death through day 28b 5 (0.72)* 44 (6.45) − 5.81 (− 7.78, − 3.84) 88.9 COVID-19-related hospitalization by day 28 5 (0.72) 44 (6.45) Death from any cause by day 28 0 9 (1.32) COVID-19 coronavirus disease 2019, pts patients, RRR relative risk reduction *p < 0.001 vs placebo aModified intention-to-treat population bPrimary endpoint Among patients whose treatment began within 5 days of symptom onset (n = 2085), eight (0.77%) nirmatrelvir plus ritonavir recipients were hospitalized for COVID-19 or died from any cause through day 28 compared with 66 (6.31%) placebo recipients (difference − 5.62%; 95% CI − 7.21, − 4.03; p < 0.001), corresponding to an 87.8% RRR (key secondary endpoint) [13]. When data from all patients (n = 2224) were analysed, including those who received or were expected to receive mAbs, COVID-19 hospitalization or death from any cause occurred in nine (0.81%) patients in the nirmatrelvir plus ritonavir group and 68 (6.10%) patients in the placebo group (difference − 5.36%; 95% CI − 6.88, − 3.84; p < 0.0001) [13]. At day 5 of treatment, the viral load was significantly (p < 0.001) lower with nirmatrelvir plus ritonavir than with placebo [13]. After adjusting for baseline viral load, serology status and geographic region, the mean difference was − 0.868 log10 copies/mL when treatment was initiated within 3 days of symptom onset and − 0.695 log10 copies/mL when treatment was initiated within 5 days of symptom onset. Similar results were seen when patients who received or were expected to receive mAbs were included in the analysis. Results from subgroup analyses were consistent with those in the overall population, regardless of baseline serology status and viral load [13]. Real-world experience with nirmatrelvir plus ritonavir supports the efficacy results observed during the EPIC-HR trial [14–20]. Reductions in the risk of severe COVID-19 [17], mortality [15–19], hospitalization [14–16, 19, 20] and disease progression (mortality, invasive mechanical ventilation or intensive care unit admission) [15, 18] were reported across studies in Hong Kong [15, 18, 19], Israel [16, 17] and the USA [14, 20] that each included ≥ 1000 patients treated with nirmatrelvir plus ritonavir. What is the tolerability profile of nirmatrelvir plus ritonavir? Nirmatrelvir plus ritonavir is generally well tolerated in adult patients with symptomatic SARS-CoV-2 infection, based on data from the EPIC-HR trial [13]. Adverse events (AEs) reported during or after the treatment period occurred in 22.6% of nirmatrelvir plus ritonavir recipients and 23.9% of placebo recipients. The most common (incidence ≥ 2%) AEs with nirmatrelvir plus ritonavir were dysgeusia (5.6% vs 0.3% with placebo), diarrhoea (3.1 vs 1.6%), increased fibrin D-dimer (1.9 vs 2.8%) and increased alanine aminotransferase (1.5 vs 2.4%). Most AEs were of mild or moderate severity. Serious AEs occurred at a lower incidence with nirmatrelvir plus ritonavir than with placebo (1.6 vs 6.6%). AEs led to discontinuation of study medication in 2.1% of nirmatrelvir plus ritonavir recipients and 4.2% of placebo recipients. AEs considered to be related to study medication occurred in 7.8% of nirmatrelvir plus ritonavir recipients and 3.8% of placebo recipients and, in nirmatrelvir plus ritonavir recipients, were most commonly dysgeusia (4.5 vs 0.2%) and diarrhoea (1.3 vs 0.2%) [13]. What is the current clinical role of nirmatrelvir plus ritonavir in COVID-19? Nirmatrelvir plus ritonavir is an effective treatment option for COVID-19 in patients who are at increased risk for progression to severe disease. Administered orally for 5 days as soon as possible after COVID-19 diagnosis, nirmatrelvir plus ritonavir reduces the risk of hospitalization or death in symptomatic, unvaccinated, non-hospitalized adults with mild-to-moderate disease. The efficacy of nirmatrelvir plus ritonavir has also been demonstrated in the real-world setting. Nirmatrelvir plus ritonavir is generally well tolerated, with most AEs being mild or moderate in severity. Despite the proven efficacy of nirmatrelvir plus ritonavir, there have been reports of virological rebound (i.e. recurrent COVID-19 symptoms and/or increased viral load) following treatment [21–25]. In a longitudinal cohort study (POSITIVES), virological rebound after treatment with nirmatrelvir plus ritonavir was associated with high viral load and, in some patients, culturable virus [21]. In the EPIC-HR trial, viral load rebound occurred in 2.3% of nirmatrelvir plus ritonavir recipients and 1.7% of placebo recipients, but was not associated with recurrence of severe COVID-19 symptoms [24]. The occurrence of viral load rebound in the placebo group suggests that some patients with COVID-19 may experience the phenomenon as part of the disease’s natural progression [24]. Virological rebound has also been reported after other COVID-19 treatments such as molnupiravir, indicating that the phenomenon is not unique to nirmatrelvir plus ritonavir [25]. Further studies are required to determine the mechanisms, incidence and clinical implications of virological rebound with COVID-19 antivirals, including nirmatrelvir plus ritonavir [21, 22, 25]. A number of treatment options are currently available for patients with non-severe COVID-19 who are at the highest risk of hospitalization, with the choice of treatment depending on factors such as drug availability, route of administration, duration of treatment and time from symptom onset [3]. The most recent update of the living WHO guideline for the treatment of COVID-19 strongly recommends nirmatrelvir plus ritonavir as an option for patients with non-severe COVID-19 at highest risk of hospitalization. The guideline also conditionally recommends against the use of nirmatrelvir plus ritonavir for patients with non-severe COVID-19 at low risk of hospitalization [3]. The European Society of Clinical Microbiology and Infectious Diseases (ESCMID) recommends the use of nirmatrelvir plus ritonavir in adult, unvaccinated, ambulatory patients with mild-to-moderate COVID-19 at high risk of progression to severe disease within 5 days of symptom onset [26]. Draft guidance from the UK National Institute for Health and Care Excellence (NICE) recommends the use of nirmatrelvir plus ritonavir in the non-hospital setting to treat adults with COVID-19 who do not need supplemental oxygen, but who are at risk of progression to severe disease [27]. To date, no randomized clinical trials have directly compared the efficacy of nirmatrelvir plus ritonavir with other oral antivirals for COVID-19, such as remdesivir and molnupiravir. A recent meta-analysis found that nirmatrelvir plus ritonavir, molnupiravir and fluvoxamine were all effective in reducing the risk of death or hospitalization in patients with COVID-19, with odds ratios of 0.05 (95% CI 0.28–0.72), 0.22 (95% CI 0.10–0.48) and 0.45 (95% CI 0.28–0.72), respectively [28]. However, results of such indirect comparisons should be interpreted with caution. Data evaluating the efficacy and tolerability of nirmatrelvir plus ritonavir relative to other agents in head-to-head clinical trials would be of interest. High epidemic resurgence of COVID-19 is expected to have a considerable burden on health system capacity [29]. In Hong Kong, both nirmatrelvir plus ritonavir and molnupiravir were associated with significant cost savings [19]. In the outpatient setting, nirmatrelvir plus ritonavir cost $US331,105 to prevent one death, but saved $US5,503 to prevent one death relative to standard care [19]. In Korea, treatment of symptomatic COVID-19 with oral nirmatrelvir plus ritonavir is likely to be cost effective, with incremental cost-effectiveness ratios (ICERs) of $US8878, $US8964 and $US1454 per prevented severe case when targeting all adults, adults with underlying diseases and elderly patients, respectively [29]. Oral molnupiravir is likely to be less cost effective, with ICERs of US$28,492, $US29,575 and $US7915, respectively [29]. According to NICE, nirmatrelvir plus ritonavir is likely a cost-effective use of National Health Service (NHS) resources in the non-hospital setting compared with standard care [27]. ‘Test-and-treat’ approaches to COVID-19 and equitable access to effective oral antivirals are crucial, particularly for high-risk populations [30]. Although affordable generic versions of nirmatrelvir plus ritonavir and molnupiravir are available in ≈ 100 low- and middle-income countries, some middle-income countries (e.g. Argentina, Brazil, Malaysia, Thailand) are excluded from the agreements. Moreover, access to nirmatrelvir plus ritonavir, which must be started within 5 days of symptom onset, will remain limited in low-income countries due to reduced testing availability. Conversely, high-income countries with good access to testing and nirmatrelvir plus ritonavir have very small high-risk populations due to high vaccination rates [30]. Clinical decision support systems have been shown to improve prescribing practice and patient outcomes [31]. For example, in a US academic health system, a best practice advisory model was developed to facilitate equitable access to nirmatrelvir plus ritonavir for patients with COVID-19 who are at high risk of clinical deterioration (Fig. 1) [31].Fig. 1 BPA workflow for prescription of nirmatrelvir plus ritonavir to treat COVID-19, as suggested by Millstein et al. [31]. BPA best practice advisory, COVID-19 coronavirus disease 2019, DDI drug–drug interaction, eGFR estimated glomerular filtration rate Supplementary Information Below is the link to the electronic supplementary material.Supplementary file 1 (PDF 200 KB) Acknowledgements The manuscript was reviewed by: Ye Htut Linn, FAME Pharmaceuticals Industry Co., Ltd., Yangon, Myanmar; A. Singh, Department of Pharmacology, All India Institute of Medical Sciences, Raipur, Chhattisgarh, India. During the peer review process, Pfizer, the marketing authorization holder of nirmatrelvir plus ritonavir, was also offered an opportunity to provide a scientific accuracy review of their data. Changes resulting from comments received were made on the basis of scientific and editorial merit. Declarations Funding The preparation of this review was not supported by any external funding. Authorship and conflict of interest H. A. Blair is a salaried employee of Adis International Ltd/Springer Nature and declares no relevant conflicts of interest. All authors contributed to the review and are responsible for the article content. Ethics approval, Consent to participate, Consent for publication, Availability of data and material, Code availability Not applicable. ==== Refs References 1. Islam T Hasan M Rahman MS Comparative evaluation of authorized drugs for treating Covid-19 patients Health Sci Rep. 2022 5 4 e671 10.1002/hsr2.671 35734340 2. Sanyaolu A Okorie C Marinkovic A Current advancements and future prospects of COVID-19 vaccines and therapeutics: a narrative review Ther Adv Vaccines Immunother. 2022 10.1177/25151355221097559 3. World Health Organization. Therapeutics and COVID-19: living guideline. 2022. https://apps.who.int/iris/bitstream/handle/10665/362843/WHO-2019-nCoV-therapeutics-2022.5-eng.pdf. Accessed 15 Nov 2022. 4. Fenton C Keam SJ Emerging small molecule antivirals may fit neatly into COVID-19 treatment Drugs Ther Perspect. 2022 38 3 112 126 10.1007/s40267-022-00897-8 35250258 5. Lamb YN Nirmatrelvir plus ritonavir: first approval Drugs 2022 82 5 585 591 10.1007/s40265-022-01692-5 35305258 6. European Medicines Agency. Paxlovid 150 mg + 100 mg film-coated tablets: EU summary of product characteristics. 2022. https://www.ema.europa.eu/en/medicines/human/EPAR/paxlovid. Accessed 15 Nov 2022. 7. Pfizer Inc. PAXLOVID—nirmatrelvir and ritonavir: US prescribing information. 2022. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=7bdddfba-bd31-44cb-ba9e-23a4e17a4691. Accessed 15 Nov 2022. 8. Duveau DY Thomas CJ The remarkable selectivity of nirmatrelvir ACS Pharmacol Transl Sci. 2022 5 6 445 447 10.1021/acsptsci.2c00065 35702394 9. Ullrich S Ekanayake KB Otting G Main protease mutants of SARS-CoV-2 variants remain susceptible to nirmatrelvir Bioorg Med Chem Lett 2022 10.1016/j.bmcl.2022.128629 10. Vangeel L Chiu W De Jonghe S Remdesivir, molnupiravir and nirmatrelvir remain active against SARS-CoV-2 Omicron and other variants of concern Antiviral Res 2022 10.1016/j.antiviral.2022.105252 11. Rai DK, Yurgelonis I, McMonagle P, et al. Nirmatrelvir, an orally active Mpro inhibitor, is a potent inhibitor of SARS-CoV-2 variants of concern. bioRxiv. 2022. 10.1101/2022.01.17.476644. 12. Singh RSP Toussi SS Hackman F Innovative randomized phase I study and dosing regimen selection to accelerate and inform pivotal COVID-19 trial of nirmatrelvir Clin Pharmacol Ther 2022 112 1 101 111 10.1002/cpt.2603 35388471 13. Hammond J Leister-Tebbe H Gardner A Oral nirmatrelvir for high-risk, nonhospitalized adults with Covid-19 N Engl J Med 2022 386 15 1397 1408 10.1056/NEJMoa2118542 35172054 14. Dryden-Peterson S, Kim A, Kim AY, et al. Nirmatrelvir plus ritonavir for early COVID-19 and hospitalization in a large US health system. medRxiv. 2022. 10.1101/2022.06.14.22276393. 15. Wong CKH, Au ICH, Lau KTK, et al. Real-world effectiveness of molnupiravir and nirmatrelvir/ritonavir against mortality, hospitalization, and in-hospital outcomes among community-dwelling, ambulatory COVID-19 patients during the BA.2.2 wave in Hong Kong: an observational study. medRxiv. 2022. 10.1101/2022.05.26.22275631. 16. Arbel R Sagy YW Hoshen M Oral nirmatrelvir and severe Covid-19 outcomes during the Omicron surge Res Sq. 2022 10.21203/rs.3.rs-1705061/v1 17. Najjar-Debbiny R Gronich N Weber G Effectiveness of Paxlovid in reducing severe COVID-19 and mortality in high risk patients Clin Infect Dis 2022 10.1093/cid/ciac443 18. Wong CKH, Au ICH, Lau KTK, et al. Real-world effectiveness of molnupiravir and nirmatrelvir/ritonavir among COVID-19 inpatients during Hong Kong’s Omicron BA.2 wave: an observational study. medRxiv. 2022. 10.1101/2022.05.19.22275291. 19. Wai AK Chan CY Cheung AW Association of molnupiravir and nirmatrelvir-ritonavir with preventable mortality, hospital admissions and related avoidable healthcare system cost among high-risk patients with mild to moderate COVID-19 Lancet Reg Health West Pac. 2022 10.1016/j.lanwpc.2022.100602 20. Lewnard JA, Malden D, Hong V, et al. Effectiveness of nirmatrelvir-ritonavir against hospital admission: a matched cohort study in a large US healthcare system. medRxiv. 2022. 10.1101/2022.10.02.22280623. 21. Boucau J Uddin R Marino C Characterization of virologic rebound following nirmatrelvir-ritonavir treatment for COVID-19 Clin Infect Dis 2022 10.1093/cid/ciac512 22. Epling BP, Rocco JM, Boswell KL, et al. COVID-19 redux: clinical, virologic, and immunologic evaluation of clinical rebound after nirmatrelvir/ritonavir. medRxiv. 2022. 10.1101/2022.06.16.22276392. 23. Ranganath N O'Horo JC Challener DW Rebound phenomenon after nirmatrelvir/ritonavir treatment of coronavirus disease-2019 in high-risk persons Clin Infect Dis 2022 10.1093/cid/ciac481 24. Soares H Baniecki M Cardin RD Viral load rebound in placebo and nirmatrelvir-ritonavir treated COVID-19 patients is not associated with recurrence of severe disease or mutations Res Sq. 2022 10.21203/rs.3.rs-1720472/v2 25. Wang L, Berger NA, Davis PB, et al. COVID-19 rebound after paxlovid and molnupiravir during January-June 2022. medRxiv. 2022. 10.1101/2022.06.21.22276724. 26. Bartoletti M Azap O Barac A European society of clinical microbiology and infectious diseases guidelines for coronavirus disease 2019: an update on treatment of patients with mild/moderate disease Clin Microbiol Infect 2022 10.1016/j.cmi.2022.08.013 27. National Institute for Health and Care Excellence. Draft guidance consultation: therapeutics for people with COVID-19. 2022. https://nice.org.uk/guidance/gid-ta10936/documents/129. Accessed 18 Nov 2022. 28. Wen W Chen C Tang J Efficacy and safety of three new oral antiviral treatment (molnupiravir, fluvoxamine and Paxlovid) for COVID-19: a meta-analysis Ann Med 2022 54 1 516 523 10.1080/07853890.2022.2034936 35118917 29. Jo Y Kim SB Radnaabaatar M Model-based cost-effectiveness analysis of oral antivirals against SARS-CoV-2 in Korea Epidemiol Health. 2022 10.4178/epih.e2022034 30. Pepperrelll T Ellis L Wang J Barriers to worldwide access for Paxlovid, a new treatment for COVID-19 Open Forum Infect Dis 2022 10.1093/ofid/ofac174 31. Millstein JH Asch DA Hamilton K Decision support and centralized pharmacy consultation for nirmatrelvir-ritonavir prescribing in an academic health system-a model to promote drug access and reduce provider burden J Gen Intern Med 2022 38 4028 4031 10.1007/s11606-022-07752-6	0	PMC9734504	NO-CC CODE	2022-12-14 23:28:29	no	Drugs Ther Perspect. 2022 Dec 6;:1-7	utf-8	Drugs Ther Perspect	2,022	10.1007/s40267-022-00971-1	oa_other
==== Front Diabetol Int Diabetol Int Diabetology international 2190-1678 2190-1686 Springer Nature Singapore Singapore 609 10.1007/s13340-022-00609-7 Original Article Perceptions of diabetes management among adolescents with type 1 diabetes and their caregivers: development and validation of the Japanese version of the diabetes family responsibility questionnaire http://orcid.org/0000-0002-2552-6733 Matsumoto Hiro [email protected] 13 Nio Kaori 2 Kawamura Tomoyuki 3 Obayashi Yoko 4 Hotta Yuko 5 Yuyama Yoshihiko 3 Nishikawa Naoko 3 1 grid.260026.0 0000 0004 0372 555X Course of Nursing Science, Graduate School of Medicine, Mie University, Tsu, Mie Japan 2 Graduate School of Nursing, Osaka Metropolitan University, Habikino, Osaka Japan 3 Department of Pediatrics, Graduate School of Medicine, Osaka Metropolitan University, Abeno-Ku, Osaka, Japan 4 grid.443092.8 0000 0004 7433 9955 Department of Nursing, School of Health Sciences, Toyohashi Sozo University, Toyohashi, Aichi Japan 5 Kashiwara Municipal Hospital, Kashiwara, Osaka Japan 7 12 2022 110 24 5 2022 24 11 2022 © The Japan Diabetes 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 shift in diabetes management responsibility is critical for adolescents with type 1 diabetes (T1D). Currently, in Japan, there is insufficient progress in the development of scales for evaluating diabetes management responsibility. We developed the Japanese version of the Diabetes Family Responsibility Questionnaire (DFRQ), a scale to evaluate diabetes management responsibility, and verified its reliability and validity. We recruited 12–18-year-old adolescents with T1D and their caregivers. The DFRQ questionnaires (DFRQ-A for adolescents and DFRQ-C for caregivers) were distributed. The responses of 31 pairs were analyzed (adolescents: 9 males, 22 females; mean age: 14.8 ± 1.5 years). The median total DFRQ scores of adolescents (30.0) and caregivers (32.0) were not significantly different (p = 0.269). The internal consistencies (Cronbach’s α) were 0.784 and 0.687 for DFRQ-A and DFRQ-C, respectively. DFRQ-A scores and adolescent age demonstrated a weak statistically significant negative correlation (r = − 0.397, p = 0.027), whereas DFRQ-C scores and adolescent age demonstrated a weak negative correlation not statistically significant (r = − 0.311, p = 0.089). Both scores were significantly negatively correlated with self-efficacy for diabetes self-management scores (r = − 0.390, p = 0.030; r = − 0.478, p = 0.006, respectively). Furthermore, a significantly moderate positive correlation was found between these scores (r = 0.624, p < 0.001). We confirmed the reliability and validity of the Japanese version of DFRQ. DFRQ is expected to be used as a dyadic scale to evaluate the status of diabetes management responsibility and its transition during adolescence in Japan. Keywords Type 1 diabetes Diabetes management Adolescents Caregivers Transition JSPS (Japan Society for the Promotion of Science) KAKENHIJP18K17563 Matsumoto Hiro ==== Body pmcIntroduction Type 1 diabetes (T1D) is an idiopathic autoimmune disease caused by the destruction of pancreatic beta cells, resulting in absolute insulin deficiency and thereby requiring lifelong insulin administration [1]. The management of T1D involves several tasks, including checking blood glucose levels, administering insulin, monitoring food intake, responding to symptoms of hyper- and hypoglycemia, and attending medical visits [2, 3]. The treatment of T1D has recently become more complex. A wide variety of treatment methods have become available, including multiple daily insulin injections, continuous subcutaneous insulin infusion, continuous glucose monitoring, sensor-augmented insulin pump, and other options. Thus, the management of T1D is very complex due to the many daily tasks involved and the wide variety of potential treatment options. In the case of type 1 diabetes, adolescence is a challenging period [4, 5], where it is a period of transition from childhood to adulthood when rapid physical and psychosocial changes occur [6]. In terms of diabetes management, adolescents also experience a transition from receiving care from caregivers to self-management [7]. Therefore, the caregivers’ responsibility for diabetes management is reduced, whereas that of adolescents with T1D is increased. Thus, there is a need to find a new balance for diabetes management between adolescents and their families, particularly caregivers [8, 9]. The importance of caregivers in diabetes management responsibility has been highlighted; however, the responsibilities of adolescents with T1D and their caregivers have not been sufficiently studied in Japan. In addition, a diabetes-specific dyadic scale that can be used to assess adolescents and their caregivers has not been developed. Recently, the transition has become increasingly important for patients with chronic diseases, such as T1D. The American Diabetes Association stated that the age range of 12–18 years is the starting point for transition planning [10]. Furthermore, in Japan, the period from 12 to 18 years is considered an important period for acquiring transition readiness [11]; that is to say, the shift in diabetes management responsibility that occurs during adolescence is an important first step in the transition. This important step must be evaluated not only from the perspective of the adolescents but also from that of the caregivers. The Diabetes Family Responsibility Questionnaire (DFRQ) can be used for such evaluation [12]. Both adolescents and caregivers complete equivalent forms and assign the DFRQ. For each task item in the questionnaire, one of the following values is assigned: 1 (the adolescent takes or initiates responsibility for this task at almost all times), 2 (the caregiver and adolescent share responsibility for this task almost equally), or 3 (the caregiver takes or initiates responsibility for this task at almost all times). A lower DFRQ score indicates that adolescents bear diabetes management responsibility. DFRQ has been used in various countries [13–17] and has been further modified according to changes in diabetes treatment [7]. However, a Japanese version of DFRQ has not yet been developed, and no previous studies in Japan have objectively investigated the division of responsibility for diabetes management between adolescents and their caregivers. This study aimed to develop the Japanese version of DFRQ and investigate its reliability and validity. Materials and methods Participants This cross-sectional study’s subjects were adolescents aged 12–18 years (7th–12th grades) and their caregivers who were recruited from the outpatient department of a Japanese general hospital between July and November 2021. All the adolescents had been diagnosed with T1D for at least 6 months and were able to answer the questionnaire independently. Those who had difficulty participating in this survey or in independently answering the questionnaire were excluded. Background data The adolescent subjects provided information about their age, school age, age at T1D diagnosis, sex, insulin administration method, blood glucose measurement method, the name of the insulin they were using, the experience of attending diabetes camps, their most-recent hemoglobin A1c (HbA1c) level, and their family structure. The adolescents’ caregivers provided information about their relationship with the adolescent, their age, their family structure, and the adolescents’ demographic variables (age, school age, age at T1D diagnosis, sex, insulin administration method, blood glucose monitoring method, and the name of the insulin used by the adolescent). Diabetes-specific self-efficacy scale To measure their perceptions of self-efficacy, the adolescents completed the Japanese version of the Self-Efficacy for Diabetes Self-Management (SEDM) questionnaire developed by Iannotti et al. [18]. The Japanese version was also developed by Sekiguchi et al. [19]. The SEDM consists of 10 items that are responded to using a 10-point Likert scale. Total SEDM scores range from 0 to 100, with a higher score indicating greater self-efficacy for diabetes self-management. The SEDM questionnaire was completed by the adolescents only. The Japanese version of DFRQ To develop the Japanese version of DFRQ, we first obtained permission from its author, Dr. Anderson. We also referred to COSMIN guidelines for scale development [20]. We then translated the original version of the questionnaire. The forward translation was performed by three researchers (one pediatrician and two nurses) who were native Japanese speakers familiar with T1D. Subsequently, the content was checked among the researchers and a consensus was reached. Subsequently, the Japanese version was back-translated into English by a bilingual translator. This version was checked by the original author, after which, the preparation of the Japanese version of DFRQ was complete. Content validity The content validity of the Japanese version of DFRQ was assessed by five medical professionals (two pediatricians and three nurses) using the content validation index (CVI) [21]. This index allows the quantitative evaluation of the tested scale. A 4-point scale (1 = not relevant, 4 = very relevant) was used to evaluate the relevance of the questionnaire items to diabetes management. For each item, the mean CVI was calculated by dividing the total score for each item by the number of raters. The CVI of the whole scale was calculated as the average of each item’s CVI. The cutoff points for the CVI were 0.78 or higher for each item and 0.9 or higher for the whole scale [22]. The Japanese version of DFRQ had CVIs of ≥ 0.8 for each item and 0.98 for the whole scale, indicating content validity. The Japanese version of DFRQ was administered to adolescents with T1D and their caregivers. For convenience, their scores are hereafter referred to as DFRQ-A and DFRQ-C scores, respectively. Known-group validity and convergent validity We quantitatively assessed for known-group validity and convergent validity by testing the following three hypotheses:Adolescents bear more responsibility for diabetes management as they get older (i.e., the older adolescents have lower total DFRQ-A and DFRQ-C scores). Adolescents with lower total DFRQ-A scores have higher total SEDM scores. Adolescents with lower total DFRQ-C scores have higher total SEDM scores. Higher total scores on SEDM indicate greater self-efficacy in diabetes self-management. The scores on the DFRQ scale (translated in this study) indicate that adolescents with lower total scores have more responsibility for diabetes management, i.e., self-management. Thus, both scales provide information about the self-management status of T1D. Particularly in Japan, where the number of patients with T1D is small, a scale reflecting the self-management status of adolescents with T1D will be highly valuable. Therefore, this study used SEDM to assess the convergent validity of DFRQ. Adolescents’ and caregivers’ perceptions of diabetes management responsibility We performed a correlation analysis of the DFRQ-A and DFRQ-C scores to confirm the similarity between adolescents’ and caregivers’ perceptions of the DFRQ based on the study by Cameron et al. [23], which confirms the correlation between adolescents’ and caregivers’ DFRQ. Results The questionnaire was distributed to 48 pairs of adolescents and their caregivers who agreed to participate in the study; 33 paired responses were received, of which 31 valid responses were included in the analysis. Participant characteristics The participants’ characteristics are presented in Table 1. In terms of demographic and diabetes-related data, 9 adolescents were males and 22 were females; their average age was 14.8 ± 1.5 years. Further, 16 of them were junior high school students (5 in 7th grade, 6 in 8th grade, and 5 in 9th grade), and 15 were high school students (5 in 10th grade, 8 in 11th grade, and 2 in 12th grade). Their mean age at T1D onset was 8.4 ± 3.6 years, and the mean duration of T1D was 6.4 ± 3.7 years. The method of insulin administration was pen in 7 subjects (22.6%), insulin pump in 16 subjects (54.8%), and a combination of pen and insulin pump in 7 subjects (22.6%). The name of the insulin they were using was known by 26 subjects (83.9%). Further, 21 subjects (67.7%) had participated in a diabetes camp. Their mean HbA1c level (an indicator of glycemic control) was 8.0 ± 1.4% (range, 6.2–13.0%).Table 1 Patient characteristics (N = 31) N or mean % Adolescents Sex Male 9 29.0 Female 22 71.0 Age (years) 14.8 Grade Junior high school (7th–9th grades) 16 51.6 High school (10th–12th grades) 15 48.4 Age at onset (years) 8.4 Time of onset (years) 0–9 18 58.1 ≥ 10 13 41.9 Method of insulin administration Pen 7 22.6 Pump 17 54.8 Pen and pump 7 22.6 Knowledge of the name of the insulin Yes 26 83.9 No 5 16.1 HbA1c (%) 8.0 Glycemic control HbA1c < 7.5% 11 35.5 HbA1c ≥ 7.5% 20 64.5 Caregivers Respondent Father 2 6.5 Mother 28 90.3 Aunt 1 3.2 Age 30 s 2 6.5 40 s 21 67.7 50 s 8 25.8 Family structure Parents 27 87.1 Single parent 4 12.9 The adolescents’ caregivers included 2 fathers, 28 mothers, and 1 aunt. Two caregivers were in their 30 s, 21 were in their 40 s, and 8 were in their 50 s. There were 27 two-parent caregiving families and four single-parent caregiving families. Characteristics of the DFRQ The distributions of responses and total scores for DFRQ are shown in Table 2. Eleven of the questionnaire’s items were perceived as involving proactive management by the adolescent, including item 3, “Remembering to take insulin injections or boluses.” Three items were perceived as initiatives of the caregiver, including item 6, “Making doctor appointments.” Both the adolescents and their caregivers perceived the following items as involving collaborative management: 2, “Telling teachers about diabetes;” 5, “Checking expiration dates on medical supplies;” and 18, “Giving extra attention to diabetes management on sick days.”Table 2 Distribution of responses and total scores for DFRQ (N = 31) Median Responsibility p-value Adolescent Share Caregivers n % n % n % ① Remembering day of clinic appointment DFRQ-A 3 1 3.2 14 45.2 16 51.6 0.564 DFRQ-C 3 1 3.2 13 41.9 17 54.8 ② Telling teachers about diabetes DFRQ-A 2 4 12.9 16 51.6 11 35.5 0.791 DFRQ-C 2 3 9.7 17 54.8 11 35.5 ③ Remembering to take insulin injections or boluses DFRQ-A 1 26 83.9 5 16.1 0 0.0 0.655 DFRQ-C 1 25 80.6 6 19.4 0 0.0 ④ Giving insulin injections or boluses before eating DFRQ-A 1 29 93.5 2 6.5 0 0.0 0.157 DFRQ-C 1 31 100.0 0 0.0 0 0.0 ⑤ Checking expiration dates on medical supplies DFRQ-A 2 3 9.7 14 45.2 14 45.2 0.160 DFRQ-C 2 7 22.6 11 35.5 13 41.9 ⑥ Making doctor appointments DFRQ-A 3 5 16.1 4 12.9 22 71.0 0.102 DFRQ-C 3 5 16.1 8 25.8 18 58.1 ⑦ Telling relatives about diabetes DFRQ-A 3 1 3.2 13 41.9 17 54.8 0.096 DFRQ-C 3 1 3.2 8 25.8 22 71.0 ⑧ Taking more or less insulin according to results of blood sugar monitoring DFRQ-A 1 21 67.7 9 29.0 1 3.2 0.705 DFRQ-C 1 22 71.0 8 25.8 1 3.2 ⑨ Noticing differences in health, such as weight changes or signs of an infection DFRQ-A 1 25 80.6 5 16.1 1 3.2 0.003 DFRQ-C 2 13 41.9 15 48.4 3 9.7 ⑩ Deciding what to eat at meals or snacks DFRQ-A 1 26 83.9 2 6.5 3 9.7 0.236 DFRQ-C 1 20 64.5 9 29.0 2 6.5 ⑪ Telling friends about diabetes DFRQ-A 1 28 90.3 3 9.7 0 0.0 0.317 DFRQ-C 1 30 96.8 1 3.2 0 0.0 ⑫ Noticing the early signs of low blood sugar DFRQ-A 1 30 96.8 0 0.0 1 3.2 0.257 DFRQ-C 1 27 87.1 3 9.7 1 3.2 ⑬ Deciding what should be eaten when family meals out DFRQ-A 1 27 87.1 3 9.7 1 3.2 0.527 DFRQ-C 1 28 90.3 3 9.7 0 0.0 ⑭ Carrying some form of sugar in case of low blood sugar DFRQ-A 1 21 67.7 10 32.3 0 0.0 0.096 DFRQ-C 1 26 83.9 5 16.1 0 0.0 ⑮ Explaining absences from school to teachers or other school personnel DFRQ-A 3 9 29.0 5 16.1 17 54.8 0.396 DFRQ-C 2 3 9.7 13 41.9 15 48.4 ⑯ Rotating injection sites or pump infusion sites DFRQ-A 1 26 83.9 3 9.7 2 6.5 0.317 DFRQ-C 1 27 87.1 4 12.9 0 0.0 ⑰ Remembering times when blood sugar should be checked DFRQ-A 1 23 74.2 3 9.7 5 16.1 0.034 DFRQ-C 1 25 80.6 5 16.1 1 3.2 ⑱ Giving extra attention to diabetes management on sick days DFRQ-A 2 13 41.9 15 48.4 3 9.7 0.046 DFRQ-C 2 6 19.4 21 67.7 4 12.9 ⑲ Making insulin and/or food adjustments for exercise DFRQ-A 1 27 87.1 3 9.7 1 3.2 0.157 DFRQ-C 1 22 71.0 9 29.0 0 0.0 Total score DFRQ-A 30 IQR: 28–33 0.269 DFRQ-C 32 IQR: 29–33 Wilcoxon’s signed-rank test We used Wilcoxon’s signed-rank test to compare the adolescents’ and their caregivers’ perceptions of each item. Significant differences were observed for the following items: 9, “Noticing differences in health, such as weight changes or signs of an infection” (p = 0.003); 17, “Remembering times when blood sugar should be checked” (p = 0.034); and 18, “Giving extra attention to diabetes management on sick days” (p = 0.046). DFRQ total scores The median (interquartile range) total DFRQ-A score was 30.0 (28–33) and the median total DFRQ-C score was 32.0 (29–33), which were not significantly different (p = 0.269) (Table 2). SEDM total scores The median (interquartile range) SEDM total score was 73 (57–82). Comparison of DFRQ scores by time of onset and method of insulin administration DFRQ scores were compared by time of onset and method of insulin administration, and no significant difference was found in terms of the time of onset. Conversely, significant differences were found in the DFRQ-A scores in terms of the insulin administration method. Multiple comparisons using Bonferroni correction on the DFRQ-A scores revealed p = 0.025 for pump vs. pen and pump (Table 3).Table 3 Comparison of DFRQ scores by time of onset and method of insulin administration aMann–Whitney U test bKruskal–Wallis test †Bonferroni used for multiple comparisons Verifying the reliability and validity of DFRQ Internal consistency The internal consistencies (Cronbach’s α) were 0.784 for DFRQ-A and 0.687 for DFRQ-C. Verifying hypotheses To confirm the known-group validity and convergent validity, each factor was tested using Spearman’s rank correlation coefficient. The results of this verification are shown in Table 4.Table 4 Spearman’s rank correlation between DFRQ-A, DFRQ-C, SEDM, age, and HbA1c levels Age HbA1c SEDM DFRQ-C r p-value r p-value r p-value R p-value DFRQ-A − 0.397 0.027 0.129 0.491 − 0.390 0.030 0.624 0.000 DFRQ-C − 0.311 0.089 0.226 0.221 − 0.478 0.006 SEDM 0.380 0.035 − 0.449 0.011 Known-group validity We tested Hypothesis 1 and found a significant negative correlation between the DFRQ-A scores and the age of the adolescents (r = − 0.397, p = 0.027) and a nonsignificant negative correlation between the DFRQ-C scores and the age of the adolescents (r = − 0.311, p = 0.089). Convergent validity Hypotheses 2 and 3 were tested and a significantly weak negative correlation between DFRQ-A and SEDM scores was observed (r = − 0.390, p = 0.030). A significantly moderate negative correlation was found between DFRQ-C and SEDM scores (r = − 0.478, p = 0.006). Adolescents’ and caregivers’ perceptions of diabetes management responsibility Correlation analysis of adolescents’ and caregivers’ perceptions of diabetes management responsibility revealed a significantly moderate positive correlation between DFRQ-A and DFRQ-C scores (r = 0.624, p < 0.001). Discussion We developed the Japanese version of DFRQ and tested its reliability and validity to evaluate whether it can be used to assess Japanese adolescents with T1D and their caregivers. We obtained two main findings. First, we confirmed the reliability and validity of the Japanese version of DFRQ. Second, we confirmed the clinical usefulness of this DFRQ. Reliability and validity of the Japanese version of DFRQ We tested the reliability, content validity, known-group validity, and convergent validity of the Japanese version of DFRQ. Its internal consistency was adequate and comparable to that reported in previous studies [7, 24, 25]. Known-group validity was confirmed by testing Hypothesis 1, and we found no significant difference between DFRQ-C scores and the age of the adolescents, but both DFRQ-A and DFRQ-C scores were negatively correlated with the age of the adolescents. Campbell et al. [25] reported that the total DFRQ scores of both young adults and caregivers decreased over time. In the assessment of adolescents’ readiness for diabetes self-management from the perspective of adolescents and their caregivers, Goethals et al. [8] showed that the adolescents’ readiness increased with age. Furthermore, research has shown that parental involvement in diabetes management decreases as adolescents age [26]. The results of the present study are consistent with those of previous studies. We found that adolescents with T1D in Japan increasingly proactively manage their diabetes as they age. Therefore, the results of this study are similar to those of previous studies, indicating that Japanese adolescents with T1D become more proactive in managing their diabetes as they get older. A previous study using DFRQ [24] showed that the transition of the responsibility of diabetes management rapidly changed around the age of 10 years. Furthermore, a qualitative study by Schilling et al. [27] reported that the transition of the responsibility of diabetes management occurred during late childhood. Based on these findings, it is likely that there was no significant correlation between DFRQ-C scores and the age of the adolescents because there was less transition of the responsibility of diabetes management in adolescence. Future research on diabetes management responsibilities should include children in late childhood and their caregivers to better understand the process of transition, which will provide better insights into longitudinal changes. Convergent validity was confirmed by testing hypotheses 2 and 3. Both DFRQ-A and DFRQ-C scores were significantly negatively correlated with SEDM scores, validating an association between the transition of diabetes management responsibility and higher self-efficacy for diabetes self-management behaviors. Self-efficacy is a concept proposed by Bandura [28–30], who defined it as an individual’s belief in his or her capacity to execute the behaviors necessary to produce specific performance attainments [31]. Self-efficacy is very important in the acquisition of complex health behaviors, such as diabetes management [18]. Previous studies have shown that self-efficacy for diabetes self-management behaviors is an important factor influencing adherence and metabolic control [32, 33]. The results of this analysis revealed that both DFRQ-A and DFRQ-C were significantly negatively correlated with SEDM, suggesting that adolescents with high diabetes self-management performance not only from the adolescent’s perspective but also from the caregiver’s perspective have high self-efficacy for diabetes self-management behaviors. Our analyses confirm the reliability and validity of the Japanese version of DFRQ. Clinical significance Recently, the importance of health care transitions has been recognized in Japan as well. There are two aspects to the transition: (1) transfer from pediatric to adult medicine and (2) adolescents themselves assuming the responsibility for disease management [34]. It is particularly important that adolescents with chronic diseases, such as T1D, assume the responsibility for disease management, and the cooperation of family members, especially caregivers, is essential as adolescents assume responsibility for disease management. We found the adolescents and their caregivers equally aware of the responsibility of diabetes management. Therefore, caregivers are important individuals who regularly understand the diabetes management situation in adolescents. Therefore, we, as health care providers, need to support the adolescents, but we must not forget the need to support the caregivers who support the adolescents as well. According to studies in other countries, providing support not only to adolescents but also to their families improves family relationships with adolescents [35] and also reduces hospitalizations due to complications [36]. Therefore, in adolescence with diabetes management responsibilities, support for not only the adolescents but also their caregivers (family support) should be considered. The incidence of T1D is lower in Japan and other Asian countries than in Europe and the USA [37, 38]. Therefore, there is currently insufficient knowledge about transitions and family support in Japan. Considering this, the Japanese version of DFRQ is invaluable as a dyadic assessment scale for evaluating the first step of transition in diabetes management responsibility. In particular, the transfer of responsibility for diabetes management from caregivers to adolescents during adolescence is the first step in the transition. At this developmental stage, the use of DFRQ in a clinical setting can help visualize the different transition stages, thereby facilitating the development of various strategies for transitional support, including multidisciplinary discussions and enhanced approaches. Our study revealed that although many adolescents recognized their responsibility for routine diabetes management tasks, not many recognized that they were responsible for nonroutine tasks, such as dealing with sick days and remembering medical appointments. In particular, there are challenges in the transition from pediatric medical services to adult medical services, such as a decrease in the frequency of diabetes-related visits and an increase in the rate of hospitalization due to diabetic ketoacidosis [39, 40]. Thus, enabling adolescents to learn to manage diabetes will allow them to manage their diabetes more safely when they become adults. Thus, using DFRQ, appropriate support can be provided during adolescence to ensure independence in diabetes management during adulthood. Both DFRQ-A and DFRQ-C scores were associated with self-efficacy for diabetes self-management behaviors. This suggests that the transfer of diabetes management responsibility to adolescents is associated with the adolescents’ confidence in diabetes self-management, which may manifest as increased self-efficacy for self-management behaviors. Self-efficacy does not arise spontaneously and can be enhanced through experience, modeling, social persuasion, and physiological factors. Considering that experience is the most important factor for enhancing self-efficacy, it can be considered that adolescents taking responsibility for their diabetes management is a crucial factor in enhancing self-efficacy in diabetes self-management. Although significant differences in DFRQ-A scores by insulin administration method (pen vs. pen and pump) were found in the present study, the lack of sufficient knowledge concerning the relationship between insulin administration method and responsibility for diabetes management requires further investigation based on these relationships in the future. Additionally, the Japanese version of DFRQ was not directly associated with HbA1c levels (an index of metabolic control). Thus, further comprehensive examination of such associations, including those among diabetes outcomes, such as time-in-range and hospitalization rates as well as psychosocial factors associated with diabetes, is warranted. Study limitations This study was conducted at a single medical institution when the coronavirus disease 2019 pandemic was spreading in Japan. Hence, the number of eligible participants was small. In the future, we aim to increase the number of target institutions and participants. Conclusions We developed and validated the Japanese version of DFRQ and verified its reliability and validity. We showed that it was a useful dyadic assessment scale for evaluating diabetes management responsibility from the perspectives of both adolescents with T1D and their caregivers. Acknowledgements We would like to thank Enago (www.enago.jp) for the English language review. Author contributions Conceptualization: HM; methodology: HM, KN, YO; formal analysis and investigation: HM, KN, TK, YY, YH; writing—original draft preparation: HM; writing—review and editing: HM, KN, YO, TK, YY, YH; funding acquisition: HM; resources: TK, YY, YH, NN; supervision: KN, YO, TK, YY, YH, NN. Funding This study was supported by the JSPS (Japan Society for the Promotion of Science) KAKENHI Grant Number JP18K17563. The funders had no role in study design, data collection and analysis, publishing decision, or manuscript preparation. Data availability All relevant data supporting the finding of this study are available from the corresponding author upon request. Declarations Conflicts of interest Hiro Matsumoto, Kaori Nio, Tomoyuki Kawamura, Yoko Obayashi, Yuko Hotta, Yoshihiko Yuyama, and Naoko Nishikawa declare that they have no conflict of interest. Ethical approval This study was conducted following the Declaration of Helsinki. The procedures were approved by the Institutional Review Boards of the two study sites the Osaka City University [2021-113] approval date: June 17, 2021 and the Mie University [U2021-031] approval date: September 15, 2021. The researchers explained the study to the subjects, both orally and in writing, and obtained their written consent for participation. It was explained that their participation was not obligatory and that they could withdraw at any time. Human rights All procedures followed were conducted following the ethical standards of the responsible committee on human experimentation by the Institutional Review Boards of the two study sites (Osaka City University [2021-113] and Mie University [U2021-031]) and the Helsinki Declaration of 1964 and later versions. Animal studies This article does not contain any studies with animal subjects performed by any of the authors. Informed consent Informed consent or substitute was obtained from all patients included in the study. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Hiro Matsumoto, Kaori Nio, Tomoyuki Kawamura and Yoko Obayashi have contributed equally to this work. Yuko Hotta, Yoshihiko Yuyama and Naoko Nishikawa also contributed equally to this work. ==== Refs References 1. American Diabetes Association Classification and diagnosis of diabetes: standards of medical care in diabetes-2021 Diabetes Care 2021 44 Suppl 1 S15 S33 10.2337/dc21-S002 33298413 2. Marker AM Noser AE Clements MA Patton SR Shared responsibility for Type 1 diabetes care is associated with glycemic variability and risk of glycemic excursions in youth J Pediatr Psychol 2018 43 61 71 10.1093/jpepsy/jsx081 28541572 3. Atkinson MA Eisenbarth GS Michels AW Type 1 diabetes Lancet 2014 383 69 82 10.1016/S0140-6736(13)60591-7 23890997 4. Danne T Mortensen HB Hougaard P Lynggaard H Aanstoot HJ Chiarelli F Daneman D Dorchy H Garandeau P Greene SA Hoey H Holl RW Kaprio EA Kocova M Martul P Matsuura N Robertson KJ Schoenle EJ Søvik O Swift PG Tsou RM Vanelli M Aman J Hvidøre Study Group on Childhood Diabetes. Persistent differences among centers over 3 years in glycemic control and hypoglycemia in a study of 3805 children and adolescents with type 1 diabetes from the Hvidøre Study Group Diabetes Care. 2001 24 1342 7 10.2337/diacare.24.8.1342 11473067 5. Helgeson VS Reynolds KA Siminerio L Escobar O Becker D Parent and adolescent distribution of responsibility for diabetes self-care: links to health outcomes J Pediatr Psychol 2008 33 497 508 10.1093/jpepsy/jsm081 17848390 6. Spear HJ Kulbok P Autonomy and adolescence: a concept analysis Public Health Nurs 2004 21 144 152 10.1111/j.0737-1209.2004.021208.x 14987214 7. Vesco AT Anderson BJ Laffel LM Dolan LM Ingerski LM Hood KK Responsibility sharing between adolescents with type 1 diabetes and their caregivers: importance of adolescent perceptions on diabetes management and control J Pediatr Psychol 2010 35 1168 1177 10.1093/jpepsy/jsq038 20444852 8. Goethals ER Commissariat PV Volkening LK Markowitz JT Laffel LM Assessing readiness for independent self-care in adolescents with type 1 diabetes: Introducing the RISQ Diabetes Res Clin Pract 2020 162 108110 10.1016/j.diabres.2020.108110 32194216 9. de Wit M Versloot J Zenlea I Goethals ER Using person-reported outcomes (PROs) to motivate young people with diabetes Curr Diab Rep 2020 20 23 10.1007/s11892-020-01305-z 32415346 10. Chiang JL Kirkman MS Laffel LM Peters AL Type 1 diabetes sourcebook authors. Type 1 diabetes through the life span: a position statement of the American Diabetes Association Diabetes Care. 2014 37 2034 54 10.2337/dc14-1140 24935775 11. Kato H. About the medical assistance model project for children in period of transition to adulthood and affected by infantile chronic diseases (in Japanese); 2017. https://www.mhlw.go.jp/file/05-Shingikai-12601000-Seisakutoukatsukan-Sanjikanshitsu_Shakaihoshoutantou/0000170303.pdf 12. Anderson BJ Auslander WF Jung KC Miller JP Santiago JV Assessing family sharing of diabetes responsibilities J Pediatr Psychol 1990 15 477 492 10.1093/jpepsy/15.4.477 2258796 13. Malik JA Koot HM Assessing diabetes support in adolescents: factor structure of the modified diabetes social support questionnaire (DSSQ-friends) Diabet Med 2012 29 e232 e240 10.1111/j.1464-5491.2012.03677.x 22486356 14. Law GU Walsh J Queralt V Nouwen A Adolescent and parent diabetes distress in type 1 diabetes: the role of self-efficacy, perceived consequences, family responsibility and adolescent-parent discrepancies J Psychosom Res 2013 74 334 339 10.1016/j.jpsychores.2012.12.009 23497836 15. Katz ML Volkening LK Butler DA Anderson BJ Laffel LM Family-based psychoeducation and care ambassador intervention to improve glycemic control in youth with type 1 diabetes: a randomized trial Pediatr Diabetes 2014 15 142 150 10.1111/pedi.12065 23914987 16. Christie D Thompson R Sawtell M Allen E Cairns J Smith F Jamieson E Hargreaves K Ingold A Brooks L Wiggins M Oliver S Jones R Elbourne D Santos A Wong IC O’Neil S Strange V Hindmarsh P Annan F Viner RM Effectiveness of a structured educational intervention using psychological delivery methods in children and adolescents with poorly controlled type 1 diabetes: a cluster-randomized controlled trial of the Cascade intervention BMJ Open Diabetes Res Care 2016 4 e000165 10.1136/bmjdrc-2015-000165 17. Vloemans AF Eilander MMA Rotteveel J Bakker-van Waarde WM Houdijk ECAM Nuboer R Winterdijk P Snoek FJ De Wit M youth with type 1 diabetes taking responsibility for self-management: the importance of executive functioning in achieving glycemic control: results from the longitudinal DINO study Diabetes Care 2019 42 225 231 10.2337/dc18-1143 30552132 18. Iannotti RJ Schneider S Nansel TR Haynie DL Plotnick LP Clark LM Sobel DO Simons-Morton B Self-efficacy, outcome expectations, and diabetes self-management in adolescents with type 1 diabetes J Dev Behav Pediatr 2006 27 98 105 10.1097/00004703-200604000-00003 16682872 19. Sekiguchi M Ando T Takagaki K Kawamura T Hashimoto T Kashihara Y Sakano Y Psychological factors related with self-management among childhood and adolescent with type 1 diabetes: focusing on self-efficacy (in Japanese) Jpn J Psychosom Med 2013 53 857 864 20. Mokkink LB, Prinsen CA, Patrick DL, Alonso J, Bouter LM, de Vet HC, Terwee CB. COSMIN study design checklist for patient-reported outcome measurement instruments. Version; 2019. https://www.cosmin.nl/wp-content/uploads/COSMIN-study-designing-checklist_final.pdf. 21. Lynn MR Determination and quantification of content validity Nurs Res 1986 35 382 385 10.1097/00006199-198611000-00017 3640358 22. Polit DF Beck CT Owen SV Is the CVI an acceptable indicator of content validity? Appraisal and recommendations Res Nurs Health 2007 30 459 467 10.1002/nur.20199 17654487 23. Cameron FJ Skinner TC De Beaufort CE Hoey H Swift PG Aanstoot H Åman J Martul P Chiarelli F Daneman D Danne T Dorchy H Kaprio EA Kaufman F Kocova M Mortensen HB Njølstad PR Phillip M Robertson KJ Schoenle EJ Urakami T Vanelli M Ackermann RW Skovlund SE Hvidoere study group on childhood diabetes. are family factors universally related to metabolic outcomes in adolescents with type 1 diabetes? Diabet Med. 2008 25 463 8 10.1111/j.1464-5491.2008.02399.x 18294223 24. Anderson BJ Holmbeck G Iannotti RJ McKay SV Lochrie A Volkening LK Laffel L Dyadic measures of the parent-child relationship during the transition to adolescence and glycemic control in children with type 1 diabetes Fam Syst Health 2009 27 141 152 10.1037/a0015759 19630455 25. Campbell MS Wang J Cheng Y Cogen FR Streisand R Monaghan M Diabetes-specific family conflict and responsibility among emerging adults with type 1 diabetes J Fam Psychol 2019 33 788 796 10.1037/fam0000537 31021129 26. King PS Berg CA Butner J Butler JM Wiebe DJ Longitudinal trajectories of parental involvement in Type 1 diabetes and adolescents’ adherence Health Psychol 2014 33 424 432 10.1037/a0032804 23795709 27. Schilling LS Knafl KA Grey M Changing patterns of self-management in youth with type I diabetes J Pediatr Nurs 2006 21 412 424 10.1016/j.pedn.2006.01.034 17101399 28. Bandura A Self-efficacy: toward a unifying theory of behavioral change Psychol Rev 1977 84 191 215 10.1037/0033-295X.84.2.191 847061 29. Bandura A Social foundations of thought and action: a social cognitive theory 1986 Englewood Cliffs Prentice Hall 30. Bandura A Self-efficacy the exercise of control 1997 New York W. H. Freeman 31. Carey MP, Forsyth AD. Teaching tip sheet: self-efficacy; 2009. https://www.apa.org/pi/aids/resources/education/self-efficacy. American Psychological Association 32. Berg CA King PS Butler JM Pham P Palmer D Wiebe DJ Parental involvement and adolescents’ diabetes management: the mediating role of self-efficacy and externalizing and internalizing behaviors J Pediatr Psychol 2011 36 329 339 10.1093/jpepsy/jsq088 20926405 33. Wiebe DJ Chow CM Palmer DL Butner J Butler JM Osborn P Berg CA Developmental processes associated with longitudinal declines in parental responsibility and adherence to type 1 diabetes management across adolescence J Pediatr Psychol 2014 39 532 541 10.1093/jpepsy/jsu006 24602891 34. Sable C Foster E Uzark K Bjornsen K Canobbio MM Connolly HM Graham TP Gurvitz MZ Kovacs A Meadows AK Reid GJ Reiss JG Rosenbaum KN Sagerman PJ Saidi A Schonberg R Shah S Tong E Williams RG American heart association congenital heart defects committee of the council on cardiovascular disease in the young, council on cardiovascular nursing, council on clinical cardiology, and council on Peripheral Vascular Disease. Best practices in managing transition to adulthood for adolescents with congenital heart disease: the transition process and medical and psychosocial issues: a scientific statement from the American Heart Association Circulation 2011 123 1454 85 10.1161/CIR.0b013e3182107c56 21357825 35. Wysocki T Harris MA Buckloh LM Mertlich D Lochrie AS Taylor A Sadler M White NH Randomized, controlled trial of behavioral family systems therapy for diabetes: maintenance and generalization of effects on parent-adolescent communication Behav Ther 2008 39 33 46 10.1016/j.beth.2007.04.001 18328868 36. Ellis D Naar-King S Templin T Frey M Cunningham P Sheidow A Cakan N Idalski A Multisystemic therapy for adolescents with poorly controlled type 1 diabetes: reduced diabetic ketoacidosis admissions and related costs over 24 months Diabetes Care 2008 31 1746 1747 10.2337/dc07-2094 18566340 37. Onda Y Sugihara S Ogata T Yokoya S Yokoyama T Tajima N Type 1 Diabetes (T1D) Study Group. Incidence and prevalence of childhood-onset Type 1 diabetes in Japan: the T1D study Diabet Med. 2017 34 909 15 10.1111/dme.13295 27925270 38. Mayer-Davis EJ Kahkoska AR Jefferies C Dabelea D Balde N Gong CX Aschner P Craig ME ISPAD Clinical Practice Consensus Guidelines 2018: definition, epidemiology, and classification of diabetes in children and adolescents Pediatr Diabetes 2018 19 Suppl 27 7 19 10.1111/pedi.12773 39. Shulman R Shah BR Fu L Chafe R Guttmann A Diabetes transition care and adverse events: a population-based cohort study in Ontario Canada Diabet Med 2018 35 1515 1522 10.1111/dme.13782 30022524 40. Sritharan A Osuagwu UL Ratnaweera M Simmons D Eight-year retrospective study of young adults in a diabetes transition clinic Int J Environ Res Public Health 2021 18 12667 10.3390/ijerph182312667 34886392	0	PMC9734505	NO-CC CODE	2022-12-14 23:28:29	no	Diabetol Int. 2022 Dec 7;:1-10	utf-8	Diabetol Int	2,022	10.1007/s13340-022-00609-7	oa_other
==== Front Mol Psychiatry Mol Psychiatry Molecular Psychiatry 1359-4184 1476-5578 Nature Publishing Group UK London 36473998 1839 10.1038/s41380-022-01839-6 Correspondence Reply to “Predominance of visuoconstructive impairment after mild COVID-19?” by Díez-Cirarda et al. 2022 de Paula Jonas J. 12 http://orcid.org/0000-0002-7603-6152 Duran Fabio L. S. 3 Busatto Geraldo 3 http://orcid.org/0000-0002-7081-8401 Miranda Debora M. 14 http://orcid.org/0000-0002-6558-4639 Romano-Silva Marco A. [email protected] 13 1 grid.8430.f 0000 0001 2181 4888 Centro de Tecnologia em Medicina Molecular (CTMM), Universidade Federal de Minas Gerais (UFMG), Av Alfredo Balena 190, Belo Horizonte-MG, Brazil 2 grid.8430.f 0000 0001 2181 4888 Departamento de Saúde Mental, Faculdade de Medicina da Universidade Federal de Minas Gerais (UFMG), Belo Horizonte-MG, Brazil 3 grid.11899.38 0000 0004 1937 0722 Departamento de Psiquiatria, Faculdade de Medicina da Universidade de São Paulo (USP), São Paulo-SP, Brazil 4 grid.8430.f 0000 0001 2181 4888 Departamento de Pediatria, Faculdade de Medicina da Universidade Federal de Minas Gerais (UFMG), Belo Horizonte-MG, Brazil 6 12 2022 13 20 9 2022 6 10 2022 10 10 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 Biomarkers Neuroscience https://doi.org/10.13039/501100002322 Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Brazilian Federal Agency for the Support and Evaluation of Graduate Education) ==== Body pmcTo the Editor: We read with great attention the correspondence by Díez-Cirarda et al. 2022 [1], regarding our recent published study on selective visuoconstructional impairment following mild COVID-19 with inflammatory and neuroimaging findings [2]. We thank the authors for recognizing the relevance of our work and to give us the opportunity to further discuss some of our findings. The results at the copy component of the Rey–Osterrieth Complex Figure Test (ROCFT) were surprising for our research team, which initially expected deficits in attention, memory and executive functions due to patients’ subjective cognitive complaints and initial findings in other COVID-19 studies. While assessing our patients the impairment observed in copy trials immediately called our attention due to its frequency and intensity (examples can be seen in the supplementary files of our manuscript), and were published as a preprint before being submitted as the current study. At that stage, other researchers not directly involved in the study were invited to analyze the drawings, including certified neuropsychologists from the Brazilian Neuropsychological Society, which were equally surprised by those unexpected results. Performance in the ROCFT-copy trial shows a ceiling effect in typical development adults, while the recall trials usually show a more symmetric distribution [3]. The larger range of possible recall scores might minimize the prevalence of memory deficits. The deficit observed in the copy trial remained significant both using normative data for scoring and when we compared the raw data of a subsample of our patients to an age, sex and education matched sample. In addition, an ANCOVA analysis suggested a more specific deficit in the copy trial since, when it was controlled, the group differences in the recall trials were not significant. Impairment in the ROCF-copy with subsequent normal recall performance had been previously reported, for example, in association with HTLV-1 infection [4] and in patients with cerebellar ataxia [5], suggesting compensation from initial organization and copying strategies. Thanks to the correspondence, an error came to our attention. The column titles were swapped (controls and patients): ROCF (copy) was 34.14 ± 2.95 for the control group and 29.22 ± 4.41 for the COVID-19 group. We are submitting an erratum to the journal. In addition, the number of correct pairs in switching fluency for controls should be read 8.76 ± 1.85. We used the Brazilian manual—commercially available from Pearson Clinic in Brazil—for its application, scoring and interpretation [6]. Normative data is available as mean ± standard-deviation for the following age-ranges: 16–20 years - 33.73 ± 2.94 (copy) and 20.31 ± 6.62 (recall), 21–40 years - 32.82 ± 3.19 (copy) and 18.70 ± 6.64 (recall, and 41–60 years - 32.25 ± 3.80 - for copy and 15.98 ± 5.56 for recall). Alternatively, percentiles could also be used for scoring, although it did not significantly change our results. These normative values are similar to those of international studies [3]. Indeed, the specific metrics for each neuropsychological test were not added in the published Table 1, which reports the following measures: Trail Making Test (time in seconds), ROCFT (total score based on the traditional 18 elements scoring system), verbal fluency (total words for animal and fruits and total number of pair in the switching condition; and 1 min for each condition), Digit Span (number of correct trials × maximum achieved span - Kessels et al. (2008) recommendations [7], Five Point Tests (total unique drawings) and Logical Memory (sum of correct answers).Table 1 Significant correlations between performance on the Rey–Osterrieth Complex Figure Test and neuroimaging measurements of gray and white matter volumes (MRI) and glucose metabolism (FDG-PET). Brain regiona Correlation Cluster sizeb Coordinatesc Peak Z scored pFWE cluster levele pFWE voxel levelf x y z Gray matter volume No significant correlations. – – – – – – – – White matter volume Left and right genu of the corpus callosum, extending to the cingulum bundle Negative 1426 −6 26 −2 4.32 0.000 0.062 Right fusiform gyrus Negative 93 32 −22 −26 4.01 0.262 0.186 Right lingual gyrus Negative 127 30 −44 −8 3.84 0.180 0.307 Right inferior frontal gyrus Negative 98 40 6 16 3.76 0.248 0.381 Left lingual gyrus Negative 41 −18 −52 2 3.73 0.486 0.408 Left inferior frontal gyrus Negative 20 −34 30 −2 3.48 0.633 0.685 Left inferior fronto-occipital fasciculus Negative 15 −24 4 −8 3.47 0.676 0.691 Left inferior fronto-occipital fasciculus Negative 15 −32 −10 −8 3.44 0.676 0.727 Right inferior fronto-occipital fasciculus Negative 13 32 −10 −8 3.43 0.875 0.730 Glucose metabolism (FDG-PET) Left inferior temporal gyrus Positive 34 −56 −46 −22 3.96 0.644 0.400 Left inferior occipital gyrus (superior portion) Positive 53 −48 −68 −16 3.92 0.476 0.452 Right dorsal anterior cingulate gyrus Negative 69 8 16 42 4.61 0.365 0.043 Right Rolandic operculum and opercular part of the inferior frontal gyrus Negative 57 52 8 12 4.48 0.446 0.072 Right inferior occipital gyrus Negative 62 38 −74 −6 4.15 0.410 0.234 Left calcarine and lingual gyri Negative 66 −14 −92 −8 3.88 0.384 0.494 Left superior frontal gyrus Negative 50 −16 60 −8 3.75 0.500 0.643 Left inferior occipital gyrus (inferior portion) Negative 19 −30 −82 −8 3.46 0.798 0.918 Right medial frontal and orbital frontal gyri Negative 20 18 54 −4 3.43 0.788 0.934 aFor the analysis of white matter volumes, the brain regions where voxel clusters were located were identified according to the MRI Atlas of Human White Matter [8]. For the analysis of glucose metabolism, brain regions were identified according to the Automatic Anatomical Labeling Toolbox for SPM12 [9]). bNumber of contiguous voxels in each cluster that surpassed the initial cutoff of p ≤ 0.0005 uncorrected for multiple comparisons. cMNI coordinates of the voxel of maximal statistical significance within each cluster. dZ-score for the voxel of maximal statistical significance in each region. eStatistical significance after family-wise error correction for multiple comparisons (cluster level). fStatistical significance after family-wise error correction for multiple comparisons (voxel level). In regard to the neuroimaging findings, the choice to inspect the statistical parametric maps with no correction for multiple comparisons was made due to the exploratory nature of the study, and we attempted to minimize false-positive findings by using a relatively strict threshold (p < 0.0005, two-tailed). We agree that it is important to reassess our findings with correction for multiple comparisons, and this is provided in the attached Table (applying family-wise error—FWE methods). The largest cluster of significant negative correlation between white matter volume and neuropsychological test performance retained statistical significance (FWE, cluster level), and so did one of the clusters of negative correlation between glucose metabolism and test performance (located in the right dorsal anterior cingulate cortex) (FWE, voxel-level). While addressing the comments, we are urged to consider what awaits ahead in terms of recovery and possible interventions. The catching questions are how the virus promotes these changes and why not everyone recovers after a mild acute phase. The pandemic put the spotlight on the neuroinflammation that follows, not only SARS-Cov-2 infection, but also other systemic inflammatory conditions that compromises the whitte matter resulting in cognitive dysfunctions. We hope the comments were satisfactorily answered and make ourselves available for any further discussion about this very important and timely topic. Author contributions Study conception and/or design: JJP, DMM, and MAR-S. Data analysis: JJP, FLSD, GB, DMM, MAR-S. Paper writing and/or revision: JJP, GB, DMM, MAR-S. 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. Díez-Cirarda M, Yus M, Matías-Guiu J, Matias-Guiu JA. Predominance of visuoconstructive impairment after mild COVID-19?. Mol Psychiatry. 2022. 10.1038/s41380-022-01797-z. 2. de Paula JJ, Paiva RERP, Souza-Silva NG, Rosa DV, Duran FLS, Coimbra RS, et al. Selective visuoconstructional impairment following mild COVID-19 with inflammatory and neuroimaging correlation findings. Mol Psychiatry. 2022. 10.1038/s41380-022-01632-5. 3. Carone DA. E. Strauss, E. M. S. Sherman, & O. Spreen, A Compendium of Neuropsychological Tests: Administration, Norms, and Commentary, Applied Neuropsychology, 2007;14:1,62-63, 10.1080/09084280701280502. 4. Kalil RS, Vasconcellos I, Rosadas C, Cony A, Lima DP, Gonçalves CCA, et al. Association between high proviral load, cognitive impairment, and white matter brain lesions in HTLV-1-infected individuals. J Neurovirol. 2021. 10.1007/s13365-021-00944-6. 5. Slapik M Kronemer SI Morgan O Bloes R Lieberman S Mandel J Visuospatial Organization and Recall in Cerebellar Ataxia Cerebellum 2019 18 33 46 10.1007/s12311-018-0948-z 29949096 6. Oliveira MS, Rigoni MS (ed.). Figuras Complexas de Rey - Manual (1ed). Brasil;Casa do Psicólogo/Pearson Clinical; 2014. 7. Kessels RPC van den Berg E Ruis C Brands AMA The backward span of the Corsi Block-Tapping Task and its association with the WAIS-III Digit Span Assessment 2008 15 426 34 10.1177/1073191108315611 18483192 8. Oishi K, Faria AV, van Zijl PCM, Mori S. MRI Atlas of Human White Matter. UK:Academic Press; 2010. 9. Rolls ET Huang C-C Lin C-P Feng J Joliot M Automated anatomical labelling atlas 3 Neuroimage 2020 206 116189 10.1016/j.neuroimage.2019.116189 31521825	36473998	PMC9734506	NO-CC CODE	2022-12-14 23:28:29	no	Mol Psychiatry. 2022 Dec 6;:1-3	utf-8	Mol Psychiatry	2,022	10.1038/s41380-022-01839-6	oa_other
==== Front Innov High Educ Innov High Educ Innovative Higher Education 0742-5627 1573-1758 Springer Netherlands Dordrecht 9639 10.1007/s10755-022-09639-0 Article Research and Scholarship During the COVID-19 Pandemic: A Wicked Problem Sezen-Barrie Asli [email protected] 1 Carter Lisa 2 Smith Sean 3 Saber Deborah 4 Wells Mark 5 1 grid.21106.34 0000000121820794 School of Learning and Teaching, Research in STEM Education (RiSE) Center, University of Maine, Orono, ME USA 2 grid.21106.34 0000000121820794 School of Education Leadership, Higher Education, and Human Development, University of Maine, Orono, ME USA 3 grid.21106.34 0000000121820794 School of Earth and Climate Science, Senator George J. Mitchell Center for Sustainability Solutions, University of Maine, Orono, ME USA 4 grid.21106.34 0000000121820794 School of Nursing, University of Maine, Orono, ME USA 5 grid.21106.34 0000000121820794 School of Marine Sciences, University of Maine, Orono, ME USA 8 12 2022 125 28 10 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. This study focuses on the impacts of the COVID-19 pandemic on research and scholarship at a research university in the United States. Building on studies in higher education policy, we conceptualized the COVID-19 pandemic as a ‘wicked problem’ that is complex, nonlinear, unique, and requiring urgent solutions. Wicked problems highlight pre-existing struggles, and therefore, recent challenges in higher education inform the methods and the findings of this study. Qualitative and quantitative survey data from 408 faculty, staff, and students explicate the reasons for reduced research output and adaptations made for increased or sustained productivity, health, and wellness that influenced research activities. The analysis showed that most respondents experienced reduced productivity mostly due to increased work responsibilities, limited access to research fields, and inadequate resources. Despite self-reported reduced productivity, participants from the University we studied experienced increases in funding during the pandemic. Thus, the findings also reported on the adaptations for sustained or increased productivity. These included new research pursuits, participation in conference and learning opportunities across geographic regions, and purchase of computer equipment/accessories for home offices. A small percentage of respondents mentioned improved health and well-being; however, many linked reduced research activities to health and well-being issues such as anxiety and fear about the pandemic and being overwhelmed due to work and home-life expectations. Knowledge of the challenges and opportunities presented within the first year of the pandemic can provide a basis for solutions to wicked problems higher education may face in the future. Keywords COVID-19 Wicked problems, research Scholarship Office of the Vice President for Research, University of Maine ==== Body pmcIntroduction The worldwide pandemic caused by the spread of COVID-19, a novel coronavirus, led to substantial changes in higher education activities across the globe (Crawford et al., 2020). Beginning March of 2020, many faculty had to adopt new institutional measures to maximize the health and safety of campus communities. Studies during this time showed that the pandemic conditions were challenging for faculty because of expanded work hours to support students, transition to online teaching, heightened levels of stress and anxiety, and regression of research activities (Canadian Association of University Teachers, 2021; Dalhousie Faculty Association, 2020). Despite these challenges, some studies highlighted desirable impacts from the pandemic on research and scholarship. One such impact was the increased funding and opportunities to work on pandemic-related research and some scholars shifted to focus on COVID-19 related research (e.g., Levin, 2020; Sanders, 2020). Moreover, some researchers took advantage of the time off from labs and fieldwork to explore new research directions and write proposals and manuscripts (Korbel & Stegle, 2020). This time also allowed researchers to organize and analyze data previously on hold (Flaherty, 2020). Our study aims to focus on the impacts of the COVID-19 pandemic on research and scholarship at higher education institutions, particularly non-essential and non-COVID-19 related studies, which are greatly impacted by pandemic measures (Basken, 2020). The study is inspired by questions raised at our institution’s faculty senate, promulgating a formal survey. Building on recent literature on the impacts of the COVID-19 pandemic and other recent problems higher education institutions face, we aimed to identify the pandemic's impacts on research and scholarship at a public research university. The university is recognized as the “flagship institution,” the most known university in the state (Douglas, 2016). The central campus of the university is located in the rural Northeastern region of the U.S; however, the institution strives to serve the whole state with its centers and units at all geographic locations. We recruited 408 participants (anonymously reported as 177 females, 156 males, 16 nonbinary and unidentified); in a university with 620 (258 female, 330 male) faculty and 3,025 postgraduate students, including non-thesis students (1704 female, 1250 male, 71 nonbinary and unidentified). Approximately a quarter (168) of our participants reported as tenure-track faculty (60 Pre-Tenure Asst. Professor, 50 Associate Professor, 58 Full Professor), 134 graduate students, 44 research staff, 27 non-research positions including administrators and lecturers. Among the 408 participants, 118 have children, 33 elderly relatives, and 12 disabled relatives at home as dependents. In contrast to the dominant focus on STEM education fields in previous studies, we examined the impacts of the pandemic across disciplines, including the Humanities, Social Sciences, and the Arts. In addition, we elaborate on the causes for work regression and individual faculty adjustments that were put in place to ensure research and scholarship productivity. We draw particular attention to the impact of COVID-related changes in health and well-being on research and scholarship activity. Specifically, we address the following research questions (RQs) using survey data collected from 408 faculty, research staff, grad students, and post-doctoral students:RQ1: In what ways did the COVID-19 pandemic influence research and scholarship activities at a research university in a rural area of the Northeastern U.S.?What were the reasons behind any reduced productivity? What adaptations led to increased or no change in the productivity of participants? RQ2: How do faculty, postdoctoral students, and graduate students allude to their health and well-being concerning the impacts on their research and scholarship productivity? Conceptual Background and Review of Literature Wicked Problems and Higher Education Driven by the recent policy design studies in higher education, we conceptualize the COVID-19 pandemic as a "wicked problem" for higher education institutions (e.g., El Masri & Sabzalieva, 2020). Wicked problems are described as complex, nonlinear, and unique. If the solutions are not provided quickly, wicked problems can develop harmful consequences for the community (Peters, 2017). The COVID-19 pandemic created a wicked problem for higher education institutions where decision-makers were forced to develop urgent and dramatic solutions to prevent viral spread. These solutions included, but were not limited to, travel restrictions, transition to online teaching, and safety measures for labs and classrooms. Early studies reported the consequences of these measures. According to Korbel and Stegle (2020), 57% of life scientists reported losing some work, while 25% noted significant project delays of up to six months. Another study highlighted eighty percent of the postdoctoral students could not continue their work, mostly due to laboratory shutdowns (Woolston, 2020). For some researchers, the restrictions created challenges to maintain animals, cell lines, and cultures. Solutions included freezing cell lines or tissue cultures (Redden, 2020) and assuming the responsibilities of research assistants and lab technicians to maintain functioning labs (Grimm, 2020). Peters (2017) highlighted that “every wicked problem may be a symptom of another problem” (p. 388), that might exist before and continue to exist in the future. Indeed, pandemic life during COVID-19 emphasized issues of enduring problems such as equity and diversity within higher education. For example, the negative impacts on research and scholarship were even more challenging for underrepresented groups, such as female scholars in science and engineering fields and international students (Milliken et. al., 2020; Oleschuk, 2020; Sotto-Santiago et al., 2021). In this study, enduring problems are considered as we sought to understand (a) how pandemic conditions can limit research and productivity; and (b) the strategies of resiliency that lead faculty and students to overcome the challenges of wicked problems. Enduring Problems in Higher Education in the Recent Times Because wicked problems, such as the COVID-19 pandemic, can be amplified by enduring challenges, we first provide an overview of the challenges that higher education institutions faced prior to the pandemic. Financial Instability in Higher Education Referred to as “the Great Recession,” the economic crisis in 2008–2009, was detrimental to many higher education institutions (Geiger, 2010). Wealthy private schools lost significant endowment resources due to low rates of return (Barr & Turner, 2013). Although endowments typically make up a small percentage of the operating revenue for public institutions (Rosen & Sappington, 2019), these institutions saw a dramatic decrease in state funding (Geiger, 2010). In addition, the crisis caused a domino effect that severely impacted higher education institutions and resulted in (a) widespread budget cuts, (b) significant staffing changes, and (c) changes in the student admission process. Once flourishing endowments, the mainstay of wealthier institutions, experienced significant losses, “translated into long-term budget cuts” (Geiger, 2010, p. 10). State revenues decreased across all avenues, similarly affecting public institutions (Barr & Turner, 2013). Facing these budget crises, colleges and universities were forced to address these shortfalls through staff terminations, hiring freezes, project and building cancellations, and tuition increases (Geiger, 2010; McGlynn, 2014). A few years following the Great Recession, State Higher Education Finance (SHEF) annual reports noted a steady decrease in state and local support in higher education (Carlson, 2019). Although there was a sign of recovery in 2013 and slight increase in some state’s budgets for higher education, this was mainly offset by the inflation (Carlson, 2019). Until 2019, some states continue to increase support for higher education mostly in the form of student financial aids. However, the state support was 8.7% below the pre-recession rates (Cummings et al., 2021). To offset the state reduction, four-year public institutions continued to increase student tuition and fees. Nevertheless, the increase was not sufficient to make up for the lost state funding (Webber, 2017). The higher education institution in this study also relies mostly on tuition and fees for its operating revenue source. A recent town hall on the budget reported that 57% of the FY2021 budget is expected to come from the tuition and fees, while state appropriation is expected to contribute to 27% of the revenue (University of Maine Office of the President, 2021). Many colleges and universities opted to not fill full-time faculty positions and instead hire and rely heavily on adjunct faculty (Stenerson et al., 2010). According to the National Center for Education Statistics (n.d.), from 1991 to 2011, the numbers of part-time faculty increased 162%, whereas full-time only increased 42%, highlighting this significant staffing change. Hiring adjuncts, at less cost than salaried faculty, helped higher education institutions save money during the economic downturn. Therefore, today most of the faulty with teaching responsibilities on part time appointments. (McComb, 2020). Adjunct faculty often juggle full-time careers, possibly multiple contingent positions, and do not have the additional time-resources or support to engage in activities and responsibilities typically performed by the full-time professoriate (Stenerson et al., 2010). Currently, the part-time faculty at our institution is around 30%, while it fluctuates slightly between the fall and spring semesters (University of Maine System Strategic Planning Data Book, 2021). The resultant increase in hiring contingent staff has worsened how institutions experience future wicked problems. Decreases in full-time faculty positions have led to fewer faculty members available to fulfill important administrative roles and contribute additional responsibilities, such as service to units, the institution, and the community (Stenerson et al., 2010). In addition, faculty members that are hired for full-time positions may feel overwhelmed, underappreciated, burned out, and left to balance the overflow responsibilities that at one time would have been distributed among several members of their academic community. The graduate students of this great recession period experienced the effects of less full-time faculty present in higher education, and students spent less time with professors for advising or mentoring (Stenerson et al., 2010). Inequalities in Higher Education The institution of Higher Education in the United States has been characterized as socially elite and prestigious (Brown, 2018). There are multiple facets of inequality including but not limited to gender, race, and academic rank. While inequality in higher education is a complex problem, there are apparent depictions that can be characterized through faculty stratification. Women and Black, Indigenous People of Color (BIPOC) are disproportionately represented in academia (Brown, 2018). Research finds that women are less represented in the higher faculty rank and their salaries are lower than men across all academic ranks (Fox-Cardamone, 2010). According to data, female faculty made 0.844 to every male faculty’s dollar (Fox-Cardamone, 2010). Salary discrepancies can also be linked to inequity in promotion, which may be more difficult to achieve for women in higher education, specifically when reaching the rank of full professor. Fox-Cardamone (2010) reported males are 40% more likely to be promoted to full professor than females. A more recent study looked at the gender pay gap for senior faculty in administrative positions and found a similar gap female administrators earned approximately 80 cents on every male administrator’s dollar (Bichsel & McChesney, 2017). In addition to these notable salary disparities, there are also rank inequalities regarding faculty workload of teaching, research, and service. In some cases, full professors may spend more time on research activities, including preparing manuscripts, acting as principal investigators, and participating in lab and fieldwork, while associate professors occupy more time teaching, advising, and administering courses (O’Meara et al., 2017). When gender and rank are considered at the full professor level, women faculty spend approximately seven hours less per week on research and significantly more time on campus service activities, student advising, and institutional housekeeping than their male colleagues (O’Meara et al. 2017). Because research and publication activities are heavily weighted for promotion from associate to full professor, these findings help to explain the continued practice of gender inequities and subsequent salary differences in higher education institutions (Fox-Cardamone, 2010; O’Meara et al., 2017). Some women, however, have developed strategies to increase academic productivity and resiliency while maintaining household responsibilities. A case study examined female graduate students, who identified as mothers working towards graduate degrees through online programs where participants developed strategies to increase productivity, effectively manage time, and avoid becoming overwhelmed while working on their degree programs (Fensie & Sezen-Barrie, 2021). For example, these women utilized selective attention techniques when faced with distractions, such as not looking up from their computer screen, maintaining intense focus, and reading out loud to themselves. While more studies are needed, some females’ improved productivity during the pandemic can be explained by their strategies for effective work to achieve success despite demanding responsibilities assigned to them. Challenges to Build a Strong Community of Practice in Higher Education Universities prosper when there is commitment to facilitating programs and interactions focused on developing social connections and a supportive community for faculty (Buckley, 2020; Himelein & Anderson, 2020; McCauliff, 2020). While not always consistent or successful, faculty development strategies include creating programs where small groups interact to share information and experiences to aid in community building within the institution (Himelein & Anderson, 2020). However, time availability and persistent commitment is a barrier to creating and sustaining these social connections. Exacerbating these challenges is the increasingly adjunct staffing model, where the transient nature of their employment and exclusion from many engagements hinders a community building atmosphere (Culver et al., 2020). The confluence of these issues results in higher education increasingly becoming plagued with community building issues (Thelin, 2001). To mitigate these challenges, Himelein and Anderson (2020), created small, academically focused groups to interact for short periods of time. This concept required less time commitment and went beyond a social gathering to build learning communities that were “valuable for teaching development as well as personally or professionally rewarding” (p. 28). As teaching is the dominant role of adjunct faculty, smaller community connections with a development focus could be attractive to this group. The onset of the COVID-19 pandemic and social distancing requirements decimated most community interactions. University campuses closed and Zoom meetings prevailed. Establishing and sustaining supportive communities at colleges and universities became more problematic due to working from home, a lack of casual social interactions on campus, and additional work and home related pandemic responsibilities. Faulty perceptions of organizational support during the pandemic have been an environment that combines research productivity with healthy work-life harmony (Culver et al., 2020). Due to challenges imposed by the pandemic (e.g., increased responsibilities, financial instability, and lingering inequalities), community building continues to be challenging in the current environment of the persistent pandemic. Study Design and Methods This survey study examined the contextual impacts of the COVID-19 pandemic on research and scholarship at a public research university in the Northeastern U.S. With a central campus in a rural location, the university participates in outreach efforts across the state, contributes to national and international research agendas, and provides comprehensive undergraduate and graduate programs. The university houses international scholars who lead cutting edge research across disciplines. It is important to note that the research and development funding trends at this institution have increased during the last five fiscal years, including the pandemic years of 2020–2021, with funding generated and spent during the pandemic at an all-time high (University of Maine System Strategic Planning Data Book, 2021). Despite funding trends, challenges in carrying out and completing research and scholarship activities were raised by the university community with the institution’s faulty senate. We administered a survey with qualitative and quantitative components during January–February 2021, at the beginning of a third academic semester under pandemic conditions. The rural state and the county the research university is located observed a low number of COVID-19 cases until early 2022 (COVID-19 Case Trends, 2021). While businesses, university facilities, and K-12 schools were shut down during the first six months of the pandemic, the phased opening started with precautions (e.g., mask-wearing, social distancing, regular testing) during the summer and fall of 2020. Starting at the beginning of the 2020–2021 academic year, the State Department of Education allowed in-person schooling for counties with low risk for COVID-19 spread and prepared to apply all pandemic measures. The small classrooms in the rural counties around the institution made it possible for in-person school options. However, most of the after and before-school programs were canceled and allowed limited care-free work hours for parents (Maine Department of Education, 2021). The University also had a phased opening for its research facilities, where labs and classrooms became available for research and teaching as long as COVID protocols were administered starting the 2020–2021 academic year. These facilities will be shut down if there is a sign of community spread. Students, faculty, and staff coming to campus were tested regularly for COVID-19. Despite the advantage of low COVID-19 cases, more online courses and remote work options were created to be preventative as the rural state's hospital facilities are not as extensive as in other locales (University of Maine COVID-19 Research Continuity Update, 2021). The participants responses may reflect the times of completed shut down and restricted opening thanks to low COVID-19 most of the time. Participants of the Survey We recruitted from a pool of all employees who conduct research and scholarship (see Introduction for the total numbers of faculty and grad students). We were able to recruit 408 participants. The distribution of participants across genders, university positions, and academic disciplines—as well as their home dependent responsibilities—are provided in Table 1.Table 1 The demographic distribution of the participants of the study Gender Female 50.8 (177) Male 44.6% (156) Prefer Not to Reply < 5% (12) Non-Binary Genderfluid Genderqueer < 5% (1) Transgender Female < 5% (2) Gender Identity Not Listed < 5% (1) University Position Tenure-Track Faculty: 45% Graduate Students 35.9% (134) Research Positions* 11.8% (44) Non-Research Positions** 7.2% (27) Professor 15.6% (58) Associate Professor 13.4% (50) Pre-Tenure Asst. Professor 16.1%(60) Academic Discipline STEM 56.3% (241) Social Sciences and Humanities 33.7% (149) Health and Medical Sciences 7.7% (34) The Arts 2.3% (10) Science 39.8% (176) Technology 3.4% (15) Engineering 10.2% (45) Math 1.1% (5) Home Dependents Children 32% (118) Elderly Relatives 8.9% (33) Disabled Relatives 3.3% (12) None 49.3% (182) Prefer Not to Mention 6.5% (24) Research positions include post-docs, research faculty and staff Non-research positions include lecturers and administrative staff 35 respondents left this section unanswered Some respondents may have chosen more than one discipline **Some respondents may have chosen more than one option or left this section unanswered Our participants are from various academic disciplines and positions, including graduate students and research faculty. This is because all research activities conducted by all positions in all disciplines are critical for a higher education institution working towards the goals of a “very high research” activity or R1 Carnegie Classification. To limit non-response bias (Matsuo et al., 2010), we used inclusive email listserves, made announcements at faculty senate meetings with representatives across disciplines, and used graduate school listers to access graduate students. The respondents represent their gender, disciplinary, and rank groups within 2–3% variability compared to the population (University of Maine System Strategic Planning Data Book, 2021). Survey Design and Development A web-based survey was developed in Qualtrics using the design process of Lumsden (2007). The first step was to define the affected area of research and variables (e.g., position, gender, …., etc.) using groups of experts in the faculty senate and independent research meetings. Once defined, we identified the research questions and the target participants. Our initial question of "how the COVID-19 pandemic influenced the research and scholarship activity at a research university" was distilled to more specific queries (e.g., reasons for reduced research activity). The initial plan of targeting full-time faculty and graduate students was expanded to all employees conducting research and scholarship, including postdoctoral scholars and research staff. The survey was comprised of six subcategories: demographics, research planning disruptions, logistical challenges, mentoring opportunities, workspace changes, and mental and physical health. Twenty-eight survey questions were developed to address these subcategories. Education researchers and members of the faculty senate reviewed the survey questions, and the Institutional Review Board gave their approval for participant recruitment after integrating feedback on the study design. A pilot-test was first conducted with five faculty and graduate students from other higher education institutions that resulted in further revisions of the questions. The survey then was published through Qualtrics and distributed to an email list with all university employees and graduate students. Data Analysis Approach The quantitative aspects of the survey data were analyzed after cross-tabulation and frequency analysis using Microsoft Excel (Creswell & Plano Clark, 2017). Embedded in this analysis were insights to how demographic information and dependent care at home relate to increased, reduced, or steady productivity in research and scholarship. Although the analysis of multiple-choice items provided general trends across disciplines, analysis of the qualitative data from the participants provided richer insights into research and scholarship experiences. Therefore, our findings are predominantly derived from the analysis of qualitative data. The analysis of the qualitative data started with an initial reading of all the qualitative excerpts and memos that were written to identify preliminary themes. These qualitative data then were organized in Dedoose (qualitative analysis software) by creating an entry for each participant. Guided by our initial themes and research questions, the excerpts were coded independently by two researchers. The disagreements in coding were discussed and often led to revision of the codes until both researchers reached 100% agreement. As a further check for validity, we triangulated our codes with different sources of information in the recent literature and higher education blogs in consultation with the other faculty senate members. The codes were then organized under three themes: (a) reasons for reduced productivity, (b) adaptations for sustained or improved productivity, and (c) health and well-being during the pandemic. The codes under each theme were sequenced by frequency (i.e., how many participants’ excerpts were tagged with a code) and interconnectedness to each other. The findings provided detailed descriptions of each theme with sample excerpts from the participants (Creswell & Miller, 2000; Lincoln & Guba, 1985). Findings Most participants reported reduced research and scholarship productivity during the pandemic (78%), while far fewer reported unclear effects on productivity (12%). Although low in numbers, some participants experienced sustained (5%) or even increased productivity (5%) (Fig. 1). We looked at the change in research productivity across University positions (Fig. 2). The breakdown showed that the tenure-track faculty was the group that mentioned the reduced research productivity the most (83%), followed by graduate students (77%). Respondents in research positions were the group that said reduced research productivity the least (66%). We see the documentation and consideration of these patterns and adaptation outcomes as important to community wellbeing, particularly with the juxtaposition to the institutions’ success in research and development funds during the pandemic years. Earlier studies show differences in pandemic effects on faculty research and scholarship productivity based on gender and dependent care (e.g., Viglione, 2020), so this question was examined here. The findings show similar research and scholarship productivity effects across genders and dependent care for untenured assistant professors, associate professors, research faculty, research staff, and postdoctoral students, but gender differences at the full professor rank and for graduate students and lecturers. Full professors reported increased and sustained productivity (21% for male, 14% for female), while graduate students reported increased productivity (5% for females, 0% for males). Dependent care factors appeared to only influence male lecturers in that those all reporting no dependent responsibility cited increased productivity.Fig. 1 Frequencies of researchers with reduced, unclear, sustained, and increased research productivity Fig. 2 Change in research productivity across different positions Reasons for Reduced Productivity in Research and Scholarship The analysis of the 408 qualitative responses revealed that 257 (63%) of the participants gave reasons for reduced productivity (Fig. 3). The most frequently cited reason for reduced productivity was ‘Increased Work Responsibilities’ (33%) due to the rapid transition to online teaching, increased attention to students' emotional well-being, additional administrative requirements for safety, decreased staff availability, and other increased administrative tasks. It is important to note that almost all excerpts in this category were from the faculty (tenure and nontenure track) and staff. Graduate students only made references to decreased staff availability. For example, one faculty member from the Sciences, explained how increased workload due to transition to online teaching left no time or mental energy for his research and scholarship activity as:Teaching a large course load remotely takes almost all of my time, very little time / energy / mental bandwidth / motivation / ambition left over for research. Research / scholarship is becoming increasingly trivial and decreasingly rewarding. After devoting time / attention / effort to my students/classes (not to mention home responsibilities!), I'm burnt out and have no surplus attention to devote to much else. Diminishing marginal returns to increased research. [male, associate professor, caring for children at home] Fig. 3 Reasons for reduced research and scholarship activity during COVID-12 pandemic Another frequently mentioned theme for reduced productivity was ‘Limited/Inconsistent Access to Research Field’ (21%), including fieldwork cancellations, inconsistent access to K–12 classrooms, laboratory limitations, and travel restrictions to research fields (particularly those in foreign countries) and conferences. The following is a representative excerpt by a faculty member in the Social Sciences focused on the disruption of access to K–12 schools as a research field:I had a five-year pilot project for a school-based intervention effectively shut down by the pandemic. We were going to run a quasi-experimental evaluation of the project in the schools we partner with after 4 years of work this year and are now unable to because of the disruption schools have experienced. We still had to pay our full-time project staff for this year and will not receive additional funds from our previous funding sources, so we effectively will never know if the intervention (a $750K investment) we worked on for years on had "gold standard" effects. We couldn't even collect outcomes for Year 4 because of the March shutdown in 2020. [female, associate professor, caring for children at home] Related to lack of access to the research field, participants wrote about delays in ‘Project Timeline and Funding’. A faculty member from the Sciences wrote “Project delays have tied up capacity to an extent that I have no time for proposal development, but no additional funds coming in” [female, assistant professor, with children as dependents at home]. Challenges to research field access and delays in project funding along with hardship to recruit reviewers during the pandemic, led to ‘Publication Delays’. As highlighted by a faculty member from the Sciences, “Several datasets which were to be collected were delayed, thereby delaying ultimate data analysis and publication by at least a year” [male, assistant professor]. Another theme frequently cited was ‘Inadequate Resources’ (13%) in terms of decreased library resources, internet connectivity, and shortages of needed supplies (caused by the limited access to build and shipping delays). One research staff in the Social Sciences saw some of the inadequate resources as an alarming problem, especially considering the university’s attempt to move to a higher research rank at a national classification. She wrote “…budget reductions have drastically affected research and scholarly activity given the loss of book budgeting, databases, and journals. This is alarming given the needs of us at an R2, but also as we move toward R1 status” [female, research staff, caring for children at home]. In some instances, lack of resources led to ‘Matriculation Extension’ (2%) for graduate students. One student from the Humanities commented:I've had to abandon my thesis altogether, for lack of access to research materials, and switch to completing my degree by coursework alone, which has set me back about 18 months (counting the pre-pandemic research time that was retroactively wasted by the changeover). [male, graduate student] Due to limited access to offices, scholars needed to use their home as a workspace. The home environment, however, was not conducive to an effective working environment for everyone. Therefore, one of the themes for reduced research activity was ‘Work from Home Issues and Family Responsibilities’ (9%). There were varying reasons for why working from home was not feasible for research and scholarship production. The reasons included lack of physical space, dependents being at home, difficulty separating home and work life, and other distractions, such as noise. A faculty member from the Social Sciences mentioned:I'm effectively leasing a home office for myself - it's expensive and mentally costly shift to working from home to accomplish research tasks impacted the following aspects of your family or personal life: Complete inability to separate work from life now. I am not working at home; I functionally live at work instead. [male, associate professor, caring for children at home] For this faculty member, the attempt to work from home led to family and personal problems and led him to rent a separate office space, despite the costs. Related to ‘Work From Home Issues and Family Responsibilities’, participants highlighted ‘Decreased Social Connectivity’ (7%) as another major reason for reduced research and scholarship activity. Faculty and staff who were in the institution pre-pandemic, such as the faculty member from the Sciences [female, associate professor, caring for children at home] mentioned “loss of community” in their department, which led to “loss of informal interactions” and “less opportunity to share ideas.” The new members of the institution also experienced social disconnect that created barriers in getting to know the local culture of their program. A student from the Medical Sciences said:As a first-year grad student, it's been much harder to interact with other members of my cohort and of my program in general. I've been mostly okay with this so far because I am in contact with other friends I made before grad school, but I suspect that there will be long-term consequences in terms of identifying with the program and [the university] in general. [female, graduate student] Other than social disconnect, working from home led to blurred schedules and created ‘Scheduling Difficulties (4%)’ for some, but was rarely mentioned as a reason for reduced research and activity. A faculty member from the Social Sciences wrote: “schedules have been less predictable, so it is more difficult to create and keep to a timeline for work” [female, associate professor, caring for elderly at home]. Adaptations for Sustained or Improved Productivity Among 408 survey participants, 73 (18%) of those talked about adaptations to sustain or improve their productivity. Figure 4 illustrates the frequencies by the type of adaptations with the most apparent theme being ‘New Research Opportunities’ (19%) thanks to the more frequent and efficient use of virtual tools. The virtual collaborative environments led to new networking opportunities among scholars that are geographically distant, and for some, these networking opportunities resulted in research and scholarship products. One faculty member who is affiliated with the Sciences and Social Sciences wrote:I have been leading a grad seminar to improve my understanding of equity in my field, with the ultimate goal of producing an article on the topic with the grad students and colleagues who are participating. Now that we are all accustomed to Zoom, the seminar is international in scope, with leading experts from around the US and the hemisphere leading us on particular topics and often participating weekly. We also have grad students from other institutions. I doubt this would have happened before the pandemic. [male, full professor] Fig. 4 Adaptations for increased or sustained research and scholarship activity during COVID-19 pandemic With greater use of virtual spaces as a collaborative environment, researchers did not need to be limited to their geographic region to start new projects and build new partnerships. Another related theme that emerged in participant responses was ‘Conference Accessibility’ (10%). Since the distance and the affordability of the conference venue was not a concern for the virtual conferences, some researchers were able to utilize these opportunities to improve their research and scholarship productivity. One professor from the Sciences explained this benefit as:Scientific conferences going online is a huge boon. Saves TONS of time and money and creates great opportunities for undergrad and grad students to attend whereas they otherwise would be excluded due to the high costs. I hope conferences never go back to in-person. [female, associate professor, caring for children, elderly relatives, and disabled family members] In addition, increased use of virtual environments created the need for gaining knowledge about and developing practices to navigate digital tools. Therefore, ‘Increased Knowledge and Skills’ was mentioned by 11% of those participants who have adapted to pandemic conditions to continue or improve their research and scholarship. A research staff in the Sciences explained how increased skills for using Zoom expanded opportunities to access researchers and teachers across the state:I've developed methods for remote use of our research instrumentation using Zoom remote control features. This is important not only for training but also for our outreach component of our recent NMR [Nuclear Magnetic Resonance] grant. This opens up access to our NMRs to researchers throughout the state, as well as to teachers at other colleges that offer courses in organic and other branches of chemistry. [male, research staff] Other than improved Zoom skills, participants seized learning opportunities for remote instrumentation use, and digital data collection strategies to adjust to a working from home environment. Another highly mentioned theme concerning adaptations was ‘Purchasing New Computer Equipment’ (18%) to accommodate a similar quality of work while at home and enhance virtual collaborations. The creation of effective home offices also meant a new cost to the employees. As one student in the Sciences wrote: “Costs associated with workplace shift to home: Audio/video equipment (quality web cam and mic for teaching online)” [transgender female, graduate student]. Our survey explored the type and the amount of costs associated with the design of home offices. Among respondents, 89% selected costs associated with new furniture, increase in utility bills, and an upgrade of internet access. Less than 10% of the respondents who mentioned costs to create a work environment from home mentioned costs related to construction at home (e.g., to create office space), materials for research and teaching, digital equipment such as new keyboards and a larger computer screen. While less than 1% of the respondents estimated the cost of this change being more than $10,000, 22% of the participants estimated costs between $1,000–10,000, and 69% of the participants estimated costs between $100—$1,000. Despite the low traffic, rural setting of the university, the third most common theme for increased productivity was ‘Less Commuting Time’ (17%) to prevent extra time in traveling a distance, particularly during wintertime when the roads are icier. A faculty member from the Social Sciences wrote that “No time lost to commuting to/from campus. It is somewhat easier to get small writing/thinking tasks done in between meetings because there are fewer interruptions from people stopping by my office” [female, associate professor, caring for elderly relatives at home]. Relatedly, 15% of the participants, who chose increased and sustained productivity, mentioned that working from home led to ‘Efficient Meetings.’ As highlighted by a lecturer from the Humanities and Social Sciences, people were “able to have more meetings as well as more productive meetings online (Zoom)” [male, lecturer, caring for children at home]. Although more rarely mentioned, survey respondents talked about recognition of needed ‘Digital Updates’ on campus resources (7%) and ‘Relaxed Requirements’ (3%) during the pandemic. Attending these needs helped some scholars continue or improve their scholarship and research. For example, one research faculty from the Social Sciences mentioned updates in digital library resources as helpful for remote access: “The pandemic has given our library time to do much needed updates to exhibits and cataloguing. It's also allowed us to focus on digitizing materials and created an online resource hub for researchers and teachers to still access remotely” [male, research faculty]. The ‘Relaxed Requirements’ code included situations where deadlines were extended and virtual participation in meetings was welcomed. A researcher from the Humanities talked about the benefits of relaxed requirements as: “Increased flexibility/accessibility of distance partners; increased willingness to meet virtually with colleagues; generally, more relaxed attitudes toward project difficulties, deadlines” [male, research staff, caring for children at home]. Health & Well-Being During the Pandemic: Challenges and Affordances to Productivity The analysis of the qualitative responses also showed that health and well-being issues during the pandemic were mostly challenging and often led to reduced research productivity. One hundred and eighteen of 408 participants (29%) highlighted mental health issues; 108 (26%) of these were obstacles for research productivity resulting from anxiety and fear about the pandemic, burnout/exhaustion, increased stress in working to meet the new demands, feeling overwhelmed due to unrealistic expectations, lack of motivation because of decreased social connectivity and lack of institutional support. Below is an example of how a student referred to the feeling of being worn out due to lack of social interactions during the pandemic:The current state of the pandemic has decreased social interactions via face-to-face interaction, which has severely impacted mental health and productivity. The inability to separate work from home has also impacted research quality. The output of research has been relatively high because it is constantly happening, but that has worn on individuals and is affecting the work that gets done in a laboratory setting because we crave the social interactions we get in a laboratory environment. [male, grad student] This excerpt suggests that mental health issues limited the work on research and scholarship during the pandemic. At the same time, there were rare benefits of working from home. The remaining 12 participants (3%) who highlighted mental health issues wrote about resilience for surviving through a pandemic as a motivating factor to improve or sustain research and scholarship productivity. For instance, a female full professor from the Humanities, with elderly dependents at home, felt the resilience “as time progressed” and when her students adapted to the new normal and “still achieved high proficiency levels.” Furthermore, she noticed that despite the imperfection of the Zoom platform, “there will be aspects of this platform that will be integrated into future planning.” In addition to mental health issues, a few scholars (5% of all participants) highlighted physical health changes; 4.6% of which were challenging and often led to reduced research productivity resulting from eye strain during long screen time, lack of access to fitness facilities, and inaccessibility to health care. The remaining 0.4% of the physical change were more positive such as increased time for exercise while working from home. Finally, we decided to present "Work-Life Issues" as a separate theme under health and well-being. Although work-life issues can be related to physical and mental health, we noticed that a significant number of survey respondents specifically highlighted a challenge due to work-life conflicts and/or an affordance pertaining to establishing a new work-life harmony during the pandemic that impacted their research and scholarship. Seventeen percent of the participants highlighted work-life issues, 6% of which mentioned better work-life harmony during the pandemic, leading to higher productivity. A student from the Sciences said, "By having the flexibility to work remotely, I am able to have more control over my work schedule and have more access to home amenities. These amenities and the additional flexibility have allowed me to be more productive" [male, graduate student]. For this participant, the home environment leads to efficiency. Another 11% highlighted difficulties that reduced productivity, such as lack of office space and caring for young children. Parents, particularly women, unequally experienced childcare and caregiving responsibilities during the pandemic (e.g., Langin, 2021; Lechuna-Pena, 2022). This trend didn't appear in our quantitative findings for all except full professors and lecturers. This finding might be due to low COVID-19 cases at the beginning of the pandemic that provided in-person school and childcare options to parents after the initial six months of closures. Although our quantitative findings showed that female-identified faculty were only disadvantaged at the full professor level and dependent care made a difference only for lecturers, most of the qualitative excerpts (10% out of 17% that highlighted work-life issues) highlighted work-life conflict from female-identified scholars. The majority of these females (6%) were in lecturer positions. A lecturer from the Humanities and the Arts wrote about her difficulty focusing while she had kids at home:As we negotiate my work being remote, our children being remote, and my husband still working away, our marriage has also been impacted. Because my work is so fragmented (by merit of working in the same place I live while simultaneously having my children attempt school in the same location), I find it incredibly difficult to find sustained periods of focus for detailed thinking and planning. I can, then, hobble along with highly routinized work, but innovation, analysis, and problem-solving are challenging. [female, lecturer, caring for children at home] Qualitative analysis also highlighted tenure-line faculty who experienced work-life conflicts. The excerpt below from a pre-tenure faculty talks about not being able to keep up with the demands of building a career for tenure and highlights the difficulty in completing research with two young children at home:This is a time in my career where I need to be publishing a lot, getting grants to support my research, and making crucial networking connections at conferences. None of this has happened, and it concerns me with tenure. I don't want to get a year extension to tenure because it impacts my earning potential for the rest of my career (which I don't believe is fair because of a pandemic that I had not to control over). Everyone is experiencing their own demons during this pandemic. But it can not be stated enough how hard it has been getting any research done with a 2 and 4 year old at home. It is mentally hard to see colleagues without kids being more productive and just having absolutely no way to get there. [female, Assistant Professor, caring for children at home] The excerpt above refers to the university system's guidance on the choice of stopping the tenure clock for one year during the COVID-19 pandemic (University of Maine Promotion and Tenure Guidance, 2022). Although this option is provided, the faculty mentions the unfairness because stopping tenure will have negative financial impacts. Previous research highlighted the long terms impacts of stopping the tenure policies in creating gender and racial differences in pay raises (Freund et al., 2016). A recent study looked at faculty decisions on application to a tenure during the pandemic in two research universities in Colorado. The study found that over half of the pre-tenured faculty accepted the option to stop the tenure clock. Furthermore, ethnically minoritized and women-identified faculty in Social and Behavioral Sciences were more likely to accept the offer. The burden of the work-life conflict unequally forces some faculty to accept the stopping the tenure option. While this option allows the faculty to continue their career, it might lead to long-term consequences such as lower pay. Discussion and Implications for Future Directions This study reports on data from a public research university that has state-wide and nationwide recognition and impact. Since the university’s largest campus is located in a rural town in the Northeastern region of the U.S., some results might reflect the advantages and challenges of being located in a rural geography. Before we discuss the findings of this study, it is important to note that surveys are designed to reveal respondents’ perceptions rather than their actual behavior (Rubenfeld, 2004). While this characteristic of surveys might be seen as a limitation, understanding how faculty, staff and students perceive their productivity is crucial for the wellbeing of higher education institutions. Despite increases in research and development funding during the pandemic, the results from the survey we conducted indicated that the pandemic led to perceptions of reduced research activity across all disciplines, genders, and academic positions. However, some male full professors and female graduate students reported an increase or sustained research and scholarship activity during the pandemic. Qualitative data from the survey suggested that the increased and sustained research and scholarship activity can be attributed to these participants’ efforts in adaptations to pandemic conditions, such as looking for new research and scholarship opportunities (synthesis of review papers), utilizing online resources and collaborations, and investing in equipment for effective virtual collaborations. We suggest future studies to explore why these demographics were more likely to be able to adapt to pandemic conditions than others. We draw on the concept of the “wicked” problem to understand the impact of the pandemic on research and scholarship activity in higher education institutions in preparation for future widespread societal disruptions. As a wicked problem requiring urgent solutions to life-threatening situations, the COVID-19 pandemic underscored several ongoing issues in higher education and clarified how faculty and staff experience these problems in some cases. Financial instability has been a struggle for many higher education institutions in the U.S. leading to a decrease in faculty and staff hires or leaning to part-time or adjunct positions (e.g., Stenerson et al., 2010). This trend has continued during post-recession as a mitigation strategy for budget shortfalls (Cummings et al., 2021). The decrease in faculty and staff from previous years can be one factor for why participants in our study stated “Increased Work Responsibilities” as the main reason for reduced research and scholarship activity. For faculty, transitioning to online teaching is the main reason for reduced research activity. In addition, limited administrative support staffing, researchers in higher education had to undertake new responsibilities in the form of additional meetings, managerial responsibilities, and online teaching preparation and execution. These new roles that are often recorded as service can be in a conflicting direction than faculty’s research program (Sotto-Santiago et al., 2021). Similar to findings from other recent studies (Ramlo, 2021; Sotto-Santiago et al., 2021), a substantial number of our participants mentioned that pandemic-induced additional responsibilities, such as simultaneous coordination of academic and household duties, have produced mental and physical exhaustion, burnout, and the sensation of being fundamentally overwhelmed. It will be important for higher education institutions to respond to the wicked problems created and clarified by this pandemic event. The survey outcomes suggest it would be prudent to consider strategic development of administrative support capacity, backup planning for operations related to research and scholarship activities, and scenario response evaluations for circumstances that might predictably affect research and scholarship activities. Work-life harmony was disrupted for many students, staff, and faculty during the pandemic and required individuals to create novel strategies to conduct research and scholarship activities. Multiple responses highlight struggles with childcare. The qualitative data on these work-life conflicts were mostly by female respondents. Understanding work-life issues has implications beyond the pandemic due to another enduring problem of day care/afterschool shortages that have been particularly daunting in small rural communities (Becker, 2020). Family responsibilities can limit the deep thinking needed for research and scholarship activity (Woodthorpe, 2018). On the other hand, flexibility created by virtual options can reduce stress for academic professionals that have considerable family responsibilities and limited childcare options. Studies revealed that the most demanding of the new responsibilities in higher education can be transitions to online teaching and designing effective virtual teaching strategies (Cutri et al., 2020; Sotto-Santiago et al., 2021). Our findings align with other studies showing that faculty needed a substantial amount of time to transition to online or hybrid modes of teaching. Scholars suggest that these new modes of learning will remain as critical options even when we move beyond the pandemic. Others have proposed that higher education must adapt to a new student population by remaining flexible and continuing to offer virtual and in-person options, while maintaining a community atmosphere expected in post-secondary experiences (Eringfeld, 2021). Useful modifications to teaching include virtual conferences and seminars, open access materials, and knowledge sharing through online platforms that encourage flexibility for those involved in fostering the community atmosphere in higher education moving forward post-covid (Eringfeld, 2021). Since the new demands will affect the way research and scholarship activities are conducted, it will be important for institution administrators to develop expanded guidelines and policies for hybrid working conditions for university faculty, staff, and graduate students. These guidelines or policies can be framed through forums where administrators, faculty, and graduate students provide perspectives. Second to “Increased Work Responsibilities,” the survey we conducted reveals that limited access to research fields, in science and social sciences, was a crucial reason for reduced research and scholarship activity. The COVID-19 pandemic has had startling effects on researcher mobility, decreasing their ability to access data (Woolston, 2020). Stationery labs and fieldwork were paused for many studies due to the inability to travel (Erickson, 2020). The predicament may cause “long-lasting impacts that could transform research and collaborations,” negatively disrupting younger researchers’ productivity, particularly for graduate students or pre-tenured faculty (Woolston, 2020, p. 614). Scholars argued that it is essential to shift research and collaborations further into the virtual realm by developing larger database initiatives, online library resources, and digitizing museum archives and collections (Scerri et al., 2020). Our survey responses indicate that the shifts were being initiated by some researchers, our institution, and the organizations as an adaptation to sustain scholarly activity. The results support the need for expanded investments in research equipment that can help provide virtual data transfer from remote research fields (e.g., rainforests, the Arctic). Artificial Intelligence, though applications are currently limited, can be utilized to build such instruments that will provide virtual, effective, and automated data (Amer-Yahia & Senjuti Basu, 2021). Other than international funds, federal and state agencies can create funding opportunities to build an infrastructure for remote work. Should these advancements be completed, researchers can be prepared for the next wicked problem such as the next pandemic or severe impact of climatic changes. Universities prosper when there is a commitment to facilitating programs and interactions focused on developing social connections and a supportive community for faculty (Buckley, 2020; Himelein & Anderson, 2020; McCauliff, 2020). Our participants mentioned that a major complication of the COVID-19 pandemic is the lack of in-person social engagement on the college campus and more specifically shifting social interactions to a virtual setting. According to Buckley (2020), virtual interactions can provide advantages yet create challenges. Zoom meetings can bring researchers together that may be geographically apart and connect with other faculty rarely seen on campus. Conversely, it can elicit awkward silences and bar side conversations that are less likely to occur in person leading to a decrease in faculty development. The pandemic conditions reveal a demand to create and maintain common guidelines for students and faculty to address the current lack of social connectedness that has impacted productivity (Milliken et al., 2020). Moving forward, it is critical to organize events and spaces where faculty and students can engage in informal mentoring and collaborative activities (Howley, 2020). In order to support faculty in the new normal, administrators in departments or units need to maintain a culture where faculty can feel agency to build a strong career and engage faculty in further learning (Neumann, et al., 2006). Faculty can feel stronger agency if their departments provide professional development resources, see models of balancing work-life priorities, perceive an aligned fit to the program’s goals, and transparency in guidelines, and tenure process (Campbell & O’Meara, 2014). Further related to the social connectivity, perceptions of collegiality are crucial for improved faculty agency (Daly & Dee, 2006; Lindholm, 2004). As we might work at distant geographic location at times, it is important to improve on these components of departmental communities. Findings from studies that explored the COVID-19 pandemic conditions at higher education institutions agree that resuming pre-pandemic operations is not a prudent strategy going forward (Roy, 2020). There is a need to collaboratively develop new sets of guidelines to help researchers and scholars optimize operational responses based on what we have learned from the pandemic conditions. New and yet evolving digital technologies can help create effective and equitable virtual teaching and research opportunities; however, accommodations are needed for the additional time and resources required to transition into digital spaces. Finally, it is important to note that the findings of this study are limited to experiences in one institution located in a rural region. Looking at one institution across disciplines can improve understanding of contextual factors affecting responses to modern demands on higher education and illuminate inequities among demographic groups that have unique responses to stresses imposed by logistical and funding changes. The focus on individual institutions in unique settings can also provide a basis for comparisons of institutions in varied geographic settings and demographic situations to guide holistic strategies for higher education at a national level. Not all of the findings from this study are generalizable, but the conceptual approach, interpretations from multiple lines of evidence, and identification of knowledge gaps that remain to be filled informs other investigations on research and scholarship productivity in other higher education contexts and during different stages of the pandemic and other large-scale disturbances to stability in societies. We propose a synthesis of these studies in the future to examine the impact of the pandemic across settings, demographics, and time. Author Contributions All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Asli Sezen-Barrie, Lisa Carter, and Sean Smith. The first draft of the manuscript was written by Asli Sezen-Barrie, Lisa Carter and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript. Funding Partial financial support was received from the UMaine’s Office of the Vice President for Research. The authors have no relevant financial or non-financial interests to disclose. Declarations Human Subjects Review Statement The study includes anonymous human-subject data. The study is reviewed by UMaine’ Institutional Review Board (IRB) and approved as exempt. All participants had to agree to the informed consent letter prior to completing the survey. Conflict of Interest Statement On behalf of all the authors, the corresponding author declares 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 Amer-Yahia, S., & Senjuti Basu, R. (2021). Data management to social science and back in the future of work. In: L. Guoliang, L. Zhanhuai, I. Stratos & Divesh Srivastava (Eds.), SIGMOD ’21: International Conference on Management of Data, pp 2876– 2877. Virtual Event, ACM. Barr A Turner SE Expanding enrollments and contracting state budgets: The effect of the great recession on higher education The Annuals of the American Academy of Political and Social Science 2013 650 1 168 193 10.1177/0002716213500035 Becker, A. (2020). Middle-income and rural families disproportionately grapple with child-care deserts, new analysis shows. Washington Post. https://www.washingtonpost.com/nation/2020/06/22/middle-income-rural-families-disproportionately-grapple-with-child-care-deserts-new-analysis-shows/ Bichsel, J. and McChesney, J. (2017). The gender pay gap and the representation of women in higher education administrative positions: The century so far. CUPAHR (College and University Professional Association for Human Resources), http://www.cupahr.org/surveys/briefs.aspx Brown, R. (2018). Higher education and inequality. Perspectives: Policy and Practice in Higher Education, 22 (2), 37–43. 10.1080/13603108.2017.1375442 Buckley H Faculty development in the COVID-19 pandemic: So close - yet so far Medical Education 2020 54 12 1189 1190 10.1111/medu.14250 32436272 Campbell CM O’Meara K Faculty agency: Departmental contexts that matter in faculty careers Research in Higher Education 2014 55 1 49 74 10.1007/s11162-013-9303-x Canadian Association of University Teachers (2021). Impacts of pandemic. https://www.caut.ca/sites/default/files/covid_release-impacts_of_pandemic-en-final-08-19-20.pdf Carlson, A. (2019). SHEEO releases state higher education finance FY 2017. State Higher Education Finance. https://sheeo.org/sheeo-releases-state-higher-education-finance-fy-2017/ Creswell JW Plano Clark VL Designing and conducting mixed methods research 2017 Sage Publications Creswell JW Miller DL Determining validity in qualitative inquiry Theory into Practice 2000 39 3 124 130 10.1207/s15430421tip3903_2 Crawford, J., Butler-Henderson, K., Rudolph, J., Malkawi, B., Glowatz, M., Burton, R., Magni, P., & Lam, S. (2020). COVID-19: 20 countries' higher education intra-period digital pedagogy responses. Journal of Applied Learning & Teaching 3(1), 1–20. 10.37074/jalt.2020.3.1.7 Culver KC Young RL Barnhardt CL Communicating support: Examining perceived organizational support among faculty members with differing appointment types Innovative Higher Education 2020 45 4 299 315 10.1007/s10755-020-09503-z Cummings, K., Laderman, S., Lee, J., Tandberg, D., & Weeden, D. (2021). Investigating the impacts of state higher education appropriations and financial aid. State Higher Education Executive Officers. https://sheeo.org/wp-content/uploads/2021/05/SHEEO_ImpactAppropationsFinancialAid.pdf Cutri RM Mena J Whiting EF Faculty readiness for online crisis teaching: Transitioning to online teaching during the COVID-19 pandemic European Journal of Teacher Education 2020 43 4 523 541 10.1080/02619768.2020.1815702 Dalhousie Faculty Association (2020). DFA Workload survey summary report. https://immediac.blob.core.windows.net/dfawebsite/2020%20Bargaining%20PDFs/DFA.Workload.Survey.Summary.Report.Sept.2020.pdf Daly CJ Dee JR Greener pastures The Journal of Higher Education 2006 77 776 803 10.1080/00221546.2006.11778944 Douglas, J. A. (Ed.). (2016). The new flagship university: Changing the paradigm from global ranking to national relevancy. Palgrave Macmillan. El Masri A Sabzalieva E Dealing with disruption, rethinking recovery: Policy responses to the COVID-19 pandemic in higher education Policy Design and Practice 2020 3 3 312 333 10.1080/25741292.2020.1813359 Eringfeld, S. (2021). Higher education and its post-coronial future: Utopian hopes and dystopian fears at Cambridge University during covid-19. Studies in Higher Education (Dorchester-on-Thames), 46(1), 146–157. 10.1080/03075079.2020.1859681 Fensie, A., & Sezen-Barrie, A. (2021). Embodied discourse analysis of online student study sessions: A novel method of screen recording research. Proceedings of the 15th International Conference of the Learning Sciences-ICLS 2021, Germany, 931–932. https://drive.google.com/file/d/1NuYhdOKDgpp_omNH6qXKYmAh2G5_c9iv/view Flaherty, C. (2020). Extraordinary measures. Inside Higher Education. https://www.insidehighered.com/news/2020/03/18/how-institutions-are-approaching-scientific-research-during-covid-19 Fox-Cardamone L Inequity in the academy: A Case study of factors influencing promotion and compensation in American universities Forum on Public Policy Online 2010 5 1 18 Geiger, R. (2010). Impact of the financial crisis on higher education in the United States. International Higher Education, (59). 10.6017/ihe.2010.59.8486 Grimm D Respirators, quarantines, and worst-case scenarios: Lab animal facilities grapple with the pandemic Science (american Association for the Advancement of Science) 2020 10.1126/science.abb7864 Himelein, M. J., & Anderson, H. L. (2020). Developing community among faculty: Can learning circles provide psychosocial benefits? The Journal of Faculty Development, 34(1), 23–29. https://go.exlibris.link/rp5L6lTY Howley, H. A. (2020). Success amid ruins: Onboarding new faculty during financial crisis and institutional change. The Journal of Faculty Development, 34(2), 73–80. https://go.exlibris.link/JWtFwRhr Korbel JO Stegle O Effects of the COVID-19 pandemic on life scientists Genome Biology 2020 21 1 113 113 10.1186/s13059-020-02031-1 32393316 Langin K Pandemic hit academic mothers hard, data show Science 2021 371 6530 660 660 10.1126/science.371.6530.660 33574191 Levin, D. (2020). Researchers across Tufts pivot to fight COVID-19. Tufts Now. https://now.tufts.edu/articles/researchers-across-tufts-pivot-fight-covid-19 Lincoln, Y. S., & Guba, E. G. (1985). Naturalistic inquiry. Sage Publications. Lindholm JA Pathways to the professoriate: The role of self, others, and environment in shaping academic career aspirations Journal of Higher Education 2004 75 603 636 10.1080/00221546.2004.11773579 Maine Department of Education. (2021). COVID-19 toolkit. Coronavirus (COVID-19) Resources for Schools. https://www.maine.gov/doe/covid-19/toolkit Maine State Data (2021). COVID-19 case trends. https://www.maine.gov/dhhs/mecdc/infectious-disease/epi/airborne/coronavirus/data.shtml Matsuo, H., Billiet, J., Loosveldt, G., Berglund, F., & Kleven, Ø. (2010). Measurement and adjustment of non-response bias based on non-response surveys: The case of Belgium and Norway in the European social survey round 3. In Survey Research Methods 4(3): 165–178. McCauliff, K. L. (2020). Changing the conversation: How small revisions to new faculty orientation can lead to big changes in culture. The Journal of Faculty Development, 34(2), 47–52. https://go.exlibris.link/zDSNl01W McGlynn, A. P. (2014). The economic story of higher education: Stratification and inequities. The Education Digest, 80(1), 52. https://go.exlibris.link/rTRT09RS Milliken FJ Kneeland MK Flynn E Implications of the COVID-19 pandemic for gender equity issues at work Journal of Management Studies 2020 57 8 1767 1772 10.1111/joms.12628 Neumann, A., Terosky, A. L., & Schell, J. (2006). Agents of learning: Strategies for assuming agency, for learning, in tenured faculty careers. In S. J. Bracken, J. K. Allen, & D. R. Dean (Eds.), The balancing act: Gendered perspectives in faculty roles and work-lives (pp. 91–120). Stylus. Oleschuk M Gender equity considerations for tenure and promotion during COVID-19 The Canadian Review of Sociology 2020 57 3 502 515 10.1111/cars.12295 32779307 O'Meara K Kuvaeva A Nyunt G Waugaman C Jackson R Asked more often: Gender differences in faculty workload in research universities and the work interactions that shape them American Educational Research Journal 2017 54 6 1154 1186 10.3102/0002831217716767 National Center for Education Statistics (n.d.). https://nces.ed.gov Peters BG What is so wicked about wicked problems? A conceptual analysis and a research program Policy & Society 2017 36 3 385 396 10.1080/14494035.2017.1361633 Ramlo S The coronavirus and higher education: Faculty viewpoints about universities moving online during a worldwide pandemic Innovative Higher Education 2021 46 3 241 259 10.1007/s10755-020-09532-8 33437113 Redden, E. (2020). Nonessential research has halted on many campuses. Inside Higher Ed. https://www.insidehighered.com/news/2020/03/30/nonessential-research-has-halted-many-campuses Rosen HS Sappington AJ The impact of endowment shocks on payouts The Journal of Higher Education 2019 90 5 690 716 10.1080/00221546.2019.1621115 Roy, A. (2020). The pandemic is a portal. Financial Times 3(4). https://go.exlibris.link/B2ZZf2X9 Rubenfeld GD Surveys: An introduction Respiratory Care 2004 49 10 1181 1185 15447800 Sanders, R. (2020). Scientists pivot to COVID-19 research, hoping for quick result to deal with pandemic. Berkeley News. https://news.berkeley.edu/2020/05/19/scientists-pivot-to-covid-19-research-hoping-for-quick-results-to-deal-with-pandemic/ Scerri EML Kühnert D Blinkhorn J Groucutt HS Roberts P Nicoll K Zerboni A Orijemie EA Barton H Candy I Goldstein ST Hawks J Niang K N'Dah D Petraglia MD Vella NC Field-based sciences must transform in response to COVID-19 Nature Ecology & Evolution 2020 4 12 1571 1574 10.1038/s41559-020-01317-8 32929241 Sotto-Santiago, S., Dilly, C., O'Leary, H., Craven, H., Kara, A., Brown, C., Kressel, A., Rohr-Kirchgraber, T., & DiMeglio, L. (2021). Reframing academic productivity, promotion and tenure as a result of the COVID-19 pandemic. The Journal of Faculty Development, 35(1), 10–18. https://go.exlibris.link/kDYlXzbz Stenerson, J., Blanchard, L., Fassiotto, M., Hernandez, M., & Muth, A. (2010). The role of adjuncts in the professoriate. Peer Review: Emerging Trends and Key Debates in Undergraduate Education, 12(3), 23. https://go.exlibris.link/JzgQyh9j Thelin JR Campus and community: A world turned inside out? Review of Higher Education 2001 25 1 1 14 10.1353/rhe.2001.0016 University of Maine System Strategic Planning Data Book (2021). Fast facts about the University of Maine System. https://www.maine.edu/databook/ University of Maine Office of the President (2021). University of Maine budget town hall presentation. https://umaine.edu/president/wp-content/uploads/sites/8/2021/05/town-hall-budget-presentations-05-04-2021.pdf University of Maine (2021). COVID-19 research continuity update. https://umaine.edu/research/category/covid-19/ University of Maine Promotion and Tenure Guidance. (2022). Prior credit and stopping the tenure clock. https://umaine.edu/provost/wp-content/uploads/sites/14/2022/03/UMaine-PT-Guidance-2022.pdf Viglione G Are women publishing less during the pandemic? Here's what the data say Nature 2020 581 7809 365 366 10.1038/d41586-020-01294-9 32433639 Webber DA State divestment and tuition at public institutions Economics of Education Review 2017 60 1 4 10.1016/j.econedurev.2017.07.007 Woodthorpe K Survive and thrive in academia: The new academic’s pocket mentor 2018 Routledge Woolston C Pandemic darkens postdocs' work and career hopes Nature 2020 585 7824 309 312 10.1038/d41586-020-02548-2 32901125	0	PMC9734507	NO-CC CODE	2022-12-14 23:28:29	no	Innov High Educ. 2022 Dec 8;:1-25	utf-8	Innov High Educ	2,022	10.1007/s10755-022-09639-0	oa_other
==== Front Indian J Pediatr Indian J Pediatr Indian Journal of Pediatrics 0019-5456 0973-7693 Springer India New Delhi 36482236 4385 10.1007/s12098-022-04385-9 Original Article Multisystem Inflammatory Syndrome in Children (MIS-C) Related to SARS-CoV-2 and 1-Year Follow-up http://orcid.org/0000-0002-9822-7187 Kapoor Rashmi [email protected] Chandra Tarun Singh Chandra Prakash Singh Ruchira Pandey Ishita Department of Pediatrics, Regency Hospital, A-2, Sarvodaya Nagar, Kanpur, Uttar Pradesh 208005 India 9 12 2022 15 9 6 2022 18 8 2022 12 9 2022 © The Author(s), under exclusive licence to Dr. K C Chaudhuri Foundation 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 To study the demographics, clinical profile, management, outcome and 1-y follow-up of children with multisystem inflammatory syndrome in children (MIS-C). Methods This was a retrospective observational study of 54 Children satisfying the WHO MIS-C criteria admitted during the study period. Results Fifty-four children were included in the study, median age was 5.5 (IQR 8.75), 68.5% were males. PICU admissions were 77%. Most involved organ was gastrointestinal (92%), followed by cardiovascular 85%, central nervous system (CNS) 74%, respiratory 72%, mucocutaneous 59%, and renal 31%, and hypotension was the presenting symptom in 43%. Coronary artery dilatation was seen in 1 (1.8%) child. All patients presented with more than three organs involvement. Raised procalcitonin was seen in 100%, raised BNP in 31.5%, low ejection fraction in 83.3%, and abnormal radiograph in 59%. All children were positive for anti-SARS-CoV-2 antibodies and negative for cultures. Methylprednisolone or intravenous immunoglobulin (IVIg) was used in 77%, mechanical ventilation in 18.5%, and inotropic support in 77%. Aspirin was used in 48% and low molecular weight heparin (LMWH) in 54%. The median stay in hospital was 7 d (IQR 2). There was 1 mortality (1.8%). On 7-d follow-up, 98% children had a normal echocardiography; on 6 mo and 1-y follow-up, all children had normal echocardiography. Conclusion MIS-C is an important complication of COVID-19 infection. Cardiac involvement resolves completely. Coronary artery involvement is not common. Keywords COVID-19 IVIg Kawasaki disease MIS-C Methylprednisolone ==== Body pmcIntroduction When COVID-19 was first recognized, it was believed that it was almost benign and of little consequence in the pediatric population. Globally, the initial reports indicated that children have lower rates of hospitalization and death than adults [1]. Subsequently, since the emergence of multisystem inflammatory syndrome in children (MIS-C), COVID-19 infections can have serious consequences in children as well [2]. In April 2020, during the peak of the COVID-19 pandemic in Europe, a cluster of children with hyperinflammatory shock was reported in England [3]. Soon after, several European countries, United States, Asia, and Latin America reported the occurrence of such hyperinflammatory process in children, possibly related to SARS-CoV-2 infection [3–7]. The patients’ signs and symptoms were temporally associated with COVID-19 but presumed to have developed 2–4 wk after acute COVID-19 [8]. MIS-C is characterized by fever, elevated inflammatory markers, and high levels of both pro- and anti-inflammatory cytokines. According to the available literature till now, the spectrum of MIS-C is a combination of typical/atypical Kawasaki disease (KD), toxic shock syndrome (TSS), and macrophage activation syndrome (MAS)/hemophagocytic lymphohistiocytosis (HLH) with prominent involvement of mucocutaneous, gastrointestinal, cardiovascular, or neurological systems [9]. The prevailing hypotheses for the pathogenesis of MIS-C are: 1) a postinfectious autoimmune-mediated inflammatory process, 2) a cytokine storm instigated by a superantigen response, and 3) a dysregulated immune response to chronic exposure to SARS-CoV-2 viral antigens [10]. Material and Methods The two MIS-C peaks had a temporal relation with two COVID-19 waves in India (1st wave from August to October in 2020 and the 2nd sharp wave in April/May in 2021). The data of 54 children from 30th September 2020 to June 6th 2021, conforming to the case definition of MIS-C by WHO [11] were collected. Patients with hemodynamic instability (requiring ionotropic support or fluid resuscitation), and or organ failure were admitted to pediatric intensive care unit (PICU) and were categorized as severe MIS-C, and all those hemodynamically stable, but needing hospitalization were categorized as moderate MIS-C. Bedside echocardiography and cardiac triaging is done for every critically ill child coming to the authors’ emergency department, suspected to be having sepsis. If MIS-C or any cardiac abnormality is suspected, patients are subsequently seen by a pediatric cardiologist. Cases of tropical fevers like dengue, scrub typhus, leptospirosis, malaria, enteric, and viral exanthems are endemic during the monsoon season, and therefore, the authors screened for these diseases. A positive dengue IgM ELISA/RT-PCR, scrub typhus card test, leptospira IgM, malaria card test/peripheral smear, blood culture for salmonella, were taken as diagnostic for specific tropical disease and were excluded from the study. Patients with a positive blood or urine culture, negative SARS-CoV-2 antibodies, and newborns were excluded from the study. The data were collected on the following parameters: demographics, clinical findings, echocardiography, radiology, laboratory investigations, treatment received including intensive care interventions and outcome. After discharge 7-d, 6-mo, and 1-y follow-ups in the outpatient department with echocardiography were noted in the datasheet. For statistics, MS Excel was used for descriptive analysis: count, mean/median, standard deviation (SD)/(IQR). Continuous variables were expressed as mean and SD, medians and interquartile ranges, and categorical variables were expressed as counts and percentages. Results The present study included 54 children; median age was 5.5 y (IQR 8.75), 68.5% were males. Children less than 5 y were 50%. PICU admissions were 77%. Most involved organ was gastrointestinal (92%), followed by cardiovascular 85% (marked by reduced ejection fraction, abnormal echocardiography, or raised cardiac enzymes). CNS involvement in the form of headache, irritability and seizures was present in 74%, 1 presented as acute encephalitis syndrome and 1 as Guillain–Barré syndrome (Table 1). Hypotension was the presenting symptom in 43% children. Coronary artery dilatation was seen in 1 (1.8%) child. All patients presented with more than three organ involvement. Raised procalcitonin was seen in 100%, raised BNP in 31.5%, 74% children came with thrombocytopenia, low ejection fraction in 83.3%, and abnormal radiograph in 59%. All children were positive for anti-SARS-CoV-2 antibodies (Table 2). Methylprednisolone or IVIg was used in 77%; only methylprednisolone was given in 18%; both methylprednisolone and IVIg were given in 58% of children; a repeat dose of IVIg was given to 5.6%. Mechanical ventilation was needed in 18.5%, whereas noninvasive ventilation was given in 27.8%. Inotropic support was given in 77%. Aspirin was used in 48% and LMWH in 54% (Table 3). Median stay in hospital was 7 d (IQR 2). There was 1 mortality (1.8%). On 7-d follow-up after discharge, 98% children had a normal echocardiography and inflammatory markers, except for one who had a coronary artery dilatation > 2 z score; on 6-mo and 1-y follow-ups, all children had complete cardiac recovery. Table 1 Children fulfilling the WHO criteria for MIS-C (N = 54) WHO criteria for MIS-C [11] (N) % Children with fever And (54) 100% A Any two of the following: A1 Rash or bilateral non-purulent conjunctivitis or muco-cutaneous inflammation signs (oral, hands or feet) (31) 59.3% A2 Hypotension or shock (23) 42.6% A3 Features of myocardial dysfunction, pericarditis, valvulitis, or coronary abnormalities (including ECHO findings or elevated troponin/NT-proBNP). (46) 85% A4 Evidence of coagulopathy (by PT, PTT, elevated D-dimers) (15) 28% A5 Acute gastrointestinal problems (diarrhea, vomiting, or abdominal pain) (50) 92.6% And B1 Elevated markers of inflammation such as elevated C-reactive protein (43) 79.6% Or B2 Elevated markers of inflammation such as elevated procalcitonin (54) 100% And C No other obvious microbial cause of inflammation including bacterial sepsis, staphylococcal or streptococcal shock syndromes None And D Evidence of COVID-19 (serology) (54) 100% Table 2 Abnormal laboratory parameters and other investigations Parameters % Median (IQR), Mean (SD) Raised procalcitonin (> 0.02 ng/mL) 100 Median 3.7 (31.1) Raised CRP (> 0.5 mg/dL) 79.6 Median 12.0 (26.9) Raised troponin I (> 0.02 ng/mL) 37.0 Median 0.07 (0.16) Raised BNP (> 100 pg/mL) 31.5 Median 220.0 (831) Raised CKMB (> 4.3 ng/mL) 51.9 Median 12.0 (29.6) Neutrophilia 20.4 Mean 73.4 (15.4) Lymphopenia 13.0 Mean 19.4 (8.4) Thrombocytopenia < 150000 74.1 INR > 1.49 27.8 Median 1.6 (0.24) Raised SGOT Raised SGPT 61 48 EF < 54% 83.3 Median 50.0 (15.0) Abnormal echo (low EF, MR, dilated RA, dilated RV, TR) 46.3 Dilated coronaries 1.8 Abnormal radiograph (b/l patchy opacity in chest radiograph, ground glass opacity in chest radiograph, or mild pleural effusion) 59.3 Raised anti-SARS-CoV-2 spike protein antibody IgG > 0.8 U/mL 100 Median 75.2 (118.0) BNP Brain natriuretic peptide, CKMB Creatinine kinase myocardial band, CRP C-reactive protein, EF Ejection fraction, IgG Immunoglobulin G, INR International normalized ratio, MR Mitral regurgitation, RA Right atrium, RV Right ventricle, SARS-CoV-2 Severe acute respiratory syndrome coronavirus 2, SGOT Serum glutamic oxaloacetic transaminase, SGPT Serum glutamic pyruvic transaminase, TR Tricuspid regurgitation Table 3 Treatment modalities Modality n (%) Immunomodulators used 42 (77%) Only methylprednisolone 10 (18%) Only IVIg 1 (1.8%) Both IVIg + methylprednisolone (combination) 31 (57.4%) Repeat dose of IVIg & escalation of methylprednisolone 3 ( 5.6%) Mechanical ventilation 10 (18.5%) Noninvasive ventilation 15 (27.8%) Inotropic support 42 (77%) Aspirin 25 (48.1%) LMWH 28 (53.7%) IVIg Intravenous immunoglobulin, LMWH Low molecular weight heparin Discussion Feldstein et al. [12] studied a cohort of 539 MIS-C patients throughout the United States and demonstrated a nearly 75% ICU admission rate, 45% vasopressor requirement with less than 2% mortality, with the majority of patients achieving a complete recovery. In the present study, 77% were PICU admissions, 77% with inotropic support, and 1.8% mortality. In the present study, the median age was 5.5 y, IQR (8.75), which is lower than in other published reports [13, 14]. Majority of studies [2–4, 8] have reported gastrointestinal involvement as the commonest followed by cardiovascular, which is similar to the authors’ observation. The present study has reported 42% children with hypotension and shock, 85% children had cardiac involvement in the form of raised cardiac enzymes or low ejection fraction, which was observed in most studies [15, 16]. Different studies found abnormal radiographs in around 30%–35% of cases [2, 5, 15, 16]. Abnormal radiographs were found in 59% of the present cases. It was observed that 23% children in the present study admitted to the wards, did not receive any immunomodulators, and there was no difference in the outcome. Davies et al. [17] found no evidence of a difference in response in clinical markers of inflammation between patients with MIS-C, who were treated with IV immunoglobulin, steroids, or biologics, compared with those who were not. McArdle et al. [18] found no evidence that recovery from MIS-C differed after primary treatment with IVIg alone, IVIg plus glucocorticoids, or glucocorticoids alone. One study [19] found that among children and adolescents with MIS-C, initial treatment with IVIg plus glucocorticoids was associated with a lower risk of new or persistent cardiovascular dysfunction than IVIg alone. Sugunan et al. [20] reported that methylprednisolone pulse therapy was associated with favorable short-term outcomes. In the present study, methylprednisolone or IVIg was used in 77%; no biologics were used in any of the present patients. Treatment of MIS-C following COVID-19 has been instituted on the basis of experience with treatment of KD. No randomized clinical trial of long-term effects of IVIg in KD has been performed since 1990. The use of steroids has more evidence than IVIg, none of the studies show a clear evidence for use of salicylates. More studies and robust clinical trials are needed to determine the most effective and efficient therapies to treat serious and life-threatening complications of MIS-C [17]. The study has a few limitations. As the data were collected from the paper and electronic medical records, some data might have been missed. Some cases might also have been missed because the understanding of the clinicians about MIS-C was not uniform, and the tests needed to confirm the case definition might not have been ordered. Conclusion MIS-C is an emerging new syndrome, and the understanding about it is evolving. The cardiac complications that are common at the time of presentation resolve completely in most cases. Even though the younger children are asymptomatic for acute SARS-CoV-2 infection, they are at a risk for this serious post-COVID complications. Acknowledgements The following pediatricians, cardiologists, microbiologist and pathologist were involved in the clinical care of the patients studied; Drs Jaya Pandey, Prabha Verma, Saipavan, Dilip Singh, Harsh Agarwal, Abhineet Gupta, Niti Luthra and Anjali Tewari. Authors’ Contributions RK: Concept, design, preparation and finalizing the draft, review literature, critical revision; TC: Preparing the draft, collating the data, review literature; CPS: Collecting data and treatment of patients; RS, IP: Data collection and entry, care of the patients. RK will act as the guarantor for this paper. Declarations Ethics Approval Clearance from the hospital internal ethics committee has been taken. The approval no is RHL-IEC-16078. Conflict of Interest None. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Dufort EM Koumans EH Chow EJ New York State and Centers for Disease Control and Prevention Multisystem Inflammatory Syndrome in Children Investigation Team. Multisystem inflammatory syndrome in children in New York State N Engl J Med 2020 383 347 58 10.1056/NEJMoa2021756 32598830 2. Henderson LA Canna SW Friedman KG American College of Rheumatology clinical guidance for multisystem inflammatory syndrome in children associated with SARS-CoV-2 and hyperinflammation in pediatric COVID-19: version 2 Arthritis Rheumatol 2021 73 e13 29 10.1002/art.41616 33277976 3. Riphagen S Gomez X Gonzalez-Martinez C Wilkinson N Theocharis P Hyperinflammatory shock in children during COVID-19 pandemic Lancet 2020 395 1607 8 10.1016/S0140-6736(20)31094-1 32386565 4. Verdoni L Mazza A Gervasoni A An outbreak of severe kawasaki-like disease at the Italian epicentre of the SARS-CoV-2 epidemic: an observational cohort study Lancet 2020 395 1771 8 10.1016/S0140-6736(20)31103-X 32410760 5. Kest H Kaushik A DeBruin W Colletti M Goldberg D multisystem inflammatory syndrome in children (MIS-C) associated with 2019 novel coronavirus (SARS-CoV-2) infection Case Rep Pediatr 2020 2020 8875987 32733733 6. Grazioli S, Tavaglione F, Torriani G, et al. Immunological assessment of pediatric multisystem inflammatory syndrome related to coronavirus disease 2019. J Pediatric Infect Dis Soc. 2021;10:706–13. 7. Dhanalakshmi K Venkataraman A Balasubramanian S Epidemiological and clinical profile of pediatric inflammatory multisystem syndrome - temporally associated with SARS-CoV-2 (PIMS-TS) in Indian children Indian Pediatr 2020 57 1010 4 10.1007/s13312-020-2025-1 32769230 8. Feldstein LR Rose EB Horwitz SM Overcoming COVID-19 Investigators; CDC COVID-19 Response Team. Multisystem inflammatory syndrome in U.S. children and adolescents N Engl J Med 2020 383 334 46 10.1056/NEJMoa2021680 32598831 9. Gupta Dch S Chopra Md N Singh Md A Unusual clinical manifestations and outcome of multisystem inflammatory syndrome in children (MIS-C) in a tertiary care hospital of North India J Trop Pediatr 2021 67 fmaa127 10.1093/tropej/fmaa127 33513240 10. Mazer MB Bulut Y Brodsky NN Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network and BLOODNET Immunology Section. Multisystem inflammatory syndrome in children: host immunologic responses Pediatr Crit Care Med 2022 23 315 20 10.1097/PCC.0000000000002897 35050932 11. World Health Organization. Multisystem inflammatory syndrome in children and adolescents with COVID-19. Scientific brief. 2020 Available at: https://apps.who.int/iris/bitstream/handle/10665/332095/WHO-2019-nCoV-Sci_Brief-Multisystem_Syndrome_Children-2020.1-eng.pdf?sequence=1&isAllowed=y. Accessed on 30 Aug 2020. 12. Feldstein LR Tenforde MW Friedman KG Overcoming COVID-19 Investigators. Characteristics and outcomes of us children and adolescents with multisystem inflammatory syndrome in children (MIS-C) compared with severe acute COVID-19 JAMA 2021 325 1074 87 10.1001/jama.2021.2091 33625505 13. Ahmed M Advani S Moreira A Multisystem inflammatory syndrome in children: a systematic review EClinicalMedicine 2020 26 100527 10.1016/j.eclinm.2020.100527 32923992 14. Hoste L Van Paemel R Haerynck F Multisystem inflammatory syndrome in children related to COVID-19: a systematic review Eur J Pediatr 2021 180 2019 34 10.1007/s00431-021-03993-5 33599835 15. Whittaker E Bamford A Kenny J PIMS-TS Study Group and EUCLIDS andConsortia PERFORM Clinical characteristics of 58 children with a pediatric inflammatory multisystem syndrome temporally associated with SARS-CoV-2 JAMA 2020 324 259 69 10.1001/jama.2020.10369 32511692 16. Rodriguez-Smith JJ Verweyen EL Clay GM Inflammatory biomarkers in COVID-19-associated multisystem inflammatory syndrome in children, Kawasaki disease, and macrophage activation syndrome: a cohort study Lancet Rheumatol 2021 3 e574 84 10.1016/S2665-9913(21)00139-9 34124694 17. Davies P Lillie J Prayle A Association between treatments and short-term biochemical improvements and clinical outcomes in post-severe acute respiratory syndrome coronavirus-2 inflammatory syndrome Pediatr Crit Care Med 2021 22 e285 93 10.1097/PCC.0000000000002728 33767074 18. McArdle AJ Vito O Patel H BATS Consortium Treatment of multisystem inflammatory syndrome in children N Engl J Med 2021 385 11 22 10.1056/NEJMoa2102968 34133854 19. Son MBF Murray N Friedman K Overcoming COVID-19 Investigators. Multisystem inflammatory syndrome in children - initial therapy and outcomes N Engl J Med 2021 385 23 34 10.1056/NEJMoa2102605 34133855 20. Sugunan S Bindusha S Geetha S Niyas HR Kumar AS Clinical profile and short-term outcome of children with sars-cov-2 related multisystem inflammatory syndrome (MIS-C) treated with pulse methylprednisolone Indian Pediatr 2021 58 718 22 10.1007/s13312-021-2277-4 33876782	36482236	PMC9734509	NO-CC CODE	2022-12-14 23:28:29	no	Indian J Pediatr. 2022 Dec 9;:1-5	utf-8	Indian J Pediatr	2,022	10.1007/s12098-022-04385-9	oa_other
==== Front Public Choice Public Choice Public Choice 0048-5829 1573-7101 Springer US New York 1031 10.1007/s11127-022-01031-y Article The political economy of public health Furton Glenn L. [email protected] Rizzo Mario J. Harper David A. Denver, Colorado USA 9 12 2022 13 © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, 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. ==== Body pmcIn the spring of 2020, civilization briefly came to a near standstill as the world watched and anticipated, with trepidation, the potential catastrophe that might unfold from the viral outbreak now known as COVID-19. During this time, governments around the globe instituted unprecedented policy changes meant to slow the spread of the disease. Given the rapid onset of the contagion and the extreme medical and epidemiological uncertainty surrounding the pandemic, direct pharmaceutical interventions were limited. Early responses, instead, were “non-pharmaceutical” (Perra, 2021). They included event cancellations, school closures, shelter-in place orders, travel bans, remote work, curfews, and limitations on social gatherings. While some of the policies were fostered exclusively through private, voluntary institutions, many were sponsored or directly enforced by coercive, political means. The unprecedented use of state intervention in response to infectious disease provoked a host of questions concerning the role of the state and the political economy of public health. A recurring question centers around the structure of institutions. What constitutes a health-related public good, for example, is not institutionally neutral—it depends, indeed, on the rules structuring our social, political, and economic interactions. Political institutions are of particular interest, given the potential for opportunistic behavior. What incentives do those institutions foster? Furthermore, what sort of epistemic properties characterize public health institutions? From where do contemporary public health institutions originate? And how might we expect them to evolve following the COVID-19 pandemic? Public choice scholars often tackle such questions by asking what sort of rules might enhance the level of institutional robustness—especially in the face of rapidly changing conditions, such as a health crisis. Each of these questions is taken up by one or more of the following authors. Leeson and Thompson (2022, p. 2), in the opening article of this symposium, write that “public choice scholars have attended only modestly to issues in public health”. And while the authors cover a great deal of economic analysis conducted prior to the COVID-19 pandemic, the weight of the preceding scholarship does not measure up to the size and scope of contemporary public health institutions, which have grown to occupy a sizable seat at the political table. Political economists in many ways were caught off guard in the spring of 2020. The articles contained in this issue were collected in the spirit of redress, with the hope that each contribution might facilitate further contributions to fields broadly conceived under the umbrella of “the political economy of public health”, better preparing future generations of scholars for engagement in serious scholarship and policy discourse. Leeson and Thompson make an important initial contribution to the symposium by taking stock of the work on public health from the perspective of public choice. They highlight three main themes that emerge from the literature predating the pandemic. First, public health regulations often are driven by private interests—not public. Second, the allocation of public health resources also reflects private interests—not public. Third, public health policies may have perverse, unintended outcomes that undermine their stated goals. That observation should come as no surprise to those familiar with the “government failure” analogue to market failure. Their review offers a convenient and useful summary of past investigations for future scholars in the field. The private–public distinction in public health policy is taken up by Anomaly (2022), who argues in favor of relegating public health to those types of goods and services that constitute genuine (or “pure”) public goods. Anomaly goes even further in excluding certain goods from the domain of public health by way of what he calls “the conversion problem”. Through conversion, governments transform what otherwise would have been a privately constrained choice into one of public concern. Thus, even if one agrees that public health should be confined to the provision of genuine public goods, what constitutes a “public-health public good” is in constant flux depending on the nature of the health hazard we face as well as the surrounding rules and institutions that influence human behavior. Similarly, Albrecht and Rajagopalan (2022) question the economic rationale behind COVID-19 vaccine mandates. Like most vaccines, COVID-19 vaccines protect individuals against the most severe symptoms of the virus. However, unlike other vaccines, it does not prevent viral contagion. The benefits of vaccination, therefore, are uniquely concentrated on the individual, rendering most COVID-19 externalities inframarginal as opposed to marginal. Insofar as externalities exist, they are Pareto-irrelevant (Buchanan & Stubblebine, 1962). The authors further maintain that existing externalities are confined to one’s immediate network and community, requiring a local rather than a global public health response, despite the push for universal vaccine mandates among public health policymakers. Congleton’s (2022) contribution bolsters the economic justification for more local public health policy. Given the innumerable margins of variation among individuals (altogether distinct from the properties of the virus itself), it should come as no surprise that the “best” policies tend to emerge differently across locations, time, and among different groups. Still, policies commonly endorsed by health professionals during the COVID-19 pandemic took insufficient account of the variation. In addition to his positive economic analysis, Congleton argues that policy variation is normatively desirable because it allows for different groups to make particularistic tradeoffs according to their unique needs and encourages experimentation among different policy approaches. “Ideal” pandemic policies, he concludes, are more likely to emerge from a polycentric system of governance than from a more centralized system. Despite the merits of decentralization, public health policies often are executed through bureaucratic institutions in conjunction with various health experts. Koppl (2022) extends the theory of “expert failure” to public health, arguing that an effective check against such failure is provided by competition among policymakers. Koppl centers his argument, not around incentives per se, but around the diversity (or lack thereof) of scientific knowledge, advocacy, and opinion. The narrower the channel of information and feedback between policymakers and experts, the more likely one is to adopt ineffective or destructive policies. Such was the nature of the British Government and the Scientific Advisory Group for Emergencies (SAGE) during the COVID-19 pandemic. Pennington (2022) advances on similar ground, using Michel Foucault’s social constructionism to shine a light on the importance of narratives and discursive formations in the governance of public health. Discursive constructions emerge through prevailing narratives and social beliefs, which enable political authorities and scientific “experts” to mobilize interest groups in pursuit of political aims. The final two papers explore the role of the state in public health and the potential tradeoff between health and liberty. Koyama (2022) argues that such a tradeoff is present in the short run, whereby government intervention can prevent catastrophes of contagion and infectious disease. In the long run, however, the tradeoff disappears as freer societies become healthier through economic development and scientific advances in medical technology. Koyama rightly recognizes such progress as a genuine challenge to liberal institutions and their apologists—one that must be taken up by serious social scientists going forward. Furton (2022) takes up the challenge, contending that the alleged tradeoff between public health and individual liberty likely is overstated. He argues that political involvement in public health reform tends to occur during health crises, wherein policies are passed under conditions of extreme urgency and uncertainty. Reforms made during crisis—where institutions are more malleable—can become embedded, leading to long run, systemic government growth. Subsequent health crises therefore must be met with larger, more cumbersome public health institutions. Acknowledgements The papers collected in this special issue were first presented at a conference held online on February 11th and 12th, 2021. The conference was hosted and funded by the Foundations of the Market Economy Program at the Department of Economics, New York University. The conference was organized by Glenn Furton, David Harper, and Mario Rizzo. We would like to thank the authors and conference participants for their stimulating discussions and contributions as well as the Classical Liberal Institute for helping to coordinate the conference sessions. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Albrecht BC Rajagopalan S Inframarginal externalities: COVID-19, vaccines, and universal mandates Public Choice 2022 10.1007/s11127-022-01006-z Anomaly J What is public health? Public goods, publicized goods, and the conversion problem Public Choice 2022 10.1007/s11127-021-00908-8 Buchanan JM Stubblebine WC Externality Economica 1962 29 116 371 384 10.2307/2551386 Congleton RD Federalism and pandemic policies: Variety as the spice of life Public Choice 2022 10.1007/s11127-021-00915-9 Furton G The pox of politics: Troesken’s tradeoff reexamined Public Choice 2022 10.1007/s11127-022-01002-3 Koppl R Public health and expert failure Public Choice 2022 10.1007/s11127-021-00928-4 Koyama M Epidemic disease and the state: Is there a tradeoff between public health and liberty? Public Choice 2022 10.1007/s11127-021-00944-4 Leeson PT Thompson HA Public choice and public health Public Choice 2022 10.1007/s11127-021-00900-2 Pennington M Foucault and Hayek on public health and the road to serfdom Public Choice 2022 10.1007/s11127-021-00926-6 Perra N Non-pharmaceutical interventions during the COVID-19 pandemic: A review Physics Reports 2021 913 1 52 10.1016/j.physrep.2021.02.001 33612922	0	PMC9734511	NO-CC CODE	2022-12-14 23:28:29	no	Public Choice. 2022 Dec 9;:1-3	utf-8	Public Choice	2,022	10.1007/s11127-022-01031-y	oa_other
==== Front Appl Biochem Microbiol Appl Biochem Microbiol Applied Biochemistry and Microbiology 0003-6838 1608-3024 Pleiades Publishing Moscow 8498 10.1134/S0003683822070067 Metrology, Standardization, and Control Rapid Assessment of Neutralizing Antibodies Using Influenza Viruses with a Luciferase Reporter Sergeeva M. V. [email protected] [email protected] 12 Pulkina A. A. 12 Romanovskaya-Romanko E. A. 1 Mustafaeva A. S. 3 Egorov A. Yu. 1 Stukova M. A. 1 1 grid.452514.3 0000 0004 0494 5466 Smorodintsev Research Institute of Influenza, Russian Ministry of Health, 197376 St. Petersburg, Russia 2 grid.32495.39 0000 0000 9795 6893 Peter the Great St. Petersburg Polytechnic University, 195251 St. Petersburg, Russia 3 grid.437869.7 0000 0004 0497 4945 St. Petersburg State Institute of Technology, 190013 St. Petersburg, Russia 6 12 2022 2022 58 7 878886 12 1 2021 19 1 2021 21 2 2021 © Pleiades Publishing, Inc. 2022, ISSN 0003-6838, Applied Biochemistry and Microbiology, 2022, Vol. 58, No. 7, pp. 878–886. © Pleiades Publishing, Inc., 2022.Russian Text © The Author(s), 2021, published in Biotekhnologiya, 2021, Vol. 37, No. 2, pp. 81–91. 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. Influenza viruses cause acute respiratory infections, especially in the autumn–winter period. They are characterized by a high mutation frequency and cause annual seasonal epidemics. The detection of antibodies that neutralize the virus is an important criterion in the assessment of population immunity and the influenza vaccine effectiveness. In this study, a method for determining the titer of virus-neutralizing antibodies in blood serum has been developed. A new test called the luciferase neutralization assay uses a bioluminescent signal for detection. The assay is based on engineered influenza reporter viruses with various surface antigens and a nanoluciferase reporter protein in the NS1 reading frame. Using the developed method, we studied paired sera of volunteers obtained before and after vaccination. The proposed assay was compared with the conventional antibody assessment methods (microneutralization and hemagglutination inhibition assay); a high degree of correlation was observed. At the same time, the use of the luciferase neutralization assay made it possible to reduce the time required for the analysis and to simplify the detection procedure. Supplementary Information The online version contains supplementary material available at 10.1134/S0003683822070067. Keywords: microneutralization reporter virus influenza virus bioluminescence nanoluciferase issue-copyright-statement© Pleiades Publishing, Inc. 2022 ==== Body pmcINTRODUCTION According to the World Health Organization, influenza virus poses a serious threat to public health, causing yearly from 3 million to 5 million severe infections with 290 000–650 000 deaths worldwide.1 Serological methods play a critical role in various aspects of influenza epidemic surveillance and in the assessment of immunogenicity of influenza vaccines. The hemagglutination inhibition assay (HI) and microneutralization assay (MNA) are used for the assessment of the individual immune status, for instance, before and after vaccination, and for retrospective studies of paired sera at the beginning and end of the epidemic season. Hemagglutination inhibition assay developed in the 1940s remains the most widespread among serological methods. It is rather easy to perform and can use both live and inactivated viruses, which makes it safer and more accessible to a wide range of laboratories. However, HI only detects antibodies that prevent the influenza virus from binding to receptors on the surface of erythrocytes. The antibody titer in HI correlates with the degree of protection from influenza [2, 3], which underlies the requirements for registration of inactivated human influenza vaccines.2 MNA evaluates the ability of antibodies to block viral reproduction in a sensitive system and, unlike HI, detects not only antibodies to the receptor-binding site of the hemagglutinin (HA) molecule, but also to other neutralizing epitopes among the viral surface antigens [4]. MNA has become critical for the detection of the A(H3N2) viruses, since beginning from 2005, due to mutations in the receptor-binding site, these viruses gradually lost the ability to agglutinate chicken red blood cells and are characterized by differences in agglutination of turkey, guinea pig, and human erythrocytes, up to a complete lack of binding [5]. Historically, MNA appeared earlier than HI and at first test animals were used as a sensitive system [6]. The modern MNA version, which was first described in 1980, is based on the use of the Madin–Darby canine kidney (MDCK) cell culture and is carried out in 96-well plate format [7]. The classic MNA variant measures viral reproduction by the cytopathic effect and in the hemagglutination reaction 3–7 days after the infection of a sensitive culture.3 The test time can be reduced by using ELISA, which determines the viral reproduction as soon as 1–2 days after infection.4 MNA can be simplified and accelerated by using, for example, reporter strains, which allows virus detection by the direct measurement of a fluorescent or bioluminescent signal. By now, a great number of recombinant influenza viruses have been created that encode reporter fluorescent or luciferase proteins [8]. As an example, an influenza reporter virus encoding green fluorescent protein (GFP) in the neuraminidase (NA) reading frame was used to detect neutralizing antibodies to high-pathogenic influenza A(H5N1) viruses. However, the incubation time of infected cells before detection is 2–4 days due to the low fluorescence level of the GFP reporter and high background cell autofluorescence, which is practically not shorter than the analysis time in MNA [9]. In addition, the low genetic stability of these reporter constructs has been shown [8]. The goal of this work was to create recombinant influenza A viruses encoding a small reporter protein nanoluciferase (NanoLuc) in the NS1 protein open reading frame and to develop an express method for evaluation of virus-neutralizing antibodies on its basis (the luciferase neutralization assay, LNA). This method, as well as the classic HI and MNA, were used to examine the blood sera of volunteers before and after immunization with the seasonal inactivated influenza vaccine. MATERIALS AND METHODS Plasmids The nucleotide sequence of the chimeric NS gene segment encoding the NS1 protein truncated to 124 amino acids (NS124) of the A/PR/8/1934 (H1N1) influenza virus and the NanoLuc protein (Promega, United States) was synthesized de novo and cloned into the pHW2006 vector by Evrogen (Russia). The pHW-PR8-NS124-Luc plasmid was designed so that the nucleotide sequence for the NanoLuc protein was optimized to correspond to the genomic sequence of the A/PR/8/1934 influenza virus in the codon frequency and GC content. Plasmids encoding internal and surface proteins of the A/PR/8/1934 (H1N1) influenza virus, as well as the surface antigens of the epidemic influenza viruses A/Mississipi/10/2013 (H1N1)pdm09, A/St. Petersburg/01/2019 (H1N1)pdm09, A/Switzerland/9715293/2013 (H3N2) and A/Kansas/14/2017 (H3N2) were received from the Laboratory of Vector Vaccines of the Smorodintsev Research Institute of Influenza. Cell Cultures Vero cells (ATCC CCL81, United States) adapted for growing in a serum-free OptiPro medium (Gibco, United States) with the addition of 2% GlutaMax (Gibco) were used in this work. MDCK cell culture, (IRR FR-58, United States), was grown in AlphaMEM medium (BioloT, Russia) with 10% SC-biol fetal serum (BioloT). A primary chicken embryo kidney culture (CEK) was obtained by extraction of the kidney tissue from 19-day-old developing chicken embryos (СE), tissue decomposition in trypsin-EDTA solution (Sigma) for 30 min at 37°C in a water bath and filtration through the syringe nozzle with a 70 μm mesh diameter (pluriSelect Life Sciences UG&Co) followed by inoculation into DMEM/F12 medium (PanEko, Russia), containing 10% SC-biol fetal serum, 1% GlutaMax, 1% sodium pyruvate (Gibco) and 1% antibiotic-antimycotic (Gibco). Viruses Epidemic influenza viruses A/Mississipi/10/2013 (H1N1)pdm09 and А/Kansas/14/2017 (H3N2) were received from the Centers for Disease Control and Prevention (CDC, United States). The influenza virus A/Switzerland/9715293/2013 (H3N2) was received from the Worldwide Influenza Center (Great Britain). The A/St. Petersburg/01/2019 (H1N1)pdm09 influenza virus was isolated in the Laboratory of Evolutionary Variability of Influenza Viruses, at the Smorodintsev Research Institute of Influenza. The viruses were cultured in 10–12-day-old CE. Recombinant influenza viruses based on the A/PR/8/1934 strain with the chimeric NS124-Luc protein and various surface antigens were obtained by reverse genetics [10]. Vero cells were transfected with a set of bidirectional plasmids encoding eight gene segments of influenza virus using a Nucleofector II electroporator (Lonza, Switzerland). After the transfection and attachment, the cells were incubated in medium supplemented with TPCK trypsin (1 μg/mL, Sigma) until the development of a specific cytopathic effect. Viruses resulting from transfection were cultivated in CE. The reporter strains with surface antigens of epidemic viruses were deposited in the Collection of Influenza and ARVI Viruses (Smorodintsev Research Institute of Influenza),5 and the A/PR8/NS124-Luc strain was deposited in the State Collection of Viruses of II–IV pathogenicity groups (no. 2906).6 Virus Infectivity Analysis The virus infectious activity was determined by the limiting dilution method. For this purpose, a series of 10-fold dilutions of a virus-containing liquid was prepared. CE infection was carried out using dilutions in Dulbecco’s phosphate-buffered saline (DPBS, BioloT) with the addition of an antibiotic–antimycotic. Embryos were infected by injecting this solution (0.2 mL) into the allantoic cavity followed by incubation at 34°С for 48 h using 3 CE for each dilution. When Vero cell culture was used for the analysis of virus infectious titer, the dilutions were prepared using OptiPro medium containing an antibiotic-antimycotic and TPCK trypsin (1 μg/mL); for MDCK cell culture, AlphaMEM medium with antibiotic-antimycotic and TPCK trypsin (2.5 μg/mL) was used; CEK was infected in full DMEM/F12 medium without serum and trypsin. Ninety six-well plates (TPP, Switzerland) with a cell monolayer were infected with virus dilutions (0.1 mL per well) using four wells for each dilution. The plates were incubated at 37°С in 5% СО2 for 72 h. At the end of the incubation, the presence of the virus was determined by hemagglutination in allantoic or culture fluid. The infection activity was calculated by the Reed and Muench method [11] and was expressed in decimal logarithms of 50% embryo/tissue infectious dose (EID50/TID50). Analysis of Luciferase Activity The activity was measured in infected cells and in allantoic fluid of infected CE. To this end, the culture media was removed, and cells were washed twice with DPBS. The substrate and the buffer from the reagent kit of the Nano-Glo Luciferase Assay System (Promega) were then added to the washed cells; the mixture was incubated for 2 min and transferred to dark-walled plates. The luminescent signal (in relative luminescent units, RLU) was evaluated using a CLARIOstar multiphotometer (BMG LABTECH, ЕС). Clinical Materials Pre- and post-vaccination sera (day 0 and day 21) from 20 volunteers over 18 years old were used. The volunteers were vaccinated once before the epidemic season 2018/2019 with a trivalent inactivated influenza vaccine that contained A/Michigan/45/2015 (H1N1)pdm09, A/Singapore/INFIMH-16-0019/2016 (H3N2) and B/Colorado/06/2017-like strains.7 Blood serum samples were obtained within the framework of the State Task for Evaluation of the Epidemiological Effectiveness of Influenza Vaccines. The serum samples were stored at –20°C. Luciferase Neutralization Assay The sera were treated with the RDE enzyme (Denka Seiken Co, Ltd, Japan) in a ratio of 1 : 3 and incubated for 16–19 h at 37°C followed by 30 min at 56°C. Serial two-fold dilutions beginning from 1 : 10 (AlphaMEM medium + 1% antibiotic–antimycotic) were prepared in a 96-well U-bottom plate. An equal volume of pre-titrated reporter virus was then added to the wells, which provided a signal detection at the level of 200–300 RLU/50 μL. The mixture of serum and virus was incubated for 1 h at 37°C, after which it was transferred to a 96-well plate (100 μL per well) with a monolayer of MDCK cells. Luciferase activity in cells was measured after 6 h of incubation. The luminescence threshold value (RLUcut_off) was determined by the formula: RLUcut_off = RLUVC/2, where RLUVC is the luciferase activity in the well of the viral control (without serum). The reciprocal of the maximal serum dilution, at which the luciferase activity was less or equal to RLUcut_off, was considered the serum titer. If the luminescent signal of the 1 : 10 dilution exceeded the threshold value, the titer was taken equal to 5. Microneutralization Assay Sera treated with RDE were used in MNA. Two-fold dilutions (1 : 10—1 : 1280) in AlphaMEM medium with a 1% antibiotic were added to an equal volume of the pre-titrated virus antigen containing 100 TID50/50 μL. After a 1-h incubation at room temperature, an aliquot (100 μL) of each mixture was transferred to 96-well plates with a pre-washed monolayer of MDCK cells. The plates were incubated at 37°C in an atmosphere of 5% CO2 for 24 h. Then medium was removed, and the cells were washed and fixed with 80% acetone for 15 min at room temperature. The wells were blocked with a 5% suspension of skimmed milk powder (TF DITOL, Russia) in PBS supplemented with 0.1% Tween-20 (PBS-T, Serva, EU) at room temperature for 2 h. The virus reproduction was detected using the murine monoclonal antibodies 6D11 to influenza A virus nucleoprotein, labeled with horse radish peroxidase (courtesy of Dr. V. Z. Krivitskaya). The antibodies in a dilution of 1 : 2000 in blocking buffer were added into plate wells (100 μL per well) and incubated for 1 h at room temperature. After washing with PBS-T, a TMB solution (100 μL, Dako, EU), which is the peroxidase substrate, was added, the plate was incubated for 15 min, and the reaction was stopped by the addition of 2 N H2SO4 (100 μL per well, VECTON, Russia). Optical density was measured at a wavelength of 450 nm (OD450) on a CLARIOstar multiphotometer. The optical density threshold (ODcut_off) was calculated by the formula: ODcut_off = (ODVC + ODCC)/2, where ODVC is the optical density of infected cells without serum, and ODСC is the optical density of uninfected cells. The final serum titer was expressed as the reciprocal of the maximal dilution, at which OD was less than or equal to ODcut_off. If OD of the 1 : 10 dilution exceeded the threshold value, the serum titer was taken equal to 5. Hemagglutination Inhibition Assay HI was carried out as described in Methodical Guidelines MU 3.3.2.1758-03.8 Epidemic influenza viruses, sera treated with RDE and 0.5% suspension of human erythrocytes of blood group I(0) were used for the analysis. Sera were additionally treated with concentrated erythrocytes (incubation for 1 h at 4°C) to prevent nonspecific binding. The HI titer of antibodies was expressed as the reciprocal of the highest serum dilution, at which agglutination inhibition was observed. If hemagglutination inhibition was not detected at the 1 : 10 dilution, the titer was taken equal to 5. Statistical Data Analysis The MS Office Excel 2016, GraphPad Prizm 6.07 and RStudio Desktop 1.0.153 software packages were used for visualization and statistical processing of the data. The results were described using such indicators as arithmetic mean, standard deviation (SD), and geometric mean. The antibody titer increase was calculated as the ratio of the antibody titer on day 21 (after vaccination) to the titer on day 0 (before vaccination). Correlation of the antibody titers and the titer increase values obtained using different methods was assessed using the Pearson test applied to the logarithms. RESULTS AND DISCUSSION Obtaining Recombinant Nanoluciferase-Encoding Influenza Viruses Recombinant viruses were obtained using the NanoLuc reporter gene. This enzyme is a genetically engineered protein obtained by directed evolution from the small subunit of luciferase from deep sea shrimp Oplophorus gracilirostris [12]. The nucleotide sequence encoding Nanoluc was optimized and inserted into the open reading frame of the influenza virus NS1 protein, which was truncated to 124 amino acids (Fig. 1). Fig. 1. A map of the chimeric A/PR/8/1934 (H1N1) NS genomic segment coding for the truncated NS1 protein, fused with the NanoLuc insert, and a nuclear export protein (NEP). The A/PR/8/1934 (H1N1) strain characterized by high reproduction in various biological systems was used as a source of genes for internal and nonstructural proteins of reporter viruses. A/PR/8/1934 (H1N1) virus and reference epidemic strains of the A(H1N1)pdm09 and A(H3N2) subtypes served as a source of surface antigens. Reverse genetics was used to obtain a set of recombinant viruses encoding the NanoLuc reporter protein (Table 1). The resulting reporter viruses were characterized by high genetic stability of the NS chimeric gene segment with the heterologous insert throughout 4 consecutive passages in CE. High luciferase activity (about 104–106 RLU) was observed in various cell cultures infected with the reporter viruses. In their level of luciferase activity the resulting reporter viruses were similar to the viruses encoding NanoLuc in the PA gene segment [13] or other small luciferase (Gaussia luciferase) in the full-length NS gene segment [14]. Table 1. Recombinant influenza viruses with a luciferase reporter and different surface antigens Reporter influenza virus* Wild type virus, a source of surface antigens Titer, logEID50/mL/logTID50/mL* CE MDCK Vero CEK A/PR8‑NS124‑Luc A/PR/8/1934 (H1N1) 8.28 7.75 ± 0.43 7.59 ± 0.63 8.24 ± 0.25 A/Miss/PR8‑NS124‑Luc A/Mississippi/10/2013 (H1N1)pdm09 6.2 <1.5 2.5 2.5 A/SPb/PR8‑NS124‑Luc A/St.‑Petersburg/01/2019 (H1N1)pdm09 6.7 6.44 ± 0.10 2.5 ± 0.00 4.4 ± 0.09 A/Switz/PR8‑NS124‑Luc A/Switzerland/9715293/2013 (H3N2) 7.2 6.53 ± 0.22 5.80 ± 0.38 8.17 ± 0.44 A/Kans/PR8‑NS124‑Luc А/Kansas/14/2017 (H3N2) 7.45 6.5 ± 0.00 3.11 ± 0.54 <1.5 * Virus titer is expressed in logEID50/mL for CE and in logTID50/mL for cell cultures. The obtained reporter viruses were characterized by high infectious activity in the CE system, regardless of the subtype of surface antigens; however, they differed in replication in different cell cultures (Table 1). The A/PR8-NS124-Luc (H1N1) strain, as was expected, replicated actively in all tested systems. The A/SPb/PR8-NS124-Luc (H1N1)pdm09 and A/Kans/PR8-NS124-Luc (H3N2) strains had lower reproduction activity in Vero and CEK cells. The A/Miss/PR8-NS124-Luc (H1N1)pdm09 strain showed a low replication level in all used cell cultures. The Kinetics of Luciferase Accumulation in Cells Infected with Reporter Viruses The luciferase activity kinetics was studied using the A/PR8-NS124-Luc influenza strain. Vero cells were infected with the virus at various doses (0.0001–0.1 TID50/cell) and luciferase activity of the expressed NanoLuc protein was evaluated after 3–24 h. The luminescent signal in the infected cells increased gradually over 24 h (Fig. 2). Using an infectious dose of 0.001 TID50/cell (about 40 TID50 per well), the luminescence was observed at the level of 102 RLU (in the absence of a background signal) as early as 6 h after infection. Thus, the bioluminescent reporter significantly reduced the time from cell infection to detection. Fig. 2. The increase in the luciferase activity over time in Vero cells infected with the recombinant influenza A/PR8-NS124-Luc virus at various infectious doses, from 0.1 TID50/cell to 0.0001 TID50/cell. In addition, there is a direct correlation between the luminescent signal in the cells and the infectious dose of the virus, which makes the parameter of luciferase activity applicable not only for virus detection, but also for assessment of its quantity. In order to clarify the nature of this correlation, the curves of the dependence of the luminescent signal in MDCK cells on the virus infectious dose were plotted for each engineered recombinant virus (Fig. 3). The graphs were linear in the RLU range from 101 to 105, where the approximation accuracy (R2) was 0.98–0.99. Fig. 3. The dependence of luciferase activity in MDCK cells 6 h after infection on the infectious dose of recombinant reporter viruses. The linear fit results in MSOffice16 Excel (R2) are shown on the diagram. The Titer of Virus-Neutralizing Antibodies in the Sera of Volunteers The engineered recombinant viruses were used for the assessment of the titer of virus-neutralizing antibodies in blood serum samples of volunteers taken before and after immunization with an inactivated vaccine against seasonal influenza virus. The neutralizing antibody assessment method was developed based on the classical MNA; however, for detection, a direct measurement of the luciferase activity of the reporter viruses in cells was used. Based on the results of the luciferase activity kinetics in infected cells, the following parameters were selected for LNA performance: incubation time, 6 h, and control signal level, 200–300 RLU. The selected parameters were validated by testing the sera of laboratory animals (rats) immunized with the reporter viruses or their combinations (the results are presented in Supplementary Materials, Table S1). The LNA antibody titers of the volunteer sera were compared with the levels of neutralizing antibodies obtained in classical MNA, as well as with the amounts of anti-hemagglutinating antibodies detected in HI. In the comparative assay, the following reporter strains were used: A/PR8-NS124-Luc (H1N1), A/SPb/PR8-NS124-Luc (H1N1)pdm09 and A/Kans/PR8-NS124-Luc (H3N2), as well as the related epidemic viruses. The results of the correlation analysis of neutralizing antibody titers are shown in Fig. 4, while the data on the comparison of LNA and HI are presented in Supplementary Materials, Fig. S1. The results demonstrate a high correlation degree between the LNA and MNA/HI data for the influenza virus of the A(H1N1)pdm09 subtype (r > 0.7) and a moderate correlation level for the A(H1N1) and A(H3N2) viruses (r > 0.5). Fig. 4. Comparative analysis of neutralizing antibody titers measured in volunteer sera by LNA (X axis) and MNA (Y axis) using (a) wild-type virus (wt) and (b) reporter virus (Luc). The graphs show the individual values (log2) of antibody titers; a darker color means the coincidence of several values at one point. The blue line represents the result of the correlation analysis; Pearson’s correlation coefficients and p values are shown on graphs: (*), p < 0.001 or (), p < 0.0001. A four-fold or more increase in the antibody titer on day 21 is an important characteristic of the immune response to vaccination (included in the three European criteria for influenza vaccines).9 In this regard, we also compared the antibody titer increase measured using LNA and MNA/HI. The results of these two methods correlated well for the A(H1N1)pdm09 and A(H1N1) influenza viruses (r > 0.7) and moderately for the A(H3N2) virus (r > 0.5) (Fig. 5). Fig. 5. Comparative analysis of the antibody titer increase in sera of vaccinated volunteers measured by LNA (X axis) and MNA (Y axis). Viruses used: (a), wild-type (wt) and (b), reporter (Luc). A darker gray color means the coincidence of several values at one point. Red lines mark the threshold of a significant increase in antibody titer (4 times or more). Blue line represents the results of the correlation analysis; Pearson’s correlation coefficients and p values are shown on the graphs: (), p < 0.001, and (***), p < 0.0001. In general, the antibody titer increase to the A(H1N1)pdm09 influenza virus after vaccination measured by both LNA and MNA was higher than that to the A(H1N1) and A(H3N2) viruses (Fig. 5). A similar dependence was observed when comparing the results of LNA and HI (Figure S2). This may be due to the fact that the A(H1N1)pdm09 influenza viruses are subject to very slow antigenic drift. According to HI with post-infectious ferret sera, the A/St.-Petersburg/01/2019 (H1N1)pdm09 virus used in the experiment does not differ antigenically from the A/Michigan/45/2015 (H1N1)pdm09 virus, which is part of the vaccine composition of the season 2018/2019.10 A(H3N2)viruses are more variable and the antigenic activity of the А/Kansas/14/2017 (H3N2) strain used in this experiment differs significantly from that of the A/Singapore/INFIMH-16-0019/2016 vaccine virus: the difference in the titers of post-infectious ferret sera in HI was 64 or more times11. It should be noted that the A/PR/8/1934 (H1N1) strain was not a component of the vaccine composition at all. Thus, we can conclude that LNA is a sensitive test for identification of antibodies to antigenically close and distant influenza viruses. In this study, recombinant influenza strains that stably expressed the luciferase reporter throughout a series of passages on CE were obtained. NanoLuc has a molecular weight of 19 kDa and is much smaller in size than fluorescent proteins, the most widespread firefly luciferase (61 kDa) and coral (Renilla) enzyme (36 kDa). In the case of the influenza vector, the smaller size provides a higher genetic stability for the reporter gene. In addition, NanoLuc differs from other luciferases in its higher activity and independence from ATP, which increases its sensitivity and facilitates the detection of the bioluminescent signal. We created reporter influenza A viruses based on the high-yield A/PR/8/1934 strain and surface antigens (HA and NA) of the epidemic A(H1N1)pdm09 and A(H3N2) influenza viruses. NanoLuc in the reporter viruses was linked to the truncated C terminus of the NS1 protein, which is actively expressed during the first hours after infection [15]; this provided the registration of the bioluminescent signal as rapidly as 6 h after the cell infection. This advantage of the created reporter viruses was used for the development of an express method for evaluating virus-neutralizing antibodies. The developed LNA method was used to assess antibody titers in blood serum samples obtained from volunteers before and after the immunization with inactivated vaccine against seasonal influenza. The LNA results correlated with the MNA results, as well as with the HI results. At the same time, the use of luciferase vectors significantly reduced the time of the neutralization test due to the higher sensitivity of bioluminescent detection and skipping the additional stages of immunoenzyme staining carried out in the classical MNA. A rapid neutralization test based on the concept of fluorescence and luminescence detection has been developed to identify antibodies to metapneumovirus, rabies virus, and several animal and avian viruses [16]. The approach based on the use of pseudoviruses with a luminescent reporter is widely used, for instance, for the assessment of neutralizing antibodies to coronavirus SARS-CoV-2 [17] and papillomovirus [18]. The method for identification of neutralizing antibodies with influenza pseudoviruses [19], in particular, its version with two reporters [20], was also described. The latter is characterized by the disadvantage of complicated process of obtaining pseudoviruses, which requires a specialized MDCK-HA cell culture expressing a specific hemagglutinin subtype. In addition, pseudoviruses only partially mimic the infection process, reducing it to the stage of virus penetration into the cell. The use of recombinant viruses with an inserted reporter gene is free from these disadvantages. CONCLUSIONS Overall within the framework of this study, we have developed a neutralization test based on influenza A reporter viruses with bioluminescent activity, and have shown that luciferase neutralization assay can be successfully used for the analysis of post-vaccination immunity to influenza A viruses. Supplementary Information 10438_2022_8498_MOESM1_ESM.pdf FUNDING The study was supported by the grant of the President of the Russian Federation for young PhDs (no. 075-15-2019-226) and the grant of the Government of Saint-Petersburg for PhD students (25.09.2018, No. 124). COMPLIANCE WITH ETHICAL STANDARDS The authors declare that they have no conflicts of interest. This article does not contain any studies involving animals performed by any of the authors. The blood samples from volunteers were taken according to the research protocol PEV-2018/2019, version 01 of September 14, 2018 approved by the Local Ethic Committee of the Smorodintsev Research Institute of Influenza (LEC meeting no. 131 dated October 10, 2018). 1 WHO Influenza FactSheet, https://www.who.int/news-room/fact-sheets/detail/influenza-(seasonal). 2 European Medicines Agency. Guideline on Influenza Vaccines, https://www.ema.europa.eu/en/documents/scientific-guideline/influenza-vaccines-non-clinical-clinical-module_en.pdf. 3 WHO Manual on Animal Influenza Diagnosis and Surveillance, https://www.who.int/csr/resources/publications/influenza/whocdscsrncs20025rev.pdf. 4 WHO Manual for the laboratory diagnosis and virological surveillance of influenza, https://www.who.int/influenza/gisrs_ laboratory/manual_diagnosis_surveillance_influenza/en/. 5 Russian Biobank of Influenza at the Smorodintsev Research Institute of Influenza, https://rubin.influenza.spb.ru/ 6 State Collection of Viruses of II–IV Pathogenicity Groups on the basis of the D.I. Ivanovsky Institute of Virology of the FSBI N.F. Gamaleya National Research Center of the Ministry of Health of Russia, https://virology.gamaleya.org/structure/otdel-gosudarstvennoy-kollektsii-virusov/. 7 WHO Recommended composition of influenza virus vaccines for use in the 2018–2019 northern hemisphere influenza season, https://www.who.int/influenza/vaccines/virus/recommendations/2018_19_north/en/. 8 Methods for determining the quality indicators of immunobiological preparations for the prevention and diagnosis of influenza: Methodical instructions MU 3.3.2.1758-03 (approved by the Chief State Sanitary Doctor of the Russian Federation on September 28, 2003). 9 European Medicines Agency. Guideline on Influenza Vaccines. https://www.ema.europa.eu/en/documents/scientific-guideline/influenza-vaccines-non-clinical-clinical-module_en.pdf. 10 Worldwide Influenza Center. Report prepared for the WHO annual consultation on the composition of influenza vaccine for the Northern Hemisphere 2019–2020. https://www.crick.ac.uk/sites/default/files/2019-04/Crick%20-VCMFeb2019%20report_toPost.pdf (Table 5–11). 11 WHO. Addendum to the recommended composition of influenza virus vaccines for use in the 2019–2020 northern hemisphere influenza season. (Table 1). https://www.who.int/influenza/vaccines/virus/recommendations/201902_recommendation_addendum.pdf?ua=1. Abbreviations: CEK, primary culture of chicken embryo kidney; CE, developing chicken embryos; DPBS, Dulbecco’s phosphate-buffered saline; EID50, 50% embryo infectious dose; ELISA, enzyme-linked immonosorbent assay; HA, hemagglutinin; HI, hemagglutination inhibition assay; LNA, luciferase neutralization assay; MNA, microneutralization assay; NA, neuraminidase; NanoLuc, nanoluciferase; NS1, influenza virus non-structural protein 1; RLU, relative luminescence units; TID50, 50% tissue culture infectious dose. Translated by I. Gordon ==== Refs REFERENCES 1 Hirst G.K. The quantitative determination of influenza virus and antibodies by means of red cell agglutination J. Exp. Med. 1942 75 49 64 10.1084/jem.75.1.49 19871167 2 Potter C.W. Oxford J.S. Determinants of immunity to influenza infection in man Br. Med. Bull. 1979 35 69 75 10.1093/oxfordjournals.bmb.a071545 367490 3 Black S. Nicolay U. Vesikari T. Hemagglutination inhibition antibody titers as a correlate of protection for inactivated influenza vaccines in children Pediatr. Infect. Dis. J. 2011 30 1081 1085 10.1097/INF.0b013e3182367662 21983214 4 Krivitskaya V.Z. Kuznetsova E.V. Mayorova V.G. Microneutralization reaction versus hemagglutination inhibition reaction in assessing the immunogenicity of influenza vaccines and diagnosing influenza Infekts. Immun. 2019 9 763 772 10.15789/2220-7619-2019-5-6-763-772 5 Lin Y.P. Xiong X. Wharton S.A. Evolution of the receptor binding properties of the influenza A(H3N2) hemagglutinin Proc. Natl. Acad. Sci. U. S. A. 2012 109 21474 21479 10.1073/pnas.1218841110 23236176 6 Francis T. Shope R.E. Neutralization tests with sera of convalescent or immunized animals and the viruses of swine and human influenza J. Exp. Med. 1936 63 645 653 10.1084/jem.63.5.645 19870494 7 Frank A.L. Puck J. Hughes B.J. Microneutralization test for influenza A and B and parainfluenza 1 and 2 viruses that uses continuous cell lines and fresh serum enhancement J. Clin. Microbiol 1980 12 426 432 10.1128/JCM.12.3.426-432.1980 6260835 8 Breen M. Nogales A. Baker S.F. Replication-competent influenza A viruses expressing reporter genes Viruses 2016 8 179 10.3390/v8070179 27347991 9 Rimmelzwaan G.F. Verburgh R.J. Nieuwkoop N.J. Use of GFP-expressing influenza viruses for the detection of influenza virus A/H5N1 neutralizing antibodies Vaccine 2011 29 3424 3430 10.1016/j.vaccine.2011.02.082 21396410 10 Hoffmann E. Neumann G. Hobom G. “Ambisense” approach for the generation of influenza A virus: vRNA and mRNA synthesis from one template Virology 2000 267 310 307 10.1006/viro.1999.0140 10662626 11 Reed L.J. Muench H. A simple method of estimating fifty per cent endpoints Am. J. Epidemiol. 1938 27 493 497 10.1093/oxfordjournals.aje.a118408 12 Hall M.P. Unch J. Binkowski B.F. Engineered luciferase reporter from a deep sea shrimp utilizing a novel imidazopyrazinone substrate ACS Chem. Biol. 2012 7 1848 1857 10.1021/cb3002478 22894855 13 Tran V. Poole D.S. Jeffery J.J. Multi-modal imaging with a toolbox of influenza A reporter viruses Viruses 2015 7 5319 5327 10.3390/v7102873 26473913 14 Eckert N. Wrensch F. Gärtner S. Influenza A virus encoding secreted Gaussia luciferase as useful tool to analyze viral replication and its inhibition by antiviral compounds and cellular proteins PLoS One 2014 9 e97695 10.1371/journal.pone.0097695 24842154 15 Young J.F. Desselberger U. Palese P. Efficient expression of influenza virus NS1 nonstructural proteins in Escherichia coli Proc. Natl. Acad. Sci. U. S. A. 1983 80 6105 6109 10.1073/pnas.80.19.6105 6310615 16 Li Y. Li L.F. Yu S. Applications of replicating-competent reporter-expressing viruses in diagnostic and molecular virology Viruses 2016 8 127 10.3390/v8050127 27164126 17 Nie J. Li Q. Wu J. Establishment and validation of a pseudovirus neutralization assay for SARS-CoV-2 Emerg. Microbes Infect. 2020 9 680 686 10.1080/22221751.2020.1743767 32207377 18 Nie J. Huang W. Wu X. Optimization and validation of a high throughput method for detecting neutralizing antibodies against human papillomavirus (HPV) based on pseudovirons J. Med. Virol. 2014 86 1542 1555 10.1002/jmv.23995 24895216 19 Martínez-Sobrido L. Cadagan R. Steel J. Hemagglutinin-pseudotyped green fluorescent proteinexpressing influenza viruses for the detection of influenza virus neutralizing antibodies J. Virol. 2010 84 2157 2163 10.1128/JVI.01433-09 19939917 20 Baker S.F. Nogales A. Santiago F.W. Competitive detection of influenza neutralizing antibodies using a novel bivalent fluorescence-based microneutralization assay (BiFMA) Vaccine 2015 33 3562 3570 10.1016/j.vaccine.2015.05.049 26044496	0	PMC9734512	NO-CC CODE	2022-12-14 23:28:29	no	Appl Biochem Microbiol. 2022 Dec 6; 58(7):878-886	utf-8	Appl Biochem Microbiol	2,022	10.1134/S0003683822070067	oa_other
==== Front Aging Clin Exp Res Aging Clin Exp Res Aging Clinical and Experimental Research 1594-0667 1720-8319 Springer International Publishing Cham 36469252 2303 10.1007/s40520-022-02303-9 Original Article Impact of COVID-19 pandemic on medication use in the older Italian population http://orcid.org/0000-0002-4851-1825 Marengoni Alessandra [email protected] 1 Cangini Agnese 2 Pierantozzi Andrea 2 Onder Graziano 3 Da Cas Roberto 4 Ippoliti Ilaria 4 Zito Simona 2 Trotta Francesco 2 The Italian Working Group on Medication Use in the ElderlyMagrini Nicola Comessatti Ivano Di Filippo Aurora Fabrizi Andrea Fontanella Marco Fortinguerra Filomena Frulio Ramon Gallinella Francesca Guerrizio Maria Alessandra Italiano Mariarosaria Marinelli Marco Marini Roberto Milozzi Federica Perna Serena Pierattini Linda Pieroni Emanuela Pistolesi Giuliano Pomponi Filippo Sacconi Matteo Settesoldi Daniela Trapanese Maurizio Traversa Giuseppe Vasta Saverio Antonio Brusaferro Silvio Ruggeri Paola Biffoli Claudia Boldrini Rosaria Brutti Chiara Patrizi Luciana Cavallo Antonietta Guerrini Sara Martino Marco Nobili Alessandro Pasina Luca Franchi Carlotta Tettamanti Mauro Novella Alessio Pietrangelo Antonello Muiesan Maria Lorenza Cricelli Claudio Aprile Pierangelo Lora Medea Gerardo Grattagliano Ignazio Michieli Raffaella Parretti Damiano Lapi Francesco Marconi Ettore 1 grid.7637.5 0000000417571846 Department of Clinical and Experimental Sciences, University of Brescia, Viale Europa 11, 25123 Brescia, Italy 2 grid.487250.c 0000 0001 0686 9987 Agenzia Italiana del Farmaco, Rome, Italy 3 grid.416651.1 0000 0000 9120 6856 Department of Cardiovascular, Endocrine-Metabolic Diseases and Aging, Istituto Superiore di Sanità, Rome, Italy 4 grid.416651.1 0000 0000 9120 6856 Pharmacoepidemiology Unit, National Centre for Drug Research and Evaluation, Italian National Institute of Health, Rome, Italy 5 12 2022 111 9 8 2022 2 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 analyse the impact of the pandemic on the amount of use and new medication dispensation for chronic diseases in the Italian population aged 65 years and older (almost 14 million inhabitants). Methods The “Pharmaceutical Prescriptions database”, which gathers data on medications, reimbursed by the National Health Service and dispensed by community pharmacies, was employed. Data were analysed as amount of use (defined daily dose—DDD per 1000 inhabitants); variation in DDD between 2020 and 2019 was calculated for the 30 categories with major consumption in 2020. Trends in prevalence and incidence of dispensations between 2020 and 2019 were calculated for four categories: antidiabetics, antihypertensives, antidepressants and drugs for respiratory diseases. Results All medications showed a negative variation in DDD/1000 inhabitants between 2020 and 2019 except for anticoagulants (+ 5%). The percentage variation ranged from − 27.7% for antibiotics to − 6.4% for antipsychotics in 85 + year-old persons, but increased for most classes in the youngest (65–69 years). On the other hand, a decrease of the dispensation incidence of antidiabetics, antihypertensives, antidepressants and drugs for pulmonary disease was high, especially in the two extreme age groups, the youngest and the oldest one. Conclusions and relevance Great variation in medication use between 2020 and 2019 was observed probably reflecting the low rate of infectious diseases due to the widespread use of protective devices and self-isolation, reduced healthcare access because of the lockdowns and the fear of going to hospital, and the reduction of screening and diagnostics due to health-care system overload. Keywords COVID-19 Older population Medications ==== Body pmcIntroduction In March 2020, the World Health Organization declared a pandemic after the coronavirus disease 2019 (COVID-19), a novel infection caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), spread worldwide in a few months [1]. Due to its high direct mortality and impact on hospitalization, COVID-19 has put pressure on the health-care systems. Primary care was drastically impacted by the pandemic, resulting in challenges in the treatment of chronic diseases which require routine care [2]. First, national lockdowns, social distancing restrictions and different organizational rules for the admission to medical ambulatories were likely to affect health-care access [3]. Second, to prevent infection, people may have decided to postpone visits with their general practitioner, both for monitoring existing health conditions and for the onset of new signs and symptoms [4]. This resulted in an increase of missed and untimely diagnoses as shown in previous studies. The incidence of hospital admissions for acute coronary syndrome decreased as well as the provision of mammograms [5–8]. Italy was one of the first Western countries greatly affected by the pandemic, particularly the Northern regions. The older population was affected in terms of hospitalizations and mortality, accounting for 20% of the total number of subjects infected with SARS-CoV-2, but representing more than 90% of deaths [9]. Further, older persons are the part of the population mostly affected by multiple non-communicable diseases [10] which require chronic pharmacological prescriptions. Different policies have been adopted in Italy to balance the need of reducing the infections and guaranteeing the therapeutic continuity for other diseases, such as the introduction of the electronic prescription, but there was a gap in knowledge regarding medications’ accessibility by patients, especially during the first year of the pandemic. From a National Health Service (NHS) perspective, it is of utmost importance to measure the effects of the pandemic on accessibility to medications to estimate the future consequences in terms of mortality and morbidity and to improve health-care accessibility during this and future pandemics. This study aims at analysing the impact of the pandemic on the use of medicines in Italy in older persons comparing medication use in the outpatient setting in 2020 to that registered in 2019. Methods Data analysis This is a quasi-experimental study based on aggregated administrative data. For the present analysis, we extracted records from the Pharmaceutical Prescriptions database (also known as the Italian Health Insurance Card database) that includes patient-level data on medicines prescribed and dispensed by community pharmacies reimbursed by Italian NHS in the Italian population aged 65 and older (almost 14 million inhabitants) [11]. Information on each drug package, identified via package unique identifier codes and the 5th level Anatomical Therapeutic Chemical (ATC) classification, was tracked at the individual level. The medicines were grouped into therapeutic classes according to the main indication, and the first 30 categories which had shown major consumption in 2020 were analysed [12]. Drug consumption was measured as the number of defined daily dose (DDD), which is the assumed average maintenance dose per day for a drug used for its main indication in adults [13]. It represents a standard in performing valid and reliable cross-national or longitudinal studies on drug consumption. The indicator calculated as “number of DDD per 1,000 inhabitants” was used and it was calculated by dividing the total number of DDDs prescribed and dispensed during 2019 and 2020 by the total number of inhabitants in the Italian population in the specific age group and sex. The result was then divided by 365 and reported per 1000 inhabitants. Prevalence and incidence of use were assessed for four commonly prescribed medications classes, i.e. antidiabetics, antihypertensives, antidepressants and drugs for chronic pulmonary disease. The prevalence of use was calculated by dividing the number of individuals receiving at least one medication in 2019 and 2020 by the total number of Italian individuals in that same age group as reported by the Italian National Institute of Statistics in January 2019 and 2020. The incidence of use was estimated dividing the number of subjects that received the first dispensation (new users) of a medication belonging to a specific therapeutic category by the subjects at risk. The new users were defined as patients who did not receive a medical dispensation in the previous 6 months. The results of prevalence and incidence were reported per 1000 inhabitants. To estimate the impact of COVID-19 on medication use, the variation between 2020 and 2019 was calculated. Ethical considerations Ethical review and approval were not required for this study in accordance with the local legislation and institutional requirements. Results In the whole population analysed, almost all medication categories showed a negative variation in DDD/1000 inhabitants between 2020 and 2019; the biggest differences were reported for the following: antibiotics (− 22.9%) without differences by sex; drugs for the genital system/sexual hormones (− 17.6%) with a greater reduction in women than in men; dermatological and non-steroidal anti-inflammatory drugs (− 13%) without differences by sex. Anticoagulants were the only class showing an increase in dispensation between 2020 and 2019 (+ 5.0%), similarly in men and women (Table 1).Table 1 Use (DDD/1000 inhabitants) in the whole population and by gender of the first 30 categories with major consumption in 2020 and comparison 2020–2019 DDD/1000 inhabitants Male Female Total 2020 Δ % 20–19 2020 Δ % 20–19 2020 Δ % 20–19 Antihypertensive 1120.0 − 5.4 1039.9 − 4.1 1.074.9 − 4.7 Lipid-lowering agents 348.0 − 0.7 249.5 0.2 292.5 − 0.2 Drugs for peptic ulcer and gastrooesophageal reflux disease 225.1 − 2.7 233.8 − 2.2 230.0 − 2.5 Antiplatelets 267.1 − 4.3 200.7 − 3.6 229.6 − 3.9 Antidiabetics 203.6 − 4.6 143.8 − 4.8 169.9 − 4.6 Drugs for genito-urinary disorders 307.2 − 2.5 0.9 7.5 134.5 − 2.2 Antidepressants 57.4 − 3.4 117.3 − 2.6 91.1 − 2.9 Drugs for osteoporosis 22.1 − 19.7 121.7 − 16.1 78.3 − 16.7 Drugs for asthma and COPD 90.2 − 6.2 63.3 − 5.0 75.1 − 5.6 Anticoagulants 82.6 4.7 66.7 5.2 73.6 5.0 Drugs for eye disorders 68.6 − 5.1 65.4 − 4.8 66.8 − 4.9 Drugs active on the cardiovascular system 73.7 − 9.4 57.3 − 8.8 64.4 − 9.1 Antianaemia preparations 59.1 − 3.2 61.1 − 3.6 60.2 − 3.4 Thyroid drugs 21.0 − 2.8 63.2 − 3.0 44.8 − 3.0 NSAIDs 29.9 − 13.0 42.3 − 13.2 36.9 − 13.2 Drugs for gout 48.6 − 3.9 26.5 − 3.0 36.1 − 3.5 Corticosteroids for systemic use 26.8 − 4.8 27.0 − 5.3 26.9 − 5.1 Anti-parkinsonian drugs 23.0 − 4.6 15.3 − 4.4 18.6 − 4.5 Pain therapy 14.2 − 6.2 21.8 − 5.6 18.5 − 5.8 Antibiotics 19.3 − 22.4 17.8 − 23.4 18.4 − 22.9 Oncological drugs 14.1 − 6.7 16.5 − 2.0 15.5 − 3.9 Antiepileptics 13.9 − 5.2 13.1 − 4.4 13.5 − 4.7 Antihistamines 10.3 − 0.6 13.0 − 1.3 11.8 − 1.0 Intestinal anti-inflammatory 10.3 − 5.0 8.1 − 3.8 9.1 − 4.4 Drugs for dementia 7.2 − 2.1 10.3 − 1.9 9,0 − 2,0 Antipsychotics 6.2 − 2.3 8.3 − 0.2 7,4 − 1,0 Hepatic and biliary therapy 6.1 − 1.7 6.8 − 1.1 6,5 − 1,3 Non-systemic antibiotics 5.0 − 8.6 6.6 − 6.7 5,9 − 7,4 Dermatologic drugs 6.6 − 12.8 3.0 − 13.6 4,5 − 13,0 Sexual hormones and modulators of the genital system 1.2 − 6.3 4.2 − 19.7 2,9 − 17,6 Table 2 describes the DDD/1000 inhabitants and the percentage variation between 2020 and 2019 for the 30 most used categories according to age groups. The age group with the widest negative difference between 2020 and 2019 was the one including persons 85-year-old and older ranging from − 27.7% for antibiotics to − 6.4% for antipsychotics; notably, this was the only age group with a negative variation of anticoagulants too (− 7.4%). The group of people aged 80–84 years was also characterized by a negative variation in dispensation of almost all drug categories, despite to a lesser extent than the oldest ones, except for anticoagulants (+ 2.3%) and antipsychotics (+ 0.7%). On the contrary, in the youngest age group, i.e. 65–69 years, the percentage variation in medication use was positive for most classes, ranging from + 0.1% for antipsychotics to + 15.8% for anticoagulants.Table 2 Use (DDD/1000 inhabitants) by age of the first 30 categories with major consumption in 2020 and comparison 2020–2019 DDD/1000 inhabitants Age, years 65–69 70–74 75–79 80–84 ≥ 85 2020 Δ % 20–19 2020 Δ % 20–19 2020 Δ % 20–19 2020 Δ% 20–19 2020 Δ% 20–19 Antihypertensives 824.7 2.4 1.039.6 − 2.9 1.198.9 0.7 1.279.1 − 7.9 1.167.2 − 16.2 Lipid-lowering agents 266.2 7.6 322.8 1.0 342.2 3.9 317.5 − 5.8 202.4 − 14.7 Drugs for peptic ulcer and gastrooesophageal reflux disease 159.5 6.9 213.3 − 0.1 256.7 3.4 284.0 − 5.6 279.7 − 14.7 Antiplatelets 152.3 6.7 213.9 − 0.6 262.1 2.0 288.7 − 7.6 276.5 − 17.2 Antidiabetics 156.1 3.1 189.3 − 3.5 196.7 − 0.7 177.6 − 10.1 122.0 − 18.4 Drugs for genito-urinary disorders 89.8 7.9 132.9 0.3 163.6 3.1 172.1 − 6.6 134.0 − 15.8 Antidepressants 69.0 3.3 81.8 − 0.5 98.8 3.9 113.2 − 5.1 108.7 − 15.0 Drugs for osteoporosis 60.8 − 12.4 76.5 − 16.0 90.3 − 11.9 94.5 − 18.8 77.6 − 25.6 Drugs for asthma and COPD 54.1 3.0 71.0 − 2.6 84.8 0.4 91.3 − 9.0 86.1 − 18.7 Anticoagulants 35.6 15.8 57.3 9.3 86.7 13.0 106.5 2.3 109.3 − 7.4 Drugs for eye disorders 42.2 4.1 60.3 − 1.7 79.2 0.9 87.6 − 8.2 79.4 − 17.3 Drugs active on the cardiovascular system 30.8 3.6 48.3 − 2.9 69.7 − 1.0 89.1 − 10.9 111.0 − 21.2 Antianaemia preparations 28.0 6.3 42.1 1.5 62.4 5.1 85.0 − 3.7 111.1 − 14.6 Thyroid drugs 44.1 1.9 48.3 − 2.9 48.2 1.5 44.8 − 6.4 36.2 − 13.8 NSAIDs 34.5 − 5.8 39.7 − 11.9 40.5 − 9.6 38.9 − 18.0 30.3 − 25.3 Drugs for gout 21.1 7.3 31.2 0.3 40.3 3.0 48.9 − 6.1 49.2 − 16.1 Corticosteroids for systemic use 21.8 1.2 25.8 − 3.5 29.2 0.1 31.2 − 8.0 29.6 − 16.0 Anti-parkinsonian drugs 10.5 7.4 16.5 0.8 23.8 2.5 27.1 − 9.5 19.9 − 19.7 Pain therapy 11.6 2.0 15.4 − 2.5 20.5 0.7 25.2 − 8.1 24.9 − 17.2 Antibiotics 17.0 − 20.0 18.2 − 23.9 18.5 − 19.7 18.9 − 24.3 20.7 − 27.7 Oncological drugs 11.4 5.1 14.1 − 0.6 17.1 2.5 19.6 − 7.2 17.8 − 18.0 Antiepileptics 12.9 0.7 13.2 − 3.4 14.1 0.7 14.4 − 7.6 12.9 − 17.1 Antihistamines 12.3 6.0 12.7 − 0.2 12.0 3.5 11.3 − 5.9 10.3 − 14.0 Intestinal anti-inflammatory 9.4 1.1 10.0 − 4.6 9.5 − 0.2 9.0 − 7.9 6.8 − 15.7 Antidementia drugs 1.7 13.4 4.7 11.2 11.3 11.3 18.2 − 3.0 14.7 − 18.2 Antipsychotics 7.0 0.1 6.2 − 2.4 6.3 4.4 7.7 0.7 10.8 − 6.4 Hepatic and biliary therapy 5.0 5.7 6.0 0.3 6.7 3.9 7.5 − 4.1 8.4 − 11.3 Non-systemic antibiotics 4.2 0.7 5.7 − 5.2 6.7 − 2.3 7.2 − 11.0 6.6 − 18.7 Dermatologic drugs 4.8 − 5.7 5.1 − 12.1 4.8 − 10.6 4.2 − 18.8 3.2 − 25.2 Sexual hormones and modulators of the genital system 4.0 − 13.6 3.1 − 18.9 2.4 − 15.3 2.1 − 20.1 2.0 − 26.0 Few medications showed a negative variation across all age groups, namely, non-steroidal anti-inflammatory drugs (from − 25.3% in 85 + to − 5.8% in 65–69 years), antibiotics (from − 27.7% in 85 + to − 19.7% in in 75–79 years), dermatologic drugs (from − 25.2% in 85 + to − 5.7% in 65–69 years), and drugs for the genital system/sexual hormones (from − 26.0% in 85 + to − 13.6% in 65–69 years). Figures 1, 2, 3, and 4 show the difference in DDD/1000 inhabitants, prevalence, and incidence between 2020 and 2019 of antidiabetics, antihypertensives, drugs for chronic pulmonary disease and antidepressants, according to age groups and sex. Incidence variations between 2020 and 2019 were very high in both men and women in all the four categories: for antidiabetics incidence ranged from − 35% in men to − 30% in women; for antihypertensives from − 47% in men to − 42% in women, for antidepressants from − 18.9% in women to − 14.6% in men, and for respiratory drugs from − 38% in women to − 35% in men (Figs. 1a, 2a, 3a, 4a). When we analysed incidence variation across age groups, we found that the highest negative trend was in the youngest persons (65–69 years) for antidiabetics (− 57%) and for antidepressants (− 35%) and in the oldest ones (85 + years) for respiratory drugs and for antihypertensives (− 43, − 75%, respectively).Fig. 1 Antidiabetics: DDD/1000 inhabitants, prevalence and incidence: variation 2020–2019. (a) By sex, (b) by age Fig. 2 Antihypertensives: DDD/1000 inhabitants, prevalence and incidence: variation 2020–2019. (a) By sex, (b) by age Fig. 3 Asthma and chronic obstructive pulmonary disease drugs: DDD/1000 inhabitants, prevalence and incidence: variation 2020–2019. (a) By sex, (b) by age Fig. 4 Antidepressants: DDD/1000 inhabitants, prevalence and incidence: variation 2020–2019. (a) By sex, (b) by age Discussion In our analysis, we found that almost all medications had a negative percentage variation in DDD/1000 inhabitants between 2020 and 2019 except for anticoagulants. However, variation varied widely according to age and was highest in 85 + year-old persons. On the contrary, in the youngest age group (65–69 years), the percentage variation in medication consumption was positive for most drug classes, reaching + 15.8% for anticoagulants. Notably, when we calculated the incidence of dispensation of four common drug classes, i.e. antidiabetics, antihypertensives, antidepressants and drugs for chronic pulmonary disease, and its difference between 2020 and 2019, we found huge negative variations (− 32.8,− 44,− 17 and− 36.9%, respectively). The impact of COVID-19 pandemic on the worldwide population has been dramatic, especially in the older age strata. According to the Italian National Health Institute, 75.000 individuals died because of COVID-19 during 2020, 10% of all deaths in Italy [14]. Beyond mortality, it is already clear that the effects of the pandemic in older persons will include symptoms of long-term COVID in those who survived the infection [15] and possibly consequences of social isolation and change in care provision. One implication of the pandemic which has been poorly analysed is the observed trend in medication consumption between 2020, when the pandemic initiated, and the previous years. Reasons for a change in the amount, prevalence and incidence of medication use may have several explanations; first, the decrease in doses used in 2020 in the very old persons may reflect the high mortality in this stratum of the population. The mean age of persons who died in Italy due to COVID-19 was 80 years (median is 82); among all deaths only 1.2% of persons were 50-year-old or younger [14]. Further, the mean number of comorbidities in those who died was 3.7, and 67.7% were affected by at least three co-existent diseases [16]. Considering that polypharmacy increases with increasing age and with the number of co-occurring diseases, it is obvious that the most negative variation amount of drug use is found in persons aged 80 + years. On the other hand, the maintenance and, for some drug classes, the increase in prevalence of dispensation in the younger adults during 2020 show the ability of the national and regional health systems to providing care supplies to patients already treated for chronic diseases even during the worst months of the pandemic. The only positive variation in DDD/1000 inhabitants concerns the class of anticoagulants. This data is not surprising considering that both low-weight heparins and oral direct anticoagulants were widely prescribed to treat COVID-19 intrinsic risk of coagulopathy and its consequences, such as deep venous thrombosis, even at home [17]. Second, underdispensation of some specific medications such as antibiotics and anti-inflammatories may reflect a lower rate of infectious diseases due to the widespread use of protective devices and self-isolation during lockdown periods as well as better appropriateness of antibiotic prescription for upper respiratory infections. Studies have reported a reduction in commonly prescribed antibiotics in several countries. A decrease in antibiotic consumption was observed in the European Union/European Economic Area-wide between 2019 and 2020 and it is the largest in European Surveillance of Antimicrobial Consumption Network’s two-decade long antimicrobial consumption surveillance history [18]. Overall, antibiotic consumption decreased by 18.3% between 2019 and 2020 and the largest decrease in terms of DDD/1,000 inhabitants between 2019 and 2020 was observed for penicillins [18]. Similar changes in community antibiotic consumption have been described at the national levels: in the adult population of primary care in Asturias in the period from March to December 2020, the consumption of antibiotics decreased by 28.6% compared to the same period in 2019 [19]. Similar data were found in primary care in Andalusia with the sharpest decline occurring in the second quarter of 2020, when the lockdown was more extensive [20]. In the USA, significant reductions in mean monthly fills of the four most prescribed outpatient agents (i.e. amoxicillin, azithromycin, amoxicillin–clavulanate, doxycycline) were found in April and persisted throughout 2020, compared to estimations based on pre-pandemic trends [21]. Considering trend in other medication use, in a very large study evaluating psychotropic drugs in comparison with statins in the 13 weeks before and after the first-known COVID-19-related death in California, in contrast to continued prescriptions, new fills for psychotropic medications were substantially lower than what would have been expected based on 2019 rates. Patients were especially less likely to fill new prescriptions for benzodiazepines, hypnotics, and statins and those aged ≥ 65 years demonstrated decreased fills for most medication classes [22]. Finally, when we evaluated the incidence of dispensation of some common medication classes and its difference between 2020 and 2019, we found negative variations in all age groups, but especially in the young-old (65–69 years) and the very old (85 + years). Different reasons may have led to a reduction in the incidence of dispensation according to age; on one hand, the lockdown, the lack of social support and the fear of going to hospitals may explain the phenomenon in persons aged 85 years or older; on the other hand, the reduction of screening and diagnostics due to the health-care system overload have probably led to a reduction of first diagnoses in the young-old. Indeed, the COVID‐19 pandemic has affected all areas of the health-care system, not just infectious disease, and critical care [23]. The first wave of COVID‐19 was associated with decreases and delays in acute care presentations for cardiovascular diseases such as myocardial infarction in the USA and Europe [24]. Primary cancer screenings (e.g. low‐dose chest CT, colonoscopies) could not be performed without significant risk leading to delays in diagnosis. Given health professionals’ reassignment, and reduced operating room capacity, it is likely that chemotherapy, radiation therapy and surgical excision of tumours were delayed [25]. Further, the COVID-19 pandemic prevented some individuals from keeping regular medical appointments even in countries with a low infection risk; in Taiwan the visits of the elderly (≥ 80 years) were the most frequent before the COVID-19 pandemic but were reduced by 44% [26]. Strengths and limitations This study has several limitations. Data on reimbursed medications exclude unfilled prescriptions and non-reimbursed medications. This might have led to an underestimation of the prevalence and incidence of use and the amount of consumption, although it is important to highlight that in Italy medicines proven to be effective for the treatment of acute or chronic diseases are covered by NHS and private purchase represents less than 30% of the outpatient consumption. It is worth noting that oncological drugs in Italy are mostly dispensed in the inpatient setting; therefore, data presented in this study related to medicines dispensed by community pharmacies could not provide a full picture for this category. Differences in the use of anti-osteoporotic drugs observed can be explained by a change in the prescription regulation, occurring at the end of 2019 [27]. Since the data referred to prescribed and dispensed medicines, it is not possible to evaluate medication adherence and the actual intake of the medications by patients, information particularly relevant in the management of chronic diseases [28]. Moreover, the analysis did not consider medication use during potential hospitalization periods of patients, for instance for the treatment of COVID-19. Finally, the annual analysis included individuals that had died within the year; however, since the analysed data referred to the entire Italian population aged 65 years and older, it is unlikely that the difference in medications use between 2020 and 2019 might have been influenced by the mortality rate. While it is true that the pandemic onset was the major acute event characterizing the period 2019–2020, there may be unmeasured events that may explain medication utilization patterns over time. Despite these limitations, our study provides important insights into the impact of the pandemic on medication utilization in older people in Italy. Conclusions Variation in drug dispensation between 2020 and 2019 is a consequence of the COVID-19 pandemic: first, underprescription of some specific drug categories such as antibiotics and anti-inflammatory drugs may reflect a lower rate of infectious diseases due to the widespread use of protective devices and self-isolation; second, the lockdown, the fear of going to hospital and the reduction of screening and diagnostics due to health-care system overload may have led to a dramatic reduction of first drug prescriptions. The consequences of underdiagnosis and underprescription should be a priority of future studies. Acknowledgements We would like to acknowledge IWGMUE, formed as follows: Italian Medicines Agency (Agenzia Italiana del Farmaco “AIFA”): Nicola Magrini, AC, Ivano Comessatti, Aurora Di Filippo, Andrea Fabrizi, Marco Fontanella, Filomena Fortinguerra, Ramon Frulio, Francesca Gallinella, Maria Alessandra Guerrizio, Mariarosaria Italiano, Marco Marinelli, Roberto Marini, Federica Milozzi, Serena Perna, AP, Linda Pierattini, Emanuela Pieroni, Giuliano Pistolesi, Filippo Pomponi, Matteo Sacconi, Daniela Settesoldi, Maurizio Trapanese, Giuseppe Traversa, FT, Saverio Antonio Vasta, and SZ. Istituto Superiore di Sanità (ISS): Silvio Brusaferro, RDC, II, GO, and Paola Ruggeri. Ministero della Salute: Claudia Biffoli, Rosaria Boldrini, and Chiara Brutti. Ministero dell’Economia e delle Finanze: Luciana Patrizi, Antonietta Cavallo, Sara Guerrini, and Marco Martino. Istituto di Ricerche Farmacologiche Mario Negri IRCCS: Alessandro Nobili, Luca Pasina, Carlotta Franchi, Mauro Tettamanti, and Alessio Novella. Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico di Milano: Pier Mannuccio Mannucci. Università di Bologna: Elisabetta Poluzzi. Università di Brescia: AM. Società Italiana di Medicina Interna (SIMI): Antonello Pietrangelo and Maria Lorenza Muiesan. Società Italiana di Medicina Generale e delle Cure Primarie (SIMG): Claudio Cricelli, Pierangelo Lora Aprile, Gerardo Medea, Ignazio Grattagliano, Raffaella Michieli, Damiano Parretti, Francesco Lapi, and Ettore Marconi. Nicola Magrini, Ivano Comessatti, Aurora Di Filippo, Andrea Fabrizi, Marco Fontanella, Filomena Fortinguerra, Ramon Frulio, Francesca Gallinella, Maria Alessandra Guerrizio, Mariarosaria Italiano, Marco Marinelli, Roberto Marini, Federica Milozzi, Serena Perna, Linda Pierattini, Emanuela Pieroni, Giuliano Pistolesi, Filippo Pomponi, Matteo Sacconi, Daniela Settesoldi, Maurizio Trapanese, Giuseppe Traversa, Saverio Antonio Vasta, Silvio Brusaferro, Paola Ruggeri, Claudia Biffoli, Rosaria Boldrini, Chiara Brutti, Luciana Patrizi, Antonietta Cavallo, Sara Guerrini, Marco Martino, Alessandro Nobili, Luca Pasina, Carlotta Franchi, Mauro Tettamanti, Alessio Novella, Antonello Pietrangelo, Maria Lorenza Muiesan, Claudio Cricelli, Pierangelo Lora Aprile, Gerardo Medea, Ignazio Grattagliano, Raffaella Michieli, Damiano Parretti, Francesco Lapi, Ettore Marconi. Author contributions Drs Cangini and Trotta had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: Drs Marengoni, Cangini, Zito, Onder, Trotta. Acquisition, analysis, or interpretation of data: Drs Cangini, Trotta, Zito, Pierantozzi. Drafting of the manuscript: Drs Marengoni, Cangini. Critical revision of the manuscript for important intellectual content: Drs Marengoni, Onder, Trotta, Cangini, Ippoliti, Zito, Pierantozzi, Da Cas. Funding None. Data availability The datasets generated during and/or analysed during the current study are not publicly available because of data sharing legal restrictions, the dataset including individual records cannot be made publicly available. However, aggregated data will be shared on reasonable request to the corresponding author. Declarations Conflict of interest No disclosures were reported. Statement of human and animal rights Thia article does not contain any studies with human participants or animals performed by any of the authors. Informed consent For this type of formal consent is not required. Disclaimer The views expressed in this article are the personal views of the authors and may not be understood or quoted as being made on behalf of or reflecting the position of the respective authors’ organizations. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. https://www.who.int/director-general/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19---11-march-2020 (last access March 11st, 2022) 2. Ismail H Marshall VD Patel M Tariq M Mohammad RA The impact of the COVID-19 pandemic on medical conditions and medication adherence in people with chronic diseases J Am Pharm Assoc 2021 10.1016/j.japh.2021.11.013 3. Palmer K Monaco A Kivipelto M Onder G Maggi S Michel JP Prieto R Sykara G Donde S The potential long-term impact of the COVID-19 outbreak on patients with non-communicable diseases in Europe: consequences for healthy ageing Aging Clin Exp Res 2020 32 1189 1194 10.1007/s40520-020-01601-4 32458356 4. Gillam MH Roughead E Tavella R Dodd T Beltrame J Ryan R O'Loughlin P Impact of COVID-19 restrictions on pathology service utilisation Intern Med J 2022 52 42 48 10.1111/imj.15501 34432345 5. De Filippo O D’Ascenzo F Angelini F Bocchino PP Conrotto F Saglietto A Reduced rate of hospital admissions for ACS during Covid-19 outbreak in Northern Italy N Engl J Med 2020 383 88 89 10.1056/NEJMc2009166 32343497 6. Song H Bergman A Chen AT Ellis D David G Friedman AB Disruptions in preventive care: mammograms during the COVID-19 pandemic Health Serv Res 2021 56 95 101 10.1111/1475-6773.13596 33146429 7. Diegoli H Magalhaes PSC Martins SCO Moro CHC França PHC Safanelli J Decrease in hospital admissions for transient ischemic attack, mild, and moderate stroke during the COVID-1 era Stroke 2020 51 2315 2321 10.1161/STROKEAHA.120.030481 32530738 8. Solomon MD McNulty EJ Rana JS Leong TK Lee C Sung S-H The Covid-19 pandemic and the incidence of acute myocardial infarction N Engl J Med 2020 383 691 693 10.1056/NEJMc2015630 32427432 9. ISTAT Impatto dell’epidemia COVID-19 sulla mortalità totale della popolazione residente. Anno 2020 e gennaio-aprile 2021. Available from: https://www.iss.it/documents/20126/0/Report_ISS_Istat_2021_10_giugno+%281%29.pdf/3df35cd3-edd5-56e1-5867-d274f5032fb5?t=1623336923282 10. Calderón-Larrañaga A Vetrano DL Onder G Gimeno-Feliu LA Coscollar-Santaliestra C Carfí A Pisciotta MS Angleman S Melis RJF Santoni G Mangialasche F Rizzuto D Welmer AK Bernabei R Prados-Torres A Marengoni A Fratiglioni L Assessing and measuring chronic multimorbidity in the older population: a proposal for its operationalization J Gerontol A Biol Sci Med Sci 2017 72 1417 1423 10.1093/gerona/glw233 28003375 11. The Medicines utilisation monitoring centre. national report on medicines use in older adults in Italy. Year 2019. Rome: Italian Medicines Agency, 2021. Rome: Italian Medicines Agency, 2021. Available from: https://www.aifa.gov.it/documents/20142/1577699/OsMed_Farmaci_anziani_13.10.2021.pdf 12. The Medicines utilisation monitoring centre. national report on medicines use in Italy. Year 2020. Rome: Italian Medicines Agency, 2021. Available from: https://www.aifa.gov.it/en/-/rapporto-nazionale-osmed-2020-sull-uso-dei-farmaci-in-italia 13. WHO collaborating centre for drug statistics methodology, guidelines for ATC classification and DDD assignment 2020. Oslo, Norway, 2019. 14. https://www.iss.it/documents/20126/0/Report_ISS_Istat_2020_5_marzo+%281%29.pdf/18f52493-6076-9ec3-7eb2-b39efed8b22f?t=1614946675778 15. Tosato M, Carfì A, Martis I, Pais C, Ciciarello F, Rota E, Tritto M, Salerno A, Zazzara MB, Martone AM, Paglionico A, Petricca L, Brandi V, Capalbo G, Picca A, Calvani R, Marzetti E, Landi F (2021) Gemelli Against COVID-19. Post-Acute Care Team. Prevalence and Predictors of Persistence of COVID-19 Symptoms in Older Adults: A Single-Center Study. J Am Med Dir Assoc 10.1016/j.jamda.2021.07.003 16. Palmieri L, Vanacore N, Donfrancesco C, Lo Noce C, Canevelli M, Punzo O, Raparelli V, Pezzotti P, Riccardo F, Bella A, Fabiani M, D'Ancona FP, Vaianella L, Tiple D, Colaizzo E, Palmer K, Rezza G, Piccioli A, Brusaferro S, Onder G (2020) Italian National Institute of Health COVID-19 Mortality Group. Clinical Characteristics of Hospitalized Individuals Dying With COVID-19 by Age Group in Italy. J Gerontol A Biol Sci Med Sci 75:1796–1800. 10.1093/gerona/glaa146 17. Hanff TC Mohareb AM Giri J Cohen JB Chirinos JA Thrombosis in COVID-19 Am J Hematol 2020 95 1578 1589 10.1002/ajh.25982 32857878 18. Högberg LD, Vlahović-Palčevski V, Pereira C, Weist K, Monnet DL; ESAC-Net study group; ESAC-Net study group participants. (2021) Decrease in community antibiotic consumption during the COVID-19 pandemic, EU/EEA, 2020. Euro Surveill 10.2807/1560-7917.ES.2021.26.46.2101020. 19. Nicieza García ML Pérez Solís P Gómez de Oña C Suárez Gil P Rolle Sóñora V Suárez MB Antibiotic consumption in primary care in the adult population of Asturias during 2014–2020 period Aten Primaria 2021 54 102261 10.1016/j.aprim.2021.102261 34922065 20. Peñalva G Benavente RS Pérez-Moreno MA Pérez-Pacheco MD Pérez-Milena A Murcia J Cisneros JM Effect of the coronavirus disease 2019 pandemic on antibiotic use in primary care Clin Microbiol Infect 2021 27 1058 1060 10.1016/j.cmi.2021.01.021 33540117 21. Buehrle DJ Wagener MM Nguyen MH Clancy CJ Trends in outpatient antibiotic prescriptions in the united states during the COVID-19 pandemic in 2020 JAMA Netw Open 2021 10.1001/jamanetworkopen.2021.26114 34550387 22. Hirschtritt ME Slama N Sterling SA Olfson M Iturralde E Psychotropic medication prescribing during the COVID-19 pandemic Medicine (Baltimore) 2021 10.1097/MD.0000000000027664 34713861 23. Chang AY Cullen MR Harrington RA Barry M The impact of novel coronavirus COVID-19 on noncommunicable disease patients and health systems: a review J Intern Med 2021 289 450 462 10.1111/joim.13184 33020988 24. Bhatt AS Moscone A McElrath EE et al Fewer hospitalizations for acute cardiovascular conditions during the COVID-19 pandemic: a multicenter tertiary care experience J Am College Cardiol 2020 76 280 288 10.1016/j.jacc.2020.05.038 25. Willan J King AJ Hayes S Collins GP Peniket A Care of haematology patients in a COVID-19 epidemic Br J Haematol 2020 189 241 243 10.1111/bjh.16620 32173855 26. Jeng Y Chen FH Jen GH Chen HC Yen AM Chen CD Kuo HW Wang ST Hsu CY Impact of COVID-19 pandemic on accessibility of Taiwanese medical care Am J Manag Care 2021 27 e330 e335 10.37765/ajmc.2021.88698 34533916 27. https://www.aifa.gov.it/documents/20142/1030827/Nota%2096.pdf 28. Doggrell SA Adherence to medicines in the older-aged with chronic conditions Drugs Aging 2010 27 239 254 10.2165/11532870-000000000-00000 20210369	36469252	PMC9734516	NO-CC CODE	2022-12-14 23:28:29	no	Aging Clin Exp Res. 2022 Dec 5;:1-11	utf-8	Aging Clin Exp Res	2,022	10.1007/s40520-022-02303-9	oa_other
==== Front Nonlinear Dyn Nonlinear Dyn Nonlinear Dynamics 0924-090X 1573-269X Springer Netherlands Dordrecht 8125 10.1007/s11071-022-08125-8 Original Paper Nonlinear dynamic epidemiological analysis of effects of vaccination and dynamic transmission on COVID-19 http://orcid.org/0000-0003-3826-1233 Kambali Prashant N. [email protected] Abbasi Amirhassan [email protected] Nataraj C. [email protected] grid.267871.d 0000 0001 0381 6134 Villanova Center for Analytics of Dynamic Systems (VCADS), Villanova University, Villanova, USA 5 12 2022 113 15 5 2022 8 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. This paper is concerned with nonlinear modeling and analysis of the COVID-19 pandemic. We are especially interested in two current topics: effect of vaccination and the universally observed oscillations in infections. We use a nonlinear Susceptible, Infected, & Immune model incorporating a dynamic transmission rate and vaccination policy. The US data provides a starting point for analyzing stability, bifurcations and dynamics in general. Further parametric analysis reveals a saddle-node bifurcation under imperfect vaccination leading to the occurrence of sustained epidemic equilibria. This work points to the tremendous value of systematic nonlinear dynamic analysis in pandemic modeling and demonstrates the dramatic influence of vaccination, and frequency, phase, and amplitude of transmission rate on the persistent dynamic behavior of the disease. Keywords COVID-19 Vaccination Dynamic transmission Bifurcation Nonlinear dynamics http://dx.doi.org/10.13039/100000006 Office of Naval Research N00014-19-1-2070 Nataraj C. ==== Body pmcIntroduction As is well known by now, the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and resulting coronavirus disease 2019 (COVID-19) was first reported in Wuhan city, Hubei province, China in early December 2019. The rapid spread of the disease to other countries led World Health Organization (WHO) to declare it a global pandemic in March 2020. Since then, COVID-19 has caused over 6.2 million fatalities and presented unprecedented challenges to healthcare systems even in well-developed countries in addition to causing major disruptions to the economy all around the world [1]. Analysis of data related to COVID-19 shows that the propagation of this disease is governed by a nonlinear multi-variable dynamic process that depends on a broad range of variables. These variables are dependent on the social, governmental, economic, and environmental situations in addition to being influenced strongly by human activities. Further exploration of worldwide data shows cumulative interactions of these variables that make it challenging to identify their individual effects from the data. Given this complexity, analytical epidemiological modeling is of high significance since it can serve as a valuable tool for uncovering how individual variables contribute to the overall dynamics. In addition, accurate modeling and dynamic analysis of the infection spreading provides deeper insight and identification of the most influential variables. Specifically, considering the variation of these variables in different target populations, e.g., cities, states, countries, and continents provide insightful demographic information about the propagation of the disease. These insights can serve as a valuable basis for designing control strategies for COVID-19 as well as for many more pandemics that will certainly arise in the future. Over the past two years, the efficacy of various strategies, both pharmaceutical and non-pharmaceutical have been evaluated. For example, it has been repeatedly shown that self-isolation and lockdown greatly limit the rate of infections thereby decreasing the peak load on the national healthcare systems. As we know, reducing the peak loads prevent the healthcare systems from getting overwhelmed leading to higher survival rates. Also, beneficial effects of massive and rapid vaccination on the spread of infection, severity of symptoms and illness duration were evidenced as well. However, clarifying the role of seasonality in modeling the disease spread and proposed control strategies are still far from mature. It would arguably improve our understating about the fluctuations, trends, and patterns that occur or repeat over a specific period of time in an epidemic [2]. The extracted information also can be used for long-term planning by determining the probability of different equilibrium points to predict whether it is surging, receding, or evolving into an endemic equilibrium. Also, as was mentioned, the demographic analysis would be insightful with regard to medical and behavioral variables such as vaccination and social distancing in each target population as well. Another important aspect of this temporal analysis concerns the biological characteristics of the disease. Indeed, observing seasonality in infectious over time can be informative about the biological parameters of the host-microbe interaction, including the prevalence of the disease among infected individuals, infection contagiousness, diagnostic capacity, and size of the susceptible population as well [3]. There has been a profusion of publications focusing on data-based modeling, too many in fact to document here. While these efforts are laudable and can lead to some understanding of immediate trends, their contribution towards gaining insight are very limited. In addition, an over emphasis on including every variable possible impedes the interpretability of the analysis results. Such an approach may also increase the risk of missing identification of causalities due to artifact effects of data and corresponding correlations. As a remarkable case in point, a recent study analyzed seasonality and corresponding oscillations in infected individuals in New York City, NY, and Los Angeles, CA, where very similar oscillations can be seen in data related to the number of new cases. This observed statistical similarity, analyzed by purely data-based modeling, would easily mislead the process of decision-making and planning into formulating the same policy, for example. In fact, further exploration actually revealed a key demographic driving variable that turned out to be the daily variation in the number of tests in each region [3]. Further, as another example, the artifact of intermittent reporting led one to believe that there were puzzling oscillations in mortality data [4]. Many other factors can adversely affect the accuracy of pure-data-based analysis as well; for example, several studies pointed out a significant local delay in reporting the number of new infections and deaths [5]. In contrast to the above instances, there is a very long history of mathematical modeling of epidemiological phenomena dating back to early studies by Bernoulli in the eighteenth century [6, 7]. These models strive to explain the relevance of real historical observations and infectious disease dynamics within a mathematical framework. In fact, most current studies use models built upon those proposed in the 1930’s by Kermack and McKendrick [8, 9]. In general, these models known as compartment models include a set of nonlinear ordinary differential equations in which state variables represent the population numbers in different stages of the infectious disease spread [10]. Various epidemiological models were developed rapidly after Kermack’s original work [11]. Several notable modern textbooks discuss fundamental phenomena in mathematics of epidemiology. We refer the reader to them for a clearer understanding of the model assumptions, derivations, and implications [10, 12–14]. Hethcote’s paper [15] is another notable resource that provides intuitive information on the dynamics of this disease. Non-periodic responses of compartmental models were analyzed in [16, 17] that represent progression toward an endemic equilibrium point. Effects of periodicity are most likely caused by actions such as opening and closure of public places, change of seasons, vacillations in social behavior, and new variants with varying infection rates. These have often been modeled using the concept of forced oscillations. In an early work by [18], the effects of seasonal fluctuations along with contact rate periodicity were studied. Another notable nonlinear phenomenon observed leading to periodic responses is Hopf bifurcation. The effects of model parameter values on the existence of locally asymptotically stable solutions and Hopf bifurcation were studied in [19]. It was shown that the occurrence of periodic solutions through Hopf bifurcations can mathematically be related to the presence of time delays and nonlinear incidence rates [20–22]. The effects of seasonal variations in the contact rate on the occurrence of infinite sub-harmonic bifurcations in epidemic models were investigated as well in [23]. Also, bifurcation was studied with the hospital resource as a varying parameter [24]. A comprehensive study of periodic and non-periodic dynamics of compartmental models is also discussed in [13]. Literature also shows that compartmental models were modified to analyze the seasonality and corresponding oscillation [15, 25, 26]. In [27] it is proved that under the assumption of a fixed population, SIR model and constant parameters, the model will show damped oscillation with a stable endemic point. In another category of study, the exogenous and endogenous mechanisms were incorporated in the modeling for temporal analysis [25]. The effects of these mechanisms on sustained oscillation were analyzed as well. It was shown that the exogenous mechanism causes oscillation by periodic forcing of the transmission term, while the endogenous mechanism can destabilize the model’s endemic equilibrium by adding delay terms [25]. Other studies tried to make the problem more realistic by canceling some of the modeling assumptions. For example, classical compartmental models are developed under the assumption of constant, random, and homogeneous mixing populations with indistinguishable individuals [28]. However, in real scenarios, each individual is in contact with a specific social network which means the number of potential contacts for each individual is significantly smaller than the total population size. Hence, in a different category of research, infectious diseases are modeled based on complex social network structures. The general idea is to abstract the population and corresponding connections in a network structure by nodes and links, respectively. Accordingly, the spread of diseases is modeled by a network of complex interactions between individuals [26, 29–32]. Other studies have investigated both temporal and spatial aspects of infectious diseases spread by developing compartmental models in a cellular automata framework, that aims to gain insight on the possibility of oscillations in different groups of the total population. Groups have been formed based on various criteria such as levels of social activity [33–36]. Another trend of research focuses on the mathematical abstraction of seasonality by considering time-dependent model parameters. A compartmental model with time-varying birth and death rates is one of the recent examples in this regard [4]. For further exploration, we suggest [37], which is a recent review that surveys the literature on seasonal dynamics. In general, prophylactic measures for the control of infectious disease spread include vaccination, treatment, and isolation. Many studies focused on analyzing the dynamics of compartmental models with these measures and control strategies including non-pharmaceutical and pharmaceutical ones [38–41]. As notable examples, in [42, 43] optimal control and parameter tuning were studied with the aim of mitigating the disease. Other studies related to control strategies focused on the effects of infectious transmission. For example, various compartmental models with different transmission formulations were studied in [44, 45]. Literature also emphatically demonstrates that the vaccination policy, which is generally adopted by governments, is the principal component in the control of disease’s spread. It should be noted that vaccination will not fully protect an individual from infection, so there is always a probability for a vaccinated person to get infected. Even in the case of releasing antibodies in a host, some pathogens can mutate. It means that the immune system might not be able to defeat the infection. To characterize these issues and hence the level of protection that is provided by a vaccine in an individual, epidemiologists define the efficacy of any vaccine. Another notable factor is a personal decision. An individual decision about complying with vaccination policy is involved with the assessment of the risk of morbidity from vaccination and the probability of infection. In addition to risk, an individual’s willingness varies with other factors such as age, health conditions, lifestyle and profession. Each of them can provide insight into analyzing the dynamics of the disease spread, vaccination and decision making [46, 47]. Another factor affecting the vaccination is heterogeneity in the population’s response to the vaccine. In this condition, it is assumed that individuals respond differently to a vaccine that affects the dynamics of the epidemic. In [48] the global stability of the infection-free equilibrium state is mathematically formulated. In this study, in order to provide a generic and interpretable abstraction of the dynamics of this disease, we consider an epidemiological model which makes a compromise between predictive accuracy and complexity. Instead of purely relying on data in modeling, we strive to understand its diversity over different target populations and seek to discover the principal parameters that contribute to this difference. In other words, instead of fitting the model simulation results to data, we use data for guiding and validating the analysis. We do believe that fundamental characteristics of dynamics of this disease along with the most influential variables can be accurately, effectively, and efficiently identified using a compact model. Such a parsimonious model provides a fundamental understanding of the dynamics of the disease that not only explains what happened so far but also predicts infectious progression and control. It is important to note that this mechanistic approach of modeling COVID-19 and emphasizing the causalities in the arguments make the analysis more easily generalizable to future pandemics and outbreaks. There is ample support in the literature to make the case that such models have the highest potential and utility. In fact, they have been used to investigate a wide range of outbreaks including the SARS epidemic in 2002-2003 [49], H5N1 influenza outbreak in 2005 [50], the H1N1 influenza pandemic of 2009 [51] and the Ebola outbreak of 2014 [52]. With this perspective, we analyze the dynamics of this COVID-19 pandemic using a compartmental model. The effects of seasonality are incorporated in modeling the disease transmission. The corresponding forced oscillation is analyzed to investigate mutual interactions between transmission and seasonality characteristics, vaccination rate and efficacy, and different sources of immunity. It should be noted that there are two major differences in developing vaccination models. The first difference concerns the vaccination class. In one approach, it is assumed that vaccination is equivalent to going through the disease and considers vaccinated individuals as recovered ones. That means, in the compartmental model there is no need for an additional class for vaccinated individuals. The second difference addresses the discrete and continuous vaccination. In the first approach, it is assumed that individuals enter the system at a point in their life upon vaccination or skipping vaccination and then are included in the susceptible class. School children are a notable example of a discrete approach. In the continuous approach, models provide for continuous vaccination of the population as long as they are in the system. In a real scenario, vaccination cannot be discrete since a vaccinated individual may get susceptible over time given the imperfection inherent in any vaccination [13]. In this study, with the continuous assumption in addition to consideration of vaccination efficacy, we study the effects of the seasonality of the disease with a periodic transmission rate as well. The remaining parts of this paper are organized as follows. In Sect. 2, the model development is outlined. In Sect. 3, nonlinear analysis for disease-free and endemic equilibrium is investigated. In Sect. 4, discussion and summary of findings are presented and finally, we present a conclusion in Sect. 5. Notes on mathematical models First, we define the key variables used in epidemiological models. Although the definitions of the terms are found universally, we define them here for completeness.Susceptible individuals. An individual is considered susceptible when there is no detectable level of pathogens in the body. Accordingly, the body’s immune system did not respond to the disease-causing pathogen, so the individual’s immune system has not yet developed a specific response to the disease-causing pathogen. Exposed individuals. An exposed person had contact with an infected person and is infected; however, no obvious symptoms can be observed and the level of the pathogen is so low that it will not sustain transmission to the other host. Infected individuals. The infected individual body has high enough amount of pathogens that he/she can transmit to other susceptible individuals. Removed individuals. The individual immune system has defeated the infection and reduced the number of parasites significantly in which case the individual has recovered. In addition, isolation from the general population or death by the disease can contribute to this class. In all of these cases, the individual is said to be removed. Immune individuals. An individual who is protected against the disease because of a vaccine or other sources of immunity. The models are typically notated using initials, for example, SIR for Susceptible-Infectious-Removed. Without loss of generality, the number of individuals in each class is usually normalized with respect to a nominal, total population (as we do in this paper). In our study, we consider the Susceptible-Infected-Immune (SII) framework. In this analysis, we assume that the total population size (denoted by N) remains relatively constant during pandemic based on the following reasoning. This analysis is carried out for a short period (a year or two), so we assume that the change in the total population due to the death of diseased and newborn babies is negligible during the analysis period. The change in the population owing to disease-related deaths and newborns is not substantial, unlike when a population used to experience diseases with decimating effects such as the plague, or when a pandemic is analyzed for a considerable period such as a decade or more. Therefore, it is reasonable to assume that the death and birth rates are not influential enough to alter the total population significantly over the period of the analysis. In this model, X, and Y represent the susceptible and infectious populations and Z represents the immune population. The immune population considered here is a combination of vaccine induced immune population and immunity of population obtained by the other sources that include natural immunity. This model (SII) is adopted and modified from a standard SIS model with vaccination as illustrated in Fig. 1 [13] and is defined as follows.1 X′=Λ-βXY-(μ+ψ)X+χγY,Y′=βXY+βδZY-(μ+γ)Y,Z′=ψX-βδZY+(1-χ)γY-μZ, where, ′ denotes derivative with respect to time. In this model, Λ is the birth rate, μ is the death rate, γ is the recovery rate, β is the transmission rate, ψ is the vaccination rate, χ is the proportion of individuals who recover to the immune class due to immunity other than vaccine-induced immunity, 1-χ is the proportion of individuals who recover to the susceptible class, δ is the vaccine reduction coefficient, where 0≤δ≤1. If δ=0 the vaccination is perfect. ϵ=1-δ is hence the vaccine efficacy. Figure 1 illustrates the Susceptible-Infected-Immune model. In practice, the vaccine is applied to healthy individuals, so only susceptible individuals get vaccinated (ψX). βXY is the number of secondary infections of susceptible individuals. βδZY is the number of secondary infections of vaccinated individuals with an imperfect vaccine. χγY is the number of individuals who recover to the immune class due to immunity other than vaccine-induced immunity. (1-χ)γY is the number of individuals who recover to the susceptible class. μX, μY and μZ are the number of individuals removed from susceptible, infected, and immune classes.Fig. 1 Susceptible-infected-immune model Although the exact reasons for the dynamic behavior of an endemic disease like COVID-19 are not precisely known, we can indulge in some rational speculation. Some of the reasons plausibly include imperfect vaccines, environmental factors, new variants of the virus (with different infection rates), social and biological factors, such as population density, social distancing policies, and governmental actions. In [53], researchers hypothesized that autonomous models exhibit dynamic behavior intrinsically and oscillate for a constant value of transmission rate, β. In practice, it is difficult to explain the dynamic behavior of the disease with a constant value of β. Recent studies suggest that the dynamic behavior can be fairly explained by taking a periodic transmission rate [54]. Therefore, to incorporate the dynamic behavior of the disease, we define the transmission rate as a combination of static and dynamic parts [54], in which the dynamic part is a periodic function of time t. The transmission rate β is consequently given by2 β=β0[1+β1cos(Ωt+ϕ))], where, Ω=2πT, β0 and β1 are amplitudes of the static and dynamic transmission rates, respectively, T is the time period and ϕ is the phase difference between static and dynamic phases. The initial infections take some time to accelerate and transmit to the masses; this delay between static and dynamic phases is captured by ϕ. Note that, in reality, we witnessed the dynamic behavior of COVID-19 with several combinations of harmonics for different countries around the world. In this paper, however, we approximate the transmission rate to the first harmonic for simplicity. Through this rather simple abstraction, we seek to explain the various reasons that affect the dynamic behavior of the disease. Our future work will perhaps consider a Fourier series-based model with higher harmonics. Nonlinear analysis This section focuses on the analysis of extinction or persistence of the disease which is determined by the stability of the disease-free equilibrium and the existence of endemic equilibrium. Investigation of the dynamic behavior and vaccination-based control of COVID-19 epidemics require a sound understanding of the mode of dissemination as well as the impacts of control strategies. In order to investigate the dynamic behavior, we need to consider various factors that affect it including perfect and imperfect vaccination. Although in this work we do not include a quantitative comparison of results to actual observations, the parameters, birth rate (Λ), death rate (μ) (per year: Λ=μ=0.0014, per day: Λ=μ=0.0014/365), vaccination rate (ψ) (per year: ψ=0.64, per day: μ=64/365), transmission rate (β0=1.6286, β1=0.8143) are chosen for a start basing them on realistic and approximate parameter values of the current COVID-19 epidemic in the U.S. [55]. The data of COVID-19 is used for high-level guidance of mathematical models while focusing on commonalities of internal transmission between different countries. To make the analysis independent of specific and local aspects of the transmission in each country, we neglect the temporary and short-lasting variance in the data. Therefore, the values of various periods and phases are selected based on analyzing the data of major countries such as the USA, China, India, the UK, and Germany [55]. In order to generalize the discussion, the statistical information of the most frequent/dominant trends across all these countries is extracted and quantified by numerical values of the phase and period. In fact, these values can be considered as principal values that drive the most frequent/dominant trend of COVID-19. Disease-free equilibrium We normalize susceptible, infected and immune in Eq. (1) such that x=XN, y=YN and z=ZN. The rescaled equations are as follows.3 x′=Λ-βxy-(μ+ψ)x+χγy,y′=βxy+βδzy-(μ+γ)y,z′=ψx-βδzy+(1-χ)γy-μz. In order to determine the equilibrium points, we set the time derivatives to zero and solve the corresponding algebraic equations. The disease-free equilibrium is obtained by setting y=0 in Eq. (3) and is given by4 E0=(Λμ+ψ,0,ψΛμ(μ+ψ)). We compute the Jacobian at the disease-free equilibrium:5 J(E0)=-(μ+ψ)-βx0+χγ00βx0+βδz0-(μ+γ)0ψ-βδz0+(1-χ)γ-μ. The eigenvalues of the Jacobian are -μ, -(μ+ψ) and βs0+βδv0-(μ+γ). For disease-free equilibrium to be stable, βx0+βδz0-(μ+γ)=0. We then define the reproduction number considering perfect vaccination (δ=0) and imperfect vaccination (δ≠0) as follows.6 R(ψp)=βΛ(μ+γ)(μ+ψ),R(ψ)=βΛ(μ+δψ)μ(μ+γ)(μ+ψ). In the absence of the vaccination rate ψ=0, the reproduction number is given by7 R0=βΛμ(μ+γ). Note that the reproduction number is an all-important parameter and determines whether or not the disease is extinguished. In particular, the reproduction number needs to be less than 1 for the disease-free equilibrium to be stable; in other words, for a disease to be eradicated. From Eq. (6), one can see that the higher the vaccination rate, the smaller the reproduction number such that limψ→∞R(ψ)=δR0. Thus, if the vaccination efficiency (ϵ=1-δ) is not high, R(ψ) cannot be less than one. The eradication of disease with imperfect vaccination exists only if δR0<1, if the vaccine efficiency satisfies ϵ>(1-1R0). If δR0<1, then there exists a critical vaccination level ψ∗ such that R(ψ∗)=1. Using Eqs. (6) and (7), critical vaccination level for eradication of the disease is given by8 ψ∗=(R0-1)μ1-δR0. Fig. 2 Response of the system for a disease-free equilibrium situation for R(ψ)=0.74, μ=3.93×10-06, γ=0.2, χ=0.8, ψ=0.0313, δ=0.05, β0=1.6286, β1=0.8143, T=360, ϕ=2.4435 Figure 2 illustrates a disease-free equilibrium situation for a very high vaccination rate of ψ=0.0313 and for the parameters μ=3.93×10-06, γ=0.2, χ=0.8, δ=0.05, β0=1.6286, β1=0.8143, T=360, ϕ=2.4435. The reproduction number evaluated for this situation is R(ψ)=0.74. Therefore, with a very high vaccination rate and nearly perfect vaccine (δ=0.05),we can eradicate the disease completely. Endemic equilibrium To compute the expression of the equilibrium points for endemic equilibrium, the time derivatives in Eq. (3) are set to zero and the resulting equations are as follows.9 0=Λ-βxy-(μ+ψ)x+χγy,0=βxy+βδzy-(μ+γ)y,0=ψx-βδzy+(1-χ)γy-μz. Evaluating x from the first equation and z from the third equation of Eq. (9), yields10 x∗=χγy∗+Λβy∗+μ+ψ,z∗=(1-χ)γy∗+ψx∗βδy∗+μ. Substituting Eq. (10) in the second equation of Eq. (9) yields a quadratic equation in y as11 βχγy+Λβδy+μ+βδyγ1-χβy+μ+ψ+βδψχγy+Λ=(μ+γ)(βδy+μ)(βy+μ+ψ). Perfect vaccination (δ=0) In this case of perfect vaccination (δ=0), Eq. (9) has a unique solution. The equilibrium points in this case are as follows.12 x∗=Λμ+γRψp-χγRψp1-χγ+μ+Λβμ+ψ-Λβ,y∗=ΛRψp-1Rψp1-χγ+μ,z∗=(1-χ)γy∗+ψx∗βδy∗+μ. The corresponding Jacobian matrix computed at E∗=(x∗,y∗,z∗) is given by13 J(E∗)=a11a120βy∗a22βδy∗ψa32a33, witha11=-βy∗-(μ+ψ),a12=-βx∗+χγ,a22=βx∗+βδz∗-(μ+γ),a32=-βδz∗+(1-χ)γ,a33=-βδy∗-μ. The system is stable if the roots of the characteristic polynomial (∣J(E∗)-λI∣=0) are all negative. It can be noticed from the reproduction number (Rψp) of perfect vaccination that if the vaccination rate (ψ) is very high, the endemic equilibrium transfers into a disease-free equilibrium.Fig. 3 Response of the system approaching a disease-free equilibrium situation under perfect vaccination for R(ψp)=0.0341, μ=3.93×10-06, γ=0.2, χ=0.8, ψ=0.0017, δ=0, β0=1.6286, β1=0.8143, T=360, ϕ=2.4435 A situation approaching a disease-free equilibrium for the case of perfect vaccination and for the parameters ψ=0.0017, μ=3.93×10-06, γ=0.2, χ=0.8, δ=0, β0=1.6286, β1=0.8143, T=360, ϕ=2.4435 and R(ψp)=0.0341 is depicted in Fig. 3. This case demonstrates that for considerably high vaccination rate and if the vaccination is perfect, the situation approaches a disease-free equilibrium in which case we would be eradicating the disease completely. Imperfect vaccination (δ≠0) In this case of imperfect vaccination (δ≠0), we investigate the equilibrium solution of Eq. (11). We rewrite Eq. (11) as follows.14 ay2+by+c=0, witha=β2δμ,b=β2δΛ+χδγμ-χγμ+δμ2+δμψ+γμ+μ2β,c=-δΛψ-μ2β+γμ2+γμψ+μ3+μ2ψ. We eliminate β from the coefficients of Eq. (14) by substituting β=ηR(ψ). The resulting coefficients as a function of R(ψ) are15 a=R(ψ)2η2δμ,b=-R(ψ)2η2δΛ+((χγ+μ+ψ)δ+(1-χ)γ+μ)μR(ψ)η,c=(δψΛ+μ2)R(ψ)η+(γ+ψ+μ)μ2+γμψ, where, η=μμ+γμ+ψΛμ+δψ.Fig. 4 Saddle-node bifurcation showing infected versus reproduction number (R(ψ)) for the case of imperfect vaccination. In this case δ=0.1, μ=0.0014, γ=0.2, χ=0.8. SN indicates the saddle-node bifurcation point. Stable solutions: blue lines, unstable solutions: red lines We solve Eq. (14) using coefficients given in Eq. (15) and plot the equilibrium values versus the reproduction number R(ψ) in Fig. 4, which displays typical saddle-node bifurcation behavior. The figure shows infected versus reproduction number for parameters δ=0.1, μ=0.0014, γ=0.2, χ=0.8 and for different vaccination rates of ψ= 0.3, 0.64 and 0.9. It can be noticed from Fig. 4 that under imperfect vaccination, R(ψ) decreases until the saddle-node point and increases after that point with an increase in infections. Therefore, under imperfect vaccination, the stable disease-free equilibrium (blue lines in Fig. 4) transfers into an unstable endemic equilibrium (red lines in Fig. 4). This phenomenon is called backward bifurcation and occurs under imperfect vaccination. The fundamental cause for backward bifurcation is the fact that imperfect vaccination creates two classes of susceptible individuals with varying susceptibilities: the naive susceptible and the vaccinated susceptible [13]. Practically speaking, this means that if we vaccinate with a less effective vaccine, controlling the disease becomes more difficult due to the presence of backward bifurcation. We postulate that this is one of the main reasons for the dynamic oscillatory behavior of the COVID-19 epidemic. Fig. 5 Response of the system under imperfect vaccination for μ=3.93×10-06, γ=0.2, χ=0.8, ψ=0.0017, β0=1.6286, β1=0.8143, T=360, ϕ=2.4435. a δ=0.1, R(ψ)=1.5 b δ=0.5, R(ψ)=7.4 Figure 5 illustrates two different situations of imperfect vaccination (a) δ=0.1 and (b) δ=0.5 and for the parameters μ=3.93×10-06, γ=0.2, χ=0.8, ψ=0.0017, β0=1.6286, β1=0.8143, T=360, ϕ=2.4435. It can be noticed from Fig. 5 that imperfect vaccination leads to an endemic equilibrium ((a)R(ψ)=1.5 and (b) R(ψ)=7.4) which is one of the causes for the dynamic behavior. It is also observed that with the increase in δ from 0.1 to 0.5 the population of infected and susceptible classes increases due to backward bifurcation. Therefore, under imperfect vaccination, the eradication of disease becomes a very hard task since the endemic will persist for a very long time. In the previous sections, we investigated the effect of vaccination on the dynamic behavior of the disease. In the following sections, the effect of other factors namely, the frequency at which the disease transmits and the amplitude of transmission rate is investigated. Frequency and phase of transmission rate The frequency at which the disease is transmitted is reflected in the time period for which the disease persists as depicted in Fig. 6 for (a) Ω=0.0349 (T=180), ϕ=2.09 (b) Ω=0.0175 (T=360), ϕ=2.4435 and for the parameters μ=3.93×10-06, γ=0.2, χ=0.8, ψ=0.0017, β0=1.6286, β1=0.8143, δ=0.1. It is observed that when the frequency at which the disease transmits is more, the phase decreases and also the bandwidth of the peaks decreases as seen in Fig. 6a. On the other hand, with the decrease in transmission frequency, the phase and the bandwidth increase as shown in Fig. 6b. The phase here represents the transformation of the disease from the static to dynamic mode; the higher the phase, the longer it takes to transform from static to dynamic mode and vice versa. The bandwidth indicates the persistence of the disease for that time period; higher bandwidth indicates longer disease persistence.Fig. 6 Response of the system for μ=3.93×10-06, γ=0.2, χ=0.8, ψ=0.0017, β0=1.6286, β1=0.8143, δ=0.1 a Ω=0.0349 (T=180), ϕ=2.09 b Ω=0.0175 (T=360), ϕ=2.4435 Amplitude of the transmission rate The amplitude of the transmission rate plays a very important role in the spread of the pandemic. Figure 7 shows two situations for (a) β0=1.6286, β1=0.8143, R(ψ)=1.5 (b) β0=2.4428, β1=0.8143, R(ψ)=2.26 and for the parameters μ=3.93×10-06, γ=0.2, χ=0.8, ψ=0.0017, δ=0.1, T=360, ϕ=2.4435. It is evident from Fig. 7 that with the increase in amplitude of the transmission rate the peak of the infected population increases and so does the reproduction number (R(ψ)).Fig. 7 Response of the system for μ=3.93×10-06, γ=0.2, χ=0.8, ψ=0.0017, δ=0.1, T=360, ϕ=2.4435. a β0=1.6286, β1=0.8143, R(ψ)=1.5 b β0=2.4428, β1=0.8143, R(ψ)=2.26 Fig. 8 Infection peak versus dynamic transmission rate (β0β1) for a constant vaccine reduction coefficient δ=0.1 b constant vaccination rate ψ=0.0017 and for the parameters μ=3.93×10-06, γ=0.2, χ=0.8, T=180, ϕ=2.09 Figure 8 shows the variation of the peak of the infected population versus dynamic transmission rate (β0β1) for (a) constant vaccine reduction coefficient δ=0.1 (b) constant vaccination rate ψ=0.0017 and for the parameters μ=3.93×10-06, γ=0.2, χ=0.8, T=180, ϕ=2.09. It is observed from Fig. 8a and b that with the increase in β0β1 there is a sharp rise in the peak of the infected population until a certain value of β0β1. After that, the peak of the infected population saturates. The sharp rise of the peak is likely attributable to many reasons including delays in identifying and reporting cases, overlap of variants like delta and omicron, and ability of the variant to evade immunity conferred by past infection or vaccination (i.e., immune evasion). The lack of hospital bed capacity, as well as the weather circumstances at different times depending on the area, all contribute to the increase in the peak. Some other factors include bigger population sizes and border closures in highly infected urban areas, and sudden spread of infection from urban to rural areas. It follows that reduction of the static and dynamic transmission rates achieved through public health measures like wearing face masks, limiting public gatherings, more rigorous testing, and boosting vaccination efforts would clearly help to flatten the peaks of the infected population. Summary of findings We summarize below the key findings of our analysis.In this paper, we defined a modified reproduction number for the Susceptible-Infected-Immune (SII) model in terms of β,μ, Λ, γ, ψ and δ. This definition embeds the importance of vaccination seamlessly into the epidemiological model and significantly expands insight for controlling the disease. There exists a critical minimum vaccination level for complete eradication of the disease. The first harmonic approximation of transmission rate incorporated in the model explains the dynamic behavior of the disease adequately. Imperfections in vaccination have a high impact on the persistence of the disease. The persistence of the disease also depends on the frequency and phase of the transmission rate. The peak values of the infected population depend on the frequency, phase and amplitude of the transmission rate. Analysis of these effects on peak values significantly expands the insight into the severity of the pandemic. The model predicts that the peak of the infected population saturates more quickly if the population is vaccinated with near perfect vaccine even when the transmission rate increases constantly. As we know, public health measures and governmental actions would reduce the transmission rate (something we did not model in this paper but has been widely documented). Hence, it follows that our analysis leads to the conclusion that a suitable combination of vaccination, public behavior and governmental actions would reduce the oscillations and hence, effectively and completely stop a pandemic such as COVID-19. Conclusion In this paper, we adapted and developed a Susceptible-Infected-Immune model for the COVID-19 pandemic including vaccination and a harmonic transmission rate. First, we used available data from the USA as base parameters. The numerical analysis is performed using the base parameters to explain the dynamic behavior of the disease and its control through vaccination. Equilibrium and stability analysis were performed revealing several situations for disease-free equilibrium and endemic equilibrium under perfect and imperfect vaccination. The effect of frequency, phase, and amplitude of the transmission rate on the dynamic behavior of the disease is investigated. We are able to show the fact that imperfect vaccination leads to a saddle-node bifurcation where the stable disease-free equilibrium transfers into an unstable endemic equilibrium. This phenomenon is called backward bifurcation and occurs under imperfect vaccination. The main reason is the fact that imperfect vaccination creates two subclasses of susceptible individuals with different susceptibilities: the naive susceptible and the vaccinated susceptible. The resulting backward bifurcation lends support to our argument that the practical task of combating COVID-19 becomes much harder if the vaccination is imperfect. This is one of the principal reasons for the persistent dynamic behavior of the COVID-19 epidemic. The paper also demonstrated that the frequency, phase and amplitude of the transmission rate greatly affect the dynamic behavior of the disease. The persistence of the disease depends on the frequency and phase of transmission rate, while the peak of the infected population depends on the amplitude of the transmission rate. These findings confirm that the dynamic behavior of an endemic disease such as COVID-19 is driven by variables such as imperfect vaccination, environmental factors, and new variants of the virus. Social and biological elements, such as population density, social distancing measures, and government initiatives also have a role in the dynamic behavior of the disease. Finally and importantly, our analysis suggests that increasing the number of vaccinations, improving the effectiveness of vaccination, and implementing public measures that reduce static and dynamic transmission rates would suppress oscillatory behavior and help to eradicate the disease completely. Acknowledgements We gratefully acknowledge the financial support from US Office of Naval Research (Grant: N00014-19-1-2070) for basic research on adaptive modeling of nonlinear dynamic systems. In particular, we appreciate the continuous encouragement from Capt. Lynn Petersen and we are humbled by his recognition of the value of our research. Data availability All data used in the study are openly available through Ref. [55]. Declarations Conflict of interest The authors declare that they 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. World Health Organization (WHO): WHO director-general’s opening remarks at the media briefing on COVID-19, March (2020) 2. D’Amico F Marmiere M Righetti B Scquizzato T Zangrillo A Puglisi R Landoni G COVID-19 seasonality in temperate countries Environ. Res. 2022 206 112614 10.1016/j.envres.2021.112614 34953888 3. Bergman A Sella Y Agre P Casadevall A Oscillations in us COVID-19 incidence and mortality data reflect diagnostic and reporting factors Msystems 2020 5 4 e00544-20 10.1128/mSystems.00544-20 32665331 4. Greer M Saha R Gogliettino A Yu C Zollo-Venecek K Emergence of oscillations in a simple epidemic model with demographic data Royal Soc. Open Sci. 2020 7 1 191187 10.1098/rsos.191187 5. Pavlíček T Rehak P Král P Oscillatory dynamics in infectivity and death rates of COVID-19 Msystems 2020 5 4 e00700-20 10.1128/mSystems.00700-20 32817387 6. Bernoulli, D.: Essai d’une nouvelle analyse de la mortalite causee par la petite verole. Mem. Math. Phys. Acad. Roy. Sci. pp 1–45 (1766) 7. Dietz K Heesterbeek J Daniel Bernoulli’s epidemiological model revisited Math. Biosci. 2002 180 1 1 21 10.1016/S0025-5564(02)00122-0 12387913 8. Kermack WO McKendrick AG Walker GT A contribution to the mathematical theory of epidemics Proc. Royal Soc. London Series A Contain. Papers Math. Phys. Char. 1927 115 772 700 721 9. Kermack WO McKendrick AG Contributions to the mathematical theory of epidemics-II. The problem of endemicity Bull. Math. Biol. 1932 53 57 87 10. Keeling MJ Rohani P Modeling Infectious Diseases in Humans and Animals 2008 Princeton Princeton University Press 11. Brauer F Mathematical epidemiology: past, present, and future Infect. Dis. Model. 2017 2 2 113 127 29928732 12. Brauer F Castillo-Chavez C Feng Z Mathematical Models in Epidemiology 2019 New York Springer 13. Martcheva M An Introduction to Mathematical Epidemiology 2015 1 US Springer 14. Vynnycky E White RG An Introduction to Infectious Disease Modelling 2010 Oxford Oxford University Press 15. Hethcote HW The mathematics of infectious diseases SIAM Rev. 2000 42 4 599 653 10.1137/S0036144500371907 16. Olsen LF Truty GL Schaffer WM Oscillations and chaos in epidemics: a nonlinear dynamic study of six childhood diseases in Copenhagen, Denmark Theor. Popul. Biol. 1988 33 344 70 10.1016/0040-5809(88)90019-6 3266037 17. Earn DJ Rohani P Grenfell BT Persistence, chaos and synchrony in ecology and epidemiology Proc. Royal Soc. London. Series B: Biol. Sci. 1998 265 7 10 10.1098/rspb.1998.0256 18. Grossman Z Gumowski I Dietz K The incidence of infectious diseases under the influence of seasonal fluctuations—analytical approach Nonlinear Systems and Applications 1977 Cambridge Academic Press 19. Hethcote HW Stech HW Van Den Driessche P Nonlinear oscillations in epidemic models SIAM J. Appl. Math. 1981 40 1 1 9 10.1137/0140001 20. Hethcote HW van den Driessche P Some epidemiological models with nonlinear incidence J. Math. Biol. 1991 29 3 271 287 10.1007/BF00160539 2061695 21. Bauch CT Earn DJD Transients and attractors in epidemics Proc. Royal Soc. London. Series B: Biol. Sci. 2003 270 12908977 1573 1578 10.1098/rspb.2003.2410 22. Abta A Laarabi H Talibi Alaoui H The Hopf bifurcation analysis and optimal control of a delayed SIR epidemic model Int. J. Anal. 2014 2014 940819 23. Schwartz IB Smith HL Infinite subharmonic bifurcation in an SEIR epidemic model J. Math. Biol. 1983 18 3 233 253 10.1007/BF00276090 6663207 24. Yan C Jia J Hopf bifurcation of a delayed epidemic model with information variable and limited medical resources Abstract Appl. Anal. 2014 2014 109372 10.1155/2014/109372 25. Allen LJ Brauer F Van den Driessche P Wu J Mathematical Epidemiology 2008 Berlin Springer 26. Zhang X Shan C Jin Z Zhu H Complex dynamics of epidemic models on adaptive networks J. Differ. Equ. 2019 266 1 803 832 10.1016/j.jde.2018.07.054 27. Bartlett M Stochastic Population Models in Ecology and Epidemiology 1960 London Methuen 28. Anderson RM May RM Infectious Diseases of Humans: Dynamics and Control 1992 Oxford Oxford University Press 29. Kuperman M Abramson G Small world effect in an epidemiological model Phys. Rev. Lett. 2001 86 13 2909 10.1103/PhysRevLett.86.2909 11290070 30. Newman ME The structure and function of complex networks SIAM Rev. 2003 45 2 167 256 10.1137/S003614450342480 31. Keeling MJ Eames KT Networks and epidemic models J. R. Soc. Interface 2005 2 4 295 307 10.1098/rsif.2005.0051 16849187 32. Kattis AA Holiday A Stoica A-A Kevrekidis IG Modeling epidemics on adaptively evolving networks: a data-mining perspective Virulence 2016 7 2 153 162 10.1080/21505594.2015.1121357 26695450 33. Ahmed E Agiza H Hassan S On modeling hepatitis b transmission using cellular automata J. Stat. Phys. 1998 92 3 707 712 10.1023/A:1023001007714 34. Doran RJ Laffan SW Simulating the spatial dynamics of foot and mouth disease outbreaks in feral pigs and livestock in Queensland, Australia, using a susceptible-infected-recovered cellular automata model Prev. Vet. Med. 2005 70 1–2 133 152 10.1016/j.prevetmed.2005.03.002 15967247 35. Silva H Monteiro L Self-sustained oscillations in epidemic models with infective immigrants Ecol. Complex. 2014 17 40 45 10.1016/j.ecocom.2013.08.002 36. Chaves L Monteiro L Oscillations in an epidemiological model based on asynchronous probabilistic cellular automaton Ecol. Complex. 2017 31 57 63 10.1016/j.ecocom.2017.03.001 37. Buonomo B Chitnis N D’Onofrio A Seasonality in epidemic models: a literature review Ricerche Mat. 2018 1 7 25 10.1007/s11587-017-0348-6 38. Zamir M Shah Z Nadeem F Memood A Alrabaiah H Kumam P Non pharmaceutical interventions for optimal control of COVID-19 Comput. Methods Programs Biomed. 2020 196 105642 10.1016/j.cmpb.2020.105642 32688137 39. Tsay C Lejarza F Stadtherr MA Baldea M Modeling, state estimation, and optimal control for the us COVID-19 outbreak Sci. Rep. 2020 10 1 1 12 10.1038/s41598-020-67459-8 31913322 40. Perkins TA España G Optimal control of the COVID-19 pandemic with non-pharmaceutical interventions Bull. Math. Biol. 2020 82 9 1 24 10.1007/s11538-020-00795-y 41. Kwuimy C Nazari F Jiao X Rohani P Nataraj C Nonlinear dynamic analysis of an epidemiological model for COVID-19 including public behavior and government action Nonlinear Dyn. 2020 101 3 1545 1559 10.1007/s11071-020-05815-z 32836814 42. Rohith G Devika K Dynamics and control of COVID-19 pandemic with nonlinear incidence rates Nonlinear Dyn. 2020 101 3 2013 2026 10.1007/s11071-020-05774-5 32836807 43. Huang J Qi G Effects of control measures on the dynamics of COVID-19 and double-peak behavior in Spain Nonlinear Dyn. 2020 101 3 1889 1899 10.1007/s11071-020-05901-2 32874016 44. Nazarimehr F Pham V-T Kapitaniak T Prediction of bifurcations by varying critical parameters of COVID-19 Nonlinear Dyn. 2020 101 3 1681 1692 10.1007/s11071-020-05749-6 32836801 45. He S Peng Y Sun K Seir modeling of the COVID-19 and its dynamics Nonlinear Dyn. 2020 101 3 1667 1680 10.1007/s11071-020-05743-y 32836803 46. Guidry JP Laestadius LI Vraga EK Miller CA Perrin PB Burton CW Ryan M Fuemmeler BF Carlyle KE Willingness to get the COVID-19 vaccine with and without emergency use authorization Am. J. Infect. Control 2021 49 2 137 142 10.1016/j.ajic.2020.11.018 33227323 47. Lazarus JV Ratzan SC Palayew A Gostin LO Larson HJ Rabin K Kimball S El-Mohandes A A global survey of potential acceptance of a COVID-19 vaccine Nat. Med. 2021 27 2 225 228 10.1038/s41591-020-1124-9 33082575 48. Kopfová J Nábělková P Rachinskii D Rouf SC Dynamics of sir model with vaccination and heterogeneous behavioral response of individuals modeled by the preisach operator J. Math. Biol. 2021 83 2 1 34 10.1007/s00285-021-01629-8 34129100 49. Chowell G Fenimore PW Castillo-Garsow MA Castillo-Chavez C Sars outbreaks in Ontario, Hong Kong and Singapore: the role of diagnosis and isolation as a control mechanism J. Theor. Biol. 2003 224 1 1 8 10.1016/S0022-5193(03)00228-5 12900200 50. Kilpatrick AM Chmura AA Gibbons DW Fleischer RC Marra PP Daszak P Predicting the global spread of H5N1 avian influenza Proc. Natl. Acad. Sci. 2006 103 51 19368 19373 10.1073/pnas.0609227103 17158217 51. Prosper O Saucedo O Thompson D Torres-Garcia G Wang X Castillo-Chavez C Modeling control strategies for concurrent epidemics of seasonal and pandemic H1N1 influenza Math. Biosci. Eng. 2011 8 1 141 10.3934/mbe.2011.8.141 21361405 52. Feng Z Zheng Y Hernandez-Ceron N Zhao H Glasser JW Hill AN Mathematical models of Ebola—consequences of underlying assumptions Math. Biosci. 2016 277 89 107 10.1016/j.mbs.2016.04.002 27130854 53. Martcheva M An evolutionary model of influenza a with drift and shift J. Biol. Dyn. 2012 6 2 299 332 10.1080/17513758.2011.573866 22873593 54. Tuncer N Martcheva M Modeling seasonality in avian influenza H5N1 J. Biol. Syst. 2013 21 04 1340004 10.1142/S0218339013400044 55. Dong E Du H Gardner L An interactive web-based dashboard to track COVID-19 in real time Lancet. Infect. Dis. 2020 20 5 533 534 10.1016/S1473-3099(20)30120-1 32087114	0	PMC9734520	NO-CC CODE	2022-12-14 23:28:29	no	Nonlinear Dyn. 2022 Dec 5;:1-13	utf-8	Nonlinear Dyn	2,022	10.1007/s11071-022-08125-8	oa_other
==== Front Biochem (Mosc) Suppl Ser A Membr Cell Biol Biochem (Mosc) Suppl Ser A Membr Cell Biol Biochemistry (Moscow) Supplement. Series A, Membrane and Cell Biology 1990-7478 1990-7494 Pleiades Publishing Moscow 5152 10.1134/S1990747822050038 Reviews Physico-Chemical Mechanisms of the Functioning of Membrane-Active Proteins of Enveloped Viruses Batishchev O. V. [email protected] grid.465278.a 0000 0004 0620 3386 Frumkin Institute of Physical Chemistry and Electrochemistry, Russian Academy of Sciences, 119071 Moscow, Russia 9 12 2022 2022 16 4 247260 4 5 2022 1 6 2022 2 6 2022 © Pleiades Publishing, Ltd. 2022, ISSN 1990-7478, Biochemistry (Moscow), Supplement Series A: Membrane and Cell Biology, 2022, Vol. 16, No. 4, pp. 247–260. © Pleiades Publishing, Ltd., 2022.Russian Text © The Author(s), 2022, published in Biologicheskie Membrany, 2022, Vol. 39, No. 5, pp. 321–336. 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. Over the past few years, the attention of the whole world has been riveted to the emergence of new dangerous strains of viruses, among which a special place is occupied by coronaviruses that have overcome the interspecies barrier in the past 20 years: SARS viruses (SARS), Middle East respiratory syndrome (MERS), as well as a new coronavirus infection (SARS-CoV-2), which caused the largest pandemic since the Spanish flu in 1918. Coronaviruses are members of a class of enveloped viruses that have a lipoprotein envelope. This class also includes such serious pathogens as human immunodeficiency virus (HIV), hepatitis, Ebola virus, influenza, etc. Despite significant differences in the clinical picture of the course of disease caused by enveloped viruses, they themselves have a number of characteristic features, which determine their commonality. Regardless of the way of penetration into the cell—by endocytosis or direct fusion with the cell membrane—enveloped viruses are characterized by the following stages of interaction with the target cell: binding to receptors on the cell surface, interaction of the surface glycoproteins of the virus with the membrane structures of the infected cell, fusion of the lipid envelope of the virion with plasma or endosomal membrane, destruction of the protein capsid and its dissociation from the viral nucleoprotein. Subsequently, within the infected cell, the newly synthesized viral proteins must self-assemble on various membrane structures to form a progeny virion. Thus, both the initial stages of viral infection and the assembly and release of new viral particles are associated with the activity of viral proteins in relation to the cell membrane and its organelles. This review is devoted to the analysis of physicochemical mechanisms of functioning of the main structural proteins of a number of enveloped viruses in order to identify possible strategies for the membrane activity of such proteins at various stages of viral infection of the cell. Keywords: enveloped viruses protein–lipid interactions membrane fusion SARS-CoV-2 coronavirus human immunodeficiency virus (HIV) influenza A virus non-structural proteins of coronaviruses matrix proteins capsid proteins issue-copyright-statement© Pleiades Publishing, Ltd. 2022 ==== Body pmcINTRODUCTION Most processes in cell membranes are mediated by specific proteins. Protein–lipid interactions in the cell regulate membrane topological rearrangements, such as fusion and division of cells and their organelles, endo- and exocytosis, which accompanied by the formation, budding and transport of membrane vesicles. These interactions require the coordinated work of a large complex of proteins that determines the complexity of their study and experimental modeling [1‒4]. Although the membrane fusion requires cooperative protein–lipid interactions, the mechanical properties of the lipid matrix surrounding membrane proteins determine the energy of the formation of intermediate structures through which membrane fusion is realized [5]. Experiments conducted by L.V. Chernomordik and G.B. Melikyan on flat lipid bilayers in the Laboratory of Bioelectrochemistry of the Institute of Electrochemistry of the Academy of Sciences of the USSR, headed by Y.A. Chizmadzhev, have demonstrated that the fusion process consists of the following sequential stages: (1) establishment of the tight contact between membranes, (2) formation of the stalk (or bridge) between adjacent monolayers, (3) expansion of the stalk, leading to the formation of a trilamellar structure (or hemifusion diaphragm) and, finally, (4) its rupture, indicating the end of the fusion [6]. It is shown that the rate of these stages essentially depends on the spontaneous curvature of the lipid monolayers. Thus, the negative spontaneous curvature of the contacting monolayers (a typical example is dioleoylphosphatidylethanolamine) promotes stalk formation, while the rupture of the hemifusion diaphragm is accelerated in the presence of positive spontaneous curvature (i.e., lysolipids) in the distal monolayers. The proof of this fusion scheme has been obtained by V.S. Markin and M.M. Kozlov, members of the Laboratory of Bioelectrochemistry [7]. The theory quantitatively describes the entire set of experiments conducted at that time. A detailed description and relevant literature can be found in the review [8]. Further studies of the membrane fusion required the consideration of the membrane protein component. The first works in this direction have been aimed at studying synaptic fusion and the action of SNARE complex proteins [9, 10]. However, the large number of proteins included in this complex makes it difficult to clarify the physical-chemical mechanisms of their function and to establish the energy of the intermediate stages. For this reason, the study of virus-induced fusion, which in a number of systems is mediated by a single protein of known structure (e.g., hemagglutinin (HA) in the case of influenza virus) has come to the fore [11]. Several HA molecules are thought to form a fusion rosette and then lipid protrusions (dimples). On their tops, after merging, monolayer junctions (stalks) are formed, which lead to the monolayer (and subsequently, to the complete) fusion. In various model systems based on the hemagglutinin-expressing cells [12], it was shown that the effect of the lipid composition of the target membrane on fusion is well described within the stalk theory. In the same works, it was shown that the role of HAs is the formation of a complex, a “rosette” of fusion proteins, within which cell membranes merge to a distance sufficient for the subsequent spontaneous membrane fusion. Thus, the fusion theory assumes that the action of hemagglutinin molecules is necessary for the local merging of lipid bilayers of cell membranes, which fuse spontaneously similar to two planar bilayer lipid membranes (BLM). However, experimental observations obtained in cellular systems, such as the absence of lipid exchange through small fusion pores, are poorly described within the proposed model. Moreover, recent electron tomography data on influenza virus fusion with lipid vesicles show the formation of structures with only the target membrane bending, while the viral membrane remains nearly undeformed [13]. These facts lead to the conclusion that the classical stalk theory, which well describes fusion in the case of compositionally symmetric and purely lipid membranes, may not be applicable to the protein-mediated fusion. In enveloped viruses, a shell of capsid or matrix proteins is located under the outer lipid membrane. These proteins constitute the majority of viral proteins. Interacting with the viral genome and surface glycoproteins, they play an important role in the assembly of viral particles and the production of progeny virions in the infected cell [14–16]. Matrix proteins are multifunctional: on the one hand, they maintain the integrity of the virus, on the other, they disassemble to release genetic material into the cytoplasm during infection and assemble to form new virions. In all these processes, matrix proteins interact with lipid membranes—the virus outer envelope or the cell plasma membrane. This interaction determines the life activity of the virus, although its molecular mechanisms remain an open question. In particular, the influenza A virus penetrates the cell by endocytosis, entering the cellular endosome as a result. The main trigger for the fusion of the viral particle and the endosomal membrane is the change of the pH of the endosomal milieu to 4–5, leading to conformational transitions in the HA and to the destruction of the capsid formed by the matrix protein M1. The drop of pH inside the virus occurs due to the action of M2 proton channels located in the viral envelope. In experiments conducted under the guidance of Yu.A. Chizmadzhev it was shown that blocking M2 channels with amantadine leads to “freezing” of viral fusion at the stage of a small fusion pore with a dimeter of about 1 nm, so that the genetic material of the virus is not released into the cell cytoplasm [17]. Thus, on the one hand, the disintegration of the envelope of matrix proteins is a critical step in the viral infection. On the other hand, matrix proteins must assemble on plasma membrane of the infected cell into the structure of a new virion. The process of assembly and budding of the progeny virion implies local deformation of cell membrane, in which matrix or capsid proteins should also play an important role. Nevertheless, the specific physico-chemical mechanisms defining the self-assembly of the viral capsid and changes in the cell membrane topology often remain unclear. VIRUS-INDUCED MEMBRANE FUSION The envelope virus genome and capsid (nucleocapsid) proteins are surrounded by an additional envelope, a bilayer lipid membrane that is acquired from the host cell during virion release from the infected cell [18] (Fig. 1). Enveloped viruses use membrane fusion to enter the cell [19]. For this process, viruses use the function of fusion proteins. Conformational rearrangements of these proteins initiated by a certain trigger (e.g., binding to a receptor on the cell surface or a change of pH inside the cell endosome) initiate the fusion between the virus lipid membrane and cell membrane [20]. These proteins do not require energy-supplying molecules for activity [21]. Fig. 1. Structure of enveloped viruses. (a) Coronavirus SARS-CoV-2; (b) influenza A virus; (c) Newcastle disease virus; (d) Human immunodeficiency virus. Abbreviations: HA, hemagglutinin; NA, neuraminidase; RNP, ribonucleoprotein; M1, M1 protein; M2, M2 proton channel. The entry of any virus into the cell begins with its binding to certain receptor molecules on the cell surface [22] (Fig. 2). Membrane fusion is the most important stage in the life cycle of enveloped viruses. Initially, the genetic material of the virus is separated from the cell cytoplasm by two membranes: the viral and the cell. As a result of the fusion of these membranes, the cell cytoplasm and the internal space of the virion containing the viral genetic material are combined. Some enveloped viruses fuse directly with the plasma membrane, while others enter the cell through endocytosis. Fig. 2. Examples of life cycles of enveloped viruses. (a) Influenza A virus; (b) Newcastle disease virus; (c) Human immunodeficiency virus. Membrane fusion is an important part of the cell life activity, mediating many vital processes. The most representative example of such a process is the synaptic transmission, during which synaptic vesicles containing neurotransmitters fuse with the presynaptic plasma membrane [23]. In addition, organelles also merge during cell fusion, for example, in the processes of fertilization, carcinogenesis, etc. [24–27]. Certain fusion proteins are involved in these processes [28, 29]. For example, during hormone and neurotransmitter secretion, the SNARE protein complex catalyzes membrane fusion [9, 10]. Intercellular fusion is also provided by specific proteins, which, however, are detected only in some cases, but for most remain unknown [30]. Fusion proteins are responsible for the deformation of interacting membranes, followed by their fusion and mixing of initially separated aqueous volumes. The function of these proteins in the fusion process in each case determines their structure, number, presence of auxiliary proteins, etc. As usual, viral membrane fusion is performed by one or two proteins; in particular, the influenza A virus has only one fusion protein—hemagglutinin [31]. Membrane fusion, as any other membrane topological rearrangements, requires energy for the mechanical deformations of the lipid matrix. Thus, protein–lipid interactions determine the energy of viral-induced membrane fusion and, therefore, regulate this stage of the infection process. Theoretical estimates show that the characteristic height of the fusion energy barriers is several tens to hundreds of kBT, and such barriers cannot be overcome only by thermal fluctuations [32]. The mechanical energy reserved in the fusion proteins covers the energy costs. Many experimental data show that membrane fusion can occur spontaneously if substances connecting the two interacting membranes (e.g., Ca2+) or dehydrating their contact (e.g., polyethylene glycol) have been added between the fusing membranes [33, 34]. These data suggest that bringing two membranes into a close contact and overcoming hydration repulsion are the most energy-consuming phases of fusion, which cannot occur spontaneously due to the energy of thermal fluctuations of lipids. As further fusion in such model systems occurs without the presence of any proteins, it can be concluded that the energy barriers on the rest of the fusion trajectory should not be measurably affected. Thus, the function of fusion proteins is largely similar to enzymes: they can reduce the basic energy activation barrier for fusion and possibly accelerate the rate constant of the reaction. Type I viral fusion proteins (e.g., influenza A virus HA and HIV Env gp120/gp41 protein) are the most well studied [21]. At the beginning of the infection viral fusion proteins bind specific receptor molecules on the cell surface. Immediately thereafter (or when the pH inside the endosome changes), the fusion protein attacks the cell membrane with a fusion peptide, which is a specific N-terminus consisting of about 20 amino acids [35]. The fusion peptide releases from the hydrophobic pocket of the ectodomain of HA and then incorporates into the lipid bilayer of the target membrane as an effective anchor. In the process of ongoing conformational rearrangements, the fusion protein is refolded, attaching the target cell membrane with the embedded fusion peptide to the viral membrane [36]. These membranes come into close contact, merge, and form a fusion pore, through which the viral genome can be released into the cytoplasm. At this stage, the fusion protein folds into a “post-fusion” conformation when its transmembrane domain contacts the fusion peptide [31]. If the fusion peptide does not attain the target membrane, it can integrate into the viral membrane next to the transmembrane domain of the fusion protein, also bringing it into the post-fusion conformation. The height of the fusion protein in this state is about 10 nm [37]. Fusion peptides of viral proteins are generally amphipathic [38, 39]. They can partially integrate into the cell membrane, providing both an anchor and a lever for the forces and moments, as well as deform the target membrane locally. The depth of the fusion peptide integration regulates its fusion activity [39]. However, replacement of the transmembrane domain of the fusion protein with a lipid anchor prevents the extension of the fusion pore [40, 41]. Hence, both the fusion peptide and the transmembrane domain of the respective protein are necessary for the efficient fusion. In fact, fusion proteins form protrusions on the viral and the cell contact membranes. The intense hydration repulsion acting on the tops of these protrusions leads to the lateral shift of the polar lipid heads from the region of a tight contact [42–44]. With a certain probability, these local protrusions of the opposite membranes can touch, forming a stalk [44]. A number of theoretical models have shown that the main energy barrier for the interaction of membrane protrusions induced by fusion proteins strongly depends on the distance between these membranes [32, 45]. This distance is defined by the thickness of the layer of fusion proteins in the post-fusion conformation, whose fusion peptides have not reached the target cell membrane. For hemagglutinin, this thickness is about 10 nm, as shown by electron microscopy [37]. Thus, fusion proteins must locally overcome this distance for bringing the fusion membranes closer over a small contact area, providing a relatively low energy barrier for hemifusion. According to various estimates, the energy required for such membrane rearrangement is on the order of several tens of kBT [32, 45, 46]. For comparison, the energy stored in a single hemagglutinin trimer is about 60 kBT, i.e., 20 kBT per hemagglutinin monomer [47]. Roughly, the same energy estimation is also obtained for the HIV gp41 trimer [48]. This indicates that effective membrane fusion requires the cooperative action of several fusion proteins that form the fusion rosette. In the case of influenza A virus, the required number of HA trimers in the fusion rosette is estimated to be between 3 and 9 [49, 50]. However, in the case of HIV, it is known that one trimer of the gp41 protein may be sufficient for the fusion process [50]. A specific feature of type I fusion proteins is the intensive deformation of the target membrane. The viral membrane is deformed less due to the presence of a capsid protein envelope in contact with it [37]. The deformation of the target membrane is caused by the integration of fusion peptides into it. Such interactions have been considered in numerous theoretical works [51–53]. In particular, it is shown that the study of the cooperative fusion rosette formation should consider the dependence of the elastic energy on the mutual orientation of fusion peptides rather than on the distance between them, as positions of viral fusion proteins are rather strictly fixed because of their transmembrane domains that interact with the viral capsid. The fusion peptides released by viral fusion proteins are also anchors, through which fusion proteins can exert mechanical forces and moments on the target membrane [11, 54]. Thus, lipid matrix protrusions required for membrane fusion can be formed either by elastic deformations of embedded fusion peptides or by direct application of forces of fusion proteins, which tighten the two fusing membranes [55, 56]. The shallow integration of amphipathic fusion peptides into the target membrane is similar to the formation of positive spontaneous curvature, which is induced, for example, by lysolipids, and leads to pore formation [57], but not to stalk. Thus, the physico-chemical mechanism of amphipathic peptide functioning for fusion proteins cannot be based only on the modification of the target membrane’s contact monolayer. Apparently, it necessarily requires the application of constricting forces to the two membranes. To decrease the energy barrier of stalk formation, fusion peptides must effectively induce negative spontaneous curvature [57]. For this purpose, their insertion depth into the contact monolayer of the target membrane should be increased. Calculations made in [58, 59], which explicitly take into account the depth of fusion peptide insertion, fully support this view. Moreover, there is an experimental evidence for the HIV fusion protein peptide that its depth of integration into the contact monolayer of the target membrane increases the probability of fusion [39]. In the case of HIV, fusion can be catalyzed by several Env proteins down to a single protein trimer [50]. The key difference is that in addition to fusion, the Env proteins are also responsible for the virus reception, i.e., its binding to the cellular plasma membrane [60]. The gp120 protein recognizes the CD4 receptor and the coreceptor CXCR4 or CCR5, which are large transmembrane proteins. Binding to the receptor triggers a conformational rearrangement of the Env glycoprotein. This type of viral fusion trigger additionally limits the number of simultaneously activated gp120/gp41 trimers due to the relatively low surface density of CD4 molecules in the plasma membrane of T-lymphocytes. This differs from the case of influenza A virus, whose hemagglutinin activation is caused by a decrease of the pH in the late endosome, i.e., all virion hemagglutinin trimers are activated almost simultaneously. Two possible mechanisms of action of fusion proteins have been commonly proposed. The first one is based on the assumption that integrated fusion peptides modify the elastic properties of the target membrane, especially its spontaneous curvature [61]. It is assumed that the modified spontaneous curvature in the ring zone of the fusion rosette may be the trigger for the formation of a protrusion in the target membrane. The membranes come into close contact at the top of the protrusion, which significantly facilitates the fusion. According to the second mechanism, rather than modifying the target membrane, fusion proteins induce bending moments, resulting in the formation of protrusions. Proteins also generate forces by directly and mechanically bringing two tightly attached membranes into close contact [31]. The analysis in [62] showed that it is the second mechanism that provides the formation of a highly symmetric fusion rosette by organizing the cooperativity of the mechanical efforts of several fusion proteins. In contrast, the first mechanism uses rosette symmetry to explain protrusions on the fusion membranes but cannot explain the nature of such symmetry. Therefore, the directed mechanical activity of fusion proteins controls the entire fusion process. This activity may be the result of the cooperative effect of various fusion protein subunits, including those sites involved in the interaction with the receptor. This may explain the observed difference in the cooperative action of influenza A virus and HIV fusion proteins. Therefore, the development of elastic models of the membrane fusion has provided answers to many questions about the mechanisms of viral-induced membrane fusion, in particular, about possible triggers of the cooperative effects of viral fusion proteins. CAPSID PROTEINS OF ENVELOPED VIRUSES AND FORMATION OF PROGENY VIRIONS Under the outer lipid membrane of enveloped viruses, there is a capsid of matrix proteins. Interacting with the viral genome and surface glycoproteins, they play an important role in the assembly of viral particles and the production of progeny virions in the infected cell [14–16]. Matrix proteins are multifunctional at different stages of the virus life cycle. On the one hand, they maintain virion integrity and architecture; on the other hand, they perform a controlled disintegration of the capsid to release viral genetic material into the cytoplasm; on the third—organize the assembly of progeny virions. In all these processes, matrix proteins interact with the lipid membranes of the virus or cell. This interaction is a determinant in the virus life activity, although the molecular mechanisms are still unknown. The nature of such protein–lipid interactions, as well as their relation to the capsid self-assembly, is debated [63–67]. Despite the diversity of envelope viruses, their matrix proteins are organized in a similar way: they are either helical structures (as in the case of influenza A virus [14, 68, 69], vesicular stomatitis virus [15] and measles virus [16]) or two-dimensional lattices forming perispherical particles (for example, Newcastle disease virus [70] or HIV [71]). Evolutionarily and structurally matrix proteins of different virus families are very similar, such as the polyprotein Gag of the HIV and the M1 protein of the influenza A virus [72]. Such proximity should also indicate common mechanisms of capsid formation and virion budding from the surface of the infected cell. The relationship between the three-dimensional structure and functional properties of a protein is one of the most important basic concepts in biochemistry. To perform a particular function, a protein should have a specific amino acid sequence and be folded in a certain way [15]. This idea, however, has been reconsidered about two decades ago when many proteins with a partially or completely disordered structure have been found. These proteins, called intrinsically disordered proteins, exist as dynamic ensembles of conformations that do not have a stable folded structure, but nevertheless perform their (often very variable) biological functions [73]. Disordered domains of proteins can promote the rapid response of viruses to changing environmental conditions and, consequently, their survival and development by implementing mechanisms of adaptation and protection [74]. Disorder in protein folding prevents antibody binding that reduces the immune response. This is the key to the extraordinary ability of HIV to evade the immune response, with main contribution of the viral capsid structural features [74]. The process of viral budding is generally similar to the cellular exocytosis in the way of formation a curved structure from the cell membrane. A minimal set of proteins in the viral envelope reduces possible ways the virus can affect the membrane. There are three possible ways of such effect, which are occurring in different viruses separately or in complex combinations. The first mechanism consists in the formation of a virion due to the certain molecular geometry of matrix proteins. It allows proteins to bind the lipid bilayer and form a two-dimensional lattice on the lipid surface. X-ray analysis demonstrates the existence of such mechanism for the Newcastle disease virus [70] and Salmonella infectious anemia virus [75]. In both cases, protein monomers bind in a capsid structure with a certain angle to each other, forming hemispherical viral particles. A nucleation of such particles occurs by the formation of circle protein–lipid domains [76] followed by curving of the lipid membrane. The liquid state of the lipid bilayer implies that transfer of the curvature to the lipid membrane requires the layer of matrix proteins to minimize lipid redistribution in the areas of protein–lipid contacts. Apparently, this should be achieved by hydrophobic interactions, which is the case of the Newcastle disease virus [77]. This mechanism of interaction has been suggested in the early studies of the interaction of the influenza A virus matrix protein with the lipid bilayer [78–81]. However, it has not been proved in later studies [63], which show neither partial integration of the M1 protein into the lipid bilayer, nor its ability to force a certain curvature on the membrane. These studies suggest the predominant electrostatic nature of the interaction of matrix protein M1 with the membrane. The second possible mechanism is the condensation of lipids under the protein layer due to electrostatic interactions. Such condensation on the inner monolayer of the plasma membrane creates a local imbalance of the membrane monolayer area, leading to the membrane bending and the formation of various outgrowths to relief the excess area [82]. In particular, a formation of filamentous structures close in diameter to viral particles is shown in experiments on negatively charged giant unilamellar vesicles during the adsorption of the influenza C virus matrix protein [83]. Electrostatic interactions with anionic lipids are common to matrix proteins of viruses such as influenza virus [63, 84], Ebola virus [85, 86], and vesicular stomatitis virus [87]. For all these viruses it has been shown that their matrix proteins are able to form virus-like particles [87–90]. At the same time, additional factors contributing to the change in membrane curvature play role in this mechanism of the viral budding. For example, it has been shown that a palmitoylation of the HA of the influenza A virus, that promotes the formation of the correct curvature of the viral particle, enhances the adsorption of protein M1 at the lipid membrane and facilitates the formation of virus-like particles [91]. For some enveloped viruses it was shown that their membrane is more resistant to detergents than the cell plasma membrane [92–96], as it has a raft structure [97]. The presence of lipid rafts in the viral membrane provides another mechanism of vesicle formation and budding: by increasing the linear tension of the raft boundary with their following release from the membrane surface [98]. However, the mechanism of this process has not yet been demonstrated for any enveloped virus. It is supposed that lipid domains can be formed by condensation of negatively charged lipids with fully saturated hydrocarbon chains on the inner monolayer of the plasma membrane, followed by the formation of a sphingomyelin-enriched lipid domain on the outer part of the membrane [99]. Enveloped viruses of different families have different pathways in the cell and differ in the primary sequence of structural proteins, but nevertheless use a number of common physico-chemical mechanisms at different stages of cell infection. The evolutional similarity of the matrix and capsid proteins of different enveloped viruses, as well as the common elements in their tertiary structure, also indicate possible general mechanisms of the capsid self-assembly and disintegration. Viral fusion proteins are able to create a small fusion pore; however, blocking the dissociation of the protein envelope prevents the release of the viral genome into the cell cytoplasm [13, 41]. It is still unclear whether matrix and capsid proteins are active participants in the expansion of the fusion pore or whether they only dissociate from the lipid envelope of the virion. The multifunctionality of viral proteins indicates that matrix proteins can also perform several functions. The effect of these proteins on the cellular transport of viral genetic material has been determined [100–102], but no effect on the release of the viral genome has been demonstrated. These proteins must also self-assemble to form a viral particle and release it from the infected cell. The formation and budding of the viral particle require extensive topological rearrangements of the cell membranes. This leads to the question of how and which viral proteins can do this. The focus is on matrix proteins, since their amount is the highest for enveloped virus proteins, and they are the ones that establish and support the viral architecture. Therefore, in the process of self-organization, capsid proteins should provide deformations of cell membranes and conditions for the budding or release of virions. Limitations of membrane structure modification strategies imply general functional mechanisms that can be described within the framework of physico-chemical models, but this area is currently less well developed than the issues of virus-induced membrane fusion. CORONAVIRUSES: A NEW CHALLENGE FOR PHYSICO-CHEMICAL MODELS Coronaviruses have attracted a great attention as triggers of outbreaks of human respiratory syndromes such as severe acute respiratory syndrome (SARS), Middle East respiratory syndrome (MERS) and new coronavirus infection (COVID-19). These respiratory diseases have arisen from zoonotic transmission from animals to humans, resulting in virus strains that have not been previously circulated in the human population. The SARS-CoV-2 coronavirus, the causative agent of COVID-19, has shown rapid global spread and has led to a pandemic. Coronaviruses, like the viruses discussed above, belong to the class of enveloped viruses. The coronavirus genome contains four main structural proteins: spike (S), membrane protein (M), envelope protein (E), and nucleocapsid protein (N), which are coded at the 3'-end of the genome [103]. The S protein mediates the viral binding to the surface receptors of the host cell, which leads to the fusion and subsequent entry of the virus. The M protein is the most represented and determines the shape of the viral envelope. The E protein has the lowest amount of the major structural proteins and is involved in viral assembly and budding. The N protein is the only one that binds to the RNA genome and also participates in the assembly and budding. It should be noted that, compared to other enveloped viruses, almost all structural proteins of coronaviruses have transmembrane domains. Despite their complexity and range of functions [104, 105], structural proteins of coronaviruses cover only about one-third of the genome’s coding part. A larger site of the genome, about two-thirds, located at the 5'-end and codes two long open reading frames 1a and 1b of the viral nonstructural proteins. Each sequence is primarily translated as a polyprotein precursor, pp1a and pp1ab. Polyproteins include several viral proteases that transform pp1a and pp1ab into 16 nonstructural proteins (nsp 1–16) needed at different stages of the viral replication cycle. These proteins are the most conservative proteins of coronaviruses [103]. Many nonstructural proteins interact with membranes as coronavirus replication occurs in dedicated cellular compartments. These are created by viral proteins that modify cell membranes to create sites of viral replication hidden from cell innate immunity [106]. The combination of multiple membrane interacting factors makes coronaviruses one of the most complex models. Replication of RNA viruses with positive RNA chain, as well as DNA viruses in plant and animal cells causes the formation of a subcellular microenvironment—a replication network—viral factories, or viroplasma [107]. Although the purpose of such membrane formations in the viral life cycle is not entirely known, it is assumed to facilitate the stage of the viral RNA synthesis [108, 109]. The formation of such “mini-organelles” requires a deformation of the host cell membranes and cytoskeleton, causing a cytopathic effect, that has been used as a marker of viral infection to test the potential efficacy of drugs [110]. All positive-chain RNA viruses (+RNA viruses) that infect eukaryotic cells are supposed to form membrane-bound organelles [111]. One of the most common membrane modifications induced by +RNA viruses is the formation of double membranes, two closely spaced lipid bilayers that give a structure for double-membrane vesicles [108]. The widespread of such structures in the replication of +RNA viruses suggests that this is an effective strategy to produce new virions, and that membrane pairing can enhance the competitiveness of these viruses. The coronavirus family has a highly complex replication membrane network that originates from the endoplasmic reticulum [106], in a similar manner to the Dengue virus [112] and reovirus [113]. Its structure includes many curved membranes and double-membrane vesicles. All of them are interconnected in a continual network with each other and with the endoplasmic reticulum by lipid nanotubes [114]. This restructuring of host cell membranes is considered to be a viral strategy for replication by localizing and concentrating the essential factors and providing protection against the immune response [115]. This hypothesis is proved by the fact that the total level of viral RNA correlates with the number of double-membrane vesicles in the cell [106, 116, 117]. However, there are data that do not confirm the relationship between double-membrane vesicles in the endoplasmic reticulum and viral RNA synthesis in the infected cell [118]. It has been shown that mutations in the viral nonstructural proteins, which prevent them from forming double-membrane vesicles, do not lead to a complete cessation of the viral RNA synthesis. However, in this case the viral replication rate drops considerably. The mechanism of double-membrane vesicle formation is still unclear. An analysis of the physico-chemical mechanisms of the formation of the double-membrane vesicle from a double-membrane disk is performed in [119]. In this work, the bending energies of the membranes of the double-membrane vesicle and the double-membrane disk are calculated and compared within the framework of the Helfrich elasticity theory [120]. The bending energy of the spherical double-membrane vesicle is independent of its radius, while the bending energy of the double-membrane disk increases linearly as the disk radius increases. At some disc radius, these two energies are compared, and the double-membrane disc can transform into the double-membrane vesicle. This work has not considered the processes of the double-membrane disk formation from a flat membrane or the formation of the double-membrane vesicle from the double-membrane spherical segment. Also, the shape of the membranes is postulated rather than found by elastic energy optimization, and the proteins involved have not been described. Thus, it is still an open question whether viral replicases are capable to induce membrane curvature on the target organelle. This implies that these proteins must be capable of generating lattice-like structures. Also, they should have lipid specificity since there is no information on other types of specific targeting of these replicases in the endoplasmic reticulum. In hepatitis C, which also forms membrane replication compartments from the endoplasmic reticulum, the NS5A protein stimulates phosphatidylinositol-4-kinase-III activity. This facilitates phosphatidylinositol-4-phosphate production on the cytoplasmic side of the endoplasmic reticulum and hypothetically facilitates RNA polymerase accumulation [121]. It has been shown that enrichment with phosphatidylinositol-4-phosphate is important for the replication of enterovirus and flavivirus RNA. The fact that the nonstructural proteins of coronaviruses can form scaffolds that contribute to cell membrane deformations makes them similar to the matrix proteins of other enveloped viruses. This suggests the possibility of common mechanisms of protein–lipid and protein–protein interactions. Such double-membrane structures are not found in uninfected cells (except for autophagosomes), indicating that spontaneous “shrinkage” of the endoplasmic reticulum cannot be the driving force of the formation of double-membrane structures. Understanding this formation mechanism is critically important for clarifying the principles of organization of the viral replication compartments. Another proposed mechanism of double-membrane vesicle formation involves cellular autophagy [122, 123]. It is assumed that the DFCP1 protein (double protein 1 containing the FYVE domain) binds phosphatidylinositol phosphate, thus forming curved regions of the endoplasmic reticulum called omegasomes [124]. Based on the data obtained by some coronaviruses, it has been suggested that their nsp 6 proteins generate autophagosomes [125]. The envelope E proteins of coronavirus play multiple roles during infection, including viral morphogenesis. They are small (74–109 amino acids) hydrophobic viroporins [126]. E proteins consist of two separate structural domains: a hydrophobic domain longer than the thickness of the lipid bilayer and a charged cytoplasmic tail. The role of E proteins in assembly and release is not completely understood. The necessity for them during virus morphogenesis depends on the viral type. Removal of E proteins from transmissible gastroenteritis virus (TGEV) leads to replication of competent but proliferation-deficient viruses [127], as in the case of MERS virus [128]. SARS-CoV demonstrated a 200-fold reduction in virus release in the absence of E protein, which depended on the cell type used for infection [129]. Hence, although the membrane-active proteins of coronaviruses, such as nsp 3, 4, and 6 and E protein, are highly conservative and critical for viral replication, the mechanisms of their functioning and interaction with cell membranes and with each other are still unclear. The S protein is a glycosylated type I membrane fusion protein that consists of two subunits, S1 and S2. The N-terminal S1 subunit contains a receptor-binding domain (RBD) that mediates binding to the host cell receptor, namely angiotensin converting enzyme 2 (ACE2), for both SARS-CoV and SARS-CoV-2 [130, 131]. Biophysical structural and functional studies of the S protein fusion peptide of SARS-CoV-2 viruses [132], SARS [133], and MERS [134] provided important insights into the functional significance of specific domains of the fusion peptide, such as the large number of conserved amino acid residues in two consecutive fusion peptide fragments named FP1 and FP2 (residues 816–835 and 835–854, respectively, in SARS-CoV-2). These studies showed that collectively the F1 and F2 fragments form a platform of bilateral interactions between membranes, and the amino acid residues in both FP1 and FP2 contribute to membrane binding through their interaction with Ca2+ ions. These data on the functional role of Ca2+ are supported by the results obtained for fusion peptides from other related viruses, such as MERS [134] and Ebola virus [135]. In fact, in the MERS virus fusion peptide, one of these amino acid residues, E891 in the N-terminal (FP1) part (corresponding to E819 in the SARS-CoV-2 fusion peptide numeration), is proved to be crucial for interactions with Ca2+ and providing fusion-related cell membrane deformations [136]. In addition, it is shown that disruption of the lipid bilayer by Ca2+-dependent interactions of the fusion peptide with the membrane affects the organization of the lipid polar groups at the interaction site, but not the central hydrophobic region of the membrane [136]. Nevertheless, the Ca2+ binding sites with the fusion peptide and their role in any specific (but unknown) modes of interaction of the fusion peptide with the membrane remain uncertain. This makes difficult to explain any measurable effects of viral interactions with the membrane and, therefore, any approaches for decreasing infectivity by targeting this important region of the S protein. As mentioned above, different activation mechanisms of fusion proteins and the depth of the fusion peptide incorporation into the lipid matrix of cell membranes regulate the cooperative action of fusion proteins through deformations of the target cell lipid bilayer rather than through direct protein–protein interactions [62]. A molecular model of the action of influenza A virus and human immunodeficiency virus fusion peptides makes it possible to clarify the role of receptor structure in the cooperativity of the virus fusion proteins. The fact that the initial stages of SARS-CoV-2 coronavirus entry are close to those of HIV (binding to the receptor on the cell surface followed by fusion with the plasma membrane), and the SARS-CoV-2 protein, as the HIV gp41/gp120 protein, refers to type I fusion proteins, suggests that the assumptions of this model are valid for describing the function of the fusion peptide and S protein of the coronavirus. CONCLUSIONS The life cycle of all enveloped viruses is a complex multistage process that begins with virus entry into the cell and ends with the release of newly formed virions from its membrane. It involves a precisely determined interaction between the components of the virus and the cell. Disruption of any stage of the virion life cycle prevents the formation of infectiously competent virions and limits the spread of infection. A key feature of enveloped viruses is the presence of not only the protein shell that protects the viral genome, but also the lipid membrane, which interacts with it and contains the main structural proteins of the virion. Therefore, protein–lipid interactions play one of the defining roles in the processes of viral infection, replication, and release of new viral particles. Such interactions in most cases do not involve the formation of covalent chemical bonds, but are based on electrostatic, hydrophobic, and Van der Waals interactions. Hence, various physico-chemical mechanisms play an important role in the processes of cell infection by enveloped viruses. Despite the common nature of the described processes for most enveloped viruses, there are still many open questions about the molecular mechanisms of certain stages, about the forces that determine the functioning and interaction of the various viral components and the infected cell. It is obvious that the search for new effective antiviral drugs should be based on the fundamental mechanisms regulating the functioning of viral proteins and determining various stages of viral infection in the cell. Models of physico-chemical mechanisms of viral fusion processes, formation of progeny virions, as well as new challenges related to the replication features of coronaviruses can become the basis for such search. The creation and analysis of such models, initiated by Yu.A. Chizmadzhev and his colleagues in the 1980s, has been successfully developed to this day thanks to his unique scientific school. ACKNOWLEDGMENTS The author is grateful to M.M. Popova for technical assistance. FUNDING This work was supported by the Russian Science Foundation (project no. 22-13-00435). COMPLIANCE WITH ETHICAL STANDARDS The author declares the absence of obvious and potential conflicts of interest related to this publication. This article does not contain any studies involving animals or humans as participants performed by the author. Translated by M. Popova ==== Refs REFERENCES 1 McMahon H.T. Gallop J.L. Membrane curvature and mechanisms of dynamic cell membrane remodelling Nature. 2005 438 590 596 10.1038/nature04396 16319878 2 Zimmerberg J. Kozlov M.M. How proteins produce cellular membrane curvature Nat. Rev. Mol. Cell Biol. 2006 7 9 19 10.1038/nrm1784 16365634 3 Schekman R. Orci L. Coat proteins and vesicle budding Science 1996 271 1526 1533 10.1126/science.271.5255.1526 8599108 4 Bremser M. Nickel W. Schweikert M. Ravazzola M. Amherdt M. Hughes C.A. Söllner T.H. Rothman J.E. Wieland F.T. Coupling of coat assembly and vesicle budding to packaging of putative cargo receptors Cell. 1999 96 495 506 10.1016/S0092-8674(00)80654-6 10052452 5 Chernomordik L. Kozlov M.M. Zimmerberg J. Lipids in biological membrane fusion J. Membr. Biol. 1995 146 1 14 10.1007/BF00232676 7563032 6 Melikyan G.B. Abidor I.G. Chernomordik L.V. Chailakhyan L.M. Electrically stimulated fusion and fission of bilayer lipid membranes Dokl. Acad. Sci. USSR (Rus.) 1982 263 1009 1012 7 Kozlov M.M. Markin V.S. Possible mechanism of membrane fusion Biofizika (Rus.) 1983 28 242 247 8 Chernomordik L.V. Kozlov M.M. Melikyan G.B. Abidor I.G. Markin V.S. Chizmadzhev Yu.A. The shape of lipid molecules and monolayer membrane fusion Biochim. Biophys. Acta BBA – Biomembr. 1985 812 643 655 10.1016/0005-2736(85)90257-3 9 Wang T. Li L. Hong W. SNARE proteins in membrane trafficking Traffic. 2017 18 767 775 10.1111/tra.12524 28857378 10 Jahn R. Scheller R.H. SNAREs – engines for membrane fusion Nat. Rev. Mol. Cell Biol. 2006 7 631 643 10.1038/nrm2002 16912714 11 Skehel J.J. Wiley D.C. Receptor binding and membrane fusion in virus entry: The influenza hemagglutinin Annu. Rev. Biochem. 2000 69 531 569 10.1146/annurev.biochem.69.1.531 10966468 12 Chernomordik L.V. Frolov V.A. Leikina E. Bronk P. Zimmerberg J. The pathway of membrane fusion catalyzed by influenza hemagglutinin: Restriction of lipids, hemifusion, and lipidic fusion pore formation J. Cell Biol. 1998 140 1369 1382 10.1083/jcb.140.6.1369 9508770 13 Lee K.K. Architecture of a nascent viral fusion pore EMBO J. 2010 29 1299 1311 10.1038/emboj.2010.13 20168302 14 Calder L.J. Wasilewski S. Berriman J.A. Rosenthal P.B. Structural organization of a filamentous influenza A virus Proc. Natl. Acad. Sci. USA. 2010 107 10685 10690 10.1073/pnas.1002123107 20498070 15 Ge P. Tsao J. Schein S. Green T.J. Luo M. Zhou Z.H. Cryo-EM model of the bullet-shaped vesicular stomatitis virus Science. 2010 327 689 693 10.1126/science.1181766 20133572 16 Liljeroos L. Huiskonen J.T. Ora A. Susi P. Butcher S.J. Electron cryotomography of measles virus reveals how matrix protein coats the ribonucleocapsid within intact virions Proc. Natl. Acad. Sci. USA. 2011 108 18085 18090 10.1073/pnas.1105770108 22025713 17 Chizmadzhev Y.A. The mechanisms of lipid–protein rearrangements during viral infection Bioelectrochemistry. 2004 63 129 136 10.1016/j.bioelechem.2003.10.016 15110263 18 Pettersson R.F. 1991. Protein localization and virus assembly at intracellular membranes. In: Protein traffic in eukaryotic cells. Vol. 170. Current topics in microbiology and immunology. Ed. Compans R.W. Berlin, Heidelberg: Springer, p. 67–106. 19 White J. Kielian M. Helenius A. Membrane fusion proteins of enveloped animal viruses Q. Rev. Biophys. 1983 16 151 195 10.1017/S0033583500005072 6359230 20 Kielian M. Rey F.A. Virus membrane-fusion proteins: More than one way to make a hairpin Nat. Rev. Microbiol. 2006 4 67 76 10.1038/nrmicro1326 16357862 21 Colman P.M. Lawrence M.C. The structural biology of type I viral membrane fusion Nat. Rev. Mol. Cell Biol. 2003 4 309 319 10.1038/nrm1076 12671653 22 Poehlmann S. Simmons G. Viral entry into host cells. 2013 New York Springer Science 23 Südhof T.C. Neurotransmitter release: The last millisecond in the life of a synaptic vesicle Neuron. 2013 80 675 690 10.1016/j.neuron.2013.10.022 24183019 24 Hernandez L.D. Hoffman L.R. Wolfsberg T.G. White J.M. Virus–cell and cell–cell fusion Annu. Rev. Cell Dev. Biol. 1996 12 627 661 10.1146/annurev.cellbio.12.1.627 8970739 25 Duelli D. Lazebnik Y. Cell fusion: A hidden enemy? Cancer Cell. 2003 3 445 448 10.1016/S1535-6108(03)00114-4 12781362 26 Chen E.H. Unveiling the mechanisms of cell–cell fusion Science. 2005 308 369 373 10.1126/science.1104799 15831748 27 Chen E.H. Grote E. Mohler W. Vignery A. Cell–cell fusion FEBS Lett. 2007 581 2181 2193 10.1016/j.febslet.2007.03.033 17395182 28 White J. Membrane fusion Science. 1992 258 917 924 10.1126/science.1439803 1439803 29 Wickner W. Schekman R. Membrane fusion Nat. Struct. Mol. Biol. 2008 15 658 664 10.1038/nsmb.1451 18618939 30 Brukman N.G. Uygur B. Podbilewicz B. Chernomordik L.V. How cells fuse J. Cell Biol. 2019 218 1436 1451 10.1083/jcb.201901017 30936162 31 Harrison S.C. Viral membrane fusion Virology. 2015 479–480 498 507 10.1016/j.virol.2015.03.043 32 Kuzmin P.I. Zimmerberg J. Chizmadzhev Y.A. Cohen F.S. A quantitative model for membrane fusion based on low-energy intermediates Proc. Natl. Acad. Sci. USA. 2001 98 7235 7240 10.1073/pnas.121191898 11404463 33 Lentz B.R. PEG as a tool to gain insight into membrane fusion Eur. Biophys. J. 2007 36 315 326 10.1007/s00249-006-0097-z 17039359 34 Hong J. Yang H. Pang D. Wei L. Deng C. Effects of mono- and di-valent metal cations on the morphology of lipid vesicles Chem. Phys. Lipids. 2018 217 19 28 10.1016/j.chemphyslip.2018.09.011 30253127 35 White J.M. Delos S.E. Brecher M. Schornberg K. Structures and mechanisms of viral membrane fusion proteins: Multiple variations on a common theme Crit. Rev. Biochem. Mol. Biol. 2008 43 189 219 10.1080/10409230802058320 18568847 36 Benhaim M.A. Lee K.K. New biophysical approaches reveal the dynamics and mechanics of type I viral fusion machinery and their interplay with membranes Viruses. 2020 12 413 10.3390/v12040413 32276357 37 Chlanda P. Mekhedov E. Waters H. Schwartz C.L. Fischer E.R. Ryham R.J. Cohen F.S. Blank P.S. Zimmerberg J. The hemifusion structure induced by influenza virus haemagglutinin is determined by physical properties of the target membranes Nat. Microbiol. 2016 1 16050 10.1038/nmicrobiol.2016.50 27572837 38 Sammalkorpi M. Lazaridis T. Configuration of influenza hemagglutinin fusion peptide monomers and oligomers in membranes Biochim. Biophys. Acta BBA – Biomembr. 2007 1768 30 38 10.1016/j.bbamem.2006.08.008 39 Qiang W. Sun Y. Weliky D.P. A strong correlation between fusogenicity and membrane insertion depth of the HIV fusion peptide Proc. Natl. Acad. Sci. USA. 2009 106 15314 15319 10.1073/pnas.0907360106 19706388 40 Frolov V.A. Cho M.-S. Bronk P. Reese T.S. Zimmerberg J. Multiple local contact sites are induced by GPI-linked influenza hemagglutinin during hemifusion and flickering pore formation Traffic. 2000 1 622 630 10.1034/j.1600-0854.2000.010806.x 11208150 41 Markosyan R.M. Cohen F.S. Melikyan G.B. The lipid-anchored ectodomain of influenza virus hemagglutinin (GPI-HA) is capable of inducing nonenlarging fusion pores Mol. Biol. Cell. 2000 11 1143 1152 10.1091/mbc.11.4.1143 10749920 42 Leikin S.L. Kozlov M.M. Chernomordik L.V. Markin V.S. Chizmadzhev Y.A. Membrane fusion: Overcoming of the hydration barrier and local restructuring J. Theor. Biol. 1987 129 411 425 10.1016/S0022-5193(87)80021-8 3455469 43 Helm C. Israelachvili J. McGuiggan P. Molecular mechanisms and forces involved in the adhesion and fusion of amphiphilic bilayers Science. 1989 246 919 922 10.1126/science.2814514 2814514 44 Kozlovsky Y. Efrat A. Siegel D.A. Kozlov M.M. Stalk phase formation: Effects of dehydration and saddle splay modulus Biophys. J. 2004 87 2508 2521 10.1529/biophysj.103.038075 15454446 45 Kozlovsky Y. Chernomordik L.V. Kozlov M.M. Lipid intermediates in membrane fusion: Formation, structure, and decay of hemifusion diaphragm Biophys. J. 2002 83 2634 2651 10.1016/S0006-3495(02)75274-0 12414697 46 Akimov S. Polynkin M.A. Jiménez-Munguía I. Pavlov K.V. Batishchev O.V. Phosphatidylcholine membrane fusion is pH-dependent Int. J. Mol. Sci. 2018 19 1358 10.3390/ijms19051358 29751591 47 Kozlov M.M. Chernomordik L.V. A mechanism of protein-mediated fusion: Coupling between refolding of the influenza hemagglutinin and lipid rearrangements Biophys. J. 1998 75 1384 1396 10.1016/S0006-3495(98)74056-1 9726939 48 Melikyan G.B. Markosyan R.M. Hemmati H. Delmedico M.K. Lambert D.M. Cohen F.S. Evidence that the transition of HIV-1 Gp41 into a six-helix bundle, not the bundle configuration, induces membrane fusion J. Cell Biol. 2000 151 413 424 10.1083/jcb.151.2.413 11038187 49 Danieli T. Pelletier S.L. Henis Y.I. White J.M. Membrane fusion mediated by the influenza virus hemagglutinin requires the concerted action of at least three hemagglutinin trimers J. Cell Biol. 1996 133 559 569 10.1083/jcb.133.3.559 8636231 50 Yang X. Kurteva S. Ren X. Lee S. Sodroski J. Stoichiometry of envelope glycoprotein trimers in the entry of human immunodeficiency virus type 1 J. Virol. 2005 79 12132 12147 10.1128/JVI.79.19.12132-12147.2005 16160141 51 Kozlovsky Y. Zimmerberg J. Kozlov M.M. Orientation and interaction of oblique cylindrical inclusions embedded in a lipid monolayer: A theoretical model for viral fusion peptides Biophys. J. 2004 87 999 1012 10.1529/biophysj.104.041467 15298906 52 Pinigin K.V. Kondrashov O.V. Jiménez-Munguía I. Alexandrova V.V. Batishchev O.V. Galimzyanov T.R. Akimov S.A. Elastic deformations mediate interaction of the raft boundary with membrane inclusions leading to their effective lateral sorting Sci. Rep. 2020 10 4087 10.1038/s41598-020-61110-2 32139760 53 Kondrashov O.V. Galimzyanov T.R. Jiménez-Munguía I. Batishchev O.V. Akimov S.A. Membrane-mediated interaction of amphipathic peptides can be described by a one-dimensional approach Phys. Rev. E. 2019 99 022401 10.1103/PhysRevE.99.022401 30934249 54 Harrison J.S. Higgins C.D. O’Meara M.J. Koellhoffer J.F. Kuhlman B.A. Lai J.R. Role of electrostatic repulsion in controlling pH-dependent conformational changes of viral fusion proteins Structure. 2013 21 1085 1096 10.1016/j.str.2013.05.009 23823327 55 McMahon H.T. Kozlov M.M. Martens S. Membrane curvature in synaptic vesicle fusion and beyond Cell. 2010 140 601 605 10.1016/j.cell.2010.02.017 20211126 56 Kozlov M.M. McMahon H.T. Chernomordik L.V. Protein-driven membrane stresses in fusion and fission Trends Biochem. Sci. 2010 35 699 706 10.1016/j.tibs.2010.06.003 20638285 57 Chernomordik L. Non-bilayer lipids and biological fusion intermediates Chem. Phys. Lipids. 1996 81 203 213 10.1016/0009-3084(96)02583-2 8810049 58 Molotkovsky R. Galimzyanov T. Jiménez-Munguía I. Pavlov K. Batishchev O. Akimov S. Switching between successful and dead-end intermediates in membrane fusion Int. J. Mol. Sci. 2017 18 2598 10.3390/ijms18122598 29207481 59 Akimov S.A. Molotkovsky R.J. Galimzyanov T.R. Radaev A.V. Shilova L.A. Kuzmin P.I. Batishchev O.V. Voronina G.F. Chizmadzhev Yu.A. Model of membrane fusion: Continuous transition to fusion pore with regard of hydrophobic and hydration interactions. Biochem. (Mosc.) Suppl. Ser. A: Membr. Cell Biol. 2014 8 153 161 60 Merk A. Subramaniam S. HIV-1 envelope glycoprotein structure Curr. Opin. Struct. Biol. 2013 23 268 276 10.1016/j.sbi.2013.03.007 23602427 61 Martens S. Kozlov M.M. McMahon H.T. How synaptotagmin promotes membrane fusion Science. 2007 316 1205 1208 10.1126/science.1142614 17478680 62 Akimov S.A. Kondrashov O.V. Zimmerberg J. Batishchev O.V. Ectodomain pulling combines with fusion peptide inserting to provide cooperative fusion for influenza virus and HIV Int. J. Mol. Sci. 2020 21 5411 10.3390/ijms21155411 32751407 63 Baudin F. Petit I. Weissenhorn W. Ruigrok R.W.H. In vitro dissection of the membrane and RNP binding activities of influenza virus M1 protein Virology. 2001 281 102 108 10.1006/viro.2000.0804 11222100 64 Leser G.P. Lamb R.A. Lateral organization of influenza virus proteins in the budozone region of the plasma membrane J. Virol. 2017 91 e02104 16 10.1128/JVI.02104-16 28202765 65 Ruigrok R.W.H. Barge A. Durrer P. Brunner J. Ma K. Whittaker G.R. Membrane interaction of influenza virus M1 protein Virology. 2000 267 289 298 10.1006/viro.1999.0134 10662624 66 Ruigrok R.W.H. Schoehn G. Dessen A. Forest E. Volchkov V. Dolnik O. Klenk H.-D. Weissenhorn W. Structural characterization and membrane binding properties of the matrix protein VP40 of Ebola virus J. Mol. Biol. 2000 300 103 112 10.1006/jmbi.2000.3822 10864502 67 Wijesinghe K.J. Urata S. Bhattarai N. Kooijman E.E. Gerstman B.S. Chapagain P.P. Li S. Stahelin R.V. Detection of lipid-induced structural changes of the Marburg virus matrix protein VP40 using hydrogen/deuterium exchange-mass spectrometry J. Biol. Chem. 2017 292 6108 6122 10.1074/jbc.M116.758300 28167534 68 Shtykova E.V. Dadinova L.A. Fedorova N.V. Golanikov A.E. Bogacheva E.N. Ksenofontov A.L. Baratova L.A. Shilova L.A. Tashkin V.Yu. Galimzyanov T.R. Jeffries C.M. Svergun D.I. Batishchev O.V. Influenza virus matrix protein M1 preserves its conformation with pH, changing multimerization state at the priming stage due to electrostatics Sci Rep. 2017 8 e82431 69 Peukes J. Xiong X. Erlendsson S. Qu K. Wan W. Calder L.J. Schraidt O. Kummer S. Freund S.M.V. Kräusslich H.G. Briggs J.A.G. The native structure of the assembled matrix protein 1 of influenza A virus Nature. 2020 587 495 498 10.1038/s41586-020-2696-8 32908308 70 Battisti A.J. Meng G. Winkler D.C. McGinnes L.W. Plevka P. Steven A.C. Morrison T.G. Rossmann M.G. Structure and assembly of a paramyxovirus matrix protein Proc. Natl. Acad. Sci. USA. 2012 109 13996 4000 10.1073/pnas.1210275109 22891297 71 Briggs J.A.G. Wilk T. Welker R. Kräusslich H.G. Fuller S.D. Structural organization of authentic, mature HIV-1 virions and cores EMBO J. 2003 22 1707 1715 10.1093/emboj/cdg143 12660176 72 Harris A. Sha B. Luo M. Structural similarities between influenza virus matrix protein M1 and human immunodeficiency virus matrix and capsid proteins: An evolutionary link between negative-stranded RNA viruses and retroviruses J. Gen. Virol. 1999 80 863 869 10.1099/0022-1317-80-4-863 10211954 73 Dunker A.K. Babu M.M. Barbar E. Blackledge M. Bondos S.E. Dosztányi Z. Dyson H.J. Forman-Kay J. Fuxreiter M. Gsponer J. Han K.H. Jones D.T. Longhi S. Metallo S.J. Nishikawa K. Nussinov R. Obradovic Z. Pappu R.V. Rost B. Selenko P. Subramaniam V. Sussman J.L. Tompa P. Uversky V.N. What’s in a name? Why these proteins are intrinsically disordered: Why these proteins are intrinsically disordered Intrinsically Disord. Proteins. 2013 1 e24157 10.4161/idp.24157 28516007 74 Goh G. Dunker A.K. Uversky V.N. A comparative analysis of viral matrix proteins using disorder predictors Virol. J. 2008 5 126 10.1186/1743-422X-5-126 18947403 75 Zhang W. Zheng W. Toh Y. Betancourt-Solis M.A. Tu J. Fan Y. Vakharia V.N. Liu J. McNew J.A. Jin M. Tao Y.J. Crystal structure of an orthomyxovirus matrix protein reveals mechanisms for self-polymerization and membrane association Proc. Natl. Acad. Sci. USA. 2017 114 8550 8555 10.1073/pnas.1701747114 28739952 76 Shnyrova A.V. Ayllon J. Mikhalyov I.I. Villar E. Zimmerberg J. Frolov V.A. Vesicle formation by self-assembly of membrane-bound matrix proteins into a fluidlike budding domain J. Cell Biol. 2007 179 627 633 10.1083/jcb.200705062 18025300 77 Faaberg K.S. Peeples M.E. Association of soluble matrix protein of newcastle disease virus with liposomes is independent of ionic conditions Virology. 1988 166 123 132 10.1016/0042-6822(88)90153-5 3413981 78 Sha B. Luo M. Structure of a bifunctional membrane-RNA binding protein, influenza virus matrix protein M1 Nat. Struct. Biol. 1997 4 239 244 10.1038/nsb0397-239 9164466 79 Gregoriades A. Interaction of influenza M protein with viral lipid and phosphatidylcholine vesicles J. Virol. 1980 36 470 479 10.1128/jvi.36.2.470-479.1980 7431487 80 Gregoriades A. Frangione B. Insertion of influenza M protein into the viral lipid bilayer and localization of site of insertion J. Virol. 1981 40 323 328 10.1128/jvi.40.1.323-328.1981 7288926 81 Ye Z.P. Pal R. Fox J.W. Wagner R.R. Functional and antigenic domains of the matrix (M1) protein of influenza A virus J. Virol. 1987 61 239 246 10.1128/jvi.61.2.239-246.1987 2433462 82 Shi Z. Baumgart T. Membrane tension and peripheral protein density mediate membrane shape transitions Nat. Commun. 2015 6 5974 10.1038/ncomms6974 25569184 83 Saletti D. Radzimanowski J. Effantin G. Midtvedt D. Mangenot S. Weissenhorn W. Bassereau P. Bally M. The Matrix protein M1 from influenza C virus induces tubular membrane invaginations in an in vitro cell membrane model Sci. Rep. 2017 7 40801 10.1038/srep40801 28120862 84 Hilsch M. Goldenbogen B. Sieben C. Höfer C.T. Rabe J.P. Klipp E. Herrmann A. Chiantia S. Influenza A matrix protein M1 multimerizes upon binding to lipid membranes Biophys. J. 2014 107 912 923 10.1016/j.bpj.2014.06.042 25140426 85 Adu-Gyamfi E. Johnson K.A. Fraser M.E. Scott J.L. Soni S.P. Jones K.R. Stahelin R.V. Host cell plasma membrane phosphatidylserine regulates the assembly and budding of Ebola virus J. Virol. 2015 89 9440 9453 10.1128/JVI.01087-15 26136573 86 Gc J.B. Gerstman B.S. Stahelin R.V. Chapagain P.P. The Ebola virus protein VP40 hexamer enhances the clustering of PI(4,5)P2 lipids in the plasma membrane Phys. Chem. Chem. Phys. 2016 18 28 409 28 417 10.1039/C6CP03776C 87 Justice P.A. Sun W. Li Y. Ye Z. Grigera P.R. Wagner R.R. Membrane vesiculation function and exocytosis of wild-type and mutant matrix proteins of vesicular stomatitis virus J. Virol. 1995 69 3156 3160 10.1128/jvi.69.5.3156-3160.1995 7707543 88 Gómez-Puertas P. Albo C. Pérez-Pastrana E. Vivo A. Portela A. Influenza virus matrix protein is the major driving force in virus budding J. Virol. 2000 74 11538 11547 10.1128/JVI.74.24.11538-11547.2000 11090151 89 Latham T. Galarza J.M. Formation of wild-type and chimeric influenza virus-like particles following simultaneous expression of only four structural proteins J. Virol. 2001 75 6154 6165 10.1128/JVI.75.13.6154-6165.2001 11390617 90 Jasenosky L.D. Neumann G. Lukashevich I. Kawaoka Y. Ebola virus VP40-induced particle formation and association with the lipid bilayer J. Virol. 2001 75 5205 5214 10.1128/JVI.75.11.5205-5214.2001 11333902 91 Chlanda P. Mekhedov E. Waters H. Sodt A. Schwartz C. Nair V. Blank P.S. Zimmerberg J. Palmitoylation contributes to membrane curvature in influenza A virus assembly and hemagglutinin-mediated membrane fusion J. Virol. 2017 91 e00947 17 10.1128/JVI.00947-17 28794042 92 Scheiffele P. Rietveld A. Wilk T. Simons K. Influenza Viruses select ordered lipid domains during budding from the plasma membrane J. Biol. Chem. 1999 274 2038 2044 10.1074/jbc.274.4.2038 9890962 93 Veit M. Thaa B. Association of influenza virus proteins with membrane rafts Adv. Virol. 2011 2011 370606 10.1155/2011/370606 22312341 94 Gerl M.J. Sampaio J.L. Urban S. Kalvodova L. Verbavatz J.M. Binnington B. Simons K. Quantitative analysis of the lipidomes of the influenza virus envelope and MDCK cell apical membrane J. Cell Biol. 2012 196 213 221 10.1083/jcb.201108175 22249292 95 Nayak D.P. Balogun R.A. Yamada H. Zhou Z.H. Barman S. Influenza virus morphogenesis and budding Virus Res. 2009 143 147 161 10.1016/j.virusres.2009.05.010 19481124 96 Pohl C. Duprex W.P. Krohne G. Rima B.K. Schneider-Schaulies S. Measles virus M and F proteins associate with detergent-resistant membrane fractions and promote formation of virus-like particles J. Gen. Virol. 2007 88 1243 1250 10.1099/vir.0.82578-0 17374768 97 Simons K. Ikonen E. Functional rafts in cell membranes Nature. 1997 387 569 572 10.1038/42408 9177342 98 Baumgart T. Hess S.T. Webb W.W. Imaging coexisting fluid domains in biomembrane models coupling curvature and line tension Nature. 2003 425 821 824 10.1038/nature02013 14574408 99 Bobone S. Hilsch M. Storm J. Dunsing V. Herrmann A. Chiantia S. Phosphatidylserine lateral organization influences the interaction of influenza virus matrix protein 1 with lipid membranes J. Virol. 2017 91 e00267 17 10.1128/JVI.00267-17 28356535 100 Peeples M.E. Wang C. Gupta K.C. Coleman N. Nuclear entry and nucleolar localization of the Newcastle disease virus (NDV) matrix protein occur early in infection and do not require other NDV proteins J. Virol. 1992 66 3263 3269 10.1128/jvi.66.5.3263-3269.1992 1560547 101 Wang Y.E. Pernet O. Lee B. Regulation of the nucleocytoplasmic trafficking of viral and cellular proteins by ubiquitin and small ubiquitin-related modifiers Biol. Cell. 2012 104 121 138 10.1111/boc.201100105 22188262 102 Bui M. Wills E.G. Helenius A. Whittaker G.R. Role of the influenza virus M1 protein in nuclear export of viral ribonucleoproteins J. Virol. 2000 74 1781 1786 10.1128/JVI.74.4.1781-1786.2000 10644350 103 Masters P.S. The molecular biology of coronaviruses Adv. Virus Res 2006 66 193 292 10.1016/S0065-3527(06)66005-3 16877062 104 Perlman S. Netland J. Coronaviruses post-SARS: Update on replication and pathogenesis Nat. Rev. Microbiol. 2009 7 439 450 10.1038/nrmicro2147 19430490 105 Reguera J. Mudgal G. Santiago C. Casasnovas J.M. A structural view of coronavirus-receptor interactions Virus Res. 2014 194 3 15 10.1016/j.virusres.2014.10.005 25451063 106 Ulasli M. Verheije M.H. de Haan C.A.M. Reggiori F. Qualitative and quantitative ultrastructural analysis of the membrane rearrangements induced by coronavirus Cell. Microbiol. 2010 12 844 861 10.1111/j.1462-5822.2010.01437.x 20088951 107 Miller S. Krijnse-Locker J. Modification of intracellular membrane structures for virus replication Nat. Rev. Microbiol. 2008 6 363 374 10.1038/nrmicro1890 18414501 108 Netherton C.L. Wileman T. Virus factories, double membrane vesicles and viroplasm generated in animal cells Curr. Opin. Virol. 2011 1 381 387 10.1016/j.coviro.2011.09.008 22440839 109 Hsu N.Y. Ilnytska O. Belov G. Santiana M. Chen Y.H. Takvorian P.M. Altan-Bonnet N. Viral reorganization of the secretory pathway generates distinct organelles for RNA replication Cell. 2010 141 799 811 10.1016/j.cell.2010.03.050 20510927 110 de Wilde A.H. Raj V.S. Oudshoorn D. Bestebroer T.M. van Nieuwkoop S. Limpens R.W. van den Hoogen B.G. MERS-coronavirus replication induces severe in vitro cytopathology and is strongly inhibited by cyclosporin A or interferon-α treatment J. Gen. Virol. 2013 94 1749 1760 10.1099/vir.0.052910-0 23620378 111 Neuman B.W. Angelini M.M. Buchmeier M.J. Does form meet function in the coronavirus replicative organelle? Trends Microbiol. 2014 22 642 647 10.1016/j.tim.2014.06.003 25037114 112 Welsch S. Miller S. Romero-Brey I. Merz A. Bleck C.K. Walther P. Bartenschlager R. Composition and three-dimensional architecture of the dengue virus replication and assembly sites Cell Host Microbe. 2009 5 365 375 10.1016/j.chom.2009.03.007 19380115 113 Tenorio R. Fernández de Castro I. Knowlton J.J. Zamora P.F. Lee C.H. Mainou B.A. Dermody T.S. Risco C. Reovirus σNS and μNS proteins remodel the endoplasmic reticulum to build replication neo-organelles mBio. 2018 9 e01253 18 10.1128/mBio.01253-18 30087167 114 Knoops K. Kikkert M. Worm S.H.E. van den Zevenhoven-Dobbe J.C. van der Meer Y. Koster A.J. Mommaas A.M. Snijder E.J. SARS-coronavirus replication is supported by a reticulovesicular network of modified endoplasmic reticulum PLoS Biol. 2008 6 e226 10.1371/journal.pbio.0060226 18798692 115 Gosert R. Kanjanahaluethai A. Egger D. Bienz K. Baker S.C. RNA replication of mouse hepatitis virus takes place at double-membrane vesicles J. Virol. 2002 76 3697 3708 10.1128/JVI.76.8.3697-3708.2002 11907209 116 Verheije M.H. Raaben M. Mari M. Te Lintelo E.G. Reggiori F. van Kuppeveld F.J.M. Rottier P.J.M. de Haan C.A.M. Mouse hepatitis coronavirus RNA replication depends on GBF1-mediated ARF1 activation PLoS Pathog. 2008 4 e1000088 10.1371/journal.ppat.1000088 18551169 117 Stokes H.L. Baliji S. Hui C.G. Sawicki S.G. Baker S.C. Siddell S.G. A new cistron in the murine hepatitis virus replicase gene J. Virol. 2010 84 10148 10158 10.1128/JVI.00901-10 20668085 118 Sola I. Almazán F. Zúñiga S. Enjuanes L. Continuous and discontinuous RNA synthesis in coronaviruses Annu. Rev. Virol. 2015 2 265 288 10.1146/annurev-virology-100114-055218 26958916 119 Knorr R.L. Dimova R. Lipowsky R. Curvature of double-membrane organelles generated by changes in membrane size and composition PloS One. 2012 7 e32753 10.1371/journal.pone.0032753 22427874 120 Helfrich W. Elastic properties of lipid bilayers: Theory and possible experiments Z. Für Naturforschung C. 1973 28 693 703 10.1515/znc-1973-11-1209 121 Reiss S. Rebhan I. Backes P. Romero-Brey I. Erfle H. Matula P. Bartenschlager R. Recruitment and activation of a lipid kinase by hepatitis C virus NS5A is essential for integrity of the membranous replication compartment Cell Host Microbe. 2011 9 32 45 10.1016/j.chom.2010.12.002 21238945 122 Reggiori F. Monastyrska I. Verheije M.H. Calì T. Ulasli M. Bianchi S. Bernasconi R. de Haan C.A.M. Molinari M. Coronaviruses Hijack the LC3-I-positive EDEMosomes, ER-derived vesicles exporting short-lived ERAD regulators, for replication Cell Host Microbe. 2010 7 500 508 10.1016/j.chom.2010.05.013 20542253 123 Cottam E.M. Maier H.J. Manifava M. Vaux L.C. Chandra-Schoenfelder P. Gerner W. Britton P. Ktistakis N.T. Wileman T. Coronavirus nsp6 proteins generate autophagosomes from the endoplasmic reticulum via an omegasome intermediate Autophagy. 2011 7 1335 1347 10.4161/auto.7.11.16642 21799305 124 Axe E.L. Walker S.A. Manifava M. Chandra P. Roderick H.L. Habermann A. Griffiths G. Ktistakis N.T. Autophagosome formation from membrane compartments enriched in phosphatidylinositol 3-phosphate and dynamically connected to the endoplasmic reticulum J. Cell Biol. 2008 182 685 701 10.1083/jcb.200803137 18725538 125 Prentice E. Jerome W.G. Yoshimori T. Mizushima N. Denison M.R. Coronavirus replication complex formation utilizes components of cellular autophagy J. Biol. Chem. 2004 279 10 136 10 141 10.1074/jbc.M306124200 126 Ruch T.R. Machamer C.E. The coronavirus E protein: Assembly and beyond Viruses. 2012 4 363 382 10.3390/v4030363 22590676 127 Curtis K.M. Yount B. Baric R.S. Heterologous gene expression from transmissible gastroenteritis virus replicon particles J. Virol. 2002 76 1422 1434 10.1128/JVI.76.3.1422-1434.2002 11773416 128 Almazán F. DeDiego M.L. Sola I. Zuñiga S. Nieto-Torres J.L. Marquez-Jurado S. Andrés G. Enjuanes L. Engineering a replication-competent, propagation-defective Middle East respiratory syndrome coronavirus as a vaccine candidate mBio. 2013 4 e00650 00613 10.1128/mBio.00650-13 24023385 129 DeDiego M.L. Alvarez E. Almazán F. Rejas M.T. Lamirande E. Roberts A. Enjuanes L. A severe acute respiratory syndrome coronavirus that lacks the E gene is attenuated in vitro and in vivo J. Virol. 2007 81 1701 1713 10.1128/JVI.01467-06 17108030 130 Bosch B.J. van der Zee R. de Haan C.A.M. Rottier P.J.M. The coronavirus spike protein is a class I virus fusion protein: Structural and functional characterization of the fusion core complex J. Virol. 2003 77 8801 8811 10.1128/JVI.77.16.8801-8811.2003 12885899 131 Hoffmann M. Kleine-Weber H. Schroeder S. Krüger N. Herrler T. Erichsen S. Pöhlmann S. SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor Cell. 2020 181 271 280 10.1016/j.cell.2020.02.052 32142651 132 Lai A.L. Freed J.H. SARS-CoV-2 Fusion peptide has a greater membrane perturbating effect than SARS-CoV with highly specific dependence on Ca2+ J. Mol. Biol. 2021 433 166946 10.1016/j.jmb.2021.166946 33744314 133 Lai A.L. Millet J.K. Daniel S. Freed J.H. Whittaker G.R. The SARS-CoV fusion peptide forms an extended bipartite fusion platform that perturbs membrane order in a calcium-dependent manner J. Mol. Biol. 2017 429 3875 3892 10.1016/j.jmb.2017.10.017 29056462 134 Straus M.R. Tang T. Lai A.L. Flegel A. Bidon M. Freed J.H. Daniel S. Whittaker G.R. Ca2+ ions promote fusion of middle east respiratory syndrome coronavirus with host cells and increase infectivity J. Virol. 2020 94 e00426 20 10.1128/JVI.00426-20 32295925 135 Nathan L. Lai A.L. Millet J.K. Straus M.R. Freed J.H. Whittaker G.R. Daniel S. Calcium ions directly interact with the Ebola virus fusion peptide to promote structure-function changes that enhance infection ACS Infect. Dis. 2020 6 250 260 10.1021/acsinfecdis.9b00296 31746195 136 Madu I.G. Belouzard S. Whittaker G.R. SARS-coronavirus spike S2 domain flanked by cysteine residues C822 and C833 is important for activation of membrane fusion Virology. 2009 393 265 271 10.1016/j.virol.2009.07.038 19717178	0	PMC9734521	NO-CC CODE	2022-12-14 23:28:29	no	Biochem (Mosc) Suppl Ser A Membr Cell Biol. 2022 Dec 9; 16(4):247-260	utf-8	Biochem (Mosc) Suppl Ser A Membr Cell Biol	2,022	10.1134/S1990747822050038	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 36469276 24543 10.1007/s11356-022-24543-y Research Article Ozone catalytic oxidation of low-concentration formaldehyde over ternary Mn-Ce-Ni oxide catalysts modified with FeOx Liu Run Yu Man Trinh Minh Chuang Hsin Tzu http://orcid.org/0000-0001-8654-2975 Chang Moo Been [email protected] grid.37589.30 0000 0004 0532 3167 Graduate Institute of Environmental Engineering, National Central University, Chungli, Taiwan Responsible editor: George Z. Kyzas 5 12 2022 114 2 9 2022 25 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. Manganese oxide-based catalysts have attracted extensive attention due to their relatively low cost and remarkable performance for removing VOCs. In this research, we used the Pechini method to synthesize manganese-cerium-nickel ternary oxide catalysts (MCN) and evaluated the effectiveness of catalytic destruction of formaldehyde (HCHO) and ozone at room temperature. FeOx prepared by the impregnation method was applied to modify the catalyst. After FeOx treatment, the catalyst represented the best performance on both HCHO destruction and ozone decomposition under dry conditions and exhibited excellent water vapor resistance. The as-prepared catalysts were next characterized via H2-temperature programmed reduction (H2-TPR), temperature programmed desorption of O2 (O2-TPD), and X-ray photoelectron spectroscopy (XPS), and the results demonstrated that addition of FeOx increased Mn3+ and Ce3+ concentrations, oxygen vacancies and surface lattice oxygen species, facilitated adsorption, and redox properties. Based on the results of in situ diffuse reflectance infrared Fourier transform spectrometry (DRIFTS), possible mechanisms of ozone catalytic oxidation of HCHO were proposed. Overall, the ternary mixed-oxide catalyst developed in this study holds great promise for HCHO and ozone decomposition in the indoor environment. Supplementary Information The online version contains supplementary material available at 10.1007/s11356-022-24543-y. Keywords Mn-Ce-Ni ternary oxide catalysts FeOx Formaldehyde Ozone catalytic oxidation ==== Body pmcIntroduction In the winter of 2019, a novel coronavirus disease (COVID-19) was detected in China and rapidly spreading as a global pandemic. China was the first country which was determined to have lockdown human and industrial activities, and the consequence was enormous. Specifically, the lockdown results in people spending more time in the indoor environment, causing them to be highly exposed to indoor air pollutants (Du et al. 2020). Especially, people might have to prepare their meals at home rather than eating out. Jung et al. (2021) pointed out that a significant amount of formaldehyde (HCHO) was generated in typical home cooking procedures. Additionally, smoking was also an important source of indoor formaldehyde. During the lockdown, the frequency of nonsmokers exposed to second-hand smoke may also increase. Long-term exposure to low-concentration HCHO may increase the prevalence of nasopharyngeal cancer and leukemia (Nielsen et al. 2017). During the past few decades, researchers have developed effective methods such as adsorption, photocatalysis, and nonthermal plasma for removing HCHO from gas streams. However, these technologies have encountered some challenges, such as low efficiency, the formation of unwanted products, and catalyst deactivation. To solve these bottlenecks, ozone catalytic oxidation (OZCO) with the advantages of low cost, high removal efficiency, and less byproduct formation has been developed (Zhu et al. 2017a). With strong oxidation capability of ozone and oxygen radicals, HCHO could entirely oxidize into harmless water and CO2 at room temperature. However, ozone is a double-edged sword (Li et al. 2018). A previous study indicated that exposure to low-concentration ozone (~ 51 ppb) may cause respiratory illness and cardiovascular mortality (Cao et al. 2019). Consequently, simultaneous removal of HCHO and ozone is essential to protect human health and the environment when OZCO is applied. Noble metal catalysts show good catalytic performance, but they are expensive and tend to be deactivated (Bao et al. 2020). Among the transition metal oxides, MnOx, CeO2, and NiO-based oxides as potential alternatives have attracted increasing attention due to their low cost and high thermal stability. Zhang et al. (2019) prepared MnCeOx catalysts by the Pechini method; Mn is introduced into the CeO2 structure to form Mn-Ce solid solution, which revealed excellent catalytic activity. Solid solutions are alloy phases in which solute atoms are dissolved into solvent lattices while remaining solvent types. The phase boundaries of bimetallic interfaces can provide abundant lattice mismatches, distortions, and defects, which will provide abundant oxygen vacancies and active sites (Wang et al. 2018). Gong et al. (2019a) synthesized heterostructure Ni/NiO nano-catalysts via solgel method and displayed excellent decomposition rate for ozone since NiO promotes the transfer of active oxygen to the surface. However, these catalysts have low humidity resistance; thus, it should be further improved for real application in the field. FeOx is a common MnOx doping material due to its high specific surface area (SBET), great stability, and high capability for electron transfer. Zhang et al. (2014) indicated that addition of FeOx into Mn2O3 reveals excellent activity toward NO removal thanks to its high SBET and abundant Mn3+, Fe3+, and chemisorbed oxygen species. Besides, strong interaction between Fe and Mn facilitated the catalyst durability and humidity resistance (Lian et al. 2015, Wang et al. 2020b). At present, there are few reports on the simultaneous destruction of HCHO and ozone for indoor air quality management. In this study, we synthesized a series of novel MnCeNiOx catalysts by the simple Pechini method, and then modified it by FeOx to further improve the catalytic activity for HCHO destruction at room temperature. It was found that the addition of FeOx contributed to a better performance in HCHO oxidation. The FeOx-MnCeNiOx samples were characterized by various techniques, including N2 physisorption, XRD, SEM, XPS, H2-TPR, O2-TPD, and TGA. The high performance was correlated with a high ratio of Mn3+/Mn4+ and Ce3+/Ce4+ and abundant surface-adsorbed oxygen species. The effects of initial O3 concentration and water vapor with the catalytic performance were also investigated to demonstrate fundamental insights into the superior activity of the FeOx-MnCeNiOx catalysts. Based on in situ DRIFTS results, a viable mechanism of formaldehyde decomposition is presented. This study provides a high-activity, economical, environment-friendly, and simple catalytic method for the degradation of low-concentration HCHO. This work offers a valuable route to simultaneously catalytic destruction of HCHO and ozone for indoor air quality environmental management. Experimental Synthesis of MnxCeNiOx catalysts All reagents were of analytical grade and without purification. Porous MnxCeNiOx catalysts were prepared by the Pechini method. Stoichiometric amounts of Mn(NO3)2•6H2O, Ni(NO3)7•6H2O and Ce(NO3)3•6H2O were dissolved in deionized water under ultrasonication to obtain a homogeneous solution and placed in a beaker with magnetic stirrer; then citric acid and ethylene glycol were added to the solution as chelators. Afterwards, the solution was stirred continuously at 80 °C until it formed a stabilized pale green gel. After drying at 100 °C overnight, it turned into a puffy and porous dry solid. Finally, the fresh MnCeNiOx catalysts were obtained by calcination at 500 °C in air for 6 h at a rate of 10 °C/min. The final products are named as MCN and M2CN based on varying Mn/Ce atomic ratios (Mn/Ce = 1:1, 2:1), respectively. Pure MnO2, CeO2, and NiO was synthesized by the same Pechini method. Synthesis of FeOx-MnxCeNiOx catalysts FeOX were prepared following a previous work (Chen et al. 2014). In brief, stoichiometric amounts of Fe(NO3)3•9H2O were dissolved in deionized water under ultrasonication to obtain a homogeneous solution. Afterwards, Na2CO3 was dropwise added into the solution under vigorous stirring at 60 °C until the pH value stabilized at 8. After aging for 4 h, MCN was soaked in the resulting precipitate and dried at 60 °C for 18 h. Then, the final product was kept at 200 °C for 4 h to obtain fresh FeOx-MnCeNiOx catalyst which is denoted as F-MCN. Pure FeOx was synthesized by the same method. The catalysts prepared were then ground and sieved to a particle size of 30–70 mesh before conducting catalyst characterization and activity test. The catalysts obtained were then characterized accordingly as described in the supplementary information. Activity test The schematic of the experimental setup for evaluating HCHO removal via OZCO is shown in Fig. S1. The inlet HCHO concentration was controlled by adjusting N2 flow rate and water bath temperature. Water vapor was produced by passing N2 through a water-containing bottle, and before entering the reactor, the relative humidity (RH) was measured by a humidity sensor (Center 310 RS-232, Taiwan). HCHO was measured by Fourier transform-infrared spectroscopy (Thermo Nicolet 6700, USA). The concentration of outlet CO2 was analyzed by CO2 Analyzer (Thermo 41C, USA), and the outlet CO concentration was measured by the Testo flue gas analyzer (Testo 350, Germany). Ozone was generated by a commercial ozone generator (Dean’s OW-K2/A-O, Taiwan), and the outlet ozone concentration was analyzed using an ozone analyzer (ITRI CMS/HOA, Taiwan). The equations for calculating HCHO mineralization rate and ozone decomposition are shown in supporting information. Results and discussion Crystal structures and morphology analysis The crystal structures of the original MCN, M2CN, and FeOx-treated MCN samples were examined by XRD, as presented in Fig. 1. The diffraction peaks at 2θ = 28°, 33°, 47°, and 56° can be classified as (110), (200), (220), and (311) planes of fluorite-type CeO2 (PDF#43–1002). The position of the diffraction peak is slightly shifted, and the diffraction angle becomes larger compared with the pure CeO2 (Zhang et al. 2019), indicating that Mn ions are added into the CeO2 lattice to form Mn-Ce solid solution. In addition, no obvious peaks of Mn oxides are found in MCN and F-MCN catalysts, which means Mn-Ce solid solution was formed or Mn ions were well dispersed over the catalyst surface. The formation of solid solution will induce more surface defects and surface oxygen species, which is conducive to the adsorption and decomposition of O3 and HCHO molecules under the catalyst surface. The appearance of the peak at 35° is attributed to Mn3O4 (PDF#13–0162) as the content of the Mn ratio is increased. Moreover, the peaks of 37°, 43°, 63°, and 75° were the characteristic peaks of NiO (PDF#47–1049). The FeOx modified catalyst exhibited weak and broad diffraction lines at 33° and 36°, which could be ascribed to Fe5HO8 ·4H2O (PDF#16–0653) or ɑ-Fe2O3 (PDF#33–0664) (Daniells et al. 2005; Hutchings et al. 2006). Among them, the diffraction peak at 35°could be attributed to Fe3Mn3O8. Chen et al. (2011) indicated the Fe3Mn3O8 could greatly enhance the catalytic activity on NOx removal. The strength of the peaks was enhanced by FeOx modification, revealing that the crystallinity increased with the modification. Generally, the average grain size (D) of the catalyst is given by the following Scherrer’s formula [51].Fig. 1 XRD patterns of MCN, M2CN, and F-MCN, respectively 1 D=K×λβ×cosθ where K is a constant (0.89); λ is the X-ray wavelength (0.154056 nm); β is the half-peak width; and θ is the diffraction angle. The as-calculated grain sizes of MCN, M2CN, and F-MCN are 5.175 nm, 7.1226 nm, and 4.815 nm, respectively. According to previous studies, the F-MCN catalyst has the smallest particle size, which can have a larger SBET (Bariki et al. 2020). To understand the overall microstructural characteristics differences between as-synthesized catalyst, the N2-sorption isotherms were presented in Fig. S2, which are well-defined IV-type isotherms with a typical-type H3 hysteresis loop, indicating the presence of mesopores, along with the generated of mesopores which could produce abundant oxygen vacancies to facilitate HCHO oxidation. The SBET of MCN, M2CN, and F-MCN are 119, 70, and 140 m2/g, respectively. Compared with the MCN catalyst, introduction of FeOx increases the SBET and pore volume, which may be due to the formation of ferrihydrite and hematite under high calcination temperature (Chen et al. 2014). On the other hand, excessive addition of Mn element to catalyst results in a significant reduction of SBET and pore volume, indicating partial blocking of mesopores by Mn particles. Further, the structure of MCN, M2CN, and F-MCN catalysts was investigated by Raman spectra, as presented in Fig. 2. All three as-prepared catalysts displayed a strong peak at 642 cm−1, which is attributed to fluorite F2g mode of CeO2. The appearance of a broad band ranging from 500 to 650 cm−1 is observed after the incorporation of Ni and can be attributed to the interaction between Ce and Ni, forming a mixed crystal; thus generation of oxygen vacancies must be accompanied to (Chagas et al. 2016). The band appearance from 500 to 650 cm−1 corresponds to the strong interaction between Ce and Mn to forming the Mn-Ce solid solution. The strong interaction will benefit the formation of oxygen vacancies due to the charge neutrality. The band at 365 cm−1 for M2CN belongs to the out-of-plane bending modes and symmetric stretching of Mn2O3 groups. After FeOx modification, the appearance of the peak at 310 cm−1 can be observed and attributed to hematite (α-Fe2O3). Raman spectra and XRD results suggest that FeOx has been successfully added to the catalyst to enhance the interaction between Mn-Ce and Ce-Ni mixed crystal, resulting in more oxygen vacancies.Fig. 2 Raman spectra of the MCN, M2CN, and F-MCN catalysts To further investigate the morphology and microstructure of the as-prepared catalysts, SEM images of MCN, M2CN, and F-MCN were obtained, as shown in Fig. S3. It was observed that all three catalysts were mainly composed of block morphology with a rough surface and porous structure. And it can be seen that some fine NiO particles adhere to the surface of CeO2, and these particles have no specific morphology. The diameter of the nanoparticles was 1.5–1.9 μm accompanied by some small pieces with a diameter of 100 nm, besides a portion of the NiO particles agglomerate into spheres with a diameter of 300 nm. Meanwhile, as shown in Fig. S3b, for M2CN with the excessive doping of Mn ions, some lamellar petals attached to their surface appear, and these petals do not cluster into balls. For F-MCN, as shown in Fig. S3c, the metal oxides were highly dispersed under the surface of catalyst, which may be due to the partial sintering by secondary calcination in the preparation of FeOx. Catalyst elemental analysis The chemical composition of surface elements of MCN, M2CN, and F-MCN was determined through XPS, as shown in Fig. 3. Fe 2p spectra can be fitted into two peaks at 710.1 and 712.5 eV, which correspond to Fe2+ and Fe3+, respectively. Moreover, a satellite peak at around 718.0 eV was observed, which can be corresponding to Fe2O3 (Liu et al. 2018). Fe3+ is beneficial to the formation of surface hydroxyl which could improve the redox property of catalyst (Chen et al. 2014). The new species generated with Fe2+ was known as a main reason for the enhancement of chemisorbed oxygen (Wang et al. 2020a). Fe2+ and Fe3+ were both beneficial to the catalytic decomposition of HCHO.Fig. 3 XPS patterns of as-prepared catalyst for a Fe 2p, (ɑ) is MCN, (β) is M2CN, (γ) is F-MCN. b Mn 2p. c Ce 3d. d O1s, respectively For the Mn species in the mixed metal oxide, it can be divided into Mn3+ (about 641.8 eV) and Mn4+ (about 642.7 eV), as shown in Fig. 3b. After modification with FeOx, the Mn3+/Mn4+ ratio increased significantly from 1.23 to 1.33 (Table 1). It is generally recognized that oxygen vacancies will be generated when Mn3+ exists in mixed metal oxide due to the maintained charge neutrality, based on the following reaction:Table 1 BET surface areas, pore volumes, average pore sizes, and elemental compositions of as-prepared catalysts Catalyst SBET (m2/g) SBETa (m2/g) Dpore (nm) Dporea (nm) Vpore (cm3/g) Vporea (cm3/g) Surface atom ratio Fe/Mn atom ratio Mn/Ce/Ni atom ratio Mn3+/Mn4+ Ce3+/Ce4+ Oads/Olatt ICP-OES XPS ICP-OES XPS MCN 119 113 4.66 4.68 0.14 0.13 1.23 0.34 0.22 - - 1:1.09:0.98 1:1.07:0.96 M2CN 70 59 6.11 8.25 0.11 0.09 1.57 0.27 0.18 - - 2:1.08:1.12 2:0.97:0.99 F-MCN 140 136 4.36 5.1 0.16 0.12 1.33 0.63 0.35 0.901 0.88 1:1.03:1.07 1:0.95:0.92 a: Used catalyst 4Mn4++O2-→4Mn4++2e-/□+12O2→2Mn4++2Mn3++□+12O2 where □ represents the oxygen vacancy site (Li et al. 2018). Surface-active oxygen species derived from oxygen vacancies can directly participate in and become conducive to HCHO oxidation (Ye et al. 2022). Meanwhile, the addition of FeOx could induce the enhancement of Mn4+ amount (Wang et al. 2020b). Mn4+ are more conducive to the decomposition reaction over Mn-based catalysts (Chang et al. 2018). As a result of the formation of Fe–Mn solid solution, the incorporation of Fe affected the Mn ion electron state, thus leading the binding energy to exhibit a declining tendency. Regarding the spectra of Ce 3d, five peaks at 881 eV, 883 eV, 899 eV, and 902 eV could be ascribed to Ce3+, while characteristic peaks of Ce4+ present at 882 eV, 888 eV, 897 eV, 900 eV, 907 eV, and 916 eV (Jiang et al. 2020). The atomic ratios of Ce3+/Ce4+ over MCN and F-MCN are 0.34 and 0.63, respectively (Table 1). In general, the oxygen vacancies will be generated when the Ce4+ transforms into nonstoichiometric Ce3+.4Ce4++O2-→4Ce4++2e-/□+12O2→4Ce4++2Ce3++□+12O2 Therefore, the relative concentration of oxygen vacancies was usually proportional to the concentration of Ce3+ cations (Li et al. 2020). The Ni 2p spectra of as-prepared catalysts were divided into three peaks at 852 eV, 857 eV, and 870 eV which were attributed to Ni0 (Song et al. 2017), respectively (Fig. S4). And the peak at 878 eV and 880 eV is corresponding to Ni2+ (Fang et al. 2019), indicating the existence of both NiO and metallic Ni in catalysts. The O 1 s spectrum shows that the peaks at binding energies of 529 eV corresponded to lattice oxygen, while those at 531 eV was corresponding to the surface adsorbed oxygen species (Li et al. 2020), as shown in Fig. 3e. The atomic ratios of Oads/Olatt over MCN and F-MCN are 0.22 and 0.35, respectively (Table 1). Similarly, surface-adsorbed oxygen species usually adsorb on oxygen vacancy. These results illustrate that the F-MCN catalyst prepared possesses higher Mn3+/Mn4+, Ce3+/Ce4+ ratios, and extremely abundant surface-adsorbed oxygen species compared with MCN and M2CN, suggesting that modification with FeOx increases the content of oxygen vacancies. The Fe, Mn, Ce, and Ni contents of as-synthesized catalysts were investigated by ICP-OES, as shown in Table 1. For MCN and M2CN, the ICP results demonstrate that the Mn, Ce, and Ni ratio is close to the theoretical value but not identical, while considering the experimental errors, such a deviation is acceptable. For F-MCN, after FeOx treatment, the intensities of Mn, Ce, and Ni signals were greatly reduced, and simultaneously a new signal of Fe appeared, which proved that Fe entered the catalyst and four elements are almost uniformly distributed in the catalysts. The strong interaction of Fe generated the reduction of Mn, Ce, and Ni ratio. The uniform distribution of the four elements will facilitate the formation of Fe–Mn and Mn-Ce solid solutions. Chemisorption measurements The H2-temperature programmed reduction profiles of MCN, M2CN, and F-MCN catalysts are measured to evaluate their reducibility, as presented in Fig. 4a. The H2-TPR profile of MCN could be divided into two principal reduction peaks, i.e., a little peak at 264 °C and the other stronger peak at 417 °C. The lower-temperature peak may be attributed to the reduction of MnO2/Mn2O3 to Mn3O4 (Liu et al. 2017), in which phase was also detected in XRD results. The higher-temperature peak was corresponding to the reduction of NiO particles to Ni0 (Chagas et al. 2016). Meanwhile, Mn3O4 will further reduce to MnO at 350–400 °C, thus enhancing H2 consumption in higher-temperature peak (Ma et al. 2020). Besides, there were no obvious peaks of CeO2 observed, which indicates that the reduction of Ce4+ accompanied with the reduction of Mn3+. The results demonstrated that the mobility of O atoms was increased when Mn ions were induced into the CeO2 lattice (Zhang et al. 2019). After addition of FeOx, the catalyst reveals three principal reduction peaks. The lower-temperature demonstrates two reduction peaks located at 240 and 285 °C, which were divided into surface oxygen species and MnO2 reduction. The higher-temperature peak was attributed to the reduction of Fe and Ni species (Song et al. 2017). The reduction peaks of Ni species moved towards lower temperature due to the addition of FeOx generated additional oxygen vacancies and facilitated the redox of catalysts. In addition, the peak area for F-MCN was significantly larger than that for MCN and M2CN catalysts, which enhancement may be due to the synergistic effect between Mn and Fe oxides, which can produce oxygen defects and structural distortion (Chen et al. 2019). Moreover, it should be noted that the addition of FeOx reduces the reduction temperature from 264 to 240 °C. However, the reduction temperature over M2CN obviously shifts to a higher temperature (305 °C) due to the excessive addition of Mn. Lower reduction temperature means a higher activity of oxygen vacancies, thus enhancing the reactivity of surface oxygen species (Jiang et al. 2020). Therefore, it is reasonable to deduce that the reactivity of oxygen vacancies increased in the following sequence: F-MCN > MCN > M2CN.Fig. 4 a H2-TPR, b O2-TPD profiles of MCN, M2CN, and F-MCN catalysts The oxygen desorption behavior was investigated by O2-TPD, as presented in Fig. 4b. The peaks could be divided into three parts; i.e., below 300 °C was assigned to the desorption of physically adsorbed oxygen, the middle-temperature peak (350–600℃) is assigned to chemisorbed surface active oxygen, and the high-temperature peak (650–900℃) attributed to the desorption of lattice oxygen species (Gong et al. 2019b). Compared with MCN and M2CN, the catalyst modified with FeOx exhibits the most intensity at 200–600℃. Meanwhile, the peak intensity of M2CN is obviously less than that of MCN; it illustrates that excessive doping of Mn is unfavorable for the formation of oxygen vacancies (Zhang et al. 2020). The results indicated that between Fe and Mn, there exists a strong interaction, resulting in surface defects. Therefore, addition of Fe into MCN could enhance surface adsorbed oxygen content and surface lattice oxygen became more mobile, which is in good agreement with the XPS results. Previous studies have proven that the existence of abundant surface active oxygen species facilitates the destruction of VOC molecules, ozone molecules, and the deep oxidation of intermediates (Gong et al. 2019b). In comparison with MCN and M2CN catalysts, the F-MCN catalyst possesses rich chemisorbed surface-active oxygen species with higher reducibility, mobility, and reactivity due to the strong interaction between Fe and Mn. Therefore, the F-MCN catalyst has a great advantage in HCHO removal. Thermo-catalytic activity During the performance of as-synthesized catalysts for HCHO oxidation at different temperatures, the removal activity was investigated with the initial HCHO concentration of 15 ppm, total gas flow rate of 1200 sccm, and GHSV of 15,000 h−1, as shown in Fig. 5. CO was not detected in our experimental condition. For all three catalysts evaluated, it is observed that the HCHO conversion is less than 50% as the operating temperature is below 50 °C, and the HCHO conversion increases with increasing temperature rapidly above 50 °C. Complete conversions of HCHO over MCN and M2CN catalysts were achieved at 150 °C, while MCN exhibited a higher activity at low temperatures. Catalytic oxidation of HCHO over F-MCN exhibited 100% conversion at 100 °C, which is 50 °C lower than those of MCN and M2CN under HCHO complete conversion. Certainly, catalytic performance of F-MCN is better than that of MCN and M2CN, suggesting that addition of FeOx could enhance the catalytic conversion of HCHO. Besides, the results indicate there is also a striking improvement of catalytic HCHO conversion when a small amount of Mn is added into the catalysts; however, doping too much Mn may lead to the activity decrease.Fig. 5 HCHO conversion as a function of reaction temperature over MCN, M2CN, and F-MCN catalysts, respectively Catalytic degradation of HCHO was obeyed by the Mars van Krevelen model, and the apparent activation energy (Ea) could be calculated by Eq. (2),2 lnk=-EaRT+lnA where R is the molar gas constant; T is the temperature at reaction; and A is the pre-exponential factor. The Arrhenius plots for HCHO conversion of less than 20% over the as-synthetized catalysts are shown in Fig. S5. Under thermo-catalytic conditions, the correlation coefficients of MCN, M2CN, and F-MCN catalysts calculated by linear regression equations are 0.996, 0.991, and 0.990, respectively, showing that the experimental results meet the assumptions. Meanwhile, the calculated activation energies of MCN, M2CN, and F-MCN catalysts for HCHO oxidation are 33 kJ/mol, 34 kJ/mol, and 29 kJ/mol, respectively. The results illustrate that the Ea values for the F-MCN catalysts are reduced with the addition of FeOx, while the M2CN catalyst showed the highest Ea value. Meanwhile, as shown in Table S1, compared with other non-noble metal catalysts, F-MCN shows a reasonable Ea value, which was close to that of noble metal catalysts. In order to better evaluate the impact of modification on catalysts, we introduced the concept of the specific reaction rate (rcat (mol/(g·s)) which was defined as the moles of HCHO being converted per second per gram of catalyst (Du et al. 2018). The rcat was used to evaluate the intrinsic activity of a catalyst. Based on the activity data, we calculated the specific reaction rates (rcat) of three catalysts for HCHO oxidation. The results show that the specific reaction rates of F-MCN, M2CN, and MCN are 3.33 × 10 −6, 1.47 × 10 −6, and 2.07 × 10 −6 mol/(g·s), respectively. In addition, we can use the molar amount of the Mn (active metal) to determine the TOFMn. The results show that the TOFMn of F-MCN, M2CN, and MCN are 2.85, 2.57, and 2.77 × 10 −2 s −1, respectively. The results confirm that modification of FeOx is the main reason for improving the degradation efficiency of HCHO after excluding the effect of specific surface area. Ozone catalytic oxidation activity Three catalysts were investigated to compare their performance for HCHO oxidation with [O3]/[ HCHO] ratio = 4. As shown in Fig. 6(a), the HCHO oxidation efficiency decreased in the following order: F-MCN > MCN > M2CN. Nearly 88% of the HCHO was degraded over F-MCN, while the HCHO conversion rates achieved with MCN and M2CN reached 70% and 65%, respectively. Notably, no CO was detected in the exhaust gas of three catalysts, while CO2 was the major C-species. Previous studies indicate that decomposition of O3 was the first step for the ozone catalytic oxidation process, so the outlet O3 concentration was monitored [15,27]. Under dry conditions, it is exhibited that all three catalysts show excellent performance and complete decomposition of O3 is achieved and maintained within the 8-h test duration.Fig. 6 Catalytic performance of HCHO over as-synthesized catalysts with [O3]/[HCHO] = 4. Conditions: HCHO,15 ppm; total flow rate, 1,200 sccm; O3, 60 ppm; GHSV,15,000 h−1; temperature, 25 °C. a Dry air; b RH = 70% In the real environment, however, there exists a significant amount of water vapor which may inhibit the catalytic activity; therefore, the test was carried out with RH = 70% to explore the humidity effect, as shown in Fig. 6b. The results indicate that F-MCN and MCN exhibit significant O3 decomposition rate, illustrating good humidity resistance. The efficiency of HCHO conversion achieved with F-MCN is up to 100%. Obviously, the rate of HCHO conversion to CO2 achieved with F-MCN catalyst increases along with the presence of water vapor. The abundant oxygen vacancies are very efficient active sites to enhance the water vapor decomposition to surface hydroxyl groups and assist the regeneration of catalyst. However, ozone conversion rate decreased sharply to nearly 60% over M2CN, and HCHO mineralization rate further decreases to 40%, which emphasizes that O3 decomposition rate which is the first step for OZCO process. According to the results of catalyst characterization, the excessive Mn content blocks the surface pores to inhibit oxygen vacancies. Besides, excessive moisture may hinder the exposure of active sites, resulting in competitive adsorption of water molecules under the surface of catalyst, reducing the efficiency of ozone decomposition. It causes the reduction of the surface oxygen species (e.g., O•, •O2−) from ozone decomposition, which is not beneficial for HCHO oxidation. For better analyzing the mechanism of catalyst deactivation, the unit water content (U) to estimate the competitive adsorption can be calculated by the following formulas:3 U=WS where W is the water content in the airflow and S is the SBET of catalyst. The unit water content value of F-MCN, MCN, and M2CN can be approximately calculated to be 0.1, 0.17, and 0.2 g/m2, respectively. The results show that competitive adsorption occurs when the weight of water on the catalyst is greater than 0.2 g per m2. It is noteworthy that after being treated with FeOx, the humidity resistance is enhanced and excellent HCHO and O3 conversion rates are simultaneously achieved. For a better understanding of the kinetics of catalytic process, the degradation data of HCHO could be fitted with the pseudo-first-order kinetics, based on Eq. (4).4 InC0Ct=kapp×t where t is the time of reaction, C0 and Ct were the HCHO concentrations at time 0 and t, and kαpp is the apparent first-order rate constant (min−1). As shown in Fig. S6, the calculated k value of F-MCN is 2.12*10–1 min−1 which is significantly higher than that of MCN and M2CN. The high decomposition efficiency and rate constant of F-MCN revealed the addition of FeOx promotes its superior reactivity, i.e., the active oxygen species, as manifested from the previous characterization. Furthermore, the poor activity of M2CN adequately suggested that the catalytic performance was highly dependent on its morphological structure, physical, and chemical properties. In order to explore the OZCO performance of F-MCN with different O3/HCHO ratios, experimental tests were carried out at 15 ppm HCHO, ozone concentration varying from 30 to 60 ppm, and RH = 70% condition, with the results being shown in Fig. 7. As the [O3]/[HCHO] ratio is controlled at 4, F-MCN catalyst reveals excellent conversion for HCHO and maintains a 100% during the 8-h test period. When [O3]/[HCHO] ratio is reduced to 3, the conversion rate of HCHO decreases slightly, but still remains at 95%. As the [O3]/[HCHO] ratio is dropped to 2, although the catalyst does not show deactivation after 8 h of reaction, the efficiency gradually decreases to 65%. In summary, in the OZCO test with F-MCN, the catalyst exhibits excellent catalytic activity; when [O3]/[HCHO] ratio is reduced to 3, it still maintains a high conversion rate of HCHO (average of 95%). In the absence of a catalyst, no significant product was observed in the reactor even in the presence of ozone, implying that the ozone catalytic oxidation of formaldehyde was triggered by the F-MCN catalyst.Fig. 7 Catalytic performance of HCHO over F-MCN with different [O3]/[HCHO] ratios Conditions: HCHO,15 ppm; total flow rate, 1,200 sccm; O3, 30–60 ppm; GHSV, 15,000 h−1; RH = 70% at room temperature (25 °C) Reusability of the F-MCN catalyst Catalytic recycling of F-MCN Stability is an important factor in judging the potential value of the catalyst in application. Regarding the stability of catalyst, the HCHO concentration of 15-ppm at [O3]/[HCHO] = 4 with MCN and F-MCN as catalyst for 6 h and the average value of each test was taken from three experimental runs. As shown in Fig. 8, after 3 cycles, the final HCHO conversion achieved with MCN was significantly decreased to 73%. However, F-MCN demonstrated an excellent stable HCHO conversion of 93% even after 3 cycles. Besides, deactivation and CO weren’t observed or detected for both catalysts. These data indicated that the F-MCN exhibited excellent durability to HCHO and ozone even under relatively long-term test. After modification with FeOx, the catalyst stability increases, and FeOx provides favorable corrosion resistance, credit to the strong interaction between the Fe and Mn (Wang et al. 2020b).Fig. 8 Stability test of MCN and F-MCN catalyst OZCO of HCHO. Conditions: HCHO,15 ppm; total flow rate, 1,200 sccm; O3, 60 ppm; GHSV, 15,000 h−1; RH = 70% Figure S7 exhibited the XRD patterns of fresh and used F-MCN catalyst. After 3 cycles, the position and strength of peak were almost unchanged, which proved that after being modified by FeOx, the micromorphology of the catalyst has no obvious change during the OZCO tests, indicating that F-MCN had significant corrosion resistance and stability. SBET analysis To disclose the influence of carbon deposition on catalyst pores, the SBET analysis after OZCO reaction was conducted, as shown in Table 1. After reaction, the SBET of MCN decreased slightly from 119 to 113 m2/g, M2CN decreased from 70 to 59 m2/g, and F-MCN decreased slightly from 140 m2/g to 136 m2/g. In the reaction process, part of the intermediates and incomplete reaction products accumulated in the mesopores of the catalyst, resulting in the decrease of SBET. As for the pore size, MCN did not change significantly, while M2CN and F-MCN increased from 6.11 and 4.2 nm to 8.25 and 5.2 nm, respectively. That is generally ascribed to the carbon deposition on the catalyst which blocks some micropores (small holes), resulting in the increase of macropores proportion. Compared with MCN and M2CN catalysts, after modification with FeOx, SBET changes slightly, indicating less deposition of byproducts on catalyst. The results reveal that F-MCN has good resistance to endure the corrosion. Thermogravimetry analysis To compare the species changes and byproducts formation on catalyst surface after OZCO and thermal process, thermogravimetric analysis (TGA) was investigated, and the profiles are presented in Fig. 9. For the fresh catalyst, almost 4.4% weight loss was observed for MCN catalyst, while this value was approximately 5.2% for M2CN catalyst. After modification with FeOx, the weight loss was reduced to 3.1%, suggesting better thermal stability. After OZCO and thermo-catalytic process, the weight losses of all catalysts were increased. Additionally, after thermo-catalytic processes, the weight loss is much higher than that of OZCO process for MCN and F-MCN catalysts. In general, a higher amount of byproduct was deposited on the catalyst when operated at room temperature, while the opposite trend was observed in this study. The results verified that the addition of O3 accelerated the conversion of formate to CO2 and reduced the formation and adsorption of intermediates and byproducts on catalyst surface. Besides, addition of O3 resulted in the formation of COx species, thereby enhanced CO and CO2 desorption (Chen et al. 2020a). However, M2CN catalyst could not utilize ozone completely and then reduces the conversion rate of formate to CO2 due to its weak O3 decomposition capacity. Then, the intermediates and byproduct are easy to accumulate on M2CN surface, and higher weight loss also proves this hypothesis. The above results also verify that catalytic decomposition of O3 is the first step for OZCO. In summary, the FeOx-MCN catalyst exhibits excellent thermostability and less byproduct deposition, suggesting a more stable structure is induced after MCN treated with FeOx.Fig. 9 Weight loss curves of the thermogravimetric analysis for a MCN, b M2CN, and c F-MCN under difference processes Intermediates and mechanism Under wet condition, F-MCN exhibited an excellent HCHO conversion rate, while M2CN HCHO mineralization rate sharply decreased. To shed light on HCHO ozone catalytic oxidation mechanism of F-MCN and catalyst deactivation mechanism over M2CN, the main intermediates formed were analyzed by in situ DRIFTS. The DRIFTS spectra over the M2CN and F-MCN catalysts under exposure to HCHO for 3 h were presented in Fig. 10a and b, respectively.Fig.10 In situ DRIFT spectra of HCHO interaction with ozone at room temperature under wet air conditions over a M2CN and b F-MCN samples as a function of time The absorption bands at 1,360, 1,550, 1,650, and 3,249 cm−1 were observed on the catalysts. According to the previous studies, the characteristic peaks at 1,550 cm−1 were assigned to the asymmetric stretch [νas(COO −)] of formate species, and the weak bands at 1,360 cm−1 can be ascribed to monodentate carbonate species (Ji et al. 2020). The peaks of carbonate and formate species were hardly discriminated due to their co-existence within the wide wavenumber range of 1,300–1,500 cm−1 (Guo et al. 2019). Thus, it is speculated that the accumulation of carbonate and formate species factors is occurred. As shown in Fig. 10a, excessive doping Mn ions results in fewer holes and oxygen vacancies, leading to the formate species and carbonate species accumulating on the surface of catalyst. Zhao et al. (2012) indicated that formate species, carbonate species, and other intermediate products are adsorbed and accumulated on the surface of catalysts which cannot be dissociated from the active sites, inducing the reduction of catalytic activity or even deactivation. Similarly, along with the reaction proceeds, the peak at 3,249 cm−1 gradually increased, which illustrates the water vapor formed during the OZCO process also accumulates on the catalyst surface. Chen et al. (2020b) demonstrated that the continuous formation of H2O results in competitive adsorption. The active sites of the catalyst were firmly occupied by them and could not be desorbed, which was also responsible for a reduction of the catalytic activity. As presented in Fig. 10b, the intensity of formate and carbonate species on the surface of F-MCN catalyst remained almost unchanged during the reaction; the high reactivity of surface oxygen species enhanced HCHO promptly transformed, which indicated that the adsorption and desorption of formate species were in dynamic equilibrium. Meanwhile, the absorbance peak at 3,249 cm−1 displayed little change; the results illustrate that the surface hydroxyl groups were consumed during the reaction, which could be replenished during the reaction of HCHO and O3 under the surface of F-MCN. These results demonstrated that the enhancement of surface-active oxygen groups by FeOx modification could facilitate the intermediate species dissociation and further oxidize them to CO2 after the action of ozone. In summary, the desorption of both intermediates and products from F-MCN catalyst surface is much faster than that from M2CN. Based on the above analysis and previous studies (Zhao et al. 2012), the possible OZCO reaction mechanisms of HCHO over F-MCN catalyst were further proposed and demonstrated in Scheme 1.Scheme 1 Schematic illustration of the HCHO ozone catalytic oxidation over F-MCN Process (A) In this mechanism, O3 was initially adsorbed and activated via the catalyst; afterwards, O3 is decomposed by oxygen vacancy into the active oxygen radicals (O• and •O2−) (Wang et al. 2019). The C atoms of HCHO will be attacked by O• and •O2− under the surface of catalyst and generated to formate species. After that, formate species are further converted into carbonate species, which will quickly decompose into CO2 and H2O(g). Meanwhile, •O2− further reacts with ozone and releases the oxygen molecule from the catalyst surface to attain the reproduction of surface oxygen vacancy (Zhu et al. 2017b). Process (B) Under wet conditions, HCHO molecules firstly adsorb onto OH• groups by hydrogen bonding, while O3 molecules are adsorbed onto activated oxygen vacancy and decomposed into the active oxygen radicals (O•, •O2−) simultaneously. Afterwards, the surface oxygen species (e.g., O•, •O2−, OH•) oxidizes the HCHO adsorbed on the catalyst to unstable formate species. Finally, the •O2− further reacts with ozone, would release as the oxygen molecule from the surface to fill oxygen vacancies under the catalyst surface, and then CO2 and H2O(g) rapidly desorb from the catalyst surface. Because of the existence of the surface OH group, there was almost negligible accumulation of carbonates species under the Process (B). Based on the investigation of the characteristics of catalysts before and after the reaction, it can be found that there is a slight deactivation and micropore blockage of the catalysts. Therefore, it is speculated that in the ozone catalytic oxidation over F-MCN, Process (B) is the predominant mechanism. Conclusion The MnCeNiOx (MCN) was prepared by the Pechini method and further treated with FeOx via impregnation method (F-MCN) and introduced into the OZCO system to discuss the effect of adding FeOx for the catalytic degradation of HCHO at room temperature. Under the dry condition, F-MCN catalyst exhibited better performance of HCHO conversion and ozone decomposition than MCN and M2CN catalysts due to abundant Mn3+ and Ce3+ which were proportional to the intensity of oxygen vacancies and surface adsorbed oxygen and large SBET. The excellent redox property is resulted from to the strong interaction between Mn-Ce, Ce-Ni, and Fe–Mn mixed crystal. Besides, under the wet system, the addition of water vapor enhanced the surface hydroxyl radicals, leading to the complete oxidation of HCHO over F-MCN catalyst under room temperature. Hydroxyl radical is regenerated from the interaction between adsorbed water molecules and surface oxygen species produced from the decomposition of ozone on the catalyst surface. This process is effective for the simultaneous decomposition of HCHO and O3 in indoor air due to water vapor always existing in the real environment. However, additional of water vapor inhibited the decomposition of ozone with M2CN and then reduces the decomposition of HCHO, which confirms that O3 catalytic destruction is the first step for OZCO. In addition, after 72-h continuous operation tests, the F-MCN catalyst exhibits excellent stability, and the crystal phase was almost unchanged, indicating FeOx provides a favorable corrosion resistance. Therefore, F-MCN catalyst revealed high HCHO and ozone removal performance in high RH conditions, demonstrating its great potential for practical applications. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOC 1271 KB) Author contribution RY Liu provided and analyzed the test data and wrote the manuscript. MM Trinh and HT Chuang helped revise the manuscript. MB Chang provided conceptual and technical guidance for all aspects of the project. All authors read and approved the final manuscript. Data availability All data generated or analyzed during this study are included in this published article and its supplementary information files. Declarations Ethics approval and consent to participate Not applicable. 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 Bao W Pt nanoparticles supported on N/Ce-doped activated carbon for the catalytic oxidation of formaldehyde at room temperature ACS Applied Nano Materials 2020 3 2614 2624 10.1021/acsanm.0c00005 Bariki R et al (2020) Facile synthesis and photocatalytic efficacy of UiO-66/CdIn2S4 nanocomposites with flowerlike 3D-microspheres towards aqueous phase decontamination of triclosan and H2 evolution. Appl Catal B :270 Cao R et al (2019) Ammonium-treated birnessite-type MnO2 to increase oxygen vacancies and surface acidity for stably decomposing ozone in humid condition. Appl Surf Sci 495 Chagas CA Copper as promoter of the NiO–CeO2 catalyst in the preferential CO oxidation Appl Catal B 2016 182 257 265 10.1016/j.apcatb.2015.09.033 Chang T Post-plasma-catalytic removal of toluene using MnO2–Co3O4 catalysts and their synergistic mechanism Chem Eng J 2018 348 X15 25 10.1016/j.cej.2018.04.186 Chen Z Low temperature selective catalytic reduction of NOx with NH3 over Fe–Mn mixed-oxide catalysts containing Fe3Mn3O8 phase Ind Eng Chem Res 2011 51 202 212 10.1021/ie201894c Chen B-b et al (2014) FeOx-supported gold catalysts for catalytic removal of formaldehyde at room temperature. Appl Catal B 154–155:73–81 Chen S A novel highly active and sulfur resistant catalyst from Mn-Fe-Al layered double hydroxide for low temperature NH3-SCR Catal Today 2019 327 81 89 10.1016/j.cattod.2018.06.006 Chen G Comparative investigation on catalytic ozonation of VOCs in different types over supported MnOx catalysts J Hazard Mater 2020 391 122218 122235 10.1016/j.jhazmat.2020.122218 32044638 Chen M Facet- and defect-engineered Pt/Fe2O3 nanocomposite catalyst for catalytic oxidation of airborne formaldehyde under ambient conditions J Hazard Mater 2020 395 122628 122638 10.1016/j.jhazmat.2020.122628 32305715 Daniells S The mechanism of low-temperature CO oxidation with Au/Fe2O3 catalysts: a combined Mössbauer, FT-IR, and TAP reactor study J Catal 2005 230 52 65 10.1016/j.jcat.2004.11.020 Du J Low-temperature abatement of toluene over Mn-Ce oxides catalysts synthesized by a modified hydrothermal approach Appl Surf Sci 2018 433 1025 1035 10.1016/j.apsusc.2017.10.116 Du W Indoor air pollution was nonnegligible during COVID-19 lockdown Aerosol and Air Quality Research 2020 20 1851 1855 10.4209/aaqr.2020.06.0281 Fang S Insights into the thermo-photo catalytic production of hydrogen from water on a low-cost NiOx-loaded TiO2 catalyst ACS Catal 2019 9 5047 5056 10.1021/acscatal.9b01110 Gong S Heterostructured Ni/NiO nanocatalysts for ozone decomposition ACS Applied Nano Materials 2019 3 597 607 10.1021/acsanm.9b02143 Gong S Highly active and humidity resistive perovskite LaFeO3 based catalysts for efficient ozone decomposition Appl Catal B 2019 241 578 587 10.1016/j.apcatb.2018.09.041 Guo J Review on noble metal-based catalysts for formaldehyde oxidation at room temperature Appl Surf Sci 2019 475 237 255 10.1016/j.apsusc.2018.12.238 Hutchings GJ Role of gold cations in the oxidation of carbon monoxide catalyzed by iron oxide-supported gold J Catal 2006 242 71 81 10.1016/j.jcat.2006.06.001 Ji J et al (2020) Potassium-modulated δ-MnO2 as robust catalysts for formaldehyde oxidation at room temperature. Appl Catal B Environ 260 Jiang F Insights into the influence of CeO2 crystal facet on CO2 hydrogenation to methanol over Pd/CeO2 catalysts ACS Catal 2020 10 11493 11509 10.1021/acscatal.0c03324 Jung H Changes in acetaldehyde and formaldehyde contents in foods depending on the typical home cooking methods J Hazard Mater 2021 414 125475 125482 10.1016/j.jhazmat.2021.125475 33647618 Li X Oxygen vacancies induced by transition metal doping in gamma-MnO2 for highly efficient ozone decomposition Environ Sci Technol 2018 52 12685 12696 10.1021/acs.est.8b04294 30346750 Li X Detrimental role of residual surface acid ions on ozone decomposition over Ce-modified gamma-MnO2 under humid conditions J Environ Sci (china) 2020 91 43 53 10.1016/j.jes.2019.12.004 32172981 Lian Z Decomposition of high-level ozone under high humidity over Mn–Fe catalyst: the influence of iron precursors Catal Commun 2015 59 156 160 10.1016/j.catcom.2014.10.005 Liu Y Removing surface hydroxyl groups of Ce-modified MnO2 to significantly improve its stability for gaseous ozone decomposition J Phys Chem C 2017 121 23488 23497 10.1021/acs.jpcc.7b07931 Liu F One-step synthesis of nanocarbon-decorated MnO2 with superior activity for indoor formaldehyde removal at room temperature Appl Catal B 2018 235 158 167 10.1016/j.apcatb.2018.04.078 Ma J et al (2020) Novel CeMnaOx catalyst for highly efficient catalytic decomposition of ozone. Appl Catal B Environ 264 Nielsen GD Re-evaluation of the WHO (2010) formaldehyde indoor air quality guideline for cancer risk assessment Arch Toxicol 2017 91 35 61 10.1007/s00204-016-1733-8 27209488 Song S Heterostructured Ni/NiO composite as a robust catalyst for the hydrogenation of levulinic acid to γ-valerolactone Appl Catal B 2017 217 115 124 10.1016/j.apcatb.2017.05.073 Wang Y Role of oxygen vacancies and Mn sites in hierarchical Mn2O3/LaMnO3-δ perovskite composites for aqueous organic pollutants decontamination Appl Catal B 2019 245 546 554 10.1016/j.apcatb.2019.01.025 Wang H Highly efficient WO3-FeOx catalysts synthesized using a novel solvent-free method for NH3-SCR J Hazard Mater 2020 388 121812 10.1016/j.jhazmat.2019.121812 31836367 Wang Q et al (2018) Ultrathin two-dimensional BiOBrxI1-x solid solution with rich oxygen vacancies for enhanced visible-light-driven photoactivity in environmental remediation. 236:222–232 Wang Y et al (2020b) Study on the structure-activity relationship of Fe-Mn oxide catalysts for chlorobenzene catalytic combustion. Chem Eng J 395 Ye J et al (2022) Hierarchical Co3O4-NiO hollow dodecahedron-supported Pt for room-temperature catalytic formaldehyde decomposition. Chem Eng J 430 Zhang M Catalytic oxidation of NO with O2 over FeMnOx/TiO2: effect of iron and manganese oxides loading sequences and the catalytic mechanism study Appl Surf Sci 2014 300 58 65 10.1016/j.apsusc.2014.02.002 Zhang Y Simultaneously catalytic decomposition of formaldehyde and ozone over manganese cerium oxides at room temperature: promotional effect of relative humidity on the MnCeOx solid solution Catal Today 2019 327 323 333 10.1016/j.cattod.2018.04.027 Zhang C LaMnO3 perovskites via a facile nickel substitution strategy for boosting propane combustion performance Ceram Int 2020 46 6652 6662 10.1016/j.ceramint.2019.11.153 Zhao D Enhanced effect of water vapor on complete oxidation of formaldehyde in air with ozone over MnOx catalysts at room temperature J Hazard Mater 2012 239–240 362 369 10.1016/j.jhazmat.2012.09.009 Zhu B A novel process of ozone catalytic oxidation for low concentration formaldehyde removal Chin J Catal 2017 38 1759 1769 10.1016/S1872-2067(17)62890-0 Zhu G Surface oxygen vacancy induced α-MnO2 nanofiber for highly efficient ozone elimination Appl Catal B 2017 209 729 737 10.1016/j.apcatb.2017.02.068	36469276	PMC9734528	NO-CC CODE	2022-12-14 23:28:29	no	Environ Sci Pollut Res Int. 2022 Dec 5;:1-14	utf-8	Environ Sci Pollut Res Int	2,022	10.1007/s11356-022-24543-y	oa_other
==== Front Drugs Drugs Drugs 0012-6667 1179-1950 Springer International Publishing Cham 36479687 1817 10.1007/s40265-022-01817-w Review Article The Struggle to End a Millennia-Long Pandemic: Novel Candidate and Repurposed Drugs for the Treatment of Tuberculosis http://orcid.org/0000-0001-9608-2589 Edwards Brett D. [email protected] http://orcid.org/0000-0002-6682-4155 Field Stephen K. grid.22072.35 0000 0004 1936 7697 Division of Infectious Diseases and Tuberculosis Services, Alberta Health Services, Department of Medicine, Cumming School of Medicine, University of Calgary, Peter Lougheed Centre, 3500, 26 Avenue NE, Calgary, AB T1Y6J4 Canada 7 12 2022 2022 82 18 16951715 20 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. This article provides an encompassing review of the current pipeline of putative and developed treatments for tuberculosis, including multidrug-resistant strains. The review has organized each compound according to its site of activity. To provide context, mention of drugs within current recommended treatment regimens is made, thereafter followed by discussion on recently developed and upcoming molecules at established and novel targets. The review is designed to provide a clinically applicable understanding of the compounds that are deemed most currently relevant, including those already under clinical study and those that have shown promising pre-clinical results. An extensive review of the efficacy and safety data for key contemporary drugs already incorporated into treatment regimens, such as bedaquiline, pretomanid, and linezolid, is provided. The three levels of the bacterial cell wall (mycolic acid, arabinogalactan, and peptidoglycan layers) are highlighted and important compounds designed to target each layer are delineated. Amongst others, the highly optimistic and potent anti-mycobacterial activity of agents such as BTZ-043, PBTZ 169, and OPC-167832 are emphasized. The evolving spectrum of oxazolidinones, such as sutezolid, delpazolid, and TBI-223, all aiming to exceed the efficacy achieved with linezolid yet offer a safer alternative to the potential toxicity, are reviewed. New and exciting prospective agents with novel mechanisms of impact against TB, including 3-aminomethyl benzoxaboroles and telacebec, are underscored. We describe new diaryloquinolines in development, striving to build on the immense success of bedaquiline. Finally, we discuss some of these compounds that have shown encouraging additive or synergistic benefit when used in combination, providing some promise for the future in treating this ancient scourge. issue-copyright-statement© Springer Nature Switzerland AG 2022 ==== Body pmcKey Points The mycobacterial cell wall is composed of three layers (the outer mycolic acid, the middle branched arabinogalactan, and the inner peptidoglycan) that offer unique target sites for tuberculosis (TB) therapies that are being exploited in drug development. The discovery and development of bedaquiline was revolutionary in TB therapy and has provided the impetus for exciting further investigation into additional diarylquinolines. Pretomanid possesses potent bactericidal activity and sterilizing capability for non-replicating bacilli that is additive to that achieved with bedaquiline and linezolid. Repeated clinical studies suggest the extensive activity of linezolid against TB, and recent data indicate its toxicity may be mitigated with therapeutic drug (trough) monitoring. Benzothiazinones and their derivatives offer some of the most potent antimycobacterial activity seen in vitro, providing great optimism for further clinical study. Telacebec (Q203) and GSK656 are new and exciting therapies designed to target novel regions of the mycobacterial cell Introduction It has been 140 years since Robert Koch discovered the bacillus identified as Mycobacterium tuberculosis (MTB), yet despite major international efforts, tuberculosis (TB) remains a major health problem. There is archaeological evidence of TB in the Neolithic period, and it has reached epidemic proportions in recent centuries [1, 2]. It was also recognized and named by the ancients: schachepheth in the Old Testament and phthisis by the ancient Greeks [1]. John Bunyan described it as the Captain of all these men of death in 1680, and it was termed the White Plague in eighteenth century England [3]. Unlike the most common causes of death now, cardiovascular disease, cancer and respiratory diseases, which generally kill the elderly, TB struck people down during the prime of their lives. Apart from the COVID-19 pandemic, TB is the most common cause of death from an infectious disease worldwide and the tenth most common cause of death overall [4]. The COVID-19 pandemic has disrupted TB care globally resulting in a reduction in new case notifications in 2020, the provision of fewer treatment regimens for rifampin-resistant disease, and fewer preventative treatments, yet more than 1.5 million died from TB in 2020, an increase from 2019 [4]. Challenges with current therapy against TB include prolonged duration of therapy and intolerance due to side effects. Additionally, particularly in settings of high TB prevalence, access to healthcare and cost remain major issues. Drug resistance, which is becoming more prevalent, complicates therapy and reduces the likelihood of successful treatment since it requires longer and more complex medication regimens than drug susceptible (DS)-TB. Additionally, it is often accompanied by more major adverse effects, is considerably more expensive to treat, and success rates are lower than with regimens for DS disease [4]. After decades without newly developed antimycobacterials, or repurpose of alternative antibacterials for use in TB, novel agents were approved for use in 2012 [5] and 2014 [6]. By utilizing genome sequencing, the identification of novel sites of action has prompted the further pursuit of similar agents that may be even more potent and less toxic than traditional therapies (Table 1). Time will tell with clinical studies ongoing, but it is reasonable to state that the pipeline of anti-TB medication is promising. Helpful updates on the status of medication trials are available and updated regularly [7].Table 1 Classification and status of anti-tubercular compounds in development Drug Class Site of action Development phase References Bedaquiline Diarylquinoline Mycobacterial ATPase Marketed [18, 146, 147, 149] Sudapyridine (WX-081) Diarylquinoline Mycobacterial ATPase 2 [167] TBAJ-876 Diarylquinoline Mycobacterial ATPase 1 [18, 19] TBAJ-587 Diarylquinoline Mycobacterial ATPase 1 [160, 165] Telacebec (Q203) Imidazopyridine Cytochrome BC1 complex 2 [67, 169, 170] Delamanid Nitroimidazole Mycolic acid synthesis/nitric acid Marketed [38, 41, 42] Pretomanid Nitroimidazole Mycolic acid synthesis/nitric acid 3 [16, 41, 53] BVL-GSK098 Amido-piperidine MymA, ethionamide enzyme activator 1 [34] SQ109 Ethylenediamine Mycolic acid transport MmpL3 2 [59, 60] BTZ043 Benzothiazinone DprE1 inhibitor, arabinoglycan synthesis 2 [78, 79] PBTZ169 Benzothiazinone DprE1 inhibitor, arabinoglycan synthesis 1 [76, 78] OPC-167832 Carbostyril DprE1 inhibitor, arabinoglycan synthesis 2 [73, 78] TBA-7371 1,4-azaindole DprE1 inhibitor, arabinoglycan synthesis 2 [76, 87] Sanfetrinem Carbapenem Peptidoglycan synthesis 2 [97] Moxifloxacin Fluoroquinolone DNA gyrase Marketed [182] Fodrepodacin (SPR720) Aminobenzimidazole DNA gyrase B 2 [184] VXC-486 Aminobenzimidazole DNA gyrase B 2 [187] Rifapentine Rifamycin RNA Polymerase Marketed [12, 138] Linezolid Oxazolidinone Protein Synthesis Marketed [16, 99] Sutezolid Oxazolidinone Protein Synthesis 2 [119–121] Tedizolid Oxazolidinone Protein Synthesis No current TB trials [109, 111–113] Delpazolid Oxazolidinone Protein Synthesis 2 [125] TBI-223 Oxazolidinone Protein Synthesis 1 [40] OTB-658 Oxazolidinone Protein Synthesis Pre-clinical [130, 131] GSK656 Oxaborole leucyl t-RNA, protein synthesis 2 [134, 135, 137] Clofazimine Riminophenazine Reactive Oxygen Species Marketed [99, 176] Pyrifazimine (TBI-166) Riminophenazine Reactive Oxygen Species 1 [66, 175, 176] GSK2556286 Pyrimidine-2,4-dione Cholesterol Metabolism 1 [22, 188] FNDR-20081 Quinoline Uncertain Pre-clinical - The purpose of this review is to provide a clinically relevant perspective on several new and repurposed compounds that are deemed to show promising early results and to describe the available clinical data for these drugs. Undoubtedly, some of these compounds will not be marketed. To provide a framework for organization, we have structured the review based on sites of activity of compounds. This includes providing sufficient detail to, for example, emphasize the unique targets within the various levels of the cell wall (Fig. 1) [8]. In preparation for this review, the PubMed database was searched using treatment of TB, DR-TB, rifampin-resistant (RR)-TB, MDR-TB, and extensively drug-resistant (XDR)-TB, and for each drug individually. As well, the bibliographies of the selected articles were reviewed for relevant articles. Moreover, the clinicaltrials.gov and www.newtbdrugs.org sites were reviewed (Fig. 2).Fig. 1 The mycobacterial cell wall with target sites of current therapies and those in development. Reproduced with permission from Elsevier Masson from: Shanib Bhat Z, Ahmad Rather M, Maqbool M, Lah HU, Khalid Yousuf S, Ahmad Z. Cell wall: A versatile fountain of drug targets in Mycobacterium tuberculosis. Biomed Pharmacother. 2017;95:1520–34 [8] Fig. 2 2022 global new tuberculosis (TB) drug pipeline, updated 3 November 2022. Reproduced with permission from the: Stop TB Partnership Working Group on New TB Drugs [7]. Figure is available at: www.newtbdrugs.org/pipeline/clinical Current Key Drug Targets Current standard of care of DS-TB remains a combination of isoniazid, rifampin, pyrazinamide, and ethambutol [9, 10]. Both isoniazid and ethambutol are active against mycobacterial cell wall production, inhibiting outer mycolic acid production and the middle arabinogalactan level, respectively. Pyrazinamide is a prodrug and enzymatically converted to a highly bactericidal and acidic substrate, pyrazinoic acid, that is believed to interrupt the production of coenzyme A through aspartate decarboxylase [11]. Rifampin remains the cornerstone of DS-TB treatment due to its combined bactericidal and sterilizing effects, allowing for treatment shortening from at least 18 months to 9. It mediates its concentration-dependent effect through inhibition of the DNA-dependent RNA polymerase, which seems to interfere with the growing RNA strand [12]. For multidrug-resistant TB (MDR-TB), current guidelines advocate for multidrug treatment generally consisting of bedaquiline, linezolid, moxifloxacin or levofloxacin, clofazimine, and cycloserine or terizidone [13–15]; however, growing data on the potency of agents such as bedaquiline, oxazolidinones, and pretomanid have provided justification for shortened treatment regimens, including the BPaL and BPaL(M) regimens [16, 17]. If the isolate is susceptible, pyrazinamide is also recommended for the first 6 months [9]. Recent research has aimed to develop antimycobacterial drugs that exploit novel sites to counter growing antimycobacterial resistance. This research has identified some of the most potent chemicals by in vitro testing to date, including diarylquinolines such as bedaquiline and its derivatives, which target adenosine triphosphate (ATP) synthesis and uncoupling of the electron transport chain [18, 19]. Mycobacterial Cell Wall It is prudent to briefly deconstruct the mycobacterial cell wall as this structure offers multiple potential sites for drug targeting. The cell wall consists of an extensive lipid-carbohydrate network that serves as an impermeable defense against hydrophilic compounds and has long been recognized as an important antimicrobial target [8]. It possesses an inner phospholipid bilayer cell membrane. Beyond this lies (i) the crosslinked peptidoglycan, (ii) an intermediate, branched arabinogalactan polysaccharide, (iii) and the outer waxy long-chain mycolic acids [20]. The inner peptidoglycan component of the cell wall possesses some unique properties from Gram-positive and Gram-negative bacteria, and specific to TB is the oxidation of N-acetylmuramic acid to N-glycolylmuramic acid, which may increase pathogenicity as well as stability from host intracellular lysozyme [8]. Drugs previously recognized to interfere with mycobacterial peptidoglycan synthesis include cycloserine and carbapenems [8]. The middle arabinogalactan layer is composed of branches of arabinose and galactose and is generated through multiple steps including the use of decaprenylphosphoryl-d-arabinose (DPA). This is a recognized target for nitro-benzothiazinones (BTZs) and dinitrobenzamides, which inhibit decaprenyl-phosphoribose 2′ epimerase (DprE1) enzymatic activity in the generation of DPA [20]. Ethambutol is another agent that targets this middle layer by inhibiting the arabinosyl transferase enzyme [21]. The outer cell wall layer contains mycolic acids: long-chain fatty acids consisting of an α-alkyl and a β-hydroxy chain. It also contains lipoglycans, phospholipids, glycopeptidolipids, and trehalose mycolates [22]. The mycolic acid-containing outer layer is covalently linked to the arabinogalactan layer, which, in turn, is covalently linked to the peptidoglycan layer forming the mycolyl-arabinogalactan-peptidoglycan complex [8, 23]. Generation of the outer mycolic acid layer utilizes many enzymatic steps already targeted by multiple antimycobacterials. Isoniazid, ethionamide, delamanid, and pretomanid all target this component of the cell wall. Finally, a protein and polysaccharide capsule surrounds the cell wall [8, 23]. Outer Mycolic Acid Layer Isoniazid Isoniazid (INH) was introduced 70 years ago and remains a first-line drug for the treatment of active and latent TB infection (LTBI) [24]. It inactivates enoyl-ACP reductase [inhA] and beta-ketoacyl ACP synthase [kasA], important enzymes involved in the elongation of long chain fatty acids [25–27]. INH is a prodrug activated by the catalase-peroxidase [katG] enzyme, generating an isoniazid-nicotinamide adenine dinucleotide (NAD) adduct that impairs inhA activity, interfering with its ability to generate NAD (+) and disrupting mycolic acid chain elongation [24]. Mutation of the katG gene is the most common cause of INH resistance, followed by mutations of the inhA gene [24]. Other mycobacterial gene mutations that can cause resistance are less common and act by altering drug metabolism or by facilitating drug efflux [24]. For the years 2003–2017, the global prevalence of INH resistance was 7.4% among new TB cases and 11.4% among previously treated patients; 8.5% of all cases worldwide are INH monoresistant [28, 29]. Among isoniazid monoresistant cases, 78.6% had katG mutations and 14.6% had both katG and inhA mutations [30]. Ethionamide Ethionamide is a structural analogue of INH and also targets inhA [31]. Similar to INH, ethionamide is a prodrug but is activated by another enzyme, the mycobacterial monooxygenase, ethA, so it maintains its activity in the presence of katG mutations but not inhA mutations [32]. It is considered a second-line drug because of adverse effects. Prothionamide is another thioamide, similar to ethionamide, but may be better tolerated in MDR-TB treatment regimens [33]. BVL-GSK098 BVL-GSK098, an amido-piperidine, boosts bactericidal activity of ethionamide and prothionamide, restoring sensitivity in ethionamide-resistant MTB strains. It is believed to stimulate an additional enzymatic activator for the prodrug ethionamide, called MymA [34]. It is currently in phase I development and a phase IIa trial is planned [35]. Thiacetazone Thiacetzone is a prodrug that is activated by ethA and once activated inhibits mycolic acid cyclopropane synthesis. It is an inexpensive drug but unfortunately may cause hepatotoxicity and severe cutaneous reactions, particularly in patients coinfected with HIV [36]. Thiadiazoles-Based Chemical Inhibitors An open collaborative model for TB lead optimization (ORCHID) consortium is developing thiadiazole-based chemical inhibitors that directly inhibit inhA, circumventing the need for cellular activation [37]. GSK693 is a thiadiazole that demonstrated good activity inside and outside macrophages and is not cytotoxic to hepatocytes nor does it inhibit the human ‘Ether-a-go-go’-related gene, KCNH2 (hERG) [22]. Currently, no trials with this drug are registered with clinicaltrials.gov (accessed 6 June 2022). Delamanid (Previously OPC-67683) Delamanid is one of two recently approved novel anti-mycobacterial agents for the treatment of TB. While global experience with it appears significantly less than with the other recently approved agent bedaquiline, there are numerous ongoing studies of its use [7]. Delamanid is a first-of-its-kind bicyclic nitroimidazooxazole derivative and is a prodrug that requires metabolic activation by the mycobacterial f420 enzyme to have anti-TB activity [35]. The catalytic enzyme responsible for this activation is Rv3547 [38]. It is not active against routinely encountered bacteria [39]. Similar to INH, its primary mechanism of action is through inhibition of mycolic acid synthesis providing it bactericidal activity; however, it may also exert activity through generation of reactive oxygen intermediates, impairing respiration [38]. It is proposed to inhibit mycolic acid biosynthesis at a site distinct from other agents like isoniazid and ethionamide, specifically by inhibiting methoxy-mycolic acid and keto-mycolic acid, but not α-mycolic acid biosynthesis [40]. Delamanid has activity against both rifampin-susceptible and rifampin-resistant TB and is significantly potent in vitro, even more so than pretomanid [40–42]. The usual dose of 100 mg twice daily generates sufficient concentrations above the epidemiological cut-off values in most individuals [42]. It exerts activity against dormant, non-replicating bacilli, as well as those harboured within macrophages [40, 43]. Delamanid may manifest a low minimum inhibitory concentration (MIC) against some slow-growing nontuberculous mycobacteria, but it frequently has elevated MICs to the most commonly encountered organisms, including M. avium and M. intracellulare, suggesting it is unlikely to offer clinical utility, and it also has no activity against rapidly growing mycobacteria species [39]. Resistance to delamanid is rare so far, but when encountered is frequently due to mutations in the nitroreductase that activates it [44]. Cross-resistance to other nitroimidazoles, such as pretomanid, may occur [41, 42]. Bioavailability is increased with fatty food consumption [45, 46]. Uniquely, delamanid is metabolised by circulating albumin and does not have any notable impact on cytochrome P450 (CYP) enzymes, limiting the potential for significant drug-drug interactions [39, 43, 47]. The half-life is approximately 30 h and most excretion is through the feces, with minimal urinary excretion. Overall, delamanid appears well tolerated, with the main reported adverse events (AEs) including mild gastrointestinal symptoms or QTc prolongation [39, 48]. The sentinel sponsored phase II study that prompted application for European Medicines Agency (EMA) conditional approval of delamanid demonstrated expedited sputum-culture conversion at 2 months in comparison to placebo in patients with MDR-TB or XDR-TB, when combined with an optimal background regimen (45.4% vs. 29.6%, p = 0.008) [6, 48]. Long-term observational follow-up of these patients identified that those who received at least 6 months of delamanid (vs. none or 2 months) had more favorable outcomes (74.5% vs. 55%, relative risk (RR) 1.35; confidence interval (CI) 1.17–1.56; p < 0.001) and lower mortality (1% vs. 8.3%, p < 0.001) [49]. Additionally, some post-marketing studies have also shown promising culture conversion results [50, 51]. Nonetheless, delamanid trials have had conflicting results. In contrast to the above studies, when added to an optimized background regimen for MDR-TB, delamanid did not shorten the time to culture conversion versus placebo (51 days vs. 57 days; hazard ratio (HR) 1.17; 95% CI 0.91–1.51, p = 0.2157) [52]. Due to these discrepant results and limited data available about efficacy (including mortality) when added to background regimens, the medication was categorized as a Group C medication (for addition when other preferred medications cannot be utilized) by the World Health Organization (WHO), and was unable to be included in the most recent recommendations by the official ATS/CDC/ERS/IDSA clinical practice guideline [14]. Pretomanid Pretomanid is a pro-drug nitroimidazooxazine molecule that also inhibits mycolic acid synthesis in replicating bacilli. However, it is also active against non-replicating bacilli, releasing reactive oxygen species including nitric oxide under anaerobic conditions [53]. These multiple functions provide it with potent bactericidal and sterilizing activity in mouse studies [54]. Mutations mediating cross-resistance between pretomanid and delamanid are already recognized, and are most commonly located in genes required for activation of the nitroimidazoles [55], glucose-6-phosphate dehydrogenase (FGD1) or the deazaflavin cofactor f420 [56]. Activity is present against both DS- and MDR-TB (including XDR-TB) and yet it has no significant activity against Gram-positive and Gram-negative bacteria [8]. The individual contribution of pretomanid to highly effective combination treatments has been demonstrated in mouse models. Specifically, compared to the combination of bedaquiline, moxifloxacin, and pyrazinamide, adding pretomanid (BPaMZ) resulted in a greater reduction in lung colony forming unit (CFU) counts and reduced the risk of selecting bedaquiline-resistant isolates and relapse [57, 58]. Furthermore, pretomanid also affords additional sterilizing activity when added to a regimen of bedaquiline and linezolid (BPaL) [57]. The Nix-TB trial was the premier introduction to the clinical use of pretomanid in combination with bedaquiline and linezolid in XDR-TB or those with MDR-TB intolerant or not responding to standard therapy [16]. After 6 months of this combination, at least 90% of patients were found to have a favorable outcome (absence of failure, relapse, retreatment) in both the intention-to-treat and per-protocol analyses up to 6 months beyond treatment completion. This 109-person phase III study led the way for expedited approval for pretomanid by the US Food and Drug Administration (FDA) in 2019 solely as part of a three-drug regimen for complicated MDR-TB and XDR-TB [16]. Notably, a Grade 3 or higher AE was reported by 57% of the patients on the regimen, although this was likely impacted by linezolid. This included 26.6% with nervous system disorders, 18.3% with musculoskeletal disorders, 12.8% with hematologic disorders, and scattered incidence of various metabolic disorders (acidosis, elevated lipase/amylase, electrolyte abnormalities). Additionally, one of the main drawbacks of the study was the absence of a comparator group. The SimpliciTB phase II/III trial builds off the recognized high bactericidal activity of the combination of bedaquiline, pretomanid, moxifloxacin, and pyrazinamide (BPaMZ), which benefits from the synergistic activity between pyrazinamide and bedaquiline [58]. This study is ongoing and compares the efficacy and safety of the above combination for 4 months compared to standard 6-month treatment for DS-TB and this regimen for 6 months for DR-TB. TB-PRACTECAL is another completed phase II/III trial evaluating shortened regimens of bedaquiline, pretomanid, moxifloxacin, linezolid, and clofazimine, with publication pending [17]. Ethylenediamines SQ109 is an ethylenediamine with a similar structure to ethambutol but maintains good activity against ethambutol-resistant MTB strains. It inhibits mycobacterial membrane protein large 3 (MmpL3), which plays an important role in the transportation of mycolic acids, in the form of trehalose monomycolate, to the outer part of the cell wall, and may also reduce the arabinogalactan underneath [59]. It also targets other synthetic enzymes for energy production and efflux pump mechanisms [40, 60]. It has been demonstrated to have in vitro activity against both DS- and drug-resistant (DR)-TB. An in vitro study demonstrated an excellent rate of killing that was superior to single drug use of sutezolid (PNU-100480); however, combination of the two showed additive efficacy even at sub-MIC concentrations [61]. Although safe and well tolerated, in combination with rifampin SQ109 did not enhance early bactericidal activity (EBA) over the first 4 days of treatment, likely because rifampin accelerated its metabolism [62]. Additionally, treatment with SQ109 alone did not show any significant activity in vivo. This is in contrast to a prior in vitro study suggesting synergy with rifampin [63]. In comparison, in combination with bedaquiline, SQ109 improved EBA four- to eightfold [63]. In a multi-arm study, the median time to sputum culture conversion did not improve by substituting SQ109 for ethambutol in standard therapy with rifampin, isoniazid, and pyrazinamide, and this arm of the trial was stopped early for lack of efficacy [64, 65, 66]. SQ109 is generally well tolerated, and rates of adverse effects were similar to those seen in the two treatment groups. There are no active studies with SQ109 listed on the clinicaltrials.gov website (accessed 4 June 2022). Arabinogalactan Middle Layer The middle layer of the cell wall consists of branched arabinogalactan, a polysaccharide chain consisting of galactose and arabinose sugars in a furanose form, covalently linked to both the adjacent peptidoglycan layer and the outer mycolic acid layers [8, 40]. Ethambutol and its analogues, capreomycin and DprE1 enzyme inhibitors, interfere with arabinogalactan metabolism [22, 38, 63, 64]. Ethambutol disrupts arabinogalactan synthesis by inhibiting the enzyme arabinosyl transferase [67]. Capreomycin is believed to alter mycolic acid synthesis in addition to inhibiting protein synthesis [68]. DprE1 is an essential flavo-enzyme required for the formation of D-arabinofuranose, a component of arabinogalactan and lipoarabinomannan [7, 69]. In combination with DprE2, it epimerizes decaprenylphosphoryl ribose to decaprenylphosphoryl arabinose, an essential step for arabinogalactan and lipoarabinomannan biosynthesis [70, 71]. Several drug classes are undergoing development that target this essential enzyme for cell wall synthesis including the benzothiazinones, a carbostyril derivative, and azaindoles [72–74]. Drugs that target DprE1 benefit from it being situated in the periplasmic space of the cell wall, thereby avoiding efflux and cytoplasmic resistance mechanisms in resistant strains [75]. BTZ-043 BTZ-043 is one of the lead compounds of the class of medications called benzothiazinones first discovered in 2009 that are suicide inhibitors of DprE1 [76]. It acts by forming an adduct with the Cys387cysteine residue [77]. BTZ-043 inhibits mycobacterial cell wall synthesis by blocking DprE1 and the subsequent formation of arabinogalactan and arabinomannan in the middle arabinogalactan component of the TB cell wall [8]. It is one of the most highly bactericidal agents in vitro with nanomolar concentrations producing significant bacterial inhibition and it is active against drug-resistant (DR)-TB as well [78]. The bactericidal effect of BTZ-043 is additive in combination with most other TB drugs but appears to be synergistic with bedaquiline [79]. A randomized, double-blind, placebo-controlled single-ascending dose study was completed in March 2019 that evaluated the safety, tolerability, and pharmacokinetics of single doses of BTZ-043 in healthy volunteers [80]. A multiple-ascending dose phase Ib/IIa study of BTZ-043 to evaluate safety, tolerability, and EBA over 14 days was started in August 2019 and currently is active but not recruiting [81]. PBTZ 169 PBTZ-169 (piperazine-BTZ; macozinone) is a derivative of benzothiazinone, similar in structure to BTZ-043 with the addition of a piperazine moiety, and yet is even more potent at nanomolar concentrations [78]. Similar to BTZ-043, it inhibits DprE1, but may be tenfold less cytotoxic [76]. Preclinical study has demonstrated in vitro activity against both MDR- and XDR-TB, and it shows excellent additive activity when combined with agents such as isoniazid, pretomanid, moxifloxacin, rifampin, and SQ109, and synergy when combined with bedaquiline [78]. The combination with bedaquiline was further assessed in the chronic TB mouse model and found to reduce the burden of TB (as measured by the reduction in CFU in lungs and spleen) to a greater degree than the standard combination of isoniazid, rifampin, and pyrazinamide. One postulate is that through inhibition of DprE1, PBTZ-169 may weaken the cell wall and thereby improve target access for bedaquiline [78]. A phase 1 clinical trial investigating single-ascending doses of a new bio-enhanced formulation of PBTZ-169 that included 32 healthy male volunteers was completed in April 2018 [82]. Another phase I trial in healthy male volunteers, aged 18–45 years with body mass indices (BMIs) 18.5–25 kg/m2, included seven cohorts who were given doses ranging from 40 to 640 mg of PBTZ-169 daily for 14 days [83]. The primary outcome was safety and tolerability. All 40 volunteers were able to complete the 14-day study. In 2017, a phase IIa EBA study of monotherapy for 14 days was started in DS-TB patients in Russia and Belarus. It was completed in February 2018 with 16 enrolled patients [84]. The EBA of three doses of PBTZ-169, 160 mg/day, 320 mg/day, and 640 mg/day, were all greater than INH 600 mg/day at 14 days as measured by a reduction in the number of CFU/mL [84]. Solubility issues are delaying further development of the drug [7]. An analogue of PBTZ-169, TZY-584, demonstrated lower MICs against a number of DS- and DR-TB strains than INH, rifampin, and bedaquiline, and results were similar to PBTZ-169. It was also effective against intracellular organisms in infected macrophages [75]. There are no active studies currently listed at clinicaltrials.gov (accessed 6 June 2022). OPC-167832 OPC-167832 is also an inhibitor of mycobacterial cell wall synthesis. It is a carbostyril derivative that inhibits DprE1 [76]. It is highly potent, and MICs have been demonstrated that are up to 1000-fold lower than many currently used anti-TB medications [73]. Combined with its favorable distribution from plasma to lung and to TB lesions, sustained drug levels above the MIC for a prolonged period suggest it may have very promising activity against TB [76]. It is effective against intracellular TB and has preserved activity against MDR- and XDR-TB, and has no significant activity against standard Gram-positive and Gram-negative bacteria, limiting any significant impact on human bacterial flora. The combination of OPC-167832 with moxifloxacin, bedaquiline, and delamanid showed excellent sterilizing activity in the mouse model, predicting a low rate of relapse [73]. In a checkerboard assay there was no antagonism when it was combined with delamanid, bedaquiline, the fluoroquinolones, levofloxacin and moxifloxacin, or linezolid [73]. There is an ongoing phase 1/2b safety/efficacy trial [85] and phase 2b dose-finding trial in combination with bedaquiline and delamanid for DS pulmonary TB [86]. TBA-7371 TBA-7371 is another inhibitor of DprE1 developed in collaboration with TB Alliance. It is a derivative of 1,4-azaindoles with an MIC of 0.64 μg/mL against TB with excellent solubility and clearance, limiting its toxicity [76, 87]. It also inhibits phosphodiesterase VI, the enzyme that normally inactivates it [7]. It has demonstrated notable reduction in CFU in the lungs in both the acute and chronic mouse models. After 8 weeks of treatment in mice, it showed an average 1.5 log10CFU reduction in the lungs [76]. Despite having higher MICs compared to PBTZ-169 and OPC-167832, it appeared to be as efficacious as PBTZ-169. A phase 2a dose-finding and EBA study is currently underway [88]. Inner Peptidoglycan Layer The inner layer of the cell wall consists of crosslinked peptidoglycan, alternating N-acetyl glucosamine and N-acetyl muramic acid moieties, attached to short peptide side chains [40, 89]. Cycloserine, terizidone, and carbapenems are second-line antibiotics that interfere with peptidoglycan synthesis and are generally used to treat DR-TB [8]. Cycloserine and terizidone target D-alanine-D-alanine ligase, an essential enzyme that joins two D-alanine moieties together that are then attached to peptidoglycan [90]. Unfortunately, adverse effects associated with cycloserine are not uncommon, and include seizures, psychosis, and peripheral neuropathy [91]. Despite MTB containing a broad-spectrum β lactamase, BlaC, it was found to not be highly active against carbapenem antibiotics, particularly meropenem [92]. Further, BlaC is an Ambler Class A ß-lactamase that can be stably inactivated by the ß-lactamase inhibitor clavulanate, but not with others such as tazobactam and sulbactam [93]. The combination of meropenem with clavulanate has potent MTB in vitro activity [94], but unfortunately this combination is not available, therefore requiring meropenem to be co-administered along with the readily available combination of amoxicillin-clavulanate. Nevertheless, in a 14-day EBA study in patients with smear-positive pulmonary TB with meropenem, in combination with amoxicillin-clavulanate, meropenem reduced the mycobacterial sputum load by 1.5 orders of magnitude, similar to the reduction expected with rifampin and pyrazinamide [92]. Carbapenems inhibit transpeptidases, essential enzymes that crosslink peptidoglycan [95]. Meropenem is only available as an intravenous medication. A recently published trial demonstrated that it has to be given three times daily to be effective, which is impractical for outpatients requiring the medication for 6 months, and it was frequently associated with gastrointestinal side effects [96]. Sanfetrinem Cilexetil Sanfetrinem cilexetil is an oral tricyclic carbapenem that was first investigated in the 1990s but development was suspended prior to phase III studies for financial reasons, and only recently has been repurposed as a drug to treat TB [7, 97]. Sanfetrinem is effective intracellularly with a lower range of MIC90 than meropenem: 1–4 μg/mL versus 2–64 μg/mL, respectively [7]. A phase II study is underway that plans to recruit 105 patients with rifampin-susceptible pulmonary TB [98]. Sanfetrinem powder will be given as a suspension in water at a dose of 1.6 g every 12 h with various combinations of rifampin and amoxicillin–clavulanate, with the primary outcome being EBA as determined by the change in CFU from pretreatment values to day 14 on solid media and secondary outcomes will be EBA as measured by a change in time-to-positivity by day 14 in BACTEC MGIT 960 liquid culture and the number of patients with treatment-emergent adverse effects [98]. There are no current studies underway for alternative oral carbapenem therapy (clinicaltrials.gov accessed 31 October 2022). Protein Inhibition Linezolid and Other Oxazolidinones Oxazolidinones inhibit protein synthesis. The bacterial 70S ribosome comprises two subunits: a 30S and a 50S subunit with a bridging surface area that contains three binding sites for t-RNA named A, P, and E. Oxazolidinones bind the A site and block the attachment of t-RNA to the ribosome, precluding protein synthesis [40]. Oxazolidinones are active against MTB and several are currently in development, and linezolid is already being used, to treat TB. In addition to linezolid, tedizolid, sutezolid, delpazolid, and TB-223 are all members of the oxazolidinone class. Linezolid Linezolid is a repurposed drug originally developed to treat Gram-positive infections with difficult drug-resistance profiles, and is now included in oral regimens to treat MDR-TB including XDR-TB [28]. Treatment regimens for MDR-TB containing linezolid were associated with treatment success and reduced mortality [99]. The WHO has since reclassified it as a group A drug for the treatment of MDR- and XDR-TB [28]. The standard dose to treat Gram-positive infections is 600 mg twice daily, but prolonged treatment with this dose is associated with a high risk of bone marrow suppression and neuropathy, and the risk is increased if the treatment course is extended beyond 10 days. Myelosuppression is due to linezolid inhibiting mitochondrial protein synthesis in bone marrow precursor cells [100]. Linezolid can also cause optic and peripheral neuropathy, lactic acidosis, pancreatitis, and serotonin syndrome in those on selective serotonin reuptake inhibitors [101–103]. A linezolid dose of 600 mg daily is sufficient to treat TB but is potentially required for up to 18 months, and is often associated with side effects [99]. In the NIX-TB trial, 81% of the patients experienced myelosuppression, either anemia or thrombocytopenia, 41% peripheral neuropathy, and 2% optic neuritis [16]. In another cohort of 472 patients, of whom 90% received linezolid, 28.4% developed peripheral neuropathy and 5.1% myelosuppression [104]. However, the results of the ZeNix trial were recently published and highlight that in combination with bedaquiline and pretomanid, a reduced dose of linezolid (compared to the doses used in Nix-TB) of 600 mg daily for 26 weeks provided similarly excellent efficacy outcomes [105]. Additionally, the reduced dose demonstrated significantly improved safety outcomes, with only 4% developing myelosuppression at this dose and 13% requiring dose modification for any cause. Other oxazolidinones are being developed with the hope of achieving the therapeutic benefits of linezolid without its adverse effects. Alternatively, there are compelling data that mitochondrial toxicity that leads to neuropathy and myelosuppression is associated with linezolid trough concentrations >2 μg/mL and may be mitigated with therapeutic drug monitoring [106, 107]. Tedizolid Tedizolid is a newer oxazolidinone prodrug converted in serum to its active form, and subsequently inhibits the 50S subunit of the ribosome, impairing protein synthesis [108]. It exhibits excellent penetration into cutaneous and pulmonary tissue (superior to linezolid), making it an optimistic option for treatment of TB [109]. It demonstrates vast Gram-positive inhibitory activity, and in addition demonstrated activity against both rapidly and slowly growing nontuberculous mycobacteria, including TB [109]. It demonstrates the most potent in vitro activity of any of the oxazolidinones against rapidly-growing M. abscessus [110]. An in vitro study has demonstrated efficacy of tedizolid against nonreplicating persisting mycobacteria with similar efficacy to standard first-line therapy [111], with superior sterilizing activity compared to linezolid in the hollow fibre model [112, 113] and lower MICs even in linezolid-resistant isolates, highlighting the potential for greater potency [108, 114]. One in vitro study found a > 10,000-fold CFU/mL difference for intracellular activity between linezolid and tedizolid, making it a potentially superior agent for cavitary and disseminated disease in children [112]. In vitro efficacy that may be superior to linezolid has been demonstrated in both DS and DR isolates, including MDR-TB [108, 114]. Tedizolid seems to have a superior safety profile to linezolid [109]. Its more common side effects were mild AEs including headache and gastrointestinal (GI) intolerance, and it seems to have a better hematologic safety profile to that shown by linezolid [96, 102]. Mouse studies have suggested that the risk of serotonin syndrome with tedizolid in combination with other serotonergic chemicals may be lower than that identified with linezolid [103]. The main challenge with prolonged use of linezolid, as seen in TB treatment regimens, is toxicity associated with mitochondrial inhibition. Models have suggested that toxicity from tedizolid may be less and intermittent dosing strategies may be possible that maintain effectiveness, while reducing the potential for toxicity [100]. However, there are no clinical studies utilizing tedizolid for TB to date (clinicaltrials.gov accessed 1 November 2022), and data indicating the reduced toxicity of tedizolid relative to linezolid are largely limited to maximal durations of 12 weeks in non-TB bacterial infections or case series [116, 117, 118]. Therefore, it is yet to be confirmed if the toxicity of tedizolid is indeed less with the durations of treatment required for TB. Sutezolid (PNU-100480) Sutezolid is a thiomorpholinyl analog of linezolid. In vitro testing has suggested that sutezolid has the strongest activity among oxazolidinones against slowly growing nontuberculous mycobacterial species, as well as TB, and that it is superior to linezolid [119, 120]. Activity is evident against both DS- and DR-TB [106]. The sterilizing activity of sutezolid increases its potential to shorten TB therapy. Comparative studies in mice demonstrated that the addition of sutezolid to shortened (4-month) standard first-line therapy improved its efficacy and reduced the risk of relapse significantly compared to regimens without it [122]. In a small trial, 59 patients with pulmonary TB were randomized to two different doses of sutezolid for 14 days. Notable EBA was evident on sputum CFU count as well as time to positivity in blood culture inoculated with MTB for whole blood bactericidal activity. Treatment was also safe, with no subject requiring dose adjustment or discontinuation [123]. The most common reported adverse effect was asymptomatic aminotransferase elevation [123]. A phase IIb dose-finding/efficacy study is currently underway, as added therapy to bedaquiline, delamanid, and moxifloxacin [124]. Delpazolid (LCB01-0371) Delpazolid is a novel oxazolidinone that contains a cyclic amidrazone group. It is highly bactericidal, and similar to other oxazolidinones exerts its antimycobacterial function by inhibiting protein synthesis at the 23s rRNA [109]. Similar to other oxazolidinones, it possesses broad Gram-positive activity and also seems to have potent activity against M. abscessus [40]. Mixed results have been evident in in vitro studies against DR-TB. Specifically, Zong et al. identified that delpazolid had a lower MIC against MDR-TB with a lower proportion with resistance, relative to linezolid, but higher MICs in XDR-TB [125]. In those with linezolid-resistant isolates, 50% were still susceptible to delpazolid. This study also highlighted that high-level resistance to both linezolid and delpazolid was mediated by mutations in 23s rRNA. Their data identified similar anti-tuberculosis activity to that of linezolid. Notably, there were more linezolid-resistant MDR-TB strains compared to delpazolid, including a novel ribosomal protein mutation (rplD) rendering high-level resistance to linezolid but preserved low MIC to delpazolid [125]. However, a significant proportion of resistant isolates to either linezolid or delpazolid did not have mutations in recognized target genes that may implicate other alterations in the 23s rRNA or perhaps efflux pumps as the cause [125]. They identified that only 2.9% of MDR isolates were resistant to delpazolid, providing the opportunity for its use as a component in MDR-TB treatment [126]. Another in vitro study demonstrated comparable EBA to that of linezolid [110, 127]. It appears to come with an acceptable hematologic safety profile in a short study and it is argued this may be due to its shorter half-life leading to reduced impact on mitochondrial toxicity [128]. There are no longer-term clinical use data or comparative data with linezolid. TBI-223 TBI-223, another oxazolidinone, has several advantageous properties including improved stability in microsomes and hepatocytes, does not induce or inhibit the CYP enzyme system, has good oral bioavailability, and is active against both DS- and DR-TB strains [7]. Its in vitro activity is similar to linezolid, but appears to have a superior safety profile [40]. Its phase I study has been completed but the results are not yet available [129]. OTB-658 OTB-658 is another oxazolidinone in pre-clinical study that has shown promising in vitro and in vivo activity with possible improvement compared to linezolid [130]. The MIC against MTB was found to be lower than that with linezolid with retained effectiveness against intracellular bacilli. It also has exhibited low spontaneous mutation frequency, rendering it a promising agent against resistant MTB and inviting the opportunity for further exploration in clinical study [131]. Posizolid (AZD5847) Posizolid was developed by AstraZeneca and found in preclinical studies to have an improved safety profile and in vitro potency over linezolid in treating both intracellular and extracellular TB, including MDR-TB [132]. However, EBA was only modest with twice-daily dosing and there were two serious AEs, including one in a patient developing severe thrombocytopenia [133]. The drug was subsequently withdrawn from development in 2016. Aminoacyl-tRNA Synthetases Inhibition of protein production at alternative targets not yet established offers potential utility in attenuating mycobacterial growth in DR-TB. Aminoacyl-tRNA synthetases are imperative enzymes for protein synthesis/translation and one such recognized enzyme is the isoleucyl-tRNA synthetase, which is inhibited by the topical antibacterial mupirocin [134, 135]. Alternatively, leucyl-tRNA synthetases may be inhibited by boron-containing compounds known as oxaboroles. Tevaborale is an irreversible leucyl-tRNA synthetase inhibitor with activity against Candida albicans [115]. Several putative compounds have been constructed and identified using X-ray crystallography and compared in in vitro and in vivo murine models against mycobacteria [134]. The most optimistic compound to date is GSK656 (‘Compound 8’) in phase IIa trials targeting TB with highly potent in vitro activity [135–137]. Several other analogues with slightly different structure are being explored for putative use in TB [135]. RNA Polymerase Inhibition Rifampin Despite almost 60 years since its introduction, rifampin remains an indispensable molecule against DS-TB, whose value has paved the way for attempts at further optimized rifamycins such as rifapentine. By inhibiting the mycobacterial RNA polymerase, rifampin impairs RNA synthesis providing bactericidal activity in a concentration-dependent manner [12]. However, it is the sterilizing capabilities of rifampin against slowly replicating bacilli with prolonged post-antibiotic effect that prevents relapse of disease and has allowed significant shortening of the treatment duration of DS-TB [12]. The primary drawback of rifampin is its induction of the hepatic cytochrome P450 system leading to numerous drug-drug interactions. Rifapentine Rifapentine is a semi-synthetic derivative of rifampin that inhibits the DNA-dependent RNA polymerase necessary for transcription. In contrast to rifampin, rifapentine has a cyclopentyl side chain ring instead of a methyl group, which significantly increases protein binding and half-life [12]. Interestingly, increasing dose and repeated dosing have suggested decreased bioavailability and increased clearance, respectively, favoring the possibility of autoinduction [138]. Preclinical studies favored an even lower MIC in TB than rifampin, and due to its strong binding to RNA polymerase even in low enzyme activity, a promising option for LTBI [12, 139]. Rifapentine shares similar risks for AEs to rifampin, including hepatotoxicity and influenza-like reaction, which is typically more common with intermittent administration [12]. Cross-resistance between rifampin and rifapentine is predicted almost entirely and is due to point mutations in the rpoB gene [12]. The potency of rifapentine against TB combined with its long half-life has found once-weekly rifapentine combined with isoniazid to be an encouraging option for LTBI treatment that is now a first-line option [140, 141]. Additionally, rifapentine is being investigated as single-drug therapy for LTBI [142]. In TB disease, mouse model data supported the possibility of daily administration that could shorten therapy [143]. The optimism that rifapentine might lead to treatment shortening led to a recent randomized, non-inferiority trial that showed that 4 months of rifapentine combined with moxifloxacin, isoniazid, and pyrazinamide was non-inferior to standard 6-month therapy in a microbiologically eligible population for unfavorable outcomes (15.5% vs. 14.6%; 1.0% difference, 95% CI − 2.6 to 4.5). The rates of AE rates were no higher [144]. This has led to the recent recommendation of this regimen as a treatment alternative in DS-TB [145]. Impaired Energy Metabolism, Including ATP synthesis, Cytochrome Complex, and Electron Transport Chain Function, Membrane Destabilization/Reactive Oxygen Species Bedaquiline Bedaquiline is the first FDA-approved novel agent against TB since 1971 and the first-in-class of diarylquinolines inhibiting mycobacterial ATP synthesis mediated along the electron transport chain. It accomplishes this by inhibiting F-ATP synthase activity, depleting bacterial ATP, which subsequently inhibits DNA and protein synthesis and nitrogen metabolism [18]. These mechanisms produce in vitro activity against both replicating and non-replicating bacilli, reducing the number of bacilli during log phase growth and offering sterilizing activity, the ability to prevent post-treatment relapse. In 2012, it received accelerated approval from the FDA for use in MDR-TB [5]. The promotion of this agent was based on expedited sputum culture conversion (83 days vs. 125 days) when added to standard of care for MDR-TB [146]. There were also more patients found to have converted their sputum to culture negative (79% vs. 58%) by 24 weeks, and this difference was sustained at 120 weeks. The relationship of culture conversion at 24 weeks and long-term outcomes at 120 weeks have been supported and reliable [147]. Tolerance of bedaquiline was adequate, with no increased risk of AEs compared to standard of care alone. However, there was an excess number of deaths in the bedaquiline group relative to placebo (ten patients (12.6%) vs. two patients (2.5%), p = 0.02) of unclear etiology, which led to a Black Box warning being required on the product monograph by the FDA [146, 148]. Nevertheless, following these promising efficacy results there was an extensive roll-out of bedaquiline in 2015 nationally in South Africa, through combined compassionate drug access and local approval, where it rapidly was incorporated into all DR-TB oral regimens. Outcomes were significantly improved. Among more than 19,000 patients with MDR- or XDR-TB studied between 2014 and 2016, bedaquiline was given to 5.2%. All-cause mortality was significantly reduced in the bedaquiline cohort with MDR-TB (HR 0.35; 95% CI 0.28–0.46) and XDR-TB (HR 0.26; 95% CI 0.18–0.38) and overall mortality was more in line with usual rates of death in patients with DS-TB [149]. Subsequently, the WHO recommended the inclusion of bedaquiline in an all-oral regimen as preferred initial treatment for MDR- or XDR-TB [13]. A large retrospective analysis of 428 patients across 25 centres and in 15 countries found excellent rates of treatment success (71.3%) and culture conversion (91.8%) with bedaquiline-containing regimens in MDR-TB. QTc prolongation beyond 500 ms occurred in 9.7% of the patients, including one of the 33 who died, but that death was felt to unlikely be due to bedaquiline [150]. Available clinical data support the effectiveness of bedaquiline when added to guideline-based therapy or more recently in combination with other agents. The Nix-TB trial demonstrated the efficacy of the combination of bedaquiline and pretomanid in conjunction with the bactericidal activity of linezolid [16]. A phase III follow-up study, ZeNix, found that this same combination along with a reduced dose of linezolid at 600 mg for 26 weeks produced the optimal balance of efficacy and safety [105]. The NExT Study was recently published and compared a 6- to 9-month interventional regimen of bedaquiline, levofloxacin, and linezolid with two other drugs (among ethionamide, high-dose isoniazid, terizidone) to the standardized ≥ 9-month injectable-based regimen for MDR-TB [151]. At 24 months, although suboptimal outcomes overall, those in the interventional arm were twice as likely to have a favorable treatment outcome (51% vs. 22.7%; RR 2.2; 95% CI 1.2–4.1; p = 0.006). Specifically, this combination produced fewer unfavorable outcomes (HR 0.4; 95% CI 0.2–0.5; p < 0.001), greater culture conversion rates when censored for those who had bedaquiline substituted into the standardized regimen (HR 2.6; 95% CI 1.4–4.9; p = 0.003), and were more likely to complete therapy without drug substitution. There were more AEs experienced with this combination, mainly driven by linezolid toxicity (hematologic and neurologic). A systematic review and meta-analysis explored the results of eight studies (two randomized controlled trials and six observational studies) with 1,784 patients who received bedaquiline added to a background MDR-TB regimen, compared with 20,000 patients who did not [152]. Those who received bedaquiline had higher rates of culture conversion (RR 1.272; 95% CI 1.165–1389; p < 0.001). Those receiving bedaquiline demonstrated lower all-cause mortality (RR 0.529; 95% CI 0.454–0.616; p < 0.001), but no difference in treatment success rates (RR 0.98; 95% CI 0.948–1.1013; p = 0.234). There was significant heterogeneity in the included studies. In a phase IIb trial, the combination of bedaquiline and delamanid with pyrazinamide plus or minus moxifloxacin was shown to have higher EBA, as judged by time-to-culture positivity, relative to standard first-line therapy in both DS- and MDR-TB, respectively, with similar rates of AEs [153]. This creates the opportunity for further phase III exploration of this combination. Both bedaquiline and delamanid risk QTc prolongation as potential side effects. However, the mean increase in QTc is additive but typically manageable, at around 20 ms above baseline [154]. A modelling study suggested that preventative treatment of pediatric household contacts of MDR-TB cases with bedaquiline or delamanid would reduce the number of secondary cases and mortality compared to treatment with fluoroquinolones [155]. Limiting treatment to children under 5 years of age and to those co-infected with HIV would reduce the number of cases and mortality prevented but would be more cost-effectiveness [130]. With clinical experience, bedaquiline is generally well tolerated. The main AEs reported include nausea/vomiting, headache, and arthralgia [146]. As mentioned, the initial phase IIb trial had an increased number of deaths compared to placebo (12.6% vs. 2.5%), though none were clearly attributed to the drug. This unexplained increased risk of death with bedaquiline has not been borne out in subsequent studies. A subsequent multicentre, single-arm clinical trial identified 16 deaths (6.9%), and none were deemed to be related to bedaquiline [146]. Nine of the 10 deaths occurred after the 24-week bedaquiline treatment period; six were due to progression of MDR-TB, and one each was due to acute alcohol poisoning, stroke, motor vehicle accident, and cirrhosis, respectively [146]. The M2 bedaquiline metabolite inhibits KCNH2 (hERG), interfering with potassium channel function and consequently interfering with normal cardiac repolarization and prolonging the QT interval [156]. Because of the known increase of the QTc, this has led to the question of whether these deaths may have been cardiac. The mean change in QTc has been reported to be between 12 and 15 ms [146, 154]; however, this is generally in combination with other medications that may further prolong the QTc, such as clofazimine and fluoroquinolones. Nevertheless, the reported incidence of QTc beyond 500 ms was 3% [157]. However, reports of resistance have been noted and it has been suggested that acquired resistance has the opportunity to occur at low serum levels [158]. Since it has a very long half-life, stopping it well before the other anti-tubercular agents is recommended to prevent selection of bedaquiline-resistant mutants [159]. Even with the potent combination of bedaquiline and pretomanid, as use of bedaquiline increases, so too does the anticipated risk of evolving resistant pathogens. Mouse models of infection with the bedaquiline loss-of-function mutation Rv0678 are less susceptible to the drug and may be selected for along with other drug resistance, even in the presence of other highly effective drugs like pretomanid [160]. Bedaquiline is metabolized by the cytochrome P450 CPY3A4 enzyme and inducers of this enzyme such as rifampin and rifapentine reduce exposure to bedaquiline, whereas inhibitors, including azoles and some antiretrovirals, increase exposure [161, 162]. With the success and broadly recommended inclusion of bedaquiline into all regimens against MDR-TB, similar in class molecules are being explored. ATP synthase inhibition appears to provide effective sterilizing activity. A prominent site of resistance recognized with bedaquiline is the Rv0678 gene, which encodes a repressor of the MmpS5–MmpL5 efflux transporter. Mutations of the gene thereby lead to increased efflux of the molecule [163]. Two novel diarylquinolines, TBAJ-587 and TBAJ-876, are being explored that may have enhanced efficacy against bedaquiline-resistant TB strains [160, 163]. TBAJ-876 TBAJ-876 is a structurally distinct diarylquinoline analogue of bedaquiline that also inhibits the F-ATP synthase [18]. Despite not possessing the same electron transport chain uncoupling activity of its predecessor bedaquiline, it demonstrates equivalent in vitro bactericidal activity to that of bedaquiline [19]. In a murine model, TBAJ-876 at low doses demonstrated similar activity to bedaquiline and at higher doses appeared more bactericidal with greater effect at reducing the CFU burden [163]. This improved efficacy was retained over bedaquiline in strains with the Rv0678 mutation. There is currently a phase I study underway to assess its safety and tolerability [164]. TBAJ-587 Another bedaquiline analogue, TBAJ-587, has demonstrated more potent in vitro activity than bedaquiline while possibly having an improved safety profile in terms of lower cardiac risk [165]. In the mouse model, it was found to be significantly more bactericidal than bedaquiline, and offers the theoretical potential to shorten treatment durations [160]. It is currently undergoing a phase I clinical trial [166]. Sudapyridine (WX-081) Sudapyridine is another diarylquinoline with comparable efficacy against DS- and DR-TB to that of bedaquiline, but is less lipophilic with a better pharmacokinetic profile, reducing the risk of QTc prolongation [167]. An in vitro assay of the effects of sudapyridine on the human hERG ion channel was reassuring and, while data are limited, sudapyridine has also been reported to have less evidence of qualitative and quantitative impact on the ECG, potentially due to reduced generation of its cardiac-relevant metabolite (WX-081-M3) compared to bedaquiline [167]. A full phase I study has been completed and a phase II trial to determine EBA and safety is underway and is almost complete [7, 168]. The phase IIa study includes 84 patients in five different treatment arms. Patients with newly treated drug-susceptible pulmonary TB are randomized into one of four arms to receive guideline-based multidrug treatment and 150 mg, 300 mg, 450 mg, or no sudapyridine daily for 2 weeks. The fifth arm consists of patients with DR-TB receiving sudapyridine 400 mg daily for 2 weeks, followed by 150 mg daily for 6 weeks, along with multi-drug background treatment (accessed 9 July 2022). Primary outcomes in this phase IIa study are time to sputum culture positivity in liquid culture medium and EBA is measured as the rate of CFU/mL of sputum on solid medium [168]. Imidazopyridine The electron transport chain provides the energy required for ATP synthesis [169]. Electron transfer through various cytochromes is initiated by the oxidation of organic material creating an electric potential resulting in a proton flow and eventually the enzyme, ATP synthase, phosphorylates adenosine diphosphate (ADP) to ATP [22]. The cytochrome bc1 is an essential component of the electron transport chain in the oxidative phosphorylation pathway [170]. Telacebec (Q203) Ending in cebec indicates the activity of telacebec is a cytochrome bc1 complex inhibitor of tuberculosis, impairing energy metabolism [170]. It is an imidazopyridine and the mode of action of telacebec is different from other drugs undergoing investigation since it inhibits MTB growth, both DS- and DR-TB, by targeting the QcrB subunit of respiratory cytochrome bc1 complex at a low concentration [67, 169]. The MIC50 against MTB is 2.7 nM in culture medium and 0.28 nM inside macrophages [169]. Telacebec has several attractive pharmacologic features including an oral bioavailability of 90%, a terminal half-life of 23.4 h, and it does not inhibit the cytochrome P450 isoenzymes, so drug-drug interactions should not be a concern [169]. This unique mode of action of telacebec allows MDR- and XDR-TB to be easily targeted. The drug was well tolerated during a phase I trial and there were no serious AEs with single oral doses ranging from 10 to 800 mg [171]. In the phase IIa study [172], treatment-naïve patients with pulmonary DS-TB were given 100 mg, 200 mg, or 300 mg of telacebec for 14 days with a reduction in the sputum mycobacterial load proportional to the dose of telacebec, as determined by the time-to-culture positivity in MGIT 960 system liquid media [173]. Telacebec also has excellent activity against Mycobacterium ulcerans, the causative organism of Buruli ulcer, and Mycobacterium leprae [174]. Pyrifazimine (TBI-166) Clofazimine, the currently available riminophenazine, has potent antimycobacterial activity. The mechanism of action of clofazimine is not entirely clear due to its predominantly intracellular activity. It is favored to produce intracellular reactive oxygen species, competition for electrons along the mycobacterial respiratory chain, and membrane disruption [175, 176]. Pyrifazimine is a second-generation riminophenazine, similar to clofazimine, that has been shown to have potent anti-TB activity, especially in DR-TB. Pyrifazimine was derived based on optimization of other potential riminophenazine analogues. Specifically, by ensuring a substituent with a 2-methoxypyridylamino group at the C-2 position of the phenazine, it was found to have improved pharmacokinetic properties with increased antimycobacterial potency compared to clofazimine without evidence of skin discoloration, which is almost ubiquitously encountered with clofazimine [66, 177–179]. The reduction in lipophilicity is a significant driver in reducing tissue deposition and subsequent pigmentation change [179]. Following the in vitro and in vivo studies in murine models, success of this chemical prompted its promotion as a drug candidate. A phase I clinical trial was initiated in 2018 in China. A phase II bactericidal activity and comparative safety study of increasing doses of pyrifazimine is underway [180]. In synergistic testing in mouse models, the combination of pyrifazimine with bedaquiline and linezolid was highly bactericidal compared to a usual first-line control regimen and the most effective combination compared to others. The regimen was recommended for further phase II testing [177]. Synergy was also found between pyrifazimine, bedaquiline, SQ109, and PBTZ169 [181]. DNA Gyrase and Topoisomerase Fluoroquinolones The fluoroquinolones levofloxacin and moxifloxacin have been repurposed as Group A drugs for the treatment of MDR-TB [28]. The pronounced efficacy of these drugs for treatment of MDR-TB has been highlighted in an individual patient-data level meta-analysis that showed that their use is associated with significantly improved treatment success and survival [99]. Several trials assessed if substituting moxifloxacin for either INH or ethambutol or substituting ethambutol with gatifloxacin could shorten the treatment regimen for DS-pulmonary TB from 6 to 4 months, but found that the 4-month regimens were inferior to standard therapy [182, 183]. Gatifloxacin is no longer available because of its dysglycemic effects. However, more recent data have offered the first positive outcomes supporting a 4-month treatment course in DS-TB [144] and led to an updated conditional recommendation for its use from the WHO [145]. Fodrepodacin (SPR720) SPR720, renamed fodrepodacin, is a prodrug that is converted to SPR719, an oral bacterial DNA gyrase (gyrB) inhibitor that interferes with DNA replication. It demonstrated activity against both MTB and nontuberculous mycobacteria in in vitro and murine studies [184]. A regimen of SPR720, rifampin, and pyrazinamide showed comparable efficacy to a moxifloxacin, rifampin, and pyrazinamide regimen (Stop TB Partnership). The FDA placed a clinical trial hold on SPR720 following a mortality data review from a nonhuman toxicology study but subsequently lifted it in January 2022 [185]. It appears that Spero therapeutics, the company developing the molecule, will focus research on treatment of nontuberculous mycobacteria [7, 186] (accessed 29 June 2022). VXc-486 VXc-486 is a new aminobenzimidazole that also targets gyrase B and is bactericidal against both DS- and DR-TB, with MICs ranging from 0.03 to 0.30 mg/mL and 0.08 to 5.48 mg/mL, respectively [7]. This activity was preserved even in XDR-TB, which was also resistant to fluoroquinolones, inhibitory against intracellular TB, and had suggested synergy with linezolid, bedaquiline, and clofazimine. It is also effective against M. avium, M. abscessus, and M. kansasii [187]. No trials are currently listed on clinicaltrials.gov (accessed 15 August 2022). GSK2556286 GSK2556286 is a pyrimidine-2,4-dione that interferes with MTB’s ability to catabolize cholesterol from the host and utilize it as a carbon source. It selectively kills intracellular MTB in macrophages, MIC50 < 0.1 μm, and is considered to be of moderate efficacy but more effective when combined with other drugs [22]. It is efficacious in cholesterol-containing media and does not cause cross-resistance with other antitubercular agents [188]. A phase I double-blind, randomized, sequential, parallel-dose cohort study in 120 healthy adult participants is currently underway [189]. Drugs That Inhibit Drug Efflux The use of efflux-pump inhibitors to address TB drug resistance is also being investigated and there have been reports of success when an otherwise effective treatment regimen could not be provided [190]. Phenothiazines inhibit bacterial efflux and the resulting high drug concentrations are toxic to both intracellular DS-and DR-TB [191]. They would be attractive options for the management of DR-TB but their CNS effects are problematic. Prior to the introduction of the new and repurposed drugs to treat DR-TB, there were reports of successful treatment of MDR- and even XDR-TB with thioridazine [192]. Thioridazine was felt to be the phenothiazine with the best risk/benefit profile but it produces CNS and cardiac adverse effects including QT interval prolongation and, rarely, torsade de pointes [193, 194]. Subsequent reports suggest that low enough doses to avoid most adverse effects are effective against MTB, but there are no currently registered clinical trials [195]. Verapamil Verapamil is also an efflux pump inhibitor. It accelerated bacterial clearance in a murine model of treated TB [196]. It is not active against extracellular MTB but increases intracellular antibiotic concentrations and possibly reduces the risk of the development of drug resistance [24]. There are no current registered clinical studies utilizing verapamil for TB (clinical trials.gov accessed 31 October 2022). FNDR-20081 While the site of action is not entirely clear, resistance mapping for FNDR-20081 seemed to localize it to MmpL5, which regulates efflux transport, or to Rv3683, a metallophosphoesterase [197]. It is a promising first-in-class quinoline derivative combined with oxadiazole and piperazine in tandem with potent inhibitory activity against both DS and DR-TB strains. When combined with other common anti-mycobacterials, no antagonism was identified, additive activity was found with bedaquiline, pretomanid, and linezolid, and synergy identified with clofazimine and ethambutol. This has offered promise in its clinical effectiveness, inviting the opportunity for clinical exploration [197]. Conclusion In conclusion, due to the prolonged nature of TB treatment that is rife with AEs, exploration of novel and repurposed therapies that might continue to improve treatment duration and toleration is an emphasis in global TB research. Herein, we have highlighted several such compounds and their unique sites of activity. Sequencing the MTB genome has identified numerous potential targets for drug development in both DS- and DR-strains. If current treatment regimens are given with a daily observed therapy protocol, results against DS-TB strains are excellent. The success rates in patients with DR-, particularly RR-TB and more complex resistance patterns, are inferior. Treatment regimens for DR-TB are more complex, considerably longer, associated with more adverse effects, and are much more expensive to provide [99]. Collaboration and global TB foundation funding have encouraged the recent discovery of novel TB targets, including mycobacterial ATP synthase and the arabinogalactan portion of the cell wall, which offer promising opportunities. Subsequent preclinical development has provided some of the most potent compounds to date in in vitro study. Despite this, it is well recognized that many of these compounds described will be delayed or not make it to the broader therapeutic market. Experiences with agents such as posizolid and sanfetrinem highlight the fact that promising in vitro and early clinical studies do not always translate into clinical efficacy, or finances may limit further development. Despite these major advances, infection rates and mortality from TB remain stubbornly high. Cost and lack of resources, primarily in the developing world, remain major impediments to the global effort to eliminate the disease. Eliminating this disease will require many things, but the provision of effective medications that are inexpensive, orally active, allow shortening of treatment regimens for both DS-and DR-TB, have infrequent and tolerable side effects, and do not require regular laboratory monitoring will help to achieve the goal of defeating this ongoing pandemic. Therefore, ongoing pursuit and exploitation of novel and repurposed sites are imperative, with the ongoing financial support of global TB research communities to ensure the pipeline does not dry up. Declarations Funding No funding was received for this review. Conflicts of interest/competing interest Neither Brett Edwards nor Stephen Field has any conflicts of interest related to the contents of this article. Authors’ contributions Both Brett Edwards and Stephen Field shared equally in the research, composition, and revision of this article. Ethics approval Not applicable. Informed consent Not applicable. Data availability Not applicable. Consent to participate Not applicable. Consent for publication Not applicable. Code availability Not applicable. ==== Refs References 1. Daniel TM The history of tuberculosis Respir Med 2006 100 11 1862 1870 10.1016/j.rmed.2006.08.006 16949809 2. Herzog H History of tuberculosis Respiration 1998 65 1 5 15 10.1159/000029220 9523361 3. Bunyan J. Life and Death of Mr. Badman. Nathaniel Ponder Publishing; 1680. 4. Global Tuberculosis Report [Internet]. World Health Organization. 2021. https://www.who.int/publications/i/item/9789240037021. Accessed 17 Jun 2021. 5. Bedaquiline [Internet]. FDA. 2012. https://www.accessdata.fda.gov/drugsatfda_docs/label/2012/204384s000lbl.pdf. Accessed 6 Feb 2020. 6. Deltyba \| European Medicines Agency [Internet]. 2022. https://www.ema.europa.eu/en/medicines/human/EPAR/deltyba#authorisation-details-section. Accessed 2 Jun 2022. 7. Clinical Pipeline [Internet]. Working Group on New TB Drugs—Stop TB Partnership. 2022. https://www.newtbdrugs.org. Accessed 22 Apr 2022. 8. Shanib Bhat Z Ahmad Rather M Maqbool M Lah HU Khalid Yousuf S Ahmad Z Cell wall: a versatile fountain of drug targets in Mycobacterium tuberculosis Biomed Pharmacother 2017 95 1520 1534 10.1016/j.biopha.2017.09.036 28946393 9. Nahid P Dorman SE Alipanah N Barry PM Brozek JL Cattamanchi A Executive summary: official American Thoracic Society/Centers for Disease Control and Prevention/Infectious Diseases Society of America Clinical Practice Guidelines: Treatment of Drug-Susceptible Tuberculosis Clin Infect Dis 2016 63 7 853 867 10.1093/cid/ciw566 27621353 10. Johnston JC Cooper R Menzies D Chapter 5: treatment of tuberculosis disease Can J Respir Crit Care Sleep Med. 2022 6 1 66 76 11. Gopal P Sarathy JP Yee M Ragunathan P Shin J Bhushan S Pyrazinamide triggers degradation of its target aspartate decarboxylase Nat Commun 2020 11 1 1661 10.1038/s41467-020-15516-1 32245967 12. Alfarisi O Alghamdi WA Al-Shaer MH Dooley KE Peloquin CA Rifampin vs. rifapentine: what is the preferred rifamycin for tuberculosis? Expert Rev Clin Pharmacol 2017 10 10 1027 1036 10.1080/17512433.2017.1366311 28803492 13. World Health Organization. Consolidated Guidelines on Drug-Resistant Tuberculosis Treatment [Internet]. World Health Organization. 2019. p. 1–104. https://www.who.int/tb/publications/2019/consolidated-guidelines-drug-resistant-TB-treatment/en/. Accessed 1 Aug 2019. 14. Nahid P Mase SR Migliori GB Sotgiu G BothamLey GH Brozek JL Treatment of drug-resistant tuberculosis an official ATS/CDC/ERS/IDSA clinical practice guideline Am J Respir Crit Care Med 2019 200 10 93 142 10.1164/rccm.201909-1874ST 15. Brode SK Dwilow R Kunimoto D Menzies D Khan FA Drug-resistant tuberculosis Can J Respir Crit Care, Sleep Med. 2022 6 1 109 128 16. Conradie F Diacon AH Ngubane N Howell P Everitt D Crook AM Treatment of highly drug-resistant pulmonary tuberculosis N Engl J Med 2020 382 10 893 902 10.1056/NEJMoa1901814 32130813 17. Berry C, du Cros P, Fielding K, Gajewski S, Kazounis E, McHugh TD, et al. TB-PRACTECAL: study protocol for a randomised, controlled, open-label, phase II-III trial to evaluate the safety and efficacy of regimens containing bedaquiline and pretomanid for the treatment of adult patients with pulmonary multidrug-resistant tubercul. Trials [Internet]. 2022;23(1). 18. Sarathy JP Gruber G Dick T Re-understanding the mechanisms of action of the anti-mycobacterial drug bedaquiline Antibiotics 2019 8 4 261 10.3390/antibiotics8040261 31835707 19. Sarathy JP Ragunathan P Cooper CB Upton AM Grüber G Dick T TBAJ-876 Displays bedaquiline-like mycobactericidal potency without retaining the parental drug’s uncoupler activity Antimicrob Agents Chemother 2020 64 e01540–19 10.1128/AAC.01540-19 20. Abrahams KA Besra GS Mycobacterial cell wall biosynthesis: a multifaceted antibiotic target Parasitology 2018 145 116 133 10.1017/S0031182016002377 27976597 21. Zhang L Zhao Y Gao Y Wu L Gao R Zhang Q Structures of cell wall arabinosyltransferases with the anti-tuberculosis drug ethambutol Science (80-). 2020 368 6496 1211 1219 10.1126/science.aba9102 22. Fernandes GFS Thompson AM Castagnolo D Denny WA Dos SJL Tuberculosis drug discovery: challenges and new horizons J Med Chem 2022 65 11 7489 7531 10.1021/acs.jmedchem.2c00227 35612311 23. Jankute M Cox JAG Harrison J Besra GS Assembly of the mycobacterial cell wall Annu Rev Microbiol 2015 69 1 405 423 10.1146/annurev-micro-091014-104121 26488279 24. Singh V Chibale K Strategies to combat multi-drug resistance in tuberculosis Acc Chem Res 2021 54 10 2361 2376 10.1021/acs.accounts.0c00878 33886255 25. Takayama K Wang C Besra GS Pathway to synthesis and processing of mycolic acids in Mycobacterium tuberculosis Clin Microbiol Rev 2005 18 1 81 101 10.1128/CMR.18.1.81-101.2005 15653820 26. Unissa AN Subbian S Hanna LE Selvakumar N Overview on mechanisms of isoniazid action and resistance in Mycobacterium tuberculosis Infect Genet Evol 2016 45 474 492 10.1016/j.meegid.2016.09.004 27612406 27. Vilchèze C Jacobs WR The mechanism of isoniazid killing: clarity through the scope of genetics Annu Rev Microbiol 2007 61 35 50 10.1146/annurev.micro.61.111606.122346 18035606 28. World Health Organization. WHO consolidated guidelines on tuberculosis Module 4: Treatment Drug-resistant tuberculosis treatment [Internet]. 2020. https://www.who.int/publications/i/item/9789240007048. Accessed 1 Jul 2022. 29. Tornheim JA Dooley KE The global landscape of tuberculosis therapeutics Annu Rev Med 2019 70 105 120 10.1146/annurev-med-040717-051150 30403551 30. Dean AS Zignol M Cabibbe AM Falzon D Glaziou P Cirillo DM Prevalence and genetic profiles of isoniazid resistance in tuberculosis patients: a multicountry analysis of cross-sectional data PLoS Med 2020 17 1 e1003008 10.1371/journal.pmed.1003008 31961877 31. Banerjee A Dubnau E Quemard A Balasubramanian V Um KS Wilson T inhA, a gene encoding a target for isoniazid and ethionamide in Mycobacterium tuberculosis Science 1994 263 5144 227 230 10.1126/science.8284673 8284673 32. Morlock GP Metchock B Sikes D Crawford JT Cooksey RC ethA, inhA, and katG loci of ethionamide-resistant clinical Mycobacterium tuberculosis isolates Antimicrob Agents Chemother 2003 47 12 3799 3805 10.1128/AAC.47.12.3799-3805.2003 14638486 33. Scardigli A Caminero JA Sotgiu G Centis R D’Ambrosio L Migliori GB Efficacy and tolerability of ethionamide versus prothionamide: a systematic review Eur Respir J 2016 48 3 946 952 10.1183/13993003.00438-2016 27288034 34. Guieu B Jourdan JP Dreneau A Willand N Rochais C Dallemagne P Desirable drug–drug interactions or when a matter of concern becomes a renewed therapeutic strategy Drug Discov Today 2021 26 2 315 328 10.1016/j.drudis.2020.11.026 33253919 35. Clinical Trial to Investigate the Safety, Tolerability and Pharmacokinetics of BVL-GSK098 in Healthy Volunteers [Internet]. 2022. https://clinicaltrials.gov/ct2/results?cond=&term=NCT04654143&cntry=&state=&city=&dist=. Accessed 20 Aug 2022. 36. Falzon D Hill G Pal SN Suwankesawong W Jaramillo E Pharmacovigilance and tuberculosis: applying the lessons of thioacetazone Bull World Health Organ 2014 92 12 918 919 10.2471/BLT.14.142570 25552777 37. Martínez-Hoyos M Perez-Herran E Gulten G Encinas L Álvarez-Gómez D Alvarez E Antitubercular drugs for an old target: GSK693 as a promising InhA direct inhibitor EBioMedicine 2016 8 291 301 10.1016/j.ebiom.2016.05.006 27428438 38. Sasaki H Haraguchi Y Itotani M Kuroda H Hashizume H Tomishige T Synthesis and antituberculosis activity of a novel series of optically active 6-nitro-2,3-dihydroimidazo[2,1-b]oxazoles J Med Chem 2006 49 26 7854 7860 10.1021/jm060957y 17181168 39. Liu Y Matsumoto M Ishida H Ohguro K Yoshitake M Gupta R Delamanid: from discovery to its use for pulmonary multidrug-resistant tuberculosis (MDR-TB) Tuberculosis 2018 111 20 30 10.1016/j.tube.2018.04.008 30029909 40. Shetye GS Franzblau SG Cho S New tuberculosis drug targets, their inhibitors, and potential therapeutic impact Transl Res 2020 220 68 97 10.1016/j.trsl.2020.03.007 32275897 41. Wen S Jing W Zhang T Zong Z Xue Y Shang Y Comparison of in vitro activity of the nitroimidazoles delamanid and pretomanid against multidrug-resistant and extensively drug-resistant tuberculosis Eur J Clin Microbiol Infect Dis 2019 38 1293 1296 10.1007/s10096-019-03551-w 30953211 42. Nguyen TVA Anthony RM Cao TTH Bañuls AL Nguyen VAT Vu DH Delamanid resistance: update and clinical management Clin Infect Dis 2020 71 12 3252 3259 10.1093/cid/ciaa755 32521000 43. Matsumoto M Hashizume H Tomishige T Kawasaki M Tsubouchi H Sasaki H OPC-67683, a nitro-dihydro-imidazooxazole derivative with promising action against tuberculosis in vitro and in mice PLoS Med 2006 3 2131 2144 10.1371/journal.pmed.0030466 44. Fujiwara M Kawasaki M Hariguchi N Liu Y Matsumoto M Mechanisms of resistance to delamanid, a drug for Mycobacterium tuberculosis Tuberculosis 2018 108 186 194 10.1016/j.tube.2017.12.006 29523322 45. Ryan NJ Lo JH Delamanid: first global approval Drugs 2014 74 9 1041 1045 10.1007/s40265-014-0241-5 24923253 46. Khoshnood S Taki E Sadeghifard N Kaviar VH Haddadi MH Farshadzadeh Z Mechanism of action, resistance, synergism, and clinical implications of delamanid against multidrug-resistant Mycobacterium tuberculosis Front Microbiol 2021 12 717045 10.3389/fmicb.2021.717045 34690963 47. Shimokawa Y Sasahara K Yoda N Mizuno K Umehara K Delamanid does not inhibit or induce cytochrome P450 enzymes in vitro Biol Pharm Bull 2014 37 11 1727 1735 10.1248/bpb.b14-00311 25366478 48. Gler MT Skripconoka V Sanchez-Garavito E Xiao H Cabrera-Rivero JL Vargas-Vasquez DE Delamanid for multidrug-resistant pulmonary tuberculosis N Engl J Med 2012 366 23 2151 2160 10.1056/NEJMoa1112433 22670901 49. Skripconoka V Danilovits M Pehme L Tomson T Skenders G Kummik T Delamanid improves outcomes and reduces mortality in multidrug-resistant tuberculosis Eur Respir J 2013 41 6 1393 1400 10.1183/09031936.00125812 23018916 50. Hafkin J Hittel N Martin A Gupta R Early outcomes in MDR-TB and XDR-TB patients treated with delamanid under compassionate use Eur Respir J 2017 50 1 1700311 10.1183/13993003.00311-2017 28751415 51. Hewison C Ferlazzo G Avaliani Z Hayrapetyan A Jonckheere S Khaidarkhanova Z Six-month response to delamanid treatment in MDR TB patients—volume 23, number 10—October 2017—emerging Infectious Diseases journal—CDC Emerg Infect Dis 2017 23 10 1746 1748 10.3201/eid2310.170468 28767036 52. Von Groote-Bidlingmaier F Patientia R Sanchez E Balanag V Jr Ticona E Segura P Efficacy and safety of delamanid in combination with an optimised background regimen for treatment of multidrug-resistant tuberculosis: a multicentre, randomised, double-blind, placebo-controlled, parallel group phase 3 trial Lancet Respir Med 2019 7 3 249 259 10.1016/S2213-2600(18)30426-0 30630778 53. Singh R Manjunatha U Boshoff HI Ha YH Niyomrattanakit P Ledwidge R PA-824 kills nonreplicating Mycobacterium tuberculosis by intracellular NO release Science (80-). 2008 322 5906 1392 1395 10.1126/science.1164571 54. Lenaerts AJ Gruppo V Marietta KS Johnson CM Driscoll DK Tompkins NM Preclinical testing of the nitroimidazopyran PA-824 for activity against Mycobacterium tuberculosis in a series of in vitro and in vivo models Antimicrob Agents Chemother 2005 49 6 2294 2301 10.1128/AAC.49.6.2294-2301.2005 15917524 55. Rifat D Li SY Ioerger T Shah K Lanoix JP Lee J Mutations in fbiD (Rv2983) as a novel determinant of resistance to pretomanid and delamanid in Mycobacterium tuberculosis Antimicrob Agents Chemother 2021 65 1 e01948-20 10.1128/AAC.01948-20 56. Haver HL Chua A Ghode P Lakshminarayana SB Singhal A Mathema B Mutations in genes for the F420 biosynthetic pathway and a nitroreductase enzyme are the primary resistance determinants in spontaneous in vitro-selected PA-824-resistant mutants of Mycobacterium tuberculosis Antimicrob Agents Chemother 2015 59 9 5316 5323 10.1128/AAC.00308-15 26100695 57. Xu J Li SY Almeida DV Tasneen R Barnes-Boyle K Converse PJ Contribution of pretomanid to novel regimens containing bedaquiline with either linezolid or moxifloxacin and pyrazinamide in murine models of tuberculosis Antimicrob Agents Chemother 2019 63 5 e00021-19 10.1128/AAC.00021-19 30833432 58. Trial to Evaluate the Efficacy, Safety and Tolerability of BPaMZ in Drug-Sensitive (DS-TB) Adult Patients and Drug-Resistant (DR-TB) Adult Patients [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03338621?term=NCT03338621&draw=2&rank=1. Accessed 13 May 2022. 59. Sacksteder KA Protopopova M Barry CE Andries K Nacy CA Discovery and development of SQ109: a new antitubercular drug with a novel mechanism of action Future Microbiol 2012 7 7 823 837 10.2217/fmb.12.56 22827305 60. Malwal SR Zimmerman MD Alvarez N Sarathy JP Dartois V Nacy CA Structure, in Vivo Detection, and Antibacterial Activity of Metabolites of SQ109, an Anti-Infective Drug Candidate ACS Infect Dis. 2021 7 8 2492 2507 10.1021/acsinfecdis.1c00259 34279904 61. Reddy VM Dubuisson T Einck L Wallis RS Jakubiec W Ladukto L SQ109 and PNU-100480 interact to kill Mycobacterium tuberculosis in vitro J Antimicrob Chemother 2012 67 1163 1166 10.1093/jac/dkr589 22258923 62. Heinrich N Dawson R Du Bois J Narunsky K Horwith G Phipps AJ Early phase evaluation of SQ109 alone and in combination with rifampicin in pulmonary TB patients J Antimicrob Chemother 2015 70 5 1558 1566 10.1093/jac/dku553 25630641 63. Reddy VM Einck L Andries K Nacy CA In vitro interactions between new antitubercular drug candidates SQ109 and TMC207 Antimicrob Agents Chemother 2010 54 7 2840 2846 10.1128/AAC.01601-09 20385864 64. Boeree MJ Heinrich N Aarnoutse R Diacon AH Dawson R Rehal S High-dose rifampicin, moxifloxacin, and SQ109 for treating tuberculosis: a multi-arm, multi-stage randomised controlled trial Lancet Infect Dis 2017 17 1 39 49 10.1016/S1473-3099(16)30274-2 28100438 65. Black TA Buchwalde UK The pipeline of new molecules and regimens against drug-resistant tuberculosis J Clin Tuberc Other Mycobact Dis. 2021 25 100285 10.1016/j.jctube.2021.100285 34816020 66. Tetali SR Kunapaeddi E Mailavaram RP Singh V Borah P Deb PK Current advances in the clinical development of anti-tubercular agents Tuberculosis 2020 125 101989 10.1016/j.tube.2020.101989 32957054 67. Oh S Trifonov L Yadav VD Barry CE Boshoff HI Tuberculosis drug discovery: a decade of hit assessment for defined targets Front Cell Infect Microbiol 2021 11 e611304 10.3389/fcimb.2021.611304 68. Man DKW Kanno T Manzo G Robertson BD Lam JKW Mason AJ Rifampin- or capreomycin-induced remodeling of the mycobacterium smegmatis mycolic acid layer is mitigated in synergistic combinations with cationic antimicrobial peptides mSphere. 2018 3 4 e00218-18 10.1128/mSphere.00218-18 30021876 69. Chikhale RV Barmade MA Murumkar PR Yadav MR Overview of the development of DprE1 inhibitors for combating the menace of tuberculosis J Med Chem 2018 61 19 8563 8593 10.1021/acs.jmedchem.8b00281 29851474 70. Mikušová K Huang H Yagi T Holsters M Vereecke D D’Haeze W Decaprenylphosphoryl arabinofuranose, the donor of the D-arabinofuranosyl residues of mycobacterial arabinan, is formed via a two-step epimerization of decaprenylphosphoryl ribose J Bacteriol 2005 187 23 8020 8025 10.1128/JB.187.23.8020-8025.2005 16291675 71. Brecik M Centárová I Mukherjee R Kolly GS Huszár S Bobovská A DprE1 Is a vulnerable tuberculosis drug target due to its cell wall localization ACS Chem Biol 2015 10 7 1631 1636 10.1021/acschembio.5b00237 25906160 72. Makarov V Manina G Mikusova K Möllmann U Ryabova O Saint-Joanis B Benzothiazinones kill Mycobacterium tuberculosis by blocking Arabinan synthesis Science 2009 324 5928 801 804 10.1126/science.1171583 19299584 73. Hariguchi N Chen X Hayashi Y Kawano Y Fujiwara M Matsuba M OPC-167832, a novel carbostyril derivative with potent antituberculosis activity as a DprE1 inhibitor Antimicrob Agents Chemother 2020 64 6 e02020–19 10.1128/AAC.02020-19 32229496 74. Chatterji M Shandil R Manjunatha MR Solapure S Ramachandran V Kumar N 1,4-azaindole, a potential drug candidate for treatment of tuberculosis Antimicrob Agents Chemother 2014 58 9 5325 5331 10.1128/AAC.03233-14 24957839 75. Guo S Fu L Wang B Chen X Zhao J Liu M In vitro and in vivo antimicrobial activities of a novel piperazine-containing benzothiazinones candidate TZY-5-84 against Mycobacterium tuberculosis Biomed Pharmacother 2020 131 110777 10.1016/j.biopha.2020.110777 33152936 76. Robertson GT Ramey ME Massoudi LM Carter CL Zimmerman M Kaya F Comparative analysis of pharmacodynamics in the c3heb/fej mouse tuberculosis model for DprE1 inhibitors TBA-7371, PBTZ169, and OPC-167832 Antimicrob Agents Chemother 2021 65 11 e00583–21 10.1128/AAC.00583-21 34370580 77. Makarov V Neres J Hartkoorn RC Ryabova OB Kazakova E Šarkan M The 8-pyrrole-benzothiazinones are noncovalent inhibitors of DprE1 from Mycobacterium tuberculosis Antimicrob Agents Chemother 2015 59 8 4446 4452 10.1128/AAC.00778-15 25987616 78. Makarov V Lechartier B Zhang M Neres J van der Sar AM Raadsen SA Towards a new combination therapy for tuberculosis with next generation benzothiazinones EMBO Mol Med 2014 6 3 372 383 10.1002/emmm.201303575 24500695 79. Lechartier B Hartkoorn RC Cole ST In vitro combination studies of benzothiazinone lead compound BTZ043 against Mycobacterium tuberculosis Antimicrob Agents Chemother 2012 56 11 5790 5793 10.1128/AAC.01476-12 22926573 80. A Single Ascending Dose Study of BTZ043 [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03590600?term=NCT03590600&draw=2&rank=1. Accessed 13 Aug 2022. 81. BTZ-043—Multiple Ascending Dose (MAD) to Evaluate Safety, Tolerability and Early Bactericidal Activity (EBA) [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT04044001?term=NCT04044001&draw=2&rank=1. Accessed 13 Aug 2022. 82. Study to Evaluate the Safety, Tolerability, Pharmacokinetics and Ex-vivo Antitubercular Activity of PBTZ169 Formulation [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03423030?term=NCT03423030&draw=2&rank=1. Accessed 13 Aug 2022. 83. Phase 1 Study of PBTZ169 [Internet]. 2022. https://clinicaltrials.gov/ct2/results?cond=&term=NCT03036163&cntry=&state=&city=&dist=. Accessed 13 Aug 2022. 84. Phase 2a Study of PBTZ169 [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03334734?term=NCT03334734&draw=2&rank=1. Accessed 13 Aug 2022. 85. A Phase 1/2 Trial of Multiple Oral Doses of OPC-167832 for Uncomplicated Pulmonary Tuberculosis [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03678688?term=NCT03678688&draw=2&rank=1. Accessed 13 Aug 2022. 86. Safety and Efficacy Evaluation of 4-month Regimen of OPC-167832, Delamanid and Bedaquiline in Participants With Drug-Susceptible Pulmonary TB [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT05221502?term=NCT05221502&draw=2&rank=1. Accessed 13 Aug 2022. 87. Shirude PS Shandil R Sadler C Naik M Hosagrahara V Hameed S Azaindoles: noncovalent DprE1 inhibitors from scaffold morphing efforts, kill Mycobacterium tuberculosis and are efficacious in vivo J Med Chem 2013 56 23 9701 9708 10.1021/jm401382v 24215368 88. Early Bactericidal Activity of TBA-7371 in Pulmonary Tuberculosis [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT04176250?term=NCT04176250&draw=2&rank=1. Accessed 13 Aug 2022. 89. van Heijenoort J. Formation of the glycan chains in the synthesis of bacterial peptidoglycan. Glycobiology. 2001;11(3):25R–36R. 90. Batson S De Chiara C Majce V Lloyd AJ Gobec S Rea D Inhibition of D-Ala:D-Ala ligase through a phosphorylated form of the antibiotic D-cycloserine Nat Commun 2017 8 1 1939 10.1038/s41467-017-02118-7 29208891 91. Lan Z Ahmad N Baghaei P Barkane L Benedetti A Brode SK Drug-associated adverse events in the treatment of multidrug-resistant tuberculosis: an individual patient data meta-analysis Lancet Respir Med 2020 8 4 383 394 10.1016/S2213-2600(20)30047-3 32192585 92. Diacon AH van der Merwe L Barnard M von Groote-Bidlingmaier F Lange C García-Basteiro AL β-Lactams against Tuberculosis-new trick for an old dog? N Engl J Med 2016 375 4 393 394 10.1056/NEJMc1513236 27433841 93. Hugonnet JE Blanchard JS Irreversible inhibition of the Mycobacterium tuberculosis β-lactamase by clavulanate Biochemistry 2007 46 43 11998 12004 10.1021/bi701506h 17915954 94. Hugonnet JE Tremblay LW Boshoff HI Barry CE Blanchard JS Meropenem-Clavulanate is effective against extensively drug-resistant Mycobacterium tuberculosis Science 2009 323 5918 1215 1218 10.1126/science.1167498 19251630 95. Cordillot M Dubée V Triboulet S Dubost L Marie A Hugonnet JE In vitro cross-linking of Mycobacterium tuberculosis peptidoglycan by L, D-transpeptidases and inactivation of these enzymes by carbapenems Antimicrob Agents Chemother 2013 57 12 5940 5945 10.1128/AAC.01663-13 24041897 96. De Jager V Gupte N Nunes S Barnes GL van Wijk RC Mostert J Early bactericidal activity of meropenem plus clavulanate (with or without rifampin) for tuberculosis: the COMRADE randomized, phase 2A clinical trial Am J Respir Crit Care Med 2022 205 10 1228 1235 10.1164/rccm.202108-1976OC 35258443 97. Doern GV Pierce G Brueggemann AB In vitro activity of sanfetrinem (GV104326), a new trinem antimicrobial agent, versus Streptococcus pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis Diagn Microbiol Infect Dis 1996 26 1 39 42 10.1016/S0732-8893(96)00173-3 8950528 98. EBA, Safety and Tolerability of Sanfetrinem Cilexetil [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT05388448?term=NCT05388448&draw=2&rank=1. Accessed 13 Aug 2022. 99. Ahmad N Ahuja SD Akkerman OW Alffenaar J Anderson L Baghaei P Treatment correlates of successful outcomes in pulmonary multidrug-resistant tuberculosis: an individual patient data meta-analysis Lancet 2018 392 10150 821 834 10.1016/S0140-6736(18)31644-1 30215381 100. Oehadian A Santoso P Menzies D Ruslami R Concise clinical review of hematologic toxicity of linezolid in multidrug-resistant and extensively drug-resistant tuberculosis: role of mitochondria Tuberc Respir Dis (Seoul). 2022 85 2 111 121 10.4046/trd.2021.0122 35045688 101. Narita M Tsuji BT Yu VL Linezolid-associated peripheral and optic neuropathy, lactic acidosis, and serotonin syndrome Pharmacotherapy 2007 27 8 1189 1197 10.1592/phco.27.8.1189 17655517 102. Kishor K Dhasmana N Kamble S Sahu R Linezolid induced adverse drug reactions—an update Curr Drug Metab 2015 16 7 553 559 10.2174/1389200216666151001121004 26424176 103. Pfizer. Zyvoxam Product Monograph [Internet]. https://www.pfizer.ca/sites/default/files/202203/ZYVOXAM_PM_EN_256427_27-Jan-2022.pdf 104. Huerga H, Khan U, Bastard M, Mitnick CD, Lachenal N, Khan PY, et al. Safety and effectiveness outcomes from a 14-country cohort of patients with multi-drug resistant tuberculosis treated concomitantly with bedaquiline, delamanid and other second-line drugs. Clin Infect Dis. 2022;ciac176. 105. Conradie F Bagdasaryan TR Borisov S Howell P Mikiashvili L Ngubane N Bedaquiline–pretomanid–linezolid regimens for drug-resistant tuberculosis N Engl J Med 2022 387 9 810 823 10.1056/NEJMoa2119430 36053506 106. Brown AN Drusano GL Adams JR Rodriquez JL Jambunathan K Baluya DL Preclinical evaluations to identify optimal linezolid regimens for tuberculosis therapy MBio 2015 6 6 e01741–15 10.1128/mBio.01741-15 26530386 107. Song T Lee M Jeon HS Park Y Dodd LE Dartois V Linezolid trough concentrations correlate with mitochondrial toxicity-related adverse events in the treatment of chronic extensively drug-resistant tuberculosis EBioMedicine 2015 2 11 1627 1633 10.1016/j.ebiom.2015.09.051 26870788 108. Ruiz P Causse M Vaquero M Casal M In vitro activity of tedizolid against Mycobacterium tuberculosis Antimicrob Agents Chemother 2019 63 e01939 e2018 10.1128/AAC.01939-18 30670412 109. Iqbal K Milioudi A Wicha SG Pharmacokinetics and pharmacodynamics of tedizolid Clin Pharmacokinet 2022 61 4 489 503 10.1007/s40262-021-01099-7 35128625 110. Wen S Gao X Zhao W Huo F Jiang G Dong L Comparison of the in vitro activity of linezolid, tedizolid, sutezolid, and delpazolid against rapidly growing mycobacteria isolated in Beijing China. Int J Infect Dis. 2021 109 253 260 10.1016/j.ijid.2021.06.055 34216736 111. Srivastava S Cirrincione KN Deshpande D Gumbo T Friberg LE Van WR Tedizolid, faropenem, and moxifloxacin combination with potential activity against nonreplicating Mycobacterium tuberculosis Front Pharmacol 2021 11 616294 10.3389/fphar.2020.616294 33542690 112. Deshpande D Srivastava S Nuermberger E Koeuth T Martin KR Cirrincione KN Multiparameter responses to tedizolid monotherapy and moxifloxacin combination therapy models of children with intracellular tuberculosis Clin Infect Dis 2018 67 Suppl 3 S342 S348 10.1093/cid/ciy612 30496456 113. Srivastava S Deshpande D Nuermberger E Lee PS Cirrincione K Dheda K The sterilizing effect of intermittent tedizolid for pulmonary tuberculosis Clin Infect Dis 2018 2018 3 336 377 10.1093/cid/ciy626 114. Aono A Murase Y Chikamatsu K Igarashi Y Shimomura Y Hosoya M In vitro activity of tedizolid and linezolid against multidrug-resistant Mycobacterium tuberculosis: a comparative study using microdilution broth assay and genomics Diagnostic Microbiol Infect Dis. 2022 103 115714 10.1016/j.diagmicrobio.2022.115714 115. Mikamo H Takesue Y Iwamoto Y Tanigawa T Kato M Tanimura Y Efficacy, safety and pharmacokinetics of tedizolid versus linezolid in patients with skin and soft tissue infections in Japan—results of a randomised, multicentre phase 3 study J Infect Chemother 2018 24 6 434 442 10.1016/j.jiac.2018.01.010 29530544 116. Flanagan S Bartizal K Minassian SL Fang E Prokocimer P In vitro, in vivo, and clinical studies of tedizolid to assess the potential for peripheral or central monoamine oxidase interactions Antimicrob Agents Chemother 2013 57 7 3060 3066 10.1128/AAC.00431-13 23612197 117. Lletí MS García-Bustos V Ruiz LM Cabañero-Navalon MD Tedizolid: new data and experiences for clinical practice Rev Española Quimioter. 2021 34 Suppl1 22 25 10.37201/req/s01.06.2021 118. Poon YK, La Hoz RM, Hynan LS, Sanders J, Monogue ML. Tedizolid vs Linezolid for the Treatment of Nontuberculous Mycobacteria Infections in Solid Organ Transplant Recipients. Open Forum Infect Dis. 2021;8(4):ofab093. 119. Yu X Huo F Wang F Wen S Jiang G Xue Y In vitro antimicrobial activity comparison of linezolid, tedizolid, sutezolid and delpazolid against slowly growing mycobacteria isolated in Beijing China. Infect Drug Resist. 2021 14 4689 4697 10.2147/IDR.S332835 34785916 120. Zhang M Sala C Dhar N Vocat A Sambandamurthy VK Sharma S In vitro and in vivo activities of three oxazolidinones against nonreplicating Mycobacterium tuberculosis Antimicrob Agents Chemother 2014 58 6 3217 3223 10.1128/AAC.02410-14 24663022 121. Yip PCW Kam KM Lam ETK Chan RCY Yew WW In vitro activities of PNU-100480 and linezolid against drug-susceptible and drug-resistant Mycobacterium tuberculosis isolates Int J Antimicrob Agents 2013 42 1 96 97 10.1016/j.ijantimicag.2013.03.002 23684005 122. Williams KN Brickner SJ Stover CK Zhu T Ogden A Tasneen R Addition of PNU-100480 to first-line drugs shortens the time needed to cure murine tuberculosis Am J Respir Crit Care Med 2009 180 4 371 376 10.1164/rccm.200904-0611OC 19520903 123. Wallis RS Dawson R Friedrich SO Venter A Paige D Mycobactericidal activity of sutezolid (PNU-100480) in sputum (EBA) and blood (WBA) of Patients with pulmonary tuberculosis PLoS ONE 2014 9 4 94462 10.1371/journal.pone.0094462 124. PanACEA Sutezolid Dose-finding and Combination Evaluation (SUDOCU) [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03959566?term=NCT03959566&draw=2&rank=1. Accessed 13 Aug 2022. 125. Zong Z Jing W Shi J Wen S Zhang T Huo F Comparison of in vitro activity and MIC distributions between the novel oxazolidinone delpazolid and linezolid against multidrug-resistant and extensively drug-resistant Mycobacterium tuberculosis in China Antimicrob Agents Chemother 2018 62 8 e00165-18 10.1128/AAC.00165-18 29844043 126. PanACEA DElpazolid Dose-finding and COmbination DEvelopment (DECODE) (DECODE) [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT04550832?term=delpazolid&cond=tuberculosis&draw=2&rank=2. Accessed 13 Aug 2022. 127. Sang Kim J Kim YH Haak Lee S Hyung Kim Y Kim JW Young Kang J Early bactericidal activity of delpazolid (LCB01-0371) in patients with pulmonary tuberculosis Antimicrob Agents Chemother 2022 66 2 e01684–21 10.1128/aac.01684-21 34871098 128. Choi Y Lee SW Kim A Jang K Nam H Cho YL Safety, tolerability and pharmacokinetics of 21 day multiple oral administration of a new oxazolidinone antibiotic, LCB01-0371, in healthy male subjects J Antimicrob Chemother 2018 73 183 190 10.1093/jac/dkx367 29069400 129. A Phase 1 Study to Evaluate Safety, Tolerability, and Pharmacokinetics of TBI-223 in Healthy Adults [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03758612?term=NCT03758612&draw=2&rank=1. Accessed 19 Jun 2022. 130. Jiang J Liu Y Liu X Zhang D Huang H Wang B Simultaneous determination of a novel oxazolidinone anti-tuberculosis OTB-658 and its metabolites in monkey blood by LC-MS/MS J Chromatogr B Anal Technol Biomed Life Sci 2020 2021 1167 122552 131. Guo S Wang B Fu L Chen X Zhang W Huang H In vitro and in vivo activity of oxazolidinone candidate OTB-658 against mycobacterium tuberculosis Antimicrob Agents Chemother 2021 65 11 e00974–21 10.1128/AAC.00974-21 34398674 132. Balasubramanian V Solapure S Iyer H Ghosh A Sharma S Kaur P Bactericidal activity and mechanism of action of AZD5847, a Novel oxazolidinone for treatment of tuberculosis Antimicrob Agents Chemother 2014 58 1 495 502 10.1128/AAC.01903-13 24189255 133. Furin JJ Du Bois J Van Brakel E Chheng P Venter A Peloquin CA Early Bactericidal activity of AZD5847 in patients with pulmonary tuberculosis Antimicrob Agents Chemother 2016 60 11 6591 6599 10.1128/AAC.01163-16 27550361 134. Palencia A Li X Bu W Choi W Ding CZ Easom EE Discovery of novel oral protein synthesis inhibitors of Mycobacterium tuberculosis that target Leucyl-tRNA synthetase Antimicrob Agents Chemother 2016 60 10 6271 6280 10.1128/AAC.01339-16 27503647 135. Bouz G Zitko J Inhibitors of aminoacyl-tRNA synthetases as antimycobacterial compounds: an up-to-date review Bioorg Chem 2021 110 104806 10.1016/j.bioorg.2021.104806 33799176 136. An Early Bactericidal Activity, Safety and Tolerability of GSK3036656 in Subjects With Drug-sensitive Pulmonary Tuberculosis [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03557281?term=NCT03557281&draw=2&rank=1. Accessed 21 Jun 2022. 137. Li X Hernandez V Rock FL Choi W Mak YSL Mohan M Discovery of a potent and specific M. tuberculosis Leucyl-tRNA synthetase inhibitor: (S)-3-(Aminomethyl)-4-chloro-7-(2-hydroxyethoxy)benzo[c][1,2]oxaborol-1(3H)-ol (GSK656) J Med Chem 2017 60 19 8011 8026 10.1021/acs.jmedchem.7b00631 28953378 138. Dooley KE Bliven-Sizemore E Weiner M Lu Y Nuermberger EL Hubbard W Safety and pharmacokinetics of escalating daily doses of the antituberculosis drug rifapentine in healthy volunteers Clin Pharmacol Ther 2012 91 5 881 888 10.1038/clpt.2011.323 22472995 139. Heifets L Lindholm-Levy P Flora M Bactericidal activity in vitro of various rifamycins against Mycobacterium avium and Mycobacterium tuberculosis Am Rev Respir Dis 1990 141 3 626 630 10.1164/ajrccm/141.3.626 2155555 140. Sterling TR Villarino ME Borisov AS Shang N Gordin F Bliven-Sizemore E Three months of rifapentine and isoniazid for latent tuberculosis infection N Engl J Med 2011 365 23 2155 2166 10.1056/NEJMoa1104875 22150035 141. Sterling TR Njie G Zenner D Cohn DL Reves R Ahmed A Guidelines for the treatment of latent tuberculosis infection: recommendations from the national tuberculosis controllers association and CDC, 2020 MMWR Recomm Rep 2020 69 RR-1 1 11 10.15585/mmwr.rr6901a1 142. Assessment of the safety, tolerability, and effectiveness of rifapentine given daily for LTBI (ASTERoiD) [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03474029?term=latent+tuberculosis&cond=rifapentine&draw=2&rank=1. Accessed 5 Nov 2022. 143. Rosenthal I Zhang M Williams K Peloquin CA Tyagi S Vernon A Daily dosing of rifapentine cures tuberculosis in three months or less in the murine model PLoS Med 2007 4 12 e344 10.1371/journal.pmed.0040344 18092886 144. Dorman SE Nahid P Kurbatova EV Phillips PPJ Bryant K Dooley KE Four-month rifapentine regimens with or without moxifloxacin for Tuberculosis N Engl J Med 2021 384 18 1705 1718 10.1056/NEJMoa2033400 33951360 145. World Health Organization. WHO consolidated guidelines on tuberculosis. Module 4: drug-susceptible tuberculosis treatment. 2022. p. 1–71. 146. Diacon AH Pym A Grobusch MP de los Rios JM Gotuzzo E Vasilyeva I Multidrug-resistant tuberculosis and culture conversion with bedaquiline N Engl J Med 2014 371 8 723 732 10.1056/NEJMoa1313865 25140958 147. Meyvisch P Kambili C Andries K Lounis N Theeuwes M Dannemann B Evaluation of six months sputum culture conversion as a surrogate endpoint in a multidrug resistant-tuberculosis trial PLoS ONE 2018 13 7 e0200539 10.1371/journal.pone.0200539 30024924 148. Janssen. Bedaquiline: Product Monograph [Internet]. 2012. https://www.accessdata.fda.gov/drugsatfda_docs/label/2012/204384s000lbl.pdf. Accessed 6 Nov 2022 149. Schnippel K Ndjeka N Maartens G Meintjes G Master I Ismail N Effect of bedaquiline on mortality in South African patients with drug-resistant tuberculosis: a retrospective cohort study Lancet Respir Med 2018 6 9 699 706 10.1016/S2213-2600(18)30235-2 30001994 150. Borisov SE Dheda K Enwerem M Leyet RR D’Ambrosio L Centis R Effectiveness and safety of bedaquiline-containing regimens in the treatment of MDR- and XDR-TB: a multicentre study Eur Respir J 2017 49 5 e1700387 10.1183/13993003.00387-2017 151. Esmail A Oelofse S Lombard C Perumal R Mbuthini L Mahomed AG An all-oral 6-month regimen for multidrug-resistant tuberculosis a multicenter, randomized controlled clinical trial (the NExT study) Am J Respir Crit Care Med 2022 205 10 1214 1227 10.1164/rccm.202107-1779OC 35175905 152. Wang MG Wu SQ He JQ Efficacy of bedaquiline in the treatment of drug-resistant tuberculosis: a systematic review and meta-analysis BMC Infect Dis 2021 21 1 970 10.1186/s12879-021-06666-8 34535090 153. Tweed CD Dawson R Burger DA Conradie A Crook AM Mendel CM Bedaquiline, moxifloxacin, pretomanid, and pyrazinamide during the first 8 weeks of treatment of patients with drug-susceptible or drug-resistant pulmonary tuberculosis: a multicentre, open-label, partially randomised, phase 2b trial Lancet Respir Med 2019 7 12 1048 1058 10.1016/S2213-2600(19)30366-2 31732485 154. Dooley KE Rosenkranz SL Conradie F Moran L Hafner R von Groote-Bidlingmaier F QT effects of bedaquiline, delamanid, or both in patients with rifampicin-resistant tuberculosis: a phase 2, open-label, randomised, controlled trial Lancet Infect Dis 2021 21 7 975 983 10.1016/S1473-3099(20)30770-2 33587897 155. Dodd PJ Mafirakureva N Seddon JA McQuaid CF The global impact of household contact management for children on multidrug-resistant and rifampicin-resistant tuberculosis cases, deaths, and health-system costs in 2019: a modelling study Lancet Glob Heal 2022 10 7 e1034 e1044 10.1016/S2214-109X(22)00113-9 156. Ahmad I Jadhav H Shinde Y Jagtap V Girase R Patel H Optimizing Bedaquiline for cardiotoxicity by structure based virtual screening, DFT analysis and molecular dynamic simulation studies to identify selective MDR-TB inhibitors Silico Pharmacol. 2021 9 1 23 10.1007/s40203-021-00086-x 157. Pontali E Sotgiu G Tiberi S D’Ambrosio L Centis R Migliori GB Cardiac safety of bedaquiline: a systematic and critical analysis of the evidence Eur Respir J 2017 50 5 1701462 10.1183/13993003.01462-2017 29146605 158. Mase S Chorba T Lobue P Castro KG Provisional CDC guidelines for the use and safety monitoring of bedaquiline fumarate (sirturo) for the treatment of multidrug-resistant tuberculosis MMWR Reports Recomm. 2013 62 RR09 1 12 159. Fox GJ Menzies D A review of the evidence for using bedaquiline (TMC207) to treat multi-drug resistant tuberculosis Infect Dis Ther 2013 2 2 123 144 10.1007/s40121-013-0009-3 25134476 160. Xu J Converse PJ Upton AM Mdluli K Fotouhi N Nuermberger E Comparative efficacy of the novel diarylquinoline TBAJ-587 and bedaquiline against a resistant Rv0678 mutant in a mouse model of tuberculosis Antimicrob Agents Chemother 2021 65 4 e02418–20 10.1128/AAC.02418-20 33526488 161. Svensson EM Murray S Karlsson MO Dooley KE Rifampicin and rifapentine significantly reduce concentrations of bedaquiline, a new anti-TB drug J Antimicrob Chemother 2015 70 4 1106 1114 10.1093/jac/dku504 25535219 162. Svensson EM Dooley KE Karlsson MO Impact of lopinavir-ritonavir or nevirapine on bedaquiline exposures and potential implications for patients with tuberculosis-HIV coinfection Antimicrob Agents Chemother 2014 58 11 6406 6412 10.1128/AAC.03246-14 25114140 163. Almeida D Converse PJ Li SY Upton AM Fotouhi N Nuermberger EL Comparative efficacy of the novel diarylquinoline TBAJ-876 and bedaquiline against a resistant Rv0678 mutant in a mouse model of tuberculosis Antimicrob Agents Chemother 2021 65 12 e0141221 10.1128/AAC.01412-21 34570644 164. Evaluate Safety, Tolerability, PK of TBAJ-876 in Healthy Adults [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT04493671?term=NCT04493671&draw=2&rank=1. Accessed 13 Jun 2022. 165. Sutherland HS Tong AST Choi PJ Blaser A Franzblau SG Cooper CB Variations in the C-unit of bedaquiline provides analogues with improved biology and pharmacology Bioorg Med Chem 2020 28 1 115213 10.1016/j.bmc.2019.115213 31810890 166. Evaluation of the Safety, Tolerability, PK of TBAJ-587 in Healthy Adults [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT04890535?term=tbaj587&cond=tuberculosis&draw=2&rank=1. Accessed 12 Aug 2022. 167. Yao R Wang B Fu L Li L You K Li YG Sudapyridine (WX-081), a novel compound against Mycobacterium tuberculosis Microbiol Spectr. 2022 10 1 e02477–21 10.1128/spectrum.02477-21 35170994 168. Evaluation of Early Bactericidal Activity and Safety in Pulmonary Tuberculosis With WX-081 (WX-081) [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT04608955?term=NCT04608955&draw=2&rank=1. Accessed 28 May 2022. 169. Pethe K Bifani P Jang J Kang S Park S Ahn S Discovery of Q203, a potent clinical candidate for the treatment of tuberculosis Nat Med 2013 19 1157 1160 10.1038/nm.3262 23913123 170. Lee BS Pethe K Telacebec: an investigational antibacterial for the treatment of tuberculosis (TB) Expert Opin Investig Drugs 2022 31 2 139 144 10.1080/13543784.2022.2030309 171. Kim J Choi J Kang H Ahn J Hutchings J van Niekerk C Safety, tolerability, and pharmacokinetics of telacebec (Q203), a new antituberculosis agent, in healthy subjects Antimicrob Agents Chemother 2022 66 1 e01436–21 10.1128/AAC.01436-21 34694872 172. A Phase 2 Study to Evaluate Early Bactericidal Activity, Safety, Tolerability, and Pharmacokinetics of Multiple Oral Doses of Telacebec (Q203) [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT03563599?term=NCT03563599&draw=1&rank=1. Accessed 2 Jun 2022. 173. de Jager VR Dawson R van Niekerk C Hutchings J Kim J Vanker N Telacebec (Q203), a new antituberculosis agent N Engl J Med 2020 382 13 1280 1281 10.1056/NEJMc1913327 32212527 174. Almeida DV Converse PJ Omansen TF Tyagi S Tasneen R Kim J Telacebec for ultrashort treatment of buruli ulcer in a mouse model Antimicrob Agents Chemother 2020 64 6 e00259–20 10.1128/AAC.00259-20 32205344 175. Bvumbi MV Activity of riminophenazines against Mycobacterium tuberculosis: a review of studies that might be contenders for use as antituberculosis agents Chem Med Chem. 2020 15 23 2207 2219 10.1002/cmdc.202000580 32844566 176. Cholo MC Mothiba MT Fourie B Anderson R Mechanisms of action and therapeutic efficacies of the lipophilic antimycobacterial agents clofazimine and bedaquiline J Antimicrob Chemother 2017 72 2 338 353 10.1093/jac/dkw426 27798208 177. Zhang Y Zhu H Fu L Wang B Guo S Chen X Identifying regimens containing TBI-166, a new drug candidate against Mycobacterium tuberculosis in vitro and in vivo Antimicrob Agents Chemother 2019 63 7 e02496–18 10.1128/AAC.02496-18 31061157 178. Lu Y Wang B Fu L Zhu H Guo S Huang H Clofazimine analogs with efficacy against experimental tuberculosis and reduced potential for accumulation Antimicrob Agents Chemother 2011 55 5185 5193 10.1128/AAC.00699-11 21844321 179. Zhang D Liu Y Zhang C Zhang H Wang B Xu J Synthesis and biological evaluation of novel 2-methoxypyridylamino-substituted riminophenazine derivatives as antituberculosis agents Molecules 2014 19 4 4380 4394 10.3390/molecules19044380 24722591 180. Early Bactericidal Activity Safety Pulmonary Tuberculosis Pyrifazimine (TBI-166) [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT04670120?term=NCT04670120&draw=2&rank=1. Accessed 14 Jun 2022. 181. Liu H Fu L Wang B Wang N Li D Ding Y Study on the pharmacodynamic activity of combinations with the new anti-tuberculosis drug pyrifazimine in vitro and in vivo in mouse Zhonghua Zhong Liu Za Zhi 2022 45 6 560 566 182. Gillespie SH Crook AM McHugh TD Mendel CM Meredith SK Murray SR Four-month moxifloxacin-based regimens for drug-sensitive tuberculosis N Engl J Med 2014 371 17 1577 1587 10.1056/NEJMoa1407426 25196020 183. Merle CS Fielding K Sow OB Gninafon M Lo MB Mthiyane T A four-month gatifloxacin-containing regimen for treating tuberculosis N Engl J Med 2014 371 17 1588 1598 10.1056/NEJMoa1315817 25337748 184. Talley AK, Thurston A, Moore G, Gupta VK, Satterfield M, Manyak E, et al. First-in-human evaluation of the safety, tolerability, and pharmacokinetics of SPR720, a novel oral bacterial DNA gyrase (GyrB) inhibitor for mycobacterial infections. Antimicrob Agents Chemother. 2021;65(11):e012021. 185. SperoTherapeutics. Spero Therapeutics Announces Lifting of FDA Clinical Trial Hold on SPR720 [Internet]. 2022. https://investors.sperotherapeutics.com/news-releases/news-release-details/spero-therapeutics-announces-lifting-fda-clinical-trial-hold. Accessed 22 Jun 2022. 186. SperoTherapeutics. SPR720: An Oral Antibiotic Designed to Treat Nontuberculous Mycobacterial Pulmonary Disease, a Rare Orphan Disease [Internet]. 2022. https://sperotherapeutics.com/pipeline/spr720-non-tuberculosis-mycobacterium/. Accessed 22 Jun 2022. 187. Locher CP Jones SM Hanzelka BL Perola E Shoen CM Cynamon MH A novel inhibitor of gyrase B is a potent drug candidate for treatment of tuberculosis and nontuberculosis mycobacterial infections Antimicrob Agents Chemother 2015 59 3 1455 1465 10.1128/AAC.04347-14 25534737 188. Nuermberger EL Martínez-Martínez MS Sanz O Urones B Esquivias J Soni H GSK2556286 is a novel antitubercular drug candidate effective in vivo with the potential to shorten tuberculosis treatment Antimicrob Agents Chemother 2022 66 6 e0013222 10.1128/aac.00132-22 35607978 189. A Study to Evaluate Safety, Tolerability and Pharmacokinetics of GSK2556286 in Healthy Adult Participants [Internet]. 2022. https://clinicaltrials.gov/ct2/show/NCT04472897?term=NCT04472897&draw=2&rank=1. Accessed 2 Jun 2022. 190. Amaral L Boeree MJ Gillespie SH Udwadia ZF Van SD Thioridazine cures extensively drug-resistant (XDR-TB) and the need for global trials is now! Int J Antimicrob Agents 2010 35 524 526 10.1016/j.ijantimicag.2009.12.019 20188526 191. Vesenbeckh S Krieger D Bettermann G Schönfeld N Bauer TT Rüssmann H Neuroleptic drugs in the treatment of tuberculosis: minimal inhibitory concentrations of different phenothiazines against Mycobacterium tuberculosis Tuberculosis (Edinb) 2016 98 27 29 10.1016/j.tube.2016.02.003 27156615 192. Field SK Fisher D Jarand JM Cowie RL New treatment options for multidrug-resistant tuberculosis Ther Adv Respir Dis 2012 6 5 255 268 10.1177/1753465812452193 22763676 193. Liu Y, Xu X, Zhang Y, Li M, Guo J, Yan C, et al. Thioridazine induces cardiotoxicity via reactive oxygen species-mediated herg channel deficiency and L-type calcium channel activation. Oxid Med Cell Longev. 2020;2020:3690123 194. Martins M Schelz Z Martins A Molnar J Hajös G Riedl Z In vitro and ex vivo activity of thioridazine derivatives against Mycobacterium tuberculosis Int J Antimicrob Agents 2007 29 3 338 340 10.1016/j.ijantimicag.2006.10.013 17239569 195. Kristiansen JE Dastidar SG Palchoudhuri S Roy DS Das S Hendricks O Phenothiazines as a solution for multidrug resistant tuberculosis: from the origin to present Int Microbiol 2015 18 1 1 12 26415662 196. Gupta S Tyagi S Almeida DV Maiga MC Ammerman NC Bishai WR Acceleration of tuberculosis treatment by adjunctive therapy with verapamil as an efflux inhibitor Am J Respir Crit Care Med 2013 188 5 600 607 10.1164/rccm.201304-0650OC 23805786 197. Kaur P Potluri V Ahuja VK Naveenkumar CN Krishnamurthy RV Gangadharaiah ST A multi-targeting pre-clinical candidate against drug-resistant tuberculosis Tuberculosis 2021 129 March 102104 10.1016/j.tube.2021.102104 34214859	36479687	PMC9734533	NO-CC CODE	2022-12-14 23:28:29	no	Drugs. 2022 Dec 7; 82(18):1695-1715	utf-8	Drugs	2,022	10.1007/s40265-022-01817-w	oa_other
==== Front Nano Res Nano Res Nano Research 1998-0124 1998-0000 Tsinghua University Press Beijing 5153 10.1007/s12274-022-5153-1 Research Article Enhancing bioactivity and stability of polymer-based material-tissue interface through coupling multiscale interfacial interactions with atomic-thin TiO2 nanosheets Xu Rongchen 12 Mu Xiaodan 1 Hu Zunhan 3 Jia Chongzhi 1 Yang Zhenyu 4 Yang Zhongliang 1 Fan Yiping 1 Wang Xiaoyu 15 Wu Yuefeng 6 Lu Xiaotong 6 Chen Jihua [email protected] 4 Xiang Guolei [email protected] 6 Li Hongbo [email protected] 1 1 grid.414252.4 0000 0004 1761 8894 Department of Stomatology, The First Medical Center, Chinese PLA General Hospital, Beijing, 100853 China 2 grid.414252.4 0000 0004 1761 8894 Department of Stomatology, The Third Medical Center, Chinese PLA General Hospital, Beijing, 100039 China 3 grid.285847.4 0000 0000 9588 0960 Department of Stomatology, Kunming Medical University, Kunming, 650500 China 4 grid.233520.5 0000 0004 1761 4404 National Clinical Research Center for Oral Diseases & Shaanxi Key Laboratory of Stomatology, Department of Prosthodontics, School of Stomatology, The Fourth Military Medical University, Xi’an, 710032 China 5 Department of Stomatology, The Strategic Support Force Medical Center, Beijing, 100101 China 6 grid.48166.3d 0000 0000 9931 8406 State Key Laboratory of Chemical Resource Engineering, College of Chemistry, Beijing University of Chemical Technology, Beijing, 100029 China 5 12 2022 19 9 8 2022 30 9 2022 3 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. Stable and bioactive material—tissue interface (MTF) basically determines the clinical applications of biomaterials in wound healing, sustained drug release, and tissue engineering. Although many inorganic nanomaterials have been widely explored to enhance the stability and bioactivity of polymer-based biomaterials, most are still restricted by their stability and biocompatibility. Here we demonstrate the enhanced bioactivity and stability of polymer-matrix bio-composite through coupling multiscale material—tissue interfacial interactions with atomically thin TiO2 nanosheets. Resin modified with TiO2 nanosheets displays improved mechanical properties, hydrophilicity, and stability. Also, we confirm that this resin can effectively stimulate the adhesion, proliferation, and differentiation into osteogenic and odontogenic lineages of human dental pulp stem cells using in vitro cell—resin interface model. TiO2 nanosheets can also enhance the interaction between demineralized dentinal collagen and resin. Our results suggest an approach to effectively up-regulate the stability and bioactivity of MTFs by designing biocompatible materials at the sub-nanoscale. Electronic Supplementary Material Supplementary material (further details of fabrication and characterization of TiO2 NSs and TiO2-ARCs, the bioactivity evaluation of TiO2-ARCs on hDPSCs, and the measurement of interaction with demineralized dentin collagen) is available in the online version of this article at 10.1007/s12274-022-5153-1. Keywords material—tissue interface TiO2 nanosheets pulpo-dentinal complex biomaterial resin composite ==== Body pmcElectronic supplementary material Enhancing bioactivity and stability of polymer-based material-tissue interface through coupling multiscale interfacial interactions with atomic-thin TiO2 nanosheets Acknowledgements This work was supported by the National Natural Science Foundation of China (Nos. 82001110, 82071154, 21801012, 81720108011, 81470773, and 81571013). Rongchen Xu, Xiaodan Mu, and Zunhan Hu contributed equally to this work. ==== Refs References [1] Zhang H Zhang Z H Zhang H Chen C W Zhang D G Zhao Y J Protein-based hybrid responsive microparticles for wound healing ACS Appl. Mater. Interfaces 2021 13 18413 18422 10.1021/acsami.0c19884 33856190 [2] Wang Y A Zhang J W Zhao Y Pu M J Song X Y Yu L M Yan X F Wu J He Z Y Innovations and challenges of polyphenol-based smart drug delivery systems Nano Res. 2022 15 8156 8184 10.1007/s12274-022-4430-3 [3] Kasper M Ellenbogen B Hardy R Cydis M Mojica-Santiago J Afridi A Spearman B S Singh I Kuliasha C A Atkinson E Development of a magnetically aligned regenerative tissue-engineered electronic nerve interface for peripheral nerve applications Biomaterials 2021 279 121212 10.1016/j.biomaterials.2021.121212 34717196 [4] Nishimura T Sasaki Y Akiyoshi K Biotransporting self-assembled nanofactories using polymer vesicles with molecular permeability for enzyme prodrug cancer therapy Adv. Mater. 2017 29 1702406 10.1002/adma.201702406 [5] Van Landuyt K L Snauwaert J De Munck J Peumans M Yoshida Y Poitevin A Coutinho E Suzuki K Lambrechts P Van Meerbeek B Systematic review of the chemical composition of contemporary dental adhesives Biomaterials 2007 28 3757 3785 10.1016/j.biomaterials.2007.04.044 17543382 [6] de Paula Rodrigues M Soares P B F Gomes M A B Pereira R A Tantbirojn D Versluis A Soares C J Direct resin composite restoration of endodontically-treated permanent molars in adolescents: Bite force and patient-specific finite element analysis J. Appl. Oral. Sci. 2020 28 e20190544 10.1590/1678-7757-2019-0544 32348440 [7] Maleckis K Kamenskiy A Lichter E Z Oberley-Deegan R Dzenis Y MacTaggart J Mechanically tuned vascular graft demonstrates rapid endothelialization and integration into the porcine iliac artery wall Acta Biomater. 2021 125 126 137 10.1016/j.actbio.2021.01.047 33549808 [8] Groen W M Diloksumpan P Van Weeren P R Levato R Malda J From intricate to integrated: Biofabrication of articulating joints J. Orthop. Res. 2017 35 2089 2097 10.1002/jor.23602 28621834 [9] Gong H Y Hajizadeh S Liu W F Ye L Imprinted polymer beads loaded with silver nanoparticles for antibacterial applications ACS Appl. Bio Mater. 2021 4 2829 2838 10.1021/acsabm.1c00045 [10] Zheng C Zhang X L Liu W Liu B Yang H H Lin Z A Chen G N A selective artificial enzyme inhibitor based on nanoparticle—enzyme interactions and molecular imprinting Adv. Mater. 2013 25 5922 5927 10.1002/adma.201302064 23955936 [11] Liu T Y Li L Cheng C He B F Jiang T Y Emerging prospects of protein/peptide-based nanoassemblies for drug delivery and vaccine development Nano Res. 2022 15 7267 7285 10.1007/s12274-022-4385-4 35692441 [12] Park G Kim H O Lim J W Park C Yeom M Song D Haam S Rapid detection of influenza A (H1N1) virus by conductive polymer-based nanoparticle via optical response to virus-specific binding Nano Res. 2022 15 2254 2262 10.1007/s12274-021-3772-6 34567436 [13] Tian X J Wang Y Duan S M Hao Y J Zhao K X Li Y Dai R T Wang W H Evaluation of a novel nano-size collagenous matrix film cross-linked with gallotannins catalyzed by laccase Food Chem. 2021 351 129335 10.1016/j.foodchem.2021.129335 33662910 [14] Dhand A P Galarraga J H Burdick J A Enhancing biopolymer hydrogel functionality through interpenetrating networks Trends Biotechnol. 2021 39 519 538 10.1016/j.tibtech.2020.08.007 32950262 [15] Martin J R Howard M T Wang S Berger A G Hammond P T Oxidation-responsive, tunable growth factor delivery from polyelectrolyte-coated implants Adv. Healthc. Mater. 2021 10 2001941 10.1002/adhm.202001941 [16] Gu L S Shan T T Ma Y X Tay F R Niu L N Novel biomedical applications of crosslinked collagen Trends Biotechnol. 2019 37 464 491 10.1016/j.tibtech.2018.10.007 30447877 [17] He Q L Liao Y G Zhang J W Yao X D Zhou W Y Hong Y Ouyang H W “All-in-one” gel system for whole procedure of stem-cell amplification and tissue engineering Small 2020 16 1906539 10.1002/smll.201906539 [18] Tian K Suo Z G Vlassak J J Chemically coupled interfacial adhesion in multimaterial printing of hydrogels and elastomers ACS Appl. Mater. Interfaces 2020 12 31002 31009 10.1021/acsami.0c07468 32536152 [19] Pinnaratip R Bhuiyan M S A Meyers K Rajachar R M Lee B P Multifunctional biomedical adhesives Adv. Healthc. Mater. 2019 8 1801568 10.1002/adhm.201801568 [20] Zhou J Gao Z Xiang G L Zhai T Y Liu Z K Zhao W X Liang X Wang L Y Interfacial compatibility critically controls Ru/TiO2 metal—support interaction modes in CO2 hydrogenation Nat. Commun. 2022 13 327 10.1038/s41467-021-27910-4 35039518 [21] Tang X M Huang K Dai J Wu Z Y Cai L Yang L L Wei J Sun H L Influences of surface treatments with abrasive paper and sand-blasting on surface morphology, hydrophilicity, mineralization and osteoblasts behaviors of n-CS/PK composite Sci. Rep. 2017 7 568 10.1038/s41598-017-00571-4 28373673 [22] Cabrera I C Berlioz S Fahs A Louarn G Carriere P Chemical functionalization of Nano fibrillated cellulose by glycidyl silane coupling agents: A grafted silane network characterization study Int. J. Biol. Macromol. 2020 165 1773 1782 10.1016/j.ijbiomac.2020.10.045 33075339 [23] Xu R C Yu F Huang L Zhou W Wang Y Wang F Sun X Chang G Fang M Zhang L Isocyanate-terminated urethane-based dental adhesive bridges dentinal matrix collagen with adhesive resin Acta Biomater. 2019 83 140 152 10.1016/j.actbio.2018.11.007 30414487 [24] Arno M C Inam M Weems A C Li Z H Binch A L A Platt C I Richardson S M Hoyland J A Dove A P O’Reilly R K Exploiting the role of nanoparticle shape in enhancing hydrogel adhesive and mechanical properties Nat. Commun. 2020 11 1420 10.1038/s41467-020-15206-y 32184392 [25] Stojanović Z S Ignjatović N Wu V Žunič V Veselinović L Škapin S Miljković M Uskoković V Uskoković D Hydrothermally processed 1D hydroxyapatite: Mechanism of formation and biocompatibility studies Mater. Sci. Eng.:C 2016 68 746 757 10.1016/j.msec.2016.06.047 [26] Zhang Y G Li J P Mouser V H M Roumans N Moroni L Habibovic P Biomimetic mechanically strong one-dimensional hydroxyapatite/poly(D, L-lactide) composite inducing formation of anisotropic collagen matrix ACS Nano 2021 15 17480 17498 10.1021/acsnano.1c03905 34662097 [27] Zhang J Y Chen H L Zhao M Liu G T Wu J 2D nanomaterials for tissue engineering application Nano Res. 2020 13 2019 2034 10.1007/s12274-020-2835-4 [28] Azizi-Lalabadi M Jafari S M Bio-nanocomposites of graphene with biopolymers; fabrication, properties, and applications Adv. Colloid Interface Sci. 2021 292 102416 10.1016/j.cis.2021.102416 33872984 [29] Qu G B Xia T Zhou W H Zhang X Zhang H Y Hu L G Shi J B Yu X F Jiang G B Property—activity relationship of black phosphorus at the nano—bio interface: From molecules to organisms Chem. Rev. 2020 120 2288 2346 10.1021/acs.chemrev.9b00445 31971371 [30] Chuang Y C Chang C C Yang F Simon M Rafailovich M TiO2 nanoparticles synergize with substrate mechanics to improve dental pulp stem cells proliferation and differentiation Mater. Sci. Eng.:C 2021 118 111366 10.1016/j.msec.2020.111366 [31] Kantovitz K R Fernandes F P Feitosa I V Lazzarini M O Denucci G C Gomes O P Giovani P A Moreira K M S Pecorari V G A Borges A F S TiO2 nanotubes improve physico-mechanical properties of glass ionomer cement Dent. Mater. 2020 36 e85 e92 10.1016/j.dental.2020.01.018 32019690 [32] Wang M H Wang Q Wang K F Lu X Functionalized TiO2 surfaces facilitate selective receptor-recognition and modulate biological function of bone morphogenetic protein-2 J. Phys. Chem. C 2018 122 29319 29329 10.1021/acs.jpcc.8b09492 [33] Xiang G L Tang Y Liu Z G Zhu W Liu H T Wang J O Zhong G M Li J Wang X Probing ligand-induced cooperative orbital redistribution that dominates nanoscale molecule—surface interactions with one-unit-thin TiO2 nanosheets Nano Lett. 2018 18 7809 7815 10.1021/acs.nanolett.8b03572 30407013 [34] Xiang G L Wang Y G Exploring electronic-level principles how size reduction enhances nanomaterial surface reactivity through experimental probing and mathematical modeling Nano Res. 2022 15 3812 3817 10.1007/s12274-021-3910-1 [35] Ma S Q Zhao W X Zhou J Wang J O Chu S Q Liu Z G Xiang G L A new type of noncovalent surface—π stacking interaction occurring on peroxide-modified titania nanosheets driven by vertical π-state polarization Chem. Sci. 2021 12 4411 4417 10.1039/D0SC06601J 34163705 [36] Yamauchi S Wang X Egusa H Sun J High-performance dental adhesives containing an ether-based monomer J. Dent. Res. 2020 99 189 195 10.1177/0022034519895269 31861961 [37] Bendary I M Garcia I M Collares F M Takimi A Samuel S M W Leitune V C B Wollastonite as filler of an experimental dental adhesive J. Dent. 2020 102 103472 10.1016/j.jdent.2020.103472 32927019 [38] Montoya C Jain A Londoño J J Correa S Lelkes P I Melo M A Orrego S Multifunctional dental composite with piezoelectric nanofillers for combined antibacterial and mineralization effects ACS Appl. Mater. Interfaces 2021 13 43868 43879 10.1021/acsami.1c06331 34494813 [39] Cadenaro M Maravic T Comba A Mazzoni A Fanfoni L Hilton T Ferracane J Breschi L The role of polymerization in adhesive dentistry Dent Mater. 2019 35 e1 e22 10.1016/j.dental.2018.11.012 30554830 [40] Ye K Wang X Cao L P Li S Y Li Z H Yu L Ding J D Matrix stiffness and nanoscale spatial organization of cell-adhesive ligands direct stem cell fate Nano Lett. 2015 15 4720 4729 10.1021/acs.nanolett.5b01619 26027605 [41] Martos R Hegedüs V Szalóki M Blum I R Lynch C D Hegedüs C A randomised controlled study on the effects of different surface treatments and adhesive self-etch functional monomers on the immediate repair bond strength and integrity of the repaired resin composite interface J. Dent. 2019 85 57 63 10.1016/j.jdent.2019.04.012 31054292 [42] Yao C M Ahmed M H Li X Nedeljkovic I Vandooren J Mercelis B Zhang F Van Landuyt K L Huang C Van Meerbeek B Zinc-calcium-fluoride bioglass-based innovative multifunctional dental adhesive with thick adhesive resin film thickness ACS Appl. Mater. Interfaces 2020 12 30120 30135 10.1021/acsami.0c06865 32530270 [43] Yu F Xu R C Huang L Luo M L Li J Tay F R Niu L N Chen J H Isocyanate-terminated urethane-based methacrylate for in situ collagen scaffold modification Mater. Sci. Eng.:C 2020 112 110902 10.1016/j.msec.2020.110902 [44] Chen K Tang X K Jia B B Chao C Z Wei Y Hou J Y Dong L T Deng X L Xiao T H Goda K Graphene oxide bulk material reinforced by heterophase platelets with multiscale interface crosslinking Nat. Mater. 2022 21 1121 1129 10.1038/s41563-022-01292-4 35798946 [45] Hosseinpour S Tang F J Wang F L Livingstone R A Schlegel S J Ohto T Bonn M Nagata Y Backus E H G Chemisorbed and physisorbed water at the TiO2/water interface J. Phys. Chem. Lett. 2017 8 2195 2199 10.1021/acs.jpclett.7b00564 28447795 [46] Yoshimitsu Z Nakajima A Watanabe T Hashimoto K Effects of surface structure on the hydrophobicity and sliding behavior of water droplets Langmuir 2002 18 5818 5822 10.1021/la020088p [47] Shakir M Jolly R Khan A A Ahmed S S Alam S Rauf M A Owais M Farooqi M A Resol based chitosan/nanohydroxyapatite nanoensemble for effective bone tissue engineering Carbohydr. Polym. 2018 179 317 327 10.1016/j.carbpol.2017.09.103 29111057 [48] Chen L Suh B I Cytotoxicity and biocompatibility of resin-free and resin-modified direct pulp capping materials: A state-of-the-art review Dent. Mater. J. 2017 36 1 7 10.4012/dmj.2016-107 27928102 [49] Garcia I M Balhaddad A A Lan Y C Simionato A Ibrahim M S Weir M D Masri R Xu H H K Collares F M Melo M A S Magnetic motion of superparamagnetic iron oxide nanoparticles-loaded dental adhesives: Physicochemical/biological properties, and dentin bonding performance studied through the tooth pulpal pressure model Acta Biomater. 2021 134 337 347 10.1016/j.actbio.2021.07.031 34303014 [50] Di Cio S Gautrot J E Cell sensing of physical properties at the nanoscale: Mechanisms and control of cell adhesion and phenotype Acta Biomater. 2016 30 26 48 10.1016/j.actbio.2015.11.027 26596568 [51] Cui Y D Hameed F M Yang B Lee K Pan C Q Park S Sheetz M Cyclic stretching of soft substrates induces spreading and growth Nat. Commun. 2015 6 6333 10.1038/ncomms7333 25704457 [52] Bacakova L Filova E Parizek M Ruml T Svorcik V Modulation of cell adhesion, proliferation and differentiation on materials designed for body implants Biotechnol. Adv. 2011 29 739 767 10.1016/j.biotechadv.2011.06.004 21821113 [53] Martins J G Camargo S E A Bishop T T Popat K C Kipper M J Martins A F Pectin-chitosan membrane scaffold imparts controlled stem cell adhesion and proliferation Carbohydr. Polym. 2018 197 47 56 10.1016/j.carbpol.2018.05.062 30007637 [54] Lee, W. A.; Pernodet, N.; Li, B. Q.; Lin, C. H.; Hatchwell, E.; Rafailovich, M. H. Multicomponent polymer coating to block photocatalytic activity of TiO2 nanoparticles. Chem. Commun. 2007, 4815–4817. [55] Vevers W F Jha A N Genotoxic and cytotoxic potential of titanium dioxide (TiO2) nanoparticles on fish cells in vitro Ecotoxicology 2008 17 410 420 10.1007/s10646-008-0226-9 18491228 [56] Park E J Yi J Chung K H Ryu D Y Choi J Park K Oxidative stress and apoptosis induced by titanium dioxide nanoparticles in cultured BEAS-2B cells Toxicol. Lett. 2008 180 222 229 10.1016/j.toxlet.2008.06.869 18662754 [57] Feng Q Gao H C Wen H J Huang H H Li Q T Liang M H Liu Y Dong H Cao X D Engineering the cellular mechanical microenvironment to regulate stem cell chondrogenesis:Insights from a microgel model Acta Biomater. 2020 113 393 406 10.1016/j.actbio.2020.06.046 32629189 [58] Liu Y H Zhu Z Pei X B Zhang X Cheng X T Hu S S Gao X M Wang J Chen J Y Wan Q B ZIF-8-modified multifunctional bone-adhesive hydrogels promoting angiogenesis and osteogenesis for bone regeneration ACS Appl. Mater. Interfaces 2020 12 36978 36995 10.1021/acsami.0c12090 32814397 [59] Li J Yan J F Wan Q Q Shen M J Ma Y X Gu J T Gao P Tang X Y Yu F Chen J H Matrix stiffening by self-mineralizable guided bone regeneration Acta Biomater. 2021 125 112 125 10.1016/j.actbio.2021.02.012 33582360 [60] Majidinia M Sadeghpour A Yousefi B The roles of signaling pathways in bone repair and regeneration J. Cell. Physiol. 2018 233 2937 2948 10.1002/jcp.26042 28590066 [61] Yu W Q Qian C Jiang X Q Zhang F Q Weng W M Mechanisms of stem cell osteogenic differentiation on TiO2 nanotubes Colloids Surf. B: Biointerfaces 2015 136 779 785 10.1016/j.colsurfb.2015.10.019 26520050 [62] De Munck J Van Landuyt K Peumans M Poitevin A Lambrechts P Braem M Van Meerbeek B A critical review of the durability of adhesion to tooth tissue: Methods and results J. Dent. Res. 2005 84 118 132 10.1177/154405910508400204 15668328 [63] Freitas P H André C B Fronza B M Giannini M Rosalen P L Consani S França R Physicochemical properties, metalloproteinases inhibition, and antibiofilm activity of doxycycline-doped dental adhesive J. Dent. 2021 104 103550 10.1016/j.jdent.2020.103550 33276081 [64] Shoulders M D Raines R T Collagen structure and stability Annu. Rev. Biochem. 2009 78 929 958 10.1146/annurev.biochem.77.032207.120833 19344236 [65] Yu F Luo M L Xu R C Huang L Yu H H Meng M Jia J Q Hu Z H Wu W Z Tay F R A novel dentin bonding scheme based on extrafibrillar demineralization combined with covalent adhesion using a dry-bonding technique Bioact. Mater. 2021 6 3557 3567 10.1016/j.bioactmat.2021.03.024 33842741 [66] Cai X Wang X Y Chlorhexidine-loaded poly(amido amine) dendrimer and a dental adhesive containing amorphous calcium phosphate nanofillers for enhancing bonding durability Dent. Mater. 2022 38 824 834 10.1016/j.dental.2022.04.009 35450701	0	PMC9734535	NO-CC CODE	2022-12-14 23:28:29	no	Nano Res. 2022 Dec 5;:1-9	utf-8	Nano Res	2,022	10.1007/s12274-022-5153-1	oa_other
==== Front J Fam Violence J Fam Violence Journal of Family Violence 0885-7482 1573-2851 Springer US New York 479 10.1007/s10896-022-00479-2 Original Article Resilience is more than Nature: An Exploration of the Conditions that Nurture Resilience Among Rural Women who have Experienced IPV http://orcid.org/0000-0003-0348-7726 Mantler Tara [email protected] 1 http://orcid.org/0000-0003-1363-8778 Shillington Katie J. 2 http://orcid.org/0000-0002-9405-3642 Yates Julia 2 http://orcid.org/0000-0002-2040-119X Tryphonopoulos Panagiota 3 http://orcid.org/0000-0002-6541-6213 Jackson Kimberley T. 3 http://orcid.org/0000-0003-4328-8748 Ford-Gilboe Marilyn 3 1 grid.39381.30 0000 0004 1936 8884 School of Health Studies, Faculty of Health Sciences, The University of Western Ontario, HSB Room 332, 1151 Richmond Street, London, Ontario N6A 5B9 Canada 2 grid.39381.30 0000 0004 1936 8884 Health and Rehabilitation Sciences Program, Faculty of Health Sciences, The University of Western Ontario, London, Ontario Canada 3 grid.39381.30 0000 0004 1936 8884 Arthur Labatt Family School of Nursing, Faculty of Health Sciences, The University of Western Ontario, London, Ontario Canada 7 12 2022 111 29 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 Intimate partner violence (IPV) is a significant public health concern exacerbated by the pandemic. Experiences of violence vary based on geographic location and living in rural communities has been found, in some contexts, to amplify consequences of IPV. Resilience, the ability to survive and thrive despite facing adversity, has long been a dominant narrative within IPV literature, yet little is known about how resilience is cultivated among rural women experiencing violence. The purpose of this study was to explore how rural women experiencing IPV cultivate resilience. Methods Using Interpretive Description, in-depth qualitative interviews were conducted with 14 women who experienced IPV and 12 staff from women’s shelters across rural communities in Ontario, Canada to elicit perspectives about women’s resilience and environmental conditions that may shape resilience in the context of IPV. Results Women’s resilience was cultivated by personal changes aimed at surviving or thriving, and aspects of their environment that enabled or created barriers for resilience. Women adopted a positive, hopeful mindset and bolstered their inner strength through living from a place of integrity, being resolute in decisions, and using mental resistance when faced with doubt. Women faced barriers to resilience in the form of unhelpful help and COVID-19 public health guidelines. Paradoxically, living in a rural community both cultivated and undermined resilience. Conclusions Supporting women to cultivate resilience through modifying environmental factors to enable personal strengths to flourish is paramount in supporting women who have experienced IPV, particularly in rural contexts. Keywords Intimate partner violence Resilience Thriving Rural Stigma Isolation Shelter services COVID-19 Social Sciences and Humanities Research CouncilInsight Development Grant Mantler Tara ==== Body pmcIntroduction Gender-based violence (GBV) broadly encompasses the gendered elements within all forms of violence (Montesanti & Thurston, 2015). Intimate partner violence (IPV) is a form of GBV that transcends geographic, racial, social, and economic classes (Heise & Garcia-Moreno, 2002). IPV impacts an estimated 44% of Canadian women (Cotter, 2021) and is understood to be a pattern of physical, sexual, or emotional violence perpetrated by an intimate partner in the context of coercive control (Tjaden & Thoennes, 2000). While IPV is typically thought of as an interpersonal issue, the far-reaching social and economic consequences include broadly perpetuating the patriarchal nature of society, interrupting education/job training opportunities (Germano, 2019), and decreasing socioeconomic mobility for women experiencing violence (Flury et al., 2010; McLean & Bocinski, 2017; Montesanti & Thurston, 2015). Research on COVID-19 has provided insights into the extent and impacts of structural inequalities for marginalized groups, including women experiencing IPV (Enekwechi et al., 2020; Moffitt et al., 2020; Webb Hooper et al., 2020). As a result of the pandemic, there has been a rise in IPV and, more specifically, the severity of violence experienced (Peterman et al., 2020; Roesch et al, 2020). Throughout the COVID-19 pandemic, the United Nations has referred to IPV as a “shadow pandemic” (UN Women, 2020). While this messaging was likely intended to illustrate how IPV is common but still exists in the shadows or privacy of individuals’ homes, this terminology inadvertently creates a hierarchy wherein COVID-19 is the main pandemic, and IPV is a secondary one (Mantler et al., 2022b). As such, Khanlou et al. (2020) posit that these co-occurring pandemics be referred to as syndemic (Mendenhall, 2017). A syndemic acknowledges how diseases are provoked by socioeconomic, political, or environmental contexts, and when co-existing interact and lead to synergistic vulnerability for marginalized groups, thus amplifying social and structural inequities (Willen et al., 2017). The convergence of COVID-19 and IPV pandemics places women at a disproportionate risk of experiencing violence compared to if they had experienced either pandemic alone (Khanlou et al., 2020). As with many public health concerns, geographic location, and more specifically living in a rural community, can, in some contexts, further amplify some of the consequences of IPV. Specifically, women experiencing IPV while living in rural communities are at an increased risk of femicide (Davies et al., 2009; Logan et al., 2005, 2007; McFarlane et al., 1999; Roy & Marcellus, 2019). Results of a recent systematic review (Edwards, 2015) in the United States also suggests worse psychosocial and physical health outcomes among women who have experienced IPV and live in rural versus urban communities. These heightened experiences of violence are exacerbated by existing inequities embedded in the rural health and social systems, such as a lack of available services and qualified personnel (Lanier & Maume, 2009). The structural barriers impeding access to support for women experiencing IPV are entrenched in rural community norms that perpetuate and stigmatize experiences of violence, including hegemonic masculinity (Tyler & Fairbrother, 2013), fear of breach of patient-provider confidentiality, and traditional gender and family norms (Annan, 2008; Kaur & Garg, 2010; Merritt-Gray & Wuest, 1995; Zorn et al., 2017). Together, the heightened experiences of violence, existing inequities in health and social services, and the existence of community norms that undermine help-seeking behaviours create an environment conducive to an increased risk of continued violence. This environmental context may impede rural women’s ability to be resilient through adversities. The topic of resilience, a person’s ability to survive, grow, and thrive—despite exposure to adversity (Howell et al., 2018; Munoz et al., 2017; Prime et al., 2020), has long been a dominant narrative within IPV literature, including throughout the COVID-19 pandemic. Within these narratives, resilience has been framed primarily as a personal trait, with little attention given to how women experiencing IPV develop resilience or the contextual factors/conditions that make this possible (Goodman et al., 2003). Some studies have focussed on identifying resilience among women who have experienced IPV. For example, Humphreys (2003) found women who experienced IPV and resided in shelters exhibited resilience despite exposure to physical and psychological stressors. More recently, a study of Canadian women experiencing IPV during the first six months of the COVID-19 pandemic reported that women struggled to cope with the abuse because of the stay-at-home measures which negatively affected their resilience (Mantler et al., 2022a). Furthermore, the neoliberal positioning of the resilience narrative during the COVID-19 pandemic has emphasized a need for individual adaptability, overlooking governmental responsibility and societal structures that exacerbate existing inequities and conditions needed to support resilience (Blundell et al., 2020; Joseph, 2013). While there is evidence that urban-dwelling women who experience IPV demonstrate resilience (Anderson et al., 2012; Humphreys, 2003; Mantler et al., 2022a), there has been limited attention to unpacking how resilience develops among rural women who have experienced IPV and under what conditions. Given the uniqueness of rural communities, and the dearth of literature (Crann & Barata, 2016), there is a need to understand how resilience is cultivated among rural women experiencing violence in ways that attend to geographic realities. Purpose The purpose of this study was to explore how rural women experiencing IPV cultivate resilience, the ability to survive and thrive in spite of adversity, and the environmental of IPV. Methods Study Design This cross-sectional qualitative study used interpretive description (Thorne, 2016), a pragmatic qualitative approach that is both constructivist and naturalist, to generate knowledge relevant to applied disciplines. Sampling, Recruitment, and Eligibility A combination of purposive and snowball sampling was used to recruit rural women experiencing IPV and service providers from rural women’s shelters to gain perspectives on IPV from multiple lenses (i.e., those who have experienced IPV and those providing support). Study information was posted on Kijiji sites in rural areas and emailed to rural women’s shelters. To be eligible for the study, women needed to live in a rural area in Ontario; identify as resilient; have experienced IPV and survived; and have access to a safe computer/telephone. Service providers needed to have worked at an Ontario rural women’s shelter for a minimum of six months. Interested participants contacted the research team via email (found on the study advertisement) and were screened for eligibility. A total of 14 women and 12 service providers were eligible and verbally consented to participate in the study. Participants The women and service providers who participated in this study were from 12 different communities across Ontario, with populations ranging from 2,000 to 42,000 people. Women’s ages ranged from 18 to 57 years old (M = 34.86 years, SD = 9.31). The education of the sample was generally high, with approximately 60% of participants achieving a college or university-level education. The employment status of the sample was diverse, with five women working part-time, four working full-time, four being unemployed, and one woman identifying as self-employed. The average annual household income after taxes varied greatly among women, ranging from $15,000 to $110,000 Canadian dollars (CAD). Approximately 65% of this sample identified as heterosexual, 29% as bisexual, and 6% identified as having a fluid sexuality. Service providers had worked for their various agencies for a minimum of 6 months, with the majority of staff (n = 10) being employed full-time. All service providers were women with an age range of 27 to 59 years old (M = 42.25, SD = 11.73). The education and income of the sample was generally high, with all staff achieving college diplomas and/or university degrees and half averaging an annual household income of more than $100,000 CAD. Procedures Ethics approval was obtained from the host institution’s Research Ethics Board (< redacted for blind review >), and interviews were completed between November 2020 and June 2021. Individual, semi-structured interviews were conducted at two time-points, approximately four months apart. Phase one interviews were conducted between November 2020 and February 2021, with women (n = 14) and service providers (n = 12), from 8 shelters. Phase two (i.e., follow-up interviews) occurred between May and June 2021with six women and five service providers for the purpose of member checking to ensure accuracy and resonance with participants’ experiences (Guba & Lincoln, 1989). At the outset of each interview, resilience was defined to all participants as “a dynamic process in which psychosocial and environmental factors interact to enable an individual to survive, grow and even thrive despite exposure to adversity”. The phase one interviewers lasted approximately 60 min and were conducted by two members of the research team (i.e., KS and CD) and an optional phase two interview lasting approximately 60 min. The second interview was conducted with only six women and five service providers and was used to member check emerging findings and refine themes. Table 1 presents the questions asked during both phases of interviews. Phase 2 interview questions were based off preliminary findings from phase 1, thus affording participants the opportunity to expand upon or clarify their experiences. All interviews were audio-recorded and transcribed verbatim, and each transcript was anonymized prior to analysis. The data collection and analysis process were guided by Guba and Lincoln’s (1989) and Thorne et al. (1997) principles of auditability, fit, dependence, and transferability. To reduce barriers to participation, women and service providers were offered a $25 gift card in recognition of their time for the first interview, and a $10 gift card for the second interview.Table 1 Interview questions asked to women and service providers Phase of interview Participant group Interview questions 1 Women (n = 14) 1. What helps to support your resilience? 2. What undermines your resilience? 3. What are some challenges/barriers that you have faced to being resilient? 4. What did [do] you need to thrive over time? 5. What adaptations have you used to be resilient? 6. What has contributed [contributes] to your inner strength? 1 Service providers (n = 12) 1. What do you think helps to support women’s resilience? 2. What do you think undermines women’s resilience? 3. What are some challenges/barriers that you have seen women encounter that prevent them from being resilient? 4. How have you seen women thrive over time? 5. What adaptations have women used to be resilient? 6. What do you think contributes to women’s inner strength? 2 Women (n = 6) 1. In your relationship, what made you feel stuck? How did you overcome that feeling of “stuck-ness”? 2. How would you describe your mindset, and how do you feel your mindset has played a role in your experiences, and your resilience? 3. What enabled you to keep moving on when things were difficult? When there were moments of crisis 2 Service providers (n = 5) 1. What forces women to stay in their relationships (or keep them “stuck” there)? How do you see women overcome that feeling of “stuck-ness”? 2. How do you feel a woman’s mindset can influence their experiences and resilience? 3. What enables women to keep moving on when things are difficult? When there were moments of crisis? Data Analysis Interpretive description following Thorne’s (2016) approach guided analysis. Initial data analysis occurred after phase one interviews were completed. In line with Thorne (2016), the collection of data occurred concurrently and iteratively while researchers reflecting, asked questions, and considered options for making sense of the data. The initial analysis then informed the creation of the semi-structured interview guide for phase two interviews with the goal of clarifying, expanding, and confirming findings from the phase one analysis. Transcripts from all interviews were organized using Quirkos qualitative analysis software (Quirkos, 2021). The 37 transcripts were each independently coded by two of the five researchers involved. Initially, those who conducted the interviews (KS, TM) and the principal investigator met and created a preliminary coding structure with definitions based on field notes and what was known from the literature. Next, random coding dyads were created, and each dyad was initially assigned two transcripts to analyze using open and line-by-line coding (Blundell et al., 2020). Then, each dyad, and subsequently the larger group, met to discuss the applicability of the coding structure and code definitions, with refinements to the coding structure and definitions made, as needed. This process was repeated twice until the entire coding team was confident the coding structure sufficiently covered the data. Next, each interview was assigned to two people for final analysis. Throughout the coding process, the coding team utilized memoing to identify theoretical outliers, theorize the relationship and structure of the data, and extract meaning from the data set (Thorne, 2016). Once all transcripts were analyzed, Quirkos files from each coder were merged, and queries/reports were run on each code and associated data related to the concept of ‘resilience.’ The coding team then met to discuss and interpret the findings and to build consensus around how resilience is cultivated among women experiencing IPV and living in rural communities. Results Women and service providers (SP) described how experiences of IPV impacted resilience. A common narrative shared throughout interviews was that, despite adverse experiences, “[resilience] is just a part of being a woman” and that “[women] are resilient by nature” (SP11). However, in spite of both women and service providers framing women’s resilience as ‘natural’ or innate, their interviews also reflected a more comprehensive understanding of resilience as more than nature, but a capacity that was nurtured and shaped in multiple ways by context. Five themes were identified that reflect this connection: 1) inner strength; 2) transitioning to a positive mindset; 3) unhelpful help; 4) COVID-19: more and less time; and 5) rural communities: a double-edged sword. Inner Strength Inner strength, understood as the integrity of character, resoluteness of will and/or mental resistance to doubt/discouragement, was described by both women and service providers as a key source of resilience in the context of IPV. Integrity of character or living in a way that aligns with one’s true self was important to women’s resilience as it helped reassure them that they were on the right path to recovery. One woman described the importance of being reminded by others, as well as reminding herself, that leaving was the right decision and allowed her to begin living from a place of integrity saying, “[I remind] myself I guess, that what I did is the best choice… Because even hearing it from other people, it doesn't really, like it helps, but you have to tell yourself the same thing” (N2). This inner strength and actively living from a place of integrity was observed by many service providers as being a key component of resilience, with one service provider explaining that, “Everybody has inner strength, its whether or not they recognize how much inner strength they have, and act on it. It's there, it's having the confidence to trust your inner strength and to act on your inner strength” (SP10). For many women, ongoing connection with their informal support system was critical in helping them live from a place of integrity. One woman described her reliance on her informal support system in times of need as be strong when she was wavering:That support system outside of work, as soon as I get a break, I'll call them, and I'll say I'm having a really hard day. Y'know, this person is really getting to me and they're just making it very difficult to get through the day, and they'll just kind of like, talk me down back to a- a normal level. (N5) Beyond integrity, being resolute in decisions made is another aspect of resilience that many women drew upon to thrive despite their experiences of abuse. Many spoke of reaching the point of being resolved in their decision to leave, but needing to reach a breaking point where they felt they could not go on living in the abusive relationship any longer. One woman described this saying “and so you hit this point where you’re just rock bottom, literally. And you have no choice but to heal” (N3). The reality of “rock bottom” was described by several women in our study as a life-or-death situation, one woman explained:Because at some point, it could become you know life or death situation, so yeah, I guess it's when it became you know, physical and there was actually bruising. That's when it became ok this is it, this is the be all end all point. (N12) With an awareness that they deserved better than remaining in an abusive relationship, many women began the process of finding the inner strength to leave. One woman described that decision process saying, “You need to find that inner strength saying, I need to get out of this situation no matter what.” (N2). This sentiment was reiterated by another woman who underscored that determination or resoluteness of will had to come from within, explaining “something has to stem from yourself… like some sort of willpower” (N6). Women also described mental resistance to doubt or discouragement as important to cultivating resilience. Being able to cast aside doubt and trust in their integrity and resoluteness to change their situations was key. For example, one woman described the need to rely on her mental resistance daily to keep her moving forward:[Reassure] myself, every single day…because everything that comes out of a person's mouth, whether intentionally trying to hurt me, or unintentionally hurting me, reverses my progress, even just a smidgen, and so every day I have to reaffirm that everything I'm doing is positive, and that I've come really far compared to where I was, say, a year ago, and things are moving in a forward manner, and they're not going backwards anymore. (N5) To cultivate mental resilience, service providers underscored the importance of their role in shelter of encouraging and building women’s confidence to increase their mental fortitude and resistance to doubt. One service provider explained, “Women have shown how resilient they are… with sharing information, discussing their fears and concerns, we are able to encourage them to be resilient and to be able to thrive and not just survive” (SP10). Transitioning to a Positive Mindset Having a positive mindset and focusing on the desire to thrive and not only survive was described by both women and service providers as a factor that was important to women’s resilience. For many women, this meant overcoming a mindset where their judgement was constantly questioned or they blamed themselves for the abuse, a tool commonly used by abusers to control women and diminish their sense of worth. One woman explained this saying:For the decade that we were married… I let him convince me that my own judgement was questionable, and it took a really long time for me to understand that it wasn't, and then once I finally got that through my head, it has been pretty life changing. (N14) For many women, taking on a positive mindset meant shifting away from seeing themselves as victims and replacing that with a sense of hope. One woman explained,When making the switch from feeling or thinking or behaving like a victim, to that of a survivor…I feel like I can look back on things with less of the, like guilt and shame, and some of the things that like I would beat myself up with… and then I'm able to look at it with more of a sense of like, like strength and hope. (N4) Similarly, service providers underscored the importance of women transitioning to a positive mindset as important to their ability to sustain themselves and create the life they imagine. For example, one service provider noted, “If a woman has a positive mindset, she is going to have that energy and have that positive experience to move forward and to be resilient.” (SP8). Transitioning to a positive mindset was an active process for women that required constantly remembering their inner worth. Many women spoke about how using affirmations, a practical tool taught through counselling, helped them keep their own value and worth in mind. One woman described repeating the mantra she had learned, “I am worth it, I'm worth- I'm worth this position at my job…I've earned the right to walk down the street and say hi to the person coming towards me that's smiling” (N5). Service providers underscored that that transitioning to a positive mindset was difficult for many women whose identities were based on primarily on their experiences of abuse. In this context, service providers identified their role in helping women recreate a new identify or “making this whole new image of themselves” (SP8) to support a more positive mindset. The active process of transitioning to a positive mind set for many women was anchored to becoming future-oriented. Many women attributed this future orientation as the reason they were thriving after being in an abusive relationship. Woman described the transition to a future orientation was shaped by having community support and neighbours that served as allies meaning they could switch from survival to goal setting. The importance of this shift to a positive mindset and becoming future oriented provided women with a source of ongoing strength that helped them to thrive over time. One woman said, “I think a positive outlook really is underrated… I can’t really stress enough how much that has probably saved me over the last ten years” (N14). Unhelpful Help The sentiment of “unhelpful help” reflects the notion that help was inaccessible and/or misaligned with the needs of women, often experienced as systemic barriers experienced by those seeking help. Service providers often described these systemic barriers as stemming from a fundamental lack of understanding about IPV. “I find on a whole people don't really have a whole lot of understanding about how abuse works. And I just feel like if that was understood, then you wouldn't see… quite the same critical [negative] attitude” (SP6). A critical or negative attitude toward women was often observed in social services external to the shelter, such as police and the justice system. A lack of understanding of abuse often impeded women’s sense of safety and made many reluctant to report their experiences to police. Specifically, failure to have their safety concerns taken seriously in reporting their experiences also undermined women’s future help seeking, whether with police forces or other social services by reinforcing those systems in general could not be counted on for help. One woman explained how she often felt like service providers weren’t really interested or responsive to her situation, “they're just saying whatever they're supposed to read off to you. And then they're like, okay, well checklists, checklists, checklists” (N2). Despite the apparent barriers to seeking help from services, some women did find solace in having their experiences acknowledged by family members. For example, one woman discussed the benefits of family members being aware of her situation, stating that “they've offered support and guidance… or, you know, [tried] to lend a friendly ear and be open or honest and actually be sympathetic” (N12). COVID-19: More and Less Time Experiencing IPV during the COVID-19 pandemic further exacerbated personal and environmental barriers to women’s resilience as the public health measures enabled abusers to isolate women further and limit their ability to access services. The stay-at-home orders meant many women were confined at home with their abuser. One woman stated, “if [women] were being abused in any manner, now they're stuck” (N10). However, for some women, COVID-19 offered them newfound time to do the work of engaging with counselling and shelter services that was important in cultivating their own resilience and moving toward the life they wanted. Women discussed having more time as they were working from home, had become unemployed, and/or commitments outside the house had decreased due to stay-at-home orders. One woman spoke of the benefits she gained from the pandemic:It was so empowering because I was like when you only have one way but up, and the pandemic is also a gift too because it gave me time to say hey I am not doing this again. I am going to come out of this pandemic stronger. I'm not going to let this happen again, so you know I did use the time. I said I'm not going to waste any time. And I didn't. (N3). At the same time, closures and changes to services offered during COVID-19 impacted women’s ability to access services. This was summed up by one service provider who said, “it's harder for [women] to leave or harder for them to reach out and get the support that they need” (SP1). Many women discussed how their access to services was interrupted during the pandemic due to their abuser’s constant presence, leaving them with no opportunity to covertly engage with shelter services that had been helping them to cultivate resilience. One woman described the negative impact of service disruption for her, “it's just difficult dealing with healing having just left abusive relationship like right before the pandemic you know it was very hard to, do anything like I couldn't see my counselor anymore” (N3). Moreover, the COVID-19 restrictions such as quarantine and physical distancing further impacted service provision within the shelters. Early in the pandemic, women entering some shelters were required to quarantine for two weeks to prevent the spread of COVID-19, a policy consistent with public health guidance given to communal living organizations at the time. This isolation meant women experienced delays in accessing the typical supports received when they enter shelter. One service provider explained,[Women] have to isolate for two weeks, which makes you know, searching for housing, and getting just daily tasks completed that we would normally have no problem getting completed, but with women in isolation it makes it a lot more complicated. (SP9) Rural Communities: A Double-Edged Sword Living in rural communities was seen as both a benefit and barrier in cultivating resilience for women experiencing IPV. Women described the tensions of: how rurality led to both feelings of isolation and connection, support and stigma, and barriers and innovations in service provision. Isolation and Connection Living in a small community, often with distance between houses, was regarded by some women as a positive feature of their communities that helped bolster their resilience. This ‘benefit’ was particularly apparent in the COVID-19 context. Women described that having physical space allowed them to cope with the abuse and pandemic and to do so privately. One woman noted that “being sort of in a rural situation has really helped with coping with the pandemic, because you get that sense of being away from all of this, right, when you don't really have anybody else around” (N14). However, for some women, living in a rural community was very isolating and lonely, making it more difficult to be hopeful about the future and to see a way out. One woman explained:Once you're isolated, things become pretty bleak. It's hard to be hopeful and it's hard to think about how you're going to be able to get out because it's all you know, it's all you see, it's all you experience. (N6) Other women shared their rural locations made them feel trapped – physically, mentally, and emotionally. Women were aware that their experiences of abuse were largely invisible to people outside the home and that this contributed to hopelessness. One woman described that compared to an urban location, “[You] can't yell, nobody's gonna hear me like, you, you literally you feel way more trapped” (N10). Tight-Knit Community: Support and Stigma For some, living in a rural community meant that informal social networks were closer with fewer people in them, and this both bolstered and inhibited resilience. A close-knit community helped women cultivate resilience when they had access to support and allies beyond shelters. One woman noted that “community support and having neighbours as allies like having safe places like you know a church or a grocery store” (N5), was integral to her feelings of safety. For many women, strong relationships with neighbours brought them comfort knowing they could access help should they need it. One woman noted that this sense of community provided “a lot of stability” (N12) following abusive events. A service provider reiterated this, explaining that neighbours tend to be more likely to express concern and provide support when they suspect IPV as part of natural, day to day encounters with women, “sometimes it's- it's helping with [farming activity], or you …have a supper. Like there's ways of—it's just finding ways to present [concern for a woman’s safety] information safely” (SP2). However, living in a tight-knit community was also seen as a drawback when knowing everyone and being known created barriers to privacy and confidentiality and make it difficult for women to access services. One SP explained that in rural communities, “there's a lot of judgment I think from you know, the public in accessing our services as well as their family and probably their friends” (SP1). Women described the varied ways that stigma was enacted in their rural communities. Some understood that rural culture, such as hegemonic and patriarchal norms, created a context where abuse is passively accepted as a normal part of life. One woman described her understanding:I feel like there was a sense of like culture within like males there, whether it's the way their family structures were or family cultures were but like I feel like that type of behaviour was acceptable where you know it would be witnessed by majority of males around or no one would do anything like you shrugged it off or no one would do anything. (N5) These experiences of stigma within their communities undermined women’s resilience further by hindering their desire to access various services. One woman commented on this saying, “I know there's like shelters that women can go to get away uh but there is a stigma and like a feeling that goes with those shelters” (N10). Concurrently, norms perpetuated abuse-related stigma, undermining resilience as women felt further isolated and embarrassed to access resources. For instance, a value for privacy and not having others intrude in personal affairs meant that witnesses of violence were more inclined to look the other way. One woman described this saying:I feel like in a lot of rural communities it's a lot of hush hush you know like, you see someone scrapping at a gas station you just let them do their thing. You don't interject. Its none of your business. You don't interject - you keep to yourself. (N5) While stigma around violence was described as a common problem for women, some reported observing changes in their communities, resulting in them feeling better supported. One woman described the gains in her community in terms of knowledge surrounding IPV and how it helped her to feel the community was more supportive and services were more accessible, “We have good neighbours, and we have community organizations that are fairly accessible to us. Now in the circumstance of COVID-19 maybe a little less but we're well surrounded” (N12). Resources: Barrier and Innovations Scarcity of resources in rural communities undermined resilience of women leaving abusive relationships in tangible ways, such as lack of space for them in a shelter or the inability for women to physically access needed services. One service provider explained this reality:I think it takes a lot of strength and coordination and effort for women to access the shelter period. So, you know, when they get the courage to access the shelter, and they make that call, and then they're being denied space because there isn't there isn't room for them. I think that can really undermine their resilience. (SP1) Beyond the lack of available service, accessibility issues also undermined resilience for some women. One woman described how her lack of access to transportation as contributing to a “total isolation from the world” (N10), a particular challenge for women trying to cultivate resilience. One service provider explained, “if they don't have transportation… they rely on their partner to get them to places. And their partners aren't typically going to drive them to counselling sessions. So, I think they probably just feel defeated” (SP1). However, although service providers often tried to work around access barriers, the migration of many services to online platforms in response to the COVID-19 pandemic was a silver lining for some. Offering services online allowed women who would not have otherwise accessed services because of existing systemic barriers with a critical opportunity to access support. One woman was unaware of the many support services available to her in the rural context, explaining “I just didn’t know that there were a lot of resources out there” (N9). Discussion This cross-sectional interpretive description qualitative study explored rural women’s ability to cultivate resilience in the context of IPV. Both women and service providers articulated that women’s ability to cultivate resilience was dependent upon 1) transitioning to a positive mindset and bolstering their inner strength through living from a place of integrity, 2) being resolute in their decisions, and 3) using mental resistance when faced with doubt and/or discouragement. Barriers to resilience that were identified included unhelpful help, or rather, help that does not account for the holistic impact of abuse on a women’s life as well as the public health guidelines associated with the COVID-19 pandemic. Specifically, the need for physical distancing reduced the capacity of services to support women, and the isolation requirements in communal living organizations slowed women’s ability to engage with support services. Living in a rural community was viewed as both a source of cultivating resilience as well as undermining resilience. The isolation associated with rural living afforded women the physical space to deal with the trauma of abuse on their own terms while also offering further opportunities for abusers to control women. The closeness of rural communities meant that neighbours could watch out for warning signs of abuse and find ways to support women but also inhibited women’s ability to access shelter services anonymously. Moreover, stigma associated with hegemonic norms and lack of resources including shelter beds and transportation were barriers to women leaving and access resources, respectively. Overall, findings from this study underscore the complexity of resilience being influenced in unexpected ways by the reality of living in rural communities. In a theoretical analysis of inner strength, Lundman et al. (2010) reported an overlap between the constructs of resilience and inner strength. Through harnessing one’s inner strength, individuals stood firm and creatively adapted, enabling them to redirect their life purpose and endure adversity. While this analysis was theoretical, our study extends this understanding as it demonstrated that rural women who had experienced abuse cultivated inner strength with the support of environmental factors to build resilience. This provided a nuanced understanding of how rural women experiencing IPV are becoming resilient in the face of abuse. Rural shelter service providers further illuminated that by providing ongoing social support to women, they can support their cultivation of resilience. This is in line with a previous study by Jose and Novaco (2016), who found that social support and resilience attenuated the effects of psychological distress among women experiencing IPV; however, our study further described this phenomenon in terms of the connection of social support to building resilience among women. For many women in the current study, adapting a positive mindset meant shifting away from seeing themselves as victims and finding a sense of hope. This finding aligned with research conducted by D’Amore et al. (2021), who reported that women who experienced IPV had a new perspective on life and experienced hope for the future. Similar to women in the current study, participants “drew hope and strength from their children and their pursuit to provide them with better lives” (D’Amore et al., 2021, p. 14). Barriers to help seeking for those experiencing IPV have been well established in the literature (Calton et al., 2016; Overstreet & Quinn, 2013; Robinson et al., 2021). A systematic review by Robinson et al. (2021) reported the six most common barriers, one of which was system failures. The notion of unhelpful help found in this study aligned with the idea of system failures; however, the connection of this barrier to service and its impact on undermining resilience was unique to the study. A previous study of service providers identified minimizing experiences of abuse as a primary barrier to engagement with services (Overstreet & Quinn, 2013). While this is a form of unhelpful help, our study provided new insights into how women are grappling with various forms of unhelpful help while also highlighting the need to address existing system failures such as minimizing experiences of abuse or negating the totality of the effects violence on women’s lives. The impact of COVID-19 on experiences of IPV have been well-established as a syndemic (Khanlou et al., 2020; Mendenhall, 2017). Studies have emerged citing the increased incidence and severity of violence (Peterman et al., 2020; Roesch et al, 2020) and the negative impact on coping among women experiencing IPV (Mantler et al., 2022a) during the COVID-19 pandemic. Unique to this study was the linking of specific aspects of public health guidance such as quarantine requirements and physical distancing on the ability to access service, and in turn, the impact on resilience. Specifically, our study revealed that these public health guidelines undermined resilience impeding access to vital support services by decreasing the capacity of social services. The unintended consequence of this COVID-19 public health guidelines was that women were unable to access vital services during the pandemic, which underscores the need of the government to consider how enacting a policy to address one pandemic inadvertently made another pandemic worse. While COVID-19 restrictions hampered women's ability to cultivate resilience, for some women the slowed-down pace of life during a pandemic meant they were afforded the time to engage with support services and start doing the ‘work’ to shift from surviving to thriving. The rural sociocultural context has been linked with heightening women’s vulnerability to IPV due to a lack of resources, increased geographic isolation, stigma associated with violence, acceptance of violence within the ‘good ol’ boy’ network, and traditional gender stereotypes (Bosch & Schumm, 2004; Eastman & Bunch, 2007; Edwards, 2015; Lanier & Maume, 2009; Peek-Asa et al., 2011; Websdale, 1998). While these sociocultural contextual factors have been previously linked to both increased experiences of violence and limited community response to violence, this is the first study examining how the rural context can both support and undermine resilience. There is a tendency in previous literature to examine constructs using a dichotomy, specifically that rurality as a sociocultural contextual factor either increases or decreases the risk of IPV; however, there is a need to examine the duality of the role rurality in the context of IPV. Limitations & Future Research The methodology of this study afforded a cross-sectional look at experiences of resilience among rural Canadian women who experienced IPV; however, the results should be considered within the context of the paper’s limitations. Firstly, the demographics of women and service providers in this study do not reflect the diverse community of women that access shelters in Canada as we are missing the experiences of racialized women, women whose first language is not English, and older women. More research is needed on the topic of resilience among rural Canadian women who have experienced IPV, particularly research that examines the role resilience plays in women’s decisions to leave abusive relationships. Future studies should also further explore the tools women use to cultivate inner strength and a positive mindset, as well as how best to leverage beneficial resources and overcome limiting sociocultural contextual factors to better support women who have experienced IPV. Conclusion Women used inner strength and transitioning into a positive mind which were cultivated through environmental factors such as counselling, access to women’s shelters, and informal support networks to promote resilience in the context of continued adversity. From the perspectives of women and service providers, rural sociocultural contexts such as geographic isolation, living in a tight-knit, and resource availability within communities were all seen as both facilitators and barriers to bolstering resilience depending on how they were enacted. Stigma in rural communities further undermined women’s resilience pointing to the need to find ways to overcome existing system failures in health and support services, services designed to support rural women who have experienced IPV. Ultimately, finding ways to support rural women in cultivating resilience through modifying environmental factors to enable the personal factors to flourish is paramount in the context of IPV. Acknowledgements The authors of this manuscript would like to extend a warm thanks to all the women and service providers who generously shared their experiences with our team. Funding This work was supported by SSHRC Insight Development Grant. Declarations Conflict of Interest The authors have no conflicts of interest to declare. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Anderson KM Renner LM Danis FS Recovery: Resilience and growth in the aftermath of domestic violence Violence against Women 2012 18 11 1279 1299 10.1177/1077801212470543 23334815 Annan SL Intimate partner violence in rural environments Annual Review of Nursing Research 2008 26 85 113 10.1891/0739-6686.26.1.85 Blundell R Costa Dias M Joyce R Xu X COVID-19 and inequalities Fiscal Studies 2020 41 2 291 319 10.1111/1475-5890.12232 32836542 Bosch K Schumm WR Accessibility to resources: Helping rural women in abusive partner relationships become free to abuse Journal of Sex and Marital Therapy 2004 30 5 357 370 10.1080/00926230490465118 15672603 Calton JM Cattaneo LB Gebhard KT Barriers to help seeking for lesbian, gay, bisexual, transgender, and queer survivors of intimate partner violence Trauma, Violence & Abuse 2016 17 5 585 600 10.1177/1524838015585318 Cotter, A. (2021). Intimate partner violence in Canada, 2018: An overview. Juristat: Canadian Centre for Justice Statistics, 1–23. Crann SE Barata PC The experience of resilience for adult female survivors of intimate partner violence: A phenomenological inquiry Violence against Women 2016 22 7 853 875 10.1177/1077801215612598 26567293 D’Amore C Martin SL Wood K Brooks C Themes of healing and posttraumatic growth in women survivors’ narratives of intimate partner violence Journal of Interpersonal Violence 2021 36 5–6 NP2697 NP2724 10.1177/0886260518767909 29642769 Davies, L., Hammerton, J., & Ford-Gilboe, M. (2009). Gender inequality and patterns of abuse post leaving. Journal of Family Violence, 24(1). 10.1007/s10896-008-9204-5 Eastman B Bunch SG Providing services to survivors of domestic violence: A comparison of rural and urban service provider perceptions Journal of Interpersonal Violence 2007 22 4 465 473 10.1177/0886260506296989 17369448 Edwards KM Intimate partner violence and the rural-urban-suburban divide: Myth or reality? A critical review of the literature Trauma, Violence & Abuse 2015 16 3 359 373 10.1177/1524838014557289 Enekwechi, A., Hardeman, R., Powell, W., & Zephyrin, L. (2020). COVID-19 webinar series session 18 – Health inequities: Addressing the disease burden in Black, Indigenous, and People of Color Communities. Alliance for Health Policy. [Webinar]. https://www.allhealthpolicy.org/covid-19-webinar-series-session-18-health-inequities-addressing-the-disease-burden-in-black-indigenous-and-people-of-color-communities/. Accessed Mar 2022. Flury, M., Nyberg, E., & Riecher-Rossler, A. (2010). Domestic violence against women: Definitions, epidemiology, risk factors and consequences. The European Journal of Medical Sciences, 140. 10.4414/smw.2010.13099 Germano, M. (2019). Domestic violence has a financial impact too. Forbes. https://www.forbes.com/sites/maggiegermano/2019/10/17/domestic-violence-has-a-financial-impact-too/?sh=3eb9a1649d04. Accessed Nov 2021. Goodman L Dutton MA Weinfurt K Cook S The intimate partner violence strategies index Violence against Women 2003 9 2 163 186 10.1177/1077801202239004 Guba EG Lincoln YS Fourth generation evaluation 1989 Sage Publications Heise, L., & Garcia-Moreno, C. (2002). Violence by intimate partners. In E. G. Krug, L. L. Dahlberg, J. A. Mercy, A. B. Zwi, & R. Lozana (Eds.), World report on violence and health (1, 87–113). http://apps.who.int/iris/bitstream/handle/10665/42495/9241545615_eng.pdf;jsessionid=99E0A4F4C594EDDF23228FE1BE9A56DD?sequence=1. Accessed Nov 2021. Howell KH Thurston IB Schwartz LE Jamison LE Hasselle AJ Protective factors associated with resilience in women exposed to intimate partner violence Psychology of Violence 2018 8 4 438 447 10.1037/vio0000147 30393573 Humphreys, J. (2003). Resilience in sheltered battered women. Issues in Mental Health Nursing, 24(2), 137–152. https://psycnet.apa.org/doi/10.1080/01612840305293. Accessed Nov 2021. Joseph J Resilience as embedded neoliberalism: A governmentality approach Resilience 2013 1 1 38 52 10.1080/21693293.2013.765741 Jose R Novaco RW Intimate partner violence victims seeking a temporary restraining order: Social support and resilience attenuating psychological distress Journal of Interpersonal Violence 2016 31 20 3352 3376 10.1177/0886260515584352 25952290 Kaur R Garg S Domestic violence against women: A qualitative study in a rural community Asia Pacific Journal of Public Health 2010 22 2 242 251 10.1177/1010539509343949 19703815 Khanlou, N., Ssawe, A., Vazquez, L. M., Pashang, S., Connolly, J. A., Bohr, Y., Epstein, I., Zahraei, S., Ahmad, F., Mgwigwi, T., & Alamdar, N. (2020). COVID-19 pandemic guidelines for mental health support of racialized women at risk of gender-based violence: Knowledge synthesis report. Government of Canada. https://cihr-irsc.gc.ca/e/52062.html. Accessed Mar 2022. Lanier C Maume MO Intimate partner violence and social isolation across the rural/urban divide Violence against Women 2009 15 11 1311 1330 10.1177/1077801209346711 19755628 Logan TK Cole J Shannon L Walker R Relationship characteristics and protective orders among a diverse sample of women Journal of Family Violence 2007 22 4 237 246 10.1007/s10896-007-9077-z Logan TK Shannon L Walker R Protective orders in rural and urban areas: A multiple perspective study Violence Against Women 2005 11 876 911 10.1177/1077801205276985 16043576 Lundman B Aléx L Jonsén E Norberg A Nygren B Santamäki Fischer R Strandberg G Inner strength–a theoretical analysis of salutogenic concepts International Journal of Nursing Studies 2010 47 2 251 260 10.1016/j.ijnurstu.2009.05.020 19577237 Mantler, T., Shillington, K. J., Davidson, C. A., Yates, J., Irwin, J. D., Kaschor, B., & Jackson, K. T. (2022a). Impacts of COVID-19 on the Coping Behaviours of Canadian Women Experiencing Intimate Partner Violence. Global Social Welfare: Research, Policy & Practice, 1–16. Advance online publication. 10.1007/s40609-022-00224-z Mantler T Wathen CN Burd C MacGregor JCD McLean I Veenendaal J Navigating multiple pandemics: A critical analysis of the impact of COVID-19 policy responses on gender-based violence services Critical Social Policy 2022 10.1177/02610183221088461 McFarlane JM Campbell JC Wilt S Sachs CJ Ulrich Y Xu X Stalking and intimate partner femicide Homicide Studies 1999 3 4 300 316 10.1177/1088767999003004003 McLean, G., & Bocinski, S. G. (2017). The economic cost of intimate partner violence, sexual assault, and stalking. Institute for Women’s Policy Research. https://iwpr.org/wp-content/uploads/2020/10/B367_Economic-Impacts-of-IPV-08.14.17.pdf. Accessed Feb 2022. Mendenhall E Syndemics: A new path for global health research Lancet (london, England) 2017 389 10072 889 891 10.1016/S0140-6736(17)30602-5 28271827 Merritt-Gray M Wuest J Counteracting abuse and breaking free: The process of leaving revealed through women’s voices Health Care for Women International 1995 16 5 388 412 10.1080/07399339509516194 Moffitt, P., Aujla, W., Giesbrecht, C. J., Grant, I., & Straatman, A. L. (2020). Intimate partner violence and COVID-19 in rural, remote, and northern Canada: Relationship, vulnerability and risk. Journal of Family Violence, 37, 775–786. Montesanti, S. R., & Thurston, W. E. (2015). Mapping the role of structural and interpersonal violence in the lives of women: Implications for public health interventions and policy. BMC Women’s Health, 15(100). 10.1186/s12905-015-0256-4 Munoz RT Brady S Brown V The psychology of resilience: A model of the relationship of locus of control to hope among survivors of intimate partner violence Traumatology 2017 23 1 102 111 10.1037/trm0000102 Overstreet NM Quinn DM the intimate partner violence stigmatization model and barriers to help-seeking Basic and Applied Social Psychology 2013 35 1 109 122 10.1080/01973533.2012.746599 23524454 Peek-Asa C Wallis A Harland K Beyer K Dickey P Saftlas A Rural disparity in domestic violence prevalence and access to resources Journal of Women's Health (2002) 2011 20 11 1743 1749 10.1089/jwh.2011.2891 Peterman, P., Potts, A., O'Donnell, M., Thompson, K., Shah, N., Oertelt-Prigione, S., & van Gelder, N. (2020). Pandemics and Violence Against Women and Children. Washington, DC: Center for Global Development. https://www.cgdev.org/publication/pandemics-and-violence-against-women-and-children. Accessed Mar 2022. Prime, H., Wade, M., & Browne, D. T. Risk and resilience in family well-being during the COVID-19 pandemic. (2020). American Psycholoist, 75(5), 631–643. https://psycnet.apa.org/doi/10.1037/amp0000660 Quirkos 2.4.2. (2021). [Computer Software]. Retrieved from https://www.quirkos.com. Accessed Mar 2022. Robinson SR Ravi K Voth Schrag RJ A systematic review of barriers to formal help seeking for adult survivors of IPV in the United States, 2005–2019 Trauma, Violence & Abuse 2021 22 5 1279 1295 10.1177/1524838020916254 Roesch, E., Amin, A., Gupta, J., & Garcia-Moreno, C. (2020). Violence against women during covid-19 pandemic restrictions. BMJ, 369. 10.1136/bmj.m1712 Roy, J., & Marcellus, S. (2019). Homicide in Canada, 2018. Juristat: Canadian Centre for Justice Statistics, 4–33. https://www.lib.uwo.ca/cgi-bin/ezpauthn.cgi?url=http://search.proquest.com/scholarly-journals/homicide-canada-2018/docview/2329108701/se-2?accountid=15115. Accessed Feb 2022. Thorne S Reimer Kirkham S MacDonald-Emes J Interpretive description: A non-categorical qualitative alternative for developing nursing knowledge Research in Nursing & Health 1997 20 169 177 10.1002/(SICI)1098-240X(199704)20:2<169::AID-NUR9>3.0.CO;2-I 9100747 Thorne, S. (2016). Interpretive Description: Qualitative Research for Applied Practice. (2nd ed.). Routledge. Tjaden P Thoennes N Prevalence and consequences of male-to-female and female-to-male intimate partner violence as measured by the National Violence Against Women Survey Violence against Women 2000 6 2 142 161 10.1177/10778010022181769 Tyler M Fairbrother P Bushfires are “men's business”: The importance of gender and rural hegemonic masculinity Journal of Rural Studies 2013 30 110 119 10.1016/j.jrurstud.2013.01.002 UN Women. (2020). Violence against women and girls: the shadow pandemic. https://www.unwomen.org/en/news/stories/2020/4/statement-ed-phumzile-violence-against-women-during-pandemic Webb Hooper M Nápoles AM Pérez-Stable EJ COVID-19 and racial/ethnic disparities JAMA 2020 323 24 2466 2467 10.1001/jama.2020.8598 32391864 Websdale N Rural woman battering and the justice system: An ethnography 1998 Sage Willen SS Knipper M Abadía-Barrero CE Davidovitch N Syndemic vulnerability and the right to health Lancet (london, England) 2017 389 10072 964 977 10.1016/S0140-6736(17)30261-1 28271847 Zorn KG Wuerch MA Faller YN Hampton MR Perspectives on regional differences and intimate partner violence in Canada: A qualitative examination Journal of Family Violence 2017 32 6 633 644 10.1007/s10896-017-9911-x	0	PMC9734538	NO-CC CODE	2022-12-14 23:28:29	no	J Fam Violence. 2022 Dec 7;:1-11	utf-8	J Fam Violence	2,022	10.1007/s10896-022-00479-2	oa_other
==== Front Neural Comput Appl Neural Comput Appl Neural Computing & Applications 0941-0643 1433-3058 Springer London London 8099 10.1007/s00521-022-08099-z Original Article Stacked ensemble learning based on deep convolutional neural networks for pediatric pneumonia diagnosis using chest X-ray images Prakash J. Arun [email protected] 1 http://orcid.org/0000-0001-6873-6469 Ravi Vinayakumar [email protected] 2 Sowmya V. [email protected] 1 Soman K. P. [email protected] 1 1 grid.411370.0 0000 0000 9081 2061 Center for Computational Engineering and Networking (CEN), Amrita School of Engineering, Amrita Vishwa Vidyapeetham, Coimbatore, India 2 grid.449337.e 0000 0004 1756 6721 Center for Artificial Intelligence, Prince Mohammad Bin Fahd University, Khobar, Saudi Arabia 7 12 2022 121 23 4 2022 22 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. Pneumonia is an acute respiratory infection caused by bacteria, viruses, or fungi and has become very common in children ranging from 1 to 5 years of age. Common symptoms of pneumonia include difficulty breathing due to inflamed or pus and fluid-filled alveoli. The United Nations Children’s Fund reports nearly 800,000 deaths in children due to pneumonia. Delayed diagnosis and overpriced tests are the prime reason for the high mortality rate, especially in underdeveloped countries. A time and cost-efficient diagnosis tool: Chest X-rays, was thus accepted as the standard diagnostic test for pediatric pneumonia. However, the lower radiation levels for diagnosis in children make the task much more onerous and time-consuming. The mentioned challenges initiate the need for a computer-aided detection model that is instantaneous and accurate. Our work proposes a stacked ensemble learning of deep learning-based features for pediatric pneumonia classification. The extracted features from the global average pooling layer of the fine-tuned Xception model pretrained on ImageNet weights are sent to the Kernel Principal Component Analysis for dimensionality reduction. The dimensionally reduced features are further trained and validated on the stacking classifier. The stacking classifier consists of two stages; the first stage uses the Random-Forest classifier, K-Nearest Neighbors, Logistic Regression, XGB classifier, Support Vector Classifier (SVC), Nu-SVC, and MLP classifier. The second stage operates on Logistic Regression using the first stage predictions for the final classification with Stratified K-fold cross-validation to prevent overfitting. The model was tested on the publicly available pediatric pneumonia dataset, achieving an accuracy of 98.3%, precision of 99.29%, recall of 98.36%, F1-score of 98.83%, and an AUC score of 98.24%. The performance shows its reliability for real-time deployment in assisting radiologists and physicians. Keywords Pneumonia Chest X-rays Computer-aided diagnosis Deep learning Transfer learning Stacking classifier Principal component analysis Stratified K-fold ==== Body pmcIntroduction Over the years, the number of respiratory diseases and infections has increased drastically. Degradation in the air quality has paved the way to numerous lung-related contaminations [1]. Pneumonia is one such acute lower respiratory infection that fills the alveoli with pus and fluid, leading to reduced oxygen holding capacity in the lungs. Lack of oxygen directly impacts the standard functioning of the body. Fatigue and lethargy are a few of many symptoms caused by inadequate oxygen levels. In severe cases, it can deter the brain and heart. Symptoms of pneumonia include fever, shallow breathing, and coughing. In extreme cases, it causes sharp chest pains when breathing and coughing. Sepsis, one of the many complications of pneumonia, can lead to tissue damage, organ failure, and even death if left untreated. Studies show that people with a weaker immune system are highly susceptible to pneumonia [2]. This acute respiratory infection poses a much bigger problem in children predominantly between 1 and 5 years of age whose immunity system is in its embryonic stages of development [3]. Symptoms of severe pneumonia in children include vomiting, severe malnutrition, and the inability to consume food and water [4]. Pediatric pneumonia accounts for nearly 800,000 deaths of young children, as reported by the United Nations Children’s Fund (UNICEF). Based on factors like age group and other medical conditions, there are several diagnostic tests for pneumonia. The most widely used diagnostic tests in children include pulse oximetry to check the oxygen levels, complete blood count (CBC) to check the activity of the immune system when there is an infection, sputum test, and chest X-rays to look for inflammation in the lungs. Abnormal CBC can be due to a variety of medical conditions. The count may decrease or increase even with mild infections. Thus, CBC is not guaranteed to confirm the presence of pneumonia. Children below the age of 10 have reduced sputum production. This reduced sample quantity restricts conducting various tests and eliminates its possibility as a confirmatory diagnostic test. Though pulse oximetry might seem like the best alternative, it cannot assure the presence of pneumonia as there may be other lung contaminations causing the low oxygen levels in the body. In addition to the limitations of these tests, they are time and cost-inefficient. These two factors are critical in saving lives. Cost in specific is a prime challenge in underdeveloped countries where people scarcely avail such diagnosis measures due to its high costs. An affordable and rapid standardized test was adopted considering these issues: Chest X-rays. Chest X-rays being time and cost-effective are the most common modality of pneumonia diagnosis. Doctors and radiologists with years of expertise examine the X-rays to detect the presence of pneumonia. Radiation level for chest X-rays in children is lower in contrast to the radiation levels used in adults to eliminate the risk of developing cancer. Low radiation levels in X-rays lead to loss of important information, making the task of pediatric pneumonia detection much more laborious and strenuous. With the ongoing COVID-19 virus advancing into other variants, several doctors across the globe are being transferred to emergency wards. The current situation might place children with pneumonia in jeopardy of not getting the required medical attention and thus, motivates the need for a computer-aided diagnostic model that is accurate and immediate. Several Computer-Aided Diagnosis (CAD) methods are currently in use for various biomedical applications, such as breast cancer detection [5], heart disease detection [6], tuberculosis detection [7], Alzheimer disease detection [8], diabetes-related retinal disease detection [9], and pneumonia detection. Literature survey shows that machine learning-based pneumonia diagnosis from chest X-rays using several feature extraction techniques helped physicians automate the process of diagnosis. However, this feature extraction process requires the usage of handcrafted filters. Feature engineering in biomedical tasks requires tremendous proficiency and relies laboriously on experts, hence hindering the widescale development of CADs. The applications of computer-aided diagnosis are now limitless with the advent of deep learning. Deep learning has been rooted down firmly in different domains owing to the availability of enormous data and ample computational resources. Deep Convolutional Neural Networks (CNN) has gained lots of attention in recent years leading to state-of-the-art performances in various image classification problems. The advantage of automatic feature extraction and engineering in deep learning, which was not previously possible with machine learning, has propelled the surge in computer-aided diagnosis-based systems. Transfer learning, a splendid breakthrough in artificial intelligence, has helped researchers overcome the disadvantage of the inadequate dataset that arises due to privacy concerns. Most of the deep learning architectures used for pediatric pneumonia diagnosis perform well nonetheless, their performance is limited. The cause for this is in the learning of a neural network: huge parameters in neural networks tend to overfit and thereby limit their performance on the test data. Most of the models proposed in the literature are not generalizable and robust as their performance has not been validated on similar datasets belonging to the same disease. Possible reasons for the limited performance include poor outlier handling, training on high class imbalance datasets, models with convoluted structure, and overfitting. The existing models are not guaranteed to perform well on unseen data. Robustness and generalizability are the key factors to be considered before real-time deployment. Therefore, it is of utmost importance to validate the performance on datasets of similar lung diseases or the same disease. Accounting to the above-mentioned concerns, the major contributions of the proposed work are summarized as follows:A stacked classifier-based learning approach, leveraging the strengths of machine learning classifiers and neural networks for pediatric pneumonia detection. A comparative performance analysis of the various pretrained deep CNN architectures with the proposed method for the task at hand. Class activation maps (CAM) to visualize the area of interest pertinent to the classification of normal and pneumonia X-rays. t-distributed stochastic neighbor embedding (t-SNE) based feature visualization for layman interpretability of the features predicted by the deep CNN architecture. Investigation on the effect of the kernel PCA on the performance of the classification model. An up-to-date comparison with other recent works tested on the publicly available Kermany et al. [10] dataset. A detailed investigation on the advantages and limitations of the proposed architecture on the pediatric pneumonia dataset. Performance analysis of the proposed method on similar pneumonia datasets to prove its generalizability and robustness. The contributions made in the field of pediatric pneumonia diagnosis that motivated us to advance with the idea of stacked ensemble learning are as follows:The unprecedented research on the diagnosis of pediatric pneumonia with transfer learning, unfurled the possibilities of research along with the open-source availability of the dataset [10]. The current impediment in the performance of deep convolution layers was solved using dilated convolutions, residual structures, and transfer learning [11]. A fusion technique involving a deep CNN model with PCA and logistic regression [12]. A weighted average ensemble of deep CNN models incorporating deep transfer learning [13]. A majority voting ensemble of the predictions from deep CNN models [14]. CheXNet [15], a DenseNet121 model trained on the ChestX-ray14 dataset whose performance exceeded that of the average radiologist. The rest of this article is organized as follows: Sect. 2 describes the literature survey and discusses the existing gap in the literature and how our approach completes it. The proposed approach is discussed in Sect. 3. Section 4 contains the description of the dataset. Section 5 details the performance metrics used in this study. The experimental results are analyzed and discussed with plots in Sect. 6. In Sect. 7, we conclude our work; summarizing the problem and the limitations of our approach, along with the possible future works. Literature survey Convolution Neural Network (CNN) gets its name from the mathematical operation called convolution. CNN is widely used for feature extraction and consists of three types of layers: convolutional layer, pooling layer, and fully connected layer. The first study on pediatric pneumonia detection using deep learning facilitated the onset of pediatric pneumonia-based diagnosis research [10]. The dataset was made public, and researchers began experimenting with different neural network approaches. Multilayer Perceptron (MLP) and CNN-based approaches were proposed in [16]. As a continuation of his previous work, Saraiva et al. [17] used CNN for feature extraction, followed by cross-validation for extensive learning from the limited dataset. Several state-of-the-art deep learning models were fine-tuned for pediatric pneumonia detection on the Kermany et al. [10] dataset with competing performances. However, the performances of these models were limited. The current limitation of deep CNN architectures is the degradation of spatial information with increasing layers. In classification tasks pertinent to medical imaging, spatial information is of acute necessity. Gaobo Liang et al. [11] proposed an elegant solution to this shortcoming. They presented a deep learning framework based on dilated convolution to preserve spatial information alongside residual structures to prevent over-fitting. In addition to dilated convolutions, their study emphasizes using transfer learning for better training on the small-scale dataset. CheXNet [15], a deep CNN model built by Stanford's researchers trained on the ChestX-ray14 dataset, achieves a diagnosis capability better than the average radiologist. Additionally, they executed a secondary check on the given dataset for proper classification. In transfer learning, predefined weights are a key factor in determining the performance of a model. The knowledge of CheXNet weights was transferred to the task of pediatric pneumonia diagnosis in several studies. It is highly favorable if the weights chosen belong to the same field. The differences in performance when using CheXNet weights, ImageNet weights, and random weights are detailed in [29]. Stephen et al. [30] investigated the performance of simple CNN architecture in the absence of transfer learning. Several studies focus on existing deep CNN architectures, such as MobileNets, VGGs, DensNets, and ResNets. Rahman et al. [21] studied the performance of AlexNet, ResNet18, DenseNet201, and SqueezeNet using transfer learning for normal vs. pneumonia, bacterial vs. viral pneumonia, and normal, bacterial, and viral pneumonia classification. Novel architectures were proposed as a solution to the existing limitations in these deep CNN architectures. Deep sequential CNNs for pediatric pneumonia detection are introduced in [19]. In [20], the authors exemplify the use of depthwise separable convolutions for the task of pediatric pneumonia diagnosis. A hybrid system consisting of adaptive median filter Convolutional Neural Network (CNN) recognition model based on Random Forest (RF) for detecting pneumonia from chest X-Ray images was introduced in [35]. In addition to different architectures, several feature extraction techniques were also employed. Wavelet transform is another technique for feature extraction based on a set of predefined filters. Akgundogdu et al. [18] analyzed the performance of 2D discrete wavelet transform for feature extraction with random forest for classification. Image enhancement techniques have become a topic of interest to improve the quality of the image and highlight essential features in an image. The effect of HE, CLAHE, image complement, gamma correction, and balance contrast enhancement techniques for chest X-rays are described in Tawsifur et al. [23]. Rubini et al. [24] compared two prominent spatial processing techniques- Adaptive histogram equalization (AHE) and Contrast Adaptive histogram equalization (CLAHE) for enhancing MRI images. El Asnaoui et al. [22] compares fine-tuned deep-learning architectures' performances for binary classification in pediatric chest X-rays. Their work details the advantage of using Contrast Limited Adaptive Histogram Equalization (CLAHE) as an image enhancement technique for better learning. The class imbalance problem is a necessity that needs to be addressed in machine learning. Machine learning is heavily dependent on a balanced dataset for unbiased training. Sampling is an important solution to deal with class imbalance problems. Habib, Nahida, et al. [25] proposed the use of Random Under Sampling, Random Over Sampling, and SMOTE on ensembled features from VGG-19 and CheXNet. Luján-García et al. [26] explored random undersampling (RUS) for unbiased training and used a cost-sensitive learning approach for the Xception network. However, such approaches' performance was limited because the data generated from SMOTE was unable to capture the required features for pediatric CXRs, and no new data was generated to improve learning in RUS. The performance of a model can be increased using several techniques. Increasing the feature set is one way to improve the performance of the model. This idea applied to pneumonia diagnosis was introduced by Nahid et al. [27] where they proposed a novel two-channel CNN architecture for pneumonia diagnosis. Predictions using feature concatenations from SqueezeNet and InceptionV3 along with ANNs are detailed by Islam et al. [28]. Their work entails retraining with modified parameters in addition to redistributing the existing dataset for unbiased training. Hyperparameters are a major contributing factor to the performance of a model. The right choice of optimizers is crucial to get the best results. While most of the recent related research focused on Adam optimizer, the effect of Stochastic Gradient Descent (SGD) optimizer was explained in [31]. Ensemble approaches are another important technique to improve the predicting accuracy of a model. Chouhan et al. [14] studied the performance of a majority voting ensemble combining the predictions from AlexNet, DenseNet121, Inception V3, GoogLeNet, and ResNet18. Sagar Kora Venu [13] proposed a weighted average ensemble of these deep CNN models—MobileNetV2, Xception, DenseNet201, ResNet152V2, and InceptionResNet. Nahida et al. [12] proposed a combination of a deep convolutional neural network for feature extraction, Principal Component Analysis (PCA) for dimensionality reduction, and logistic regression for classification. Improved feature representation may increase the performance of a classification model. A graph knowledge embedded convolutional network called CGNet was proposed by Yu et al. [33]. They used the transfer learning technique for feature extraction followed by graph-based feature reconstruction for classification. Mittal et al. [34] proposed a CapsNet architecture for classifying normal and pneumonia images. The main impeding factor for the complete transition to artificial intelligence (AI) is the lack of transparency. A promising field of research called explainable AI (XAI) has been gaining momentum lately. A unique approach in integrating explainability for pneumonia detection was introduced by Nguyen, Hai, et al. [32]. They proposed a combination of custom CNN architecture and Grad-CAM for pneumonia detection. An abundance of research has been done in this field. However, there exist limitations which are discussed below:Most studies propose data augmentation techniques to increase the number of samples for training to ensure improved performance. Artificially increasing the dataset is time and space inefficient. Studies emphasize the use of CheXNet weights for custom CNN training which is a challenging task. Lot of research proposes the use of custom complex architectures that are not easily replicable and hampers the reproducibility of the work. The absence in the exploration of ensemble approaches pertinent to pediatric pneumonia diagnosis was observed. The same was witnessed concerning the use of machine learning classifiers. The pressing need for dimensionality reduction using PCA has not been stated firmly. Data sampling methods like RUS, ROS, and SMOTE lead to longer training times and over-fitting. Most of the above-mentioned studies failed to cover the aspect of feature visualization. This is very important to ensure the learned features are meaningful for predictions. Our work proposes a detection pipeline to bridge the gap in the existing literature. The dataset has been redistributed for unbiased training instead of using data sampling methods. The proposed methodology is based on the Xception architecture pretrained on the commonly available ImageNet weights for feature extraction. The extracted feature maps from the global average pooling layer are passed to the t-SNE for feature visualization. Kernel PCA is then used for dimensionality reduction. Stacking ensemble classifier approach with KNN, SVC, Random-Forest classifier, Nu-SVC, MLP classifier, and Logistic Regression was used along with Stratified K-Fold cross-validation to overcome overfitting. All additional details are discussed in the forthcoming sections. Proposed approach This section details the workflow of the proposed pediatric pneumonia detection model, from the collection of data to the final classification as illustrated in Fig. 1. The dataset contains images of varying sizes. In this study, we reshape the images according to the requirement of different deep CNN models. Each image is normalized to bring the pixel values between the range 0–1 using the Keras image generator in addition to the introduction of sheer, zoom, and flip augmentations as shown in Table 1. Image augmentations are a necessary part of modeling to prevent over-fitting. These augmentations are generated on the fly in concurrence with the training.Fig. 1 Proposed architecture for pediatric pneumonia classification Table 1 Augmentations used in our study and their corresponding values Methods Corresponding parameters Rescale 255 Shear 0.2 Zoom 0.2 Horizontal Flip True The proposed architecture is trained on a two-step process. The first step was to use train deep CNN architecture for feature extraction. The Xception network was selected among all the other existing deep CNN architectures based on its performance for this task. A global average pooling layer was added to obtain feature maps. To prevent overfitting, a dropout rate of 0.4 was used and the Xception network was trained using binary cross-entropy loss. The ImageNet weights were used for transfer learning from second half of the layers in the Xception network. This resulted in better feature extraction from the CXRs. With the model now being able to extract the required features, the second step was to train the stacking classifier using the extracted features. The extracted features from the fine-tuned Xception network are sent through Kernel PCA for dimensionality reduction. The reduced features are trained on the stacking classifier with Nu-SVC, XGB classifier, Logistic Regression, K-Nearest classifier, Support Vector classifier, Randomforest classifier and MLP classifier for the first stage. The predictions from the base estimators (first stage classifiers) are trained on a meta classifier (logistic regression) for the final binary NORMAL and PNEUMONIA classification. Transfer Learning The performance of any deep learning model relies on the amount of data available. Accessibility to large datasets is guaranteed to increase the performance of deep learning models and make them more robust. Large datasets allow the model to learn much more intrinsic patterns. However, this is not always the case in medical imaging pertinent to pediatric pneumonia due to concerns, such as patient privacy and the time-consuming task of inspecting and labeling the data. Transfer learning [36] serves as a solution to this problem. In transfer learning, we use the existing knowledge gained when trained on a similar task and apply it to our detection of pediatric pneumonia. In our study, we fine-tune models pre-trained on ImageNet weights (trained on more than 14 million images ranging across 1000 classes). Deep learning models The literature survey concludes on the observation that competing performances were obtained when using pretrained deep CNN models. A detailed investigation on existing pretrained CNN architectures was performed to find the architecture best suited to the task at hand. These models pretrained on ImageNet weights were trained and tested on the Kermany dataset [10] to understand its advantages and limitations for pediatric pneumonia diagnosis. The initial half of layers were frozen while the second half of the models were fine-tuned. Pre-trained deep CNN models, such as VGG16 [37], VGG19 [37], MobileNet [38], MobileNetV2 [38], MobileNetV3Large [50], MobileNetV3Small [50], InceptionResNetV2 [39], DenseNet121 [40], DenseNet169 [40], DenseNet201 [40], InceptionV3 [41], ResNet50 [42], ResNet101 [42], ResNet152 [42], ResNet50V2 [43], ResNet101V2 [43], ResNet152V2 [43], EfficientNetB0 [51] and Xception [44] are trained on the Kermany dataset [10] to find the best performing model. The features are then extracted using the best performing model. The extracted features are passed through the global average pooling layer to extract one feature map from each image. These features are used for further processing. Details on the parameters used to conduct all the experimentations are explained in results and discussions. Xception CNNs rely on the gradients in the image for feature retrieval. Increasing convolution layers introduces the vanishing gradient problem, hence explaining the staggering performance with the increasing number of convolutional layers. Residual connections were introduced as a solution to the vanishing gradient problem. In time, researchers began incorporating residual structure in deep CNN models. Inception made its way into the research community along with its successors- InceptionV3 and InceptionResNets. Inception was built on the hypothesis that the spatial and cross-channel correlations in feature maps can be decoupled. Xception, leveraging this hypothesis pushed it to the extreme, thereby getting the name Xception, the extreme version of Inception. The Xception architecture is a stack of 14 modules (36 convolutional layers) with linear residual connections except for the first and the last modules. The entry flow initiates the flow of data and is followed by the middle flow where the set of operations is repeated 8 times. The architecture incorporates residual structure to tackle the vanishing gradient problem. The exit flow terminates the order of convolutions. The detailed architecture flow diagram is shown in Fig. 2. Each convolution and separable convolution layer is succeeded by batch normalization. In contrast to depthwise separable convolution where depthwise convolution is followed by pointwise convolution as shown in Fig. 3, Xception follows the reverse. The process starts with pointwise convolution followed by depthwise convolution.Fig. 2 The architecture for xception deconstructed as (a) Entry flow, (b) Middle flow and (c) Exit flow Fig. 3 Illustration of the working of a depthwise separable convolution network Hyperparameters To improve the feature extraction capability of the models, the best performing deep CNN architecture was first selected among existing deep CNN architectures. This selection was based on training all the architectures with a learning rate of 0.001 with adam as the optimizer and selecting the best performing model. The sigmoid activation function was used for this binary classification task. Hyperparameter tuning is a crucial step to boost the performance of a model. In our study, we fine-tuned the models based on different combinations of optimizers and learning rates as shown in Table 2, to select the perfect composition that results in the highest validation accuracy.Table 2 List of hyperparameters and their values used in our study to finalize the perfect combination for the task at hand Hyper parameter Corresponding values Optimizer Adam, SGD, Nadam, RMSprop, Adamax, Adagrad Learning rate 0.01, 0.001, 0.0001, decay rate from 0.001 to 0.000001 Batch size 32 Binary cross-entropy is used as the loss function which calculates the difference between the expected and actual output. The value for the loss function ranges between 0 and 1 and is given by Eq. 1.1 Loss =∑i=0outputsizeyi∗logy^i+1-yi∗log1-y^i Principal component analysis Large datasets often have redundant features which makes them difficult to interpret. No additional benefit is gained by learning the redundant feature rather it burdens the time taken to train a model. An immediate solution to this complication is Principal component analysis [45]. PCA is a famous statistical method used for dimensionality reduction. The PCA algorithm reduces the dimensionality of the given input in such a way that it minimizes the loss in reduced data. The objective of the PCA algorithm is to maximize the variance by creating new uncorrelated variables. The first set of uncorrelated variables forms the first principal axes. Similarly, subsequent sets of variables form their corresponding principal axes. The first principal axes capture the maximum variance and subsequent axes that are orthogonal to the previous axes capture decreasing variances in order. PCA performs well for linearly separable data however, this is never the case in real-world data. Kernel PCA [46] was developed as a solution to deal with nonlinear dimensionality reduction. It captures much more intrinsic correlations between the given high-dimensional features. Stacking classifiers Ensemble learning has attracted considerable attention in the past years. Studies emphasize it as a promising way to improve the performance of a model. Stacking Classifier is one such ensemble learning technique. As the name suggests, it stacks the predictions from individual classifiers (base classifiers) and uses them as features. These features are trained on a final classifier called the meta classifier. Stacking exploits the strengths of individual predictions, making predictions much richer and accurate. Stratified K-fold cross-validation Cross-validation is the most widely employed technique to estimate the model’s performance on unseen data. The performance on unseen data is of utmost importance for real-world deployment. The cross-validation technique facilitates the model to learn the most out of the provided data and prevent over-fitting. The Stratified K-Fold is an extension of the K-Fold cross-validation technique developed for the purpose of dealing with imbalanced class distributions. It ensures that each fold has same class distribution as in the original dataset. The dataset used has a higher number of pneumonia CXRs than normal CXRs for training. The class distribution in the training set was preserved in each of the folds. In this study we used Stratified K-Fold cross-validation with n_splits = 10. Dataset description The Kermany et al. [10] dataset was used for all the experiments in this comparative study. The dataset comprises 5856 Chest X-Ray images belonging to two categories- Normal (1538 X-rays) and Pneumonia (4273 X-rays). The dataset was split on an 80–10-10 (train-test-validation) split ratio after recombining the train and test data of the Kermany et al. [10] dataset. The data distribution used in this study is shown in Table 3. These chest X-ray images are from routine screening in Pediatric patients between 1 – 5 years of age from the Guangzhou Women and Children’s Medical Centre. The faint white occlusions, present in the X-rays in the second row in Fig. 4 are due to the occupancy of pus and fluids in the alveoli.Table 3 Distribution of the dataset for our study Category Train Test Validation Normal 1266 159 158 Pneumonia 3418 427 428 Total 4684 586 586 Fig. 4 Samples of Normal x-rays and Pneumonia x-rays from the dataset in the first row and second row, respectively Performance metrics Performance metrics are imperative to distinguish the performance of classification models. The metrics used in this study are accuracy, precision, recall, F1-score, and the AUC value. The Confusion matrix counts the distribution of predictions across the actual labels as shown in Fig. 5. Accuracy, Precision, Recall, and F1-score are derived from the confusion matrix. Fig. 5 Confusion matrix. True Positive (TP)—number of pneumonia x-rays correctly predicted as pneumonia. False Negative (FN)—number of pneumonia x-rays wrongly predicted as normal. True Negative (TN)—number of normal x-rays correctly predicted as normal. False Positive (FP)—number of normal x-rays predicted wrongly as pneumonia The accuracy of a model is calculated as the ratio between correct predictions and total predictions as shown in Eq. 2. The precision of a model is calculated as the ratio between true positives and total positives as shown in Eq. 3. It summarizes the quality of positive predictions made by the model. For a good classifier, precision is close to 1. Recall of a model is calculated using Eq. 4, which shows how well the predictions are classified as actual positive. F1-score is the harmonic mean of precision and recall as shown in Eq. 5. The Area Under Curve (AUC) score is the area under the receiver operating characteristics (ROC) curve. It defines the ability of the model to distinguish between patients with and without pneumonia. For a good classification model, the AUC score must be close to 1. 2 Accuracy =TP+TNTP+TN+FP+FN 3 Precision =TPTP+FP 4 Recall =TPTP+FN 5 F1 score =2Precision×RecallPrecision + Recall Results and discussion Several deep-CNN models were trained on the 4684 x-ray images for 30 epochs and evaluated on the test data consisting of 586 images to determine the model best suited for the task at hand. Google Colab resourced with K80 GPU and 12 GB RAM was used to conduct all the forementioned experiments in this study. Tensorflow2 and Keras2 were used to build and evaluate the models. The following models are compared and the best performing model is used for feature extraction: VGG16, VGG19, MobileNet, MobileNetV2, MobileNetV3Small, MobileNetV3Large, InceptionResNetV2, DenseNet121, DenseNet169, DenseNet201, InceptionV3, ResNet50, ResNet101, ResNet152, ResNet50V2, ResNet101V2, ResNet152V2, Xception and EfficientNetB0. Each of the models above were pre-trained on ImageNet weights with the corresponding input image of size 224 × 224 for all architectures except for InceptionV3, ResNet152V2 and Xception with an input image of size 299 × 299. All deep CNN models were trained with a constant learning rate of 0.001 and Adam as the optimizer. The initial layers of all the deep CNN models were frozen during training. Table 4 describes the layer count at which fine-tuning commenced for each deep CNN model with the corresponding count of trainable parameters. Table 5 illustrates the performance of the existing deep CNN architectures for the binary classification of no-pneumonia vs pneumonia detection along with the proposed method.Table 4 Fine-tuning information and the number of trainable parameters associated with each model used in our study Deep CNN model Fine-tuned from Total number of trainable parameters VGG16 9 13,569,793 VGG19 11 17,699,329 MobileNet 50 2,665,473 MobileNetV2 77 2,064,769 MobileNetV3Small 117 1,371,849 MobileNetV3Large 134 4,028,273 ResNet50 87 21,364,225 ResNet50V2 95 21,352,449 ResNet101 172 30,640,129 ResNet101V2 188 30,625,793 ResNet152 257 39,855,617 ResNet152V2 282 39,836,673 InceptionV3 155 16,791,489 Xception 66 14,860,313 InceptionResNetV2 390 41,922,529 DenseNet121 213 4,632,897 DenseNet169 297 8,544,833 DenseNet201 353 12,741,185 EfficientNetB0 118 3,700,169 Table 5 Performance chart of deep learning models with values rounded off to the nearest two decimal positions Model Accuracy Precision Recall F1-score AUC VGG16 0.73 0.73 1.00 0.84 0.5 VGG19 0.73 0.73 1.00 0.84 0.5 MOBILENET 0.96 1.00 0.96 0.97 0.97 MOBILENETV2 0.94 0.99 0.91 0.95 0.95 MOBILENETV3SMALL 0.27 0.00 0.00 0.00 0.5 MOBILENETV3LARGE 0.89 0.89 0.97 0.93 0.82 RESNET50 0.69 1.00 0.57 0.73 0.79 RESNET50V2 0.96 0.99 0.96 0.98 0.97 RESNET101 0.23 0.46 0.32 0.37 0.16 RESNET101V2 0.96 1.00 0.94 0.97 0.97 RESNET152 0.75 0.75 1.00 0.86 0.54 RESNET152V2 0.97 1.00 0.96 0.98 0.98 DENSENET121 0.96 1.00 0.95 0.97 0.97 DENSENET169 0.96 1.00 0.94 0.97 0.97 DENSENET201 0.96 1.00 0.95 0.97 0.97 INCEPTIONV3 0.92 1.00 0.89 0.94 0.95 XCEPTION 0.97 0.99 0.97 0.98 0.98 EFFICIENTNETB0 0.312 1 0.05 0.11 0.53 INCEPTIONRESNETV2 0.97 1.00 0.96 0.98 0.98 PROPOSED METHOD 0.98 0.99 0.98 0.99 0.98 When noticed, the family of DenseNet models performs consistently well. The reflection of collective knowledge in DenseNets enabled it to achieve an accuracy of 0.96. InceptionResNetV2, ResNet152V2, and Xception are the best performing architectures with the highest accuracy compared to the rest of the models for the task of pediatric pneumonia detection. The residual connections are a key factor that has suppressed over-fitting and thus enabled the above models to perform well on the test data. Though ResNet152V2 and InceptionResNetV2 achieve the same accuracy of 0.97 and an AUC of 0.98 similar to that of Xception, the latter has a higher recall of 0.97 compared to the former architectures. The recall of a model is of utmost importance as we do not want X-rays with pneumonia to be classified as normal. The confusion matrix for the test data predictions from the Xception architecture is shown in Fig. 6. From the confusion matrix, we conclude that the Xception in itself is unable to deal with false positives and false negatives. Figure 7 shows the ROC curve for the test data predictions. Xception proves to be a good feature extractor with an AUC of 0.97 still, its performance can be improved by looking at the feature representations.Fig. 6 Confusion matrix for xception predictions on the test data Fig. 7 ROC curve for test data predictions made by the fine-tuned xception model The training and validation plots are shown in Fig. 8. Though the loss initially peaks at irregular intervals, it substantially decreases. It can also be inferred that the validation loss and accuracy are constrained to certain bounds from 1 to 0 and 0.75 to 1, respectively. The validation data of 586 images were used for hyperparameter tuning. The Xception model was first fine-tuned on different optimizers to find the best fit for the task at hand. The Adam optimizer performs best as seen in Fig. 9. This combination was further tested on different learning rates. Figure 10 illustrates the competing performances of these learning rates when set to a static and a continuously regressing value. Based on Figs. 9 and 10, the optimal hyperparameters with the adam optimizer and a constant learning rate of 0.001 were chosen for feature extraction.Fig. 8 Training and validation accuracy-loss history of the fine-tuned xception model Fig. 9 Xception model performance on the validation set using different optimizers Fig. 10 Xception model performance on the validation set using different learning rates with adam as the optimizer Inspection of the features learned by deep learning models is crucial especially in the biomedical domain for its adaptability as a life-saving resource. This inspection was made possible with class activation maps [57] giving an overall vision of what the Xception model has learned. Figures 11 and 12 show the pixels that contributed the most while looking at pediatric pneumonia diagnosis for misclassified and correctly classified samples, respectively.Fig. 11 Class activation maps of misclassified X-rays (row 1: normal classified as pneumonia, row 2: normal classified as pneumonia, row 3: pneumonia classified as normal) Fig. 12 Class activation maps of correctly classified X-rays (row 1: normal classified as normal, row 2: normal classified as normal, row 3: normal classified as normal) Our method uses Xception for feature extraction with adam as the optimizer, the learning rate set to a constant value of 0.001 throughout the experiment, and a batch size of 32. The extracted features are visualized using the t-SNE [58] feature representation for the layman interpretability of the features predicted by the model. The t-SNE is a nonlinear dimensionality reduction technique that tries to preserve the local structure of the data. The feature maps of the test data are visualized using the t-SNE feature representation. The two dimensions (x and y-axes) shown in Fig. 13 are the first two principal components of the test data. This approach allowed us to visualize the normal and pneumonia samples in separate clusters. The cluster formation gives an idea of how well the predictions are made. In addition, the visualization element gives an insight into the possible classifiers that can be used for the classification task.Fig. 13 t-SNE feature representation of the test data extracted from the xception model The parameter values used for visualization are n_components = 2, perplexity = 40, and n_iter = 300. The t-SNE plot of the extracted feature maps from the Xception architecture is shown in Fig. 13. Looking at the cluster formations, we conclude that the test samples are nonlinearly separable with minor overlaps between the predictions and that we need a classifier that is able to deal with such complexity. This study proposes the use of the stacking classifier to deal with the nonlinearly separable classification. Thus, finalizing Xception as the feature extractor, the next step is dimensionality reduction using PCA (Principal Component Analysis). Dimensionality reduction is an important step to prevent the model from learning redundant features. In our study, we use the RBF (radial basis function) kernel with the number of resulting components as 200. This number has been chosen based on careful examination of the cumulative variance plot with a 95% cut-off threshold, shown in Fig. 14. Several machine learning classifiers were trained on the dimensionally reduced features and validated against the stacking classifier for the binary classification of normal and pneumonia CXRs. Table 6 concludes that the stacking classifier outperforms all machine learning classifiers by leveraging the strength of individual estimators.Fig. 14 Cumulative variance plot of the extracted xception features Table 6 Performance comparison of different machine learning classifiers with the stacking classifier with values rounded off to the nearest two decimal positions Classifier Accuracy Precision Recall F1-score AUC Logistic regression 98.13 98.83 98.60 98.71 97.73 Support vector classifier 98.13 99.29 98.13 98.71 98.12 Nu- Support vector classifier 97.44 97.47 99.07 98.26 96.07 K-Nearest classifier 98.13 99.29 98.13 98.71 98.12 MLP classifier 97.10 98.81 97.20 98.00 97.03 Gaussian naïve bayes 95.91 96.12 98.36 97.23 93.84 Bernoulli NB 94.89 98.77 94.16 96.41 95.50 Gradient boosting classifier 94.72 97.37 95.33 96.34 94.20 XGB classifier 96.59 99.28 96.03 97.62 97.07 Decision Tree classifier 94.72 97.37 95.33 96.34 94.20 Random forest classifier 96.08 96.13 98.60 97.35 93.95 Extra Trees classifier 96.76 97.01 98.60 98.80 95.21 Bagging classifier 98.13 98.60 98.83 98.72 97.53 AdaBoost classifier 95.06 97.84 95.33 96.57 94.83 LGB classifier 97.10 98.58 97.43 98.00 96.83 CatBoost classifier 97.96 99.29 97.90 98.59 98.00 HistGradient boosting classifier 96.08 99.27 95.33 97.26 96.72 Proposed method 98.30 99.29 98.36 98.83 98.24 Redundant features are detrimental to the performance of a classification model. The existing correlations between the important and redundant features are the key explanation for the hampering performance. The beneficial effect of removing redundant features in the task pertinent to pediatric pneumonia diagnosis is illustrated in Table 7 (Normal vs Pneumonia classification). The cumulative variance plot describes the percentage of the total variance captured by the first n components from the entire data. Higher variance indicates better preservation of important information from the data. The cumulative variance plot, Fig. 14 shows that the first 200 components capture most of the variance and that all additional principal components henceforth are redundant. The 200-dimensional output is passed to the two-stage stacking classifier.Table 7 Performance comparison with and without PCA with values rounded off to the nearest two decimal positions Method Accuracy Precision Recall F1-score AUC Stacking classifier in the absence of PCA 97.79 99.05 97.90 98.47 97.69 Stacking classifier with PCA 98.30 99.29 98.36 98.83 98.24 The first stage in the stacking classifier leverages the RandomForestClassifier, Support Vector Classifier, KNeighborsClassifier, XGBClassifier, LogisticRegression, Nu-Support Vector Classifier, and MLPClassifier. The hyperparameters for each of these classifiers were selected using GridsearchCV and are detailed in Table 8. Individual predictions from each of the five classifiers are sent to the meta-classifier for the final classification. The meta classifier uses LogisticRegression with penalty = l2, tol = 1e-4, C = 1.0, solver = ‘lbfgs’ and max_iter = 100. Stratified K-Fold cross-validation with n_splits = 10 was employed to help the model learn the most from the existing limited dataset and prevent over-fitting.Table 8 Fine-tuning information and the number of trainable parameters associated with each model used in our study Classifier Hyperparameters RandomForest n_estimators = 100, criterion = ’gini’, min_samples_split = 2, min_samples_leaf = 1, min_weight_fraction_leaf = 0.0, min_impurity_decrease = 0.0, ccp_alpha = 0.0 Support vector C = 1.0, kernel = ’poly’, degree = 3, gamma = ’scale’, coef0 = 0.0, tol = 1e-3 Nu-Support vector kernel = ’rbf’, degree = 1, gamma = ’scale’, probability = True, nu = 0.25, tol = 1e-3 K-Neighbors n_neighbors = 5, weights = ’uniform’, leaf_size = 30, p = 2 XGB loss = ’deviance’, learning_rate = 0.1, n_estimators = 100, subsample = 1.0, criterion = ’friedman_mse’, min_samples_split = 2, min_samples_leaf = 1, max_depth = 3, min_weight_fraction_leaf = 0.0 Logistic regression penalty = ’l2’, tol = 1e-4, C = 1.0, solver = ’lbfg’, max_iter = 100 MLP Hidden_layer_sizes = (50,10,10,10), activation = ’tanh’, solver = ’adam’ The confusion matrix for Stratified K-Fold cross-validation stacking classifier predictions on the test set is shown in Fig. 15. Lesser false-positive predictions from the stacking classifier are observed compared to the raw predictions made by the Xception architecture due to the strengths of individual classifiers. Thus, the strength of a stacking classifier solely relies on the individual strengths of the predictors. Principal component analysis has facilitated in lowering the number of false positives and false negatives which can be seen as a comparison between Figs. 15 and 16. The ROC curve, shown in Fig. 17 has an AUC value of 0.98. The AUC value from the stacking classifier has a 1% increase from the previously obtained AUC value. Looking at the confusion matrix, the loss of 1.7% in the accuracy of the model might favorably be due to the imbalanced dataset or insufficient training samples for training. The proposed method achieves a much higher accuracy of 98.30%.Fig. 15 Confusion matrix for predictions made on the test dataset using the stacked classifier with kernel PCA Fig. 16 Confusion matrix for predictions made on the test dataset using the stacked classifier without kernel PCA Fig. 17 ROC curve for predictions made on the test dataset using the stacked classifier Table 9 compares the performance, technique, and classification classes of our proposed approach with other recent works. The proposed work exhibits competing performances with other literary works for the binary classification of normal and pneumonia CXRs. All the works mentioned in the Table validated their results tested on the Kermany et al. [10] dataset. Since the Xception model used as the feature extractor is based on the commonly available ImageNet weights, reproducibility is easier. In addition to stacking various machine learning classifiers for rich predictions, the proposed method was tested on unseen pneumonia datasets for model generalization and robustness which was previously absent in recent works. The limitation of the proposed model is in its heavy reliance on the correct combination of base classifiers for accurate classification. The comparison hints at a possible future direction for using feature concatenations (Islam et al. [28]) followed by a stacking classifier for better results.Table 9 Performance of other recent works on the Kermany et al. [10] dataset with values rounded off to the nearest two decimal positions Authors Classes Technique Accuracy (%) Precision (%) Recall (%) AUC (%) Kermany et al. [10] Normal and Pneumonia Inception V3 pretrained CNN model 92.8 90.1 93.2 – Nahida et al. [27] Normal and Pneumonia Two-channel CNN model 97.92 98.38 97.47 97.97 Stephen et al. [30] Normal and Pneumonia Custom CNN model without Transfer Learning 93.73 – – – Chouhan et al. [14] Normal and Pneumonia Majority voting ensemble model 96.39 93.28 99.62 99.34 Rajaraman et al. [47] Normal and Pneumonia Custom VGG-16 model 96.2 97.0 99.5 99.0 Siddiqi et al. [19] Normal and Pneumonia Deep sequential CNN model 94.39 92.0 99.0 – Hashmi et al. [48] Normal and Pneumonia Weighted classifier 98.43 – – 99.76 Yu Xiang et al. [33] Normal and Pneumonia CGNET 98.72 97.48 99.15 – El Asnaoui et al. [22] Normal and Pneumonia Deep CNN model 96.27 98.06 94.61 – Saraiva et al. [16] Normal and Pneumonia MLP and NN approach 92.16 – – – Saraiva et al. [17] Normal and Pneumonia Custom CNN 95.30 – – – Mittal et al. [34] Normal and Pneumonia CapsNet architecture 96.36 – – – Rahman et al. [21] Normal and Pneumonia Deep CNN model 98.0 97.0 99.0 98.0 Sagar Kora Venu et al. [5] Normal and Pneumonia Weighted average ensemble model 98.46 98.38 99.53 99.60 Toğaçar et al. [49] Normal and Pneumonia Deep CNN model 96.84 96.88 96.83 96.80 Nahida et al. [25] Normal and Pneumonia SMOTE on ensembled features from VGG-19 and CheXNet 98.90 – – 99.00 Islam et al. [28] Normal and Pneumonia Feature concatenations with ANN 98.99 99.18 98.90 – Proposed Work Normal and Pneumonia Stacking classifier based on features extracted from Xception 98.3 99.29 98.36 98.24 Robustness and generalization of the proposed approach for lung disease classification The generalization of a proposed approach is essential to validate its performance. The proposed stacking classifier trained on the Kermany et al. [10] pediatric pneumonia dataset was tested on other pneumonia datasets [55, 56]. The confusion matrix of the predictions made on the test data on the two pneumonia datasets is shown in Figs. 18 and 19, respectively. The misclassifications in the first [55] and second [56] datasets are 25 and 31 false positives (normal predicted as pneumonia), respectively. The proposed method shows null false negatives in both unseen datasets. Tables 10 and 11 discuss the classification report for the corresponding datasets [55, 56]. The proposed method achieves an accuracy of 88% on the unseen test dataset [55] with 100 images belonging to normal and pneumonia classes each as shown in Table 10. The model’s reliability is supported by the precision of 100%, recall of 75% for the normal class, and precision of 80%, and recall of 100% correct prediction for the pneumonia class. In unseen dataset [56], the proposed method achieves an accuracy of 95% supported by 234 X-rays belonging to class normal and 390 X-rays belonging to class pneumonia as shown in Table 11. The model’s reliability is supported by the precision of 100%, recall of 87% for the normal class, and precision of 93%, and recall of 100% correct prediction for the pneumonia class. The weighted and macro averages differ by a small margin because of the class imbalance but are limited within 93−96%.Fig. 18 Confusion matrix for predictions made on the test dataset of normal vs pneumonia classification dataset [55] Fig. 19 Confusion matrix for predictions made on the test dataset of normal vs pneumonia classification dataset [56] Table 10 Classification report on the test data of normal vs pneumonia classification dataset [55] Precision Recall F1-score Support Normal 1.00 0.75 0.86 100 Pneumonia 0.80 1.00 0.89 100 Accuracy 0.88 200 Macro avg 0.90 0.88 0.87 200 Weighted avg 0.90 0.88 0.87 200 Table 11 Classification report on the test data of normal vs pneumonia classification dataset [56] Precision Recall F1-score Support Normal 1.00 0.87 0.93 234 Pneumonia 0.93 1.00 0.96 390 Accuracy 0.95 624 Macro avg 0.96 0.93 0.95 624 Weighted avg 0.95 0.95 0.95 624 The results conclude that though the challenges of pediatric pneumonia diagnosis are characteristically different from adult pneumonia, the proposed method can be extended to aid with the diagnosis of adult pneumonia. Conclusion and future work In this work, we propose a computer-aided diagnosis tool for pneumonia detection in infants using chest X-rays. Pediatric pneumonia is one of the substantial causes of the increasing death toll among children. Lower radiation levels in chest X-rays for children make detection a cumbersome and time-consuming task. Other works in the same field include using novel architectures and an ensemble of deep CNN models with the added advantage of using an augmented dataset to increase the number of samples in each category. Our work uses the existing deep CNN models for feature extraction; visualized using t-SNE feature representations and class activation maps, followed by Kernel PCA for dimensionality reduction. The reduced features advance into the stacking classifier for the final normal or pneumonia classification. Redistribution of the dataset instead of added augmentations to ensure unbiased training was the initial dominant factor for reliable performance. Our work uses transfer learning on pre-trained models to compensate for the availability of a limited dataset and introduces data augmentations to prevent overfitting. The Xception model achieves the highest accuracy and is used as the feature extractor. The advantage of Xception for this task in specific has been studied in detail along with the addition of PCA on the performance of the classification model. Dimensionality reduction is used to eliminate the redundant features. A stacking classifier covering nearly all machine learning models and neural networks was employed. Stacking classifier with Stratified K-Fold cross-validation results in an accuracy of 98.3%. The proposed approach was tested other pneumonia datasets to validate the performance across unseen data for generalization. As for future work, we would like to explore the effects of spatial domain data pre-processing techniques like Histogram Equalization (HE), Local Histogram Equalization (LHE), and Contrast Limited Adaptive Histogram Equalization (CLAHE) for the task of pediatric pneumonia detection. Reinforcement Learning-based hyperparameter tuning is another potential area of research. In the t-SNE plot (Fig. 13), we notice a few outliers and feature overlap between normal and pneumonia chest X-rays. This visualization pinpoints a potentially better model for better classification results. Custom CNN architectures with fewer parameters specific to occlusion-based categorization can be employed. The introduction of augmentations for training might help the model perform much better and reduce the current misclassification rate of 1.7%. In addition to that, we would like to explore simple yet powerful feature extraction models. CheXNet [15] was set as a benchmark for this study for it has reached the diagnostic level of human radiologists. With our work performing better CheXNet [15], it will be of immense help to all physicians and radiologists for accurate diagnosing in a matter of seconds. This early detection will help reduce the mortality rate of children suffering from pneumonia. Funding Not applicable. Availability of data and material The data that support the findings of this study are available from the first author upon reasonable request. Code availability The code is available from the first author upon reasonable request. Declarations Conflicts of interest The authors declare no conflict of interest. Compliance with ethical standards None. Informed consent None. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Neupane B et al. (2010) Long-term exposure to ambient air pollution and risk of hospitalization with community-acquired pneumonia in older adults." American journal of respiratory and critical care medicine 181(1):47–53 2. Ramezani M, Aemmi SZ, Moghadam ZE (2015) Factors affecting the rate of pediatric pneumonia in developing countries: a review and literature study. Int J Pediatrics 3(6.2):1173–1181 3. Lee GE et al. (2010) National hospitalization trends for pediatric pneumonia and associated complications. Pediatrics 126(2):204–213 4. Dean P Florin TA Factors associated with pneumonia severity in children: a systematic review J Pediatric Infect Dis Soc 2018 7 4 323 334 29850828 5. Rahman MM Machine learning based computer aided diagnosis of breast cancer utilizing anthropometric and clinical features Irbm 2021 42 4 215 226 10.1016/j.irbm.2020.05.005 6. Cherradi B et al. (2021) Computer-aided diagnosis system for early prediction of atherosclerosis using machine learning and K-fold cross-validation. In: 2021 International congress of advanced technology and engineering (ICOTEN). IEEE 7. Qin ZZ et al. (2021) Tuberculosis detection from chest x-rays for triaging in a high tuberculosis-burden setting: an evaluation of five artificial intelligence algorithms. Lancet Digital Health 3(9):e543-e554 8. Kundaram SS, Ketki CP (2021) Deep learning-based alzheimer disease detection. In: Proceedings of the fourth international conference on microelectronics, computing and communication systems. Springer, Singapore 9. Perdomo O et al. (2019) Classification of diabetes-related retinal diseases using a deep learning approach in optical coherence tomography. Comput Methods Prog Biomed 178: 181–189 10. Kermany DS et al. (2018) Identifying medical diagnoses and treatable diseases by image-based deep learning. Cell 172(5):1122–1131 11. Liang G Zheng L A transfer learning method with deep residual network for pediatric pneumonia diagnosis Comput Methods Prog Biomed 2020 187 104964 10.1016/j.cmpb.2019.06.023 12. Habib N, Hasan MM, Rahman MM (2020) Fusion of deep convolutional neural network with PCA and logistic regression for diagnosis of pediatric pneumonia on chest X-Rays. Network Biol 76 13. Kora Venu S (2020) An ensemble-based approach by fine-tuning the deep transfer learning models to classify pneumonia from chest X-ray images. arXiv e-prints (2020): arXiv-2011 14. Chouhan V et al. (2020) A novel transfer learning based approach for pneumonia detection in chest X-ray images. Appl Sci 10(2):559 15. Rajpurkar P et al. (2017) Chexnet: Radiologist-level pneumonia detection on chest x-rays with deep learning. arXiv preprint arXiv:1711.05225 16. Saraiva AA et al. (2019) Models of learning to classify x-ray Images for the detection of pneumonia using neural networks. Bioimaging 17. Saraiva AA et al. (2019) Classification of images of childhood pneumonia using convolutional neural networks.Bioimaging 18. Akgundogdu A Detection of pneumonia in chest X-ray images by using 2D discrete wavelet feature extraction with random forest Int J Imaging Syst Technol 2021 31 1 82 93 10.1002/ima.22501 19. Siddiqi R (2019) Automated pneumonia diagnosis using a customized sequential convolutional neural network. In: Proceedings of the 2019 3rd international conference on deep learning technologies 20. Siddiqi R Efficient pediatric pneumonia diagnosis using depthwise separable convolutions SN Comput Sci 2020 1 6 1 15 10.1007/s42979-020-00361-2 21. Rahman T et al. (2020) Transfer learning with deep convolutional neural network (CNN) for pneumonia detection using chest X-ray. Appl Sci 10(9):3233 22. El Asnaoui K, Chawki Y, Idri A (2021) Automated methods for detection and classification pneumonia based on x-ray images using deep learning. Artificial intelligence and blockchain for future cybersecurity applications. Springer, Cham, pp 257–284 23. Rahman T et al. (2021) Exploring the effect of image enhancement techniques on COVID-19 detection using chest X-ray images. Comput Biol Medicine 132:104319 24. Rubini C Pavithra N Contrast enhancement of MRI images using AHE and CLAHE techniques Int J Innov Technol Explor Eng 2019 9 2 2442 2445 10.35940/ijitee.B7017.129219 25. Habib N et al. (2020) Ensemble of CheXNet and VGG-19 feature extractor with random forest classifier for pediatric pneumonia detection. SN Comput Sci 1(6):1–9 26. Luján-García JE et al. (2020) A transfer learning method for pneumonia classification and visualization. Appl Sci 10(8):2908 27. Nahid A et al. (2020) A novel method to identify pneumonia through analyzing chest radiographs employing a multichannel convolutional neural network. Sensors 20(12):3482 28. Islam KT et al. (2020) A deep transfer learning framework for pneumonia detection from chest X-ray images. VISIGRAPP (5: VISAPP) 29. Mahajan S et al. (2019)Towards evaluating performance of domain specific transfer learning for pneumonia detection from X-Ray images. In: 2019 IEEE 5th international conference for convergence in technology (I2CT). IEEE 30. Stephen O et al. (2019) An efficient deep learning approach to pneumonia classification in healthcare. J Healthcare Eng 31. Manickam A et al. (2021) Automated pneumonia detection on chest X-ray images: a deep learning approach with different optimizers and transfer learning architectures. Measurement 184:109953 32. Nguyen H et al. (2020) Explanation of the convolutional neural network classifying chest X-ray images supporting pneumonia diagnosis. EAI Endors Trans Context Aware Syst Appl 7(21) 33. Yu X Wang S-H Zhang Y-D CGNet: A graph-knowledge embedded convolutional neural network for detection of pneumonia Inf Process Manage 2021 58 1 102411 10.1016/j.ipm.2020.102411 34. Mittal A et al. (2020) Detecting pneumonia using convolutions and dynamic capsule routing for chest X-ray images. Sensors 20(4):1068 35. Wu H et al. (2020) Predict pneumonia with chest X-ray images based on convolutional deep neural learning networks. J Intell Fuzzy Syst 39(3):2893–2907 36. Pan SJ, Yang Q (2009) A survey on transfer learning. IEEE Trans Knowl Data Eng 22(10):1345–1359 37. Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 38. Howard AG et al. (2017) Mobilenets: efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861 39. Szegedy C et al. (2017) Inception-v4, inception-resnet and the impact of residual connections on learning. In: Thirty-first AAAI conference on artificial intelligence 40. Huang G et al. (2017) Densely connected convolutional networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition 41. Szegedy C et al. (2016) Rethinking the inception architecture for computer vision. In: Proceedings of the IEEE conference on computer vision and pattern recognition 42. He K et al. (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition 43. He K et al. (2016) Identity mappings in deep residual networks. In: European conference on computer vision. Springer, Cham 44. Chollet F (2017) Xception: Deep learning with depthwise separable convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition 45. Jolliffe IT Cadima J Principal component analysis: a review and recent developments Philos Trans R Soc A Math Phys Eng Sci 2016 374 2065 20150202 10.1098/rsta.2015.0202 46. Ezukwoke K, Zareian SJ (2019) Kernel methods for principal component analysis (PCA) A comparative study of classical and kernel PCA. A preprint 47. Rajaraman S et al. (2018) Visualization and interpretation of convolutional neural network predictions in detecting pneumonia in pediatric chest radiographs. Appl Sci 8(10):1715 48. Hashmi MF et al. (2020) Efficient pneumonia detection in chest xray images using deep transfer learning. Diagnostics 10(6):417 49. Toğaçar M A deep feature learning model for pneumonia detection applying a combination of mRMR feature selection and machine learning models Irbm 2020 41 4 212 222 10.1016/j.irbm.2019.10.006 50. Howard A et al. (2019) Searching for mobilenetv3. In: Proceedings of the IEEE/CVF international conference on computer vision 51. Tan M, Le Q (2019) Efficientnet: rethinking model scaling for convolutional neural networks. In: International conference on machine learning. PMLR 52. Chowdhury MEH Can AI help in screening viral and COVID-19 pneumonia? IEEE Access 2020 8 132665 132676 10.1109/ACCESS.2020.3010287 53. Yang X et al. (2020) COVID-CT-dataset: a CT scan dataset about COVID-19. arXiv preprint arXiv:2003.13865 54. Nafi’iyah N, Setyati E (2021) Lung X-ray image enhancement to identify pneumonia with CNN. In: 2021 3rd East Indonesia conference on computer and information technology (EIConCIT). IEEE 55. https://www.kaggle.com/c/detecting-pneumonia-using-cnn-in-pytorch/data 56. https://www.kaggle.com/paultimothymooney/chest-xray-pneumonia 57. Zhou B et al. (2016) Learning deep features for discriminative localization. In: Proceedings of the IEEE conference on computer vision and pattern recognition 58. Van der Maaten L, Hinton, G (2008) Visualizing data using t-SNE. J Mach Learn Res 9(11)	0	PMC9734540	NO-CC CODE	2022-12-14 23:28:29	no	Neural Comput Appl. 2022 Dec 7;:1-21	utf-8	Neural Comput Appl	2,022	10.1007/s00521-022-08099-z	oa_other
==== Front Eur Geriatr Med Eur Geriatr Med European Geriatric Medicine 1878-7649 1878-7657 Springer International Publishing Cham 36471122 726 10.1007/s41999-022-00726-1 Research Paper A geriatric re-evaluation clinic is associated with fewer unplanned returns in the Emergency Department: an observational case–control study http://orcid.org/0000-0001-7744-1267 Balzaretti P. L. [email protected] 1 Reano A. 1 Canonico S. 2 Aurucci M. L. 1 Ricotti A. 1 Pili F. G. 1 http://orcid.org/0000-0003-4303-7252 Monacelli F. 2 Vallino D. 1 1 grid.414700.6 0000 0004 0484 5983 Emergency Department, Azienda Ospedaliera “Ordine Mauriziano”, Turin, Italy 2 grid.5606.5 0000 0001 2151 3065 Department of Internal Medicine and Medical Specialties (DIMI), University of Genoa, Genoa, Italy 6 12 2022 17 28 5 2022 24 11 2022 © The Author(s), under exclusive licence to European Geriatric Medicine 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. Key summary points Aim To evaluate whether the referral to a dedicated Geriatric Revaluation Clinic after discharge from the Emergency Department is associated with fewer early unplanned returns. Findings The referral to a Geriatric Revaluation Clinic was associated with fewer early unplanned revisits in a population of older age patients discharged from the Emergency Department. Message The creation of facilities aimed to evaluate the clinical evolution, treatment compliance, and functional status of geriatric patients discharged from the Emergency Department may contribute to prevent early unplanned revisits. Purpose The increasing share of older adults is associated with heavier Emergency Health Services utilization. In this context, a significant problem is the rate of unplanned revisits of geriatric patients after discharge from the Emergency Department (ED). We aimed to evaluate whether the referral to a dedicated Geriatric Revaluation Clinic (GRC) after discharge from the ED is associated with fewer early unplanned returns. Methods We conducted an observational, retrospective, case–control study comparing patients 65 years or older evaluated in a GRC after an ED visit and a control group of same age subjects accessing the ED during the study period and discharged with one of the ICD-9-CM diagnoses used for the cases, for whom defined post-ED assessment was not arranged. The intervention at the GRC consisted of a comprehensive geriatric evaluation. We calculated unadjusted and adjusted OR for unplanned ED revisits within 30 days from ED discharge using two logistic regression models including the variables with statistically significant differences among study groups at univariate analysis. Results During the study period, 121 eligible patients were evaluated at the GRC and were matched to 242 subjects included in the control group. The median age of the study population was 85 years. The adjusted OR for unplanned return after ED discharge and unplanned hospital admission after ED discharge were 0.44 (CI 0.20–0.89) and 0.52 (CI 95% 0.18–1.74), respectively. Conclusions In a population of older patients discharged from the ED, the referral to a GRC is associated with fewer early unplanned revisits. Keywords Older adults Emergency department Unplanned revisits Comprehensive geriatric evaluation ==== Body pmcIntroduction All developed countries are experiencing progressive aging of the population. The rise in the number of older people in the population leads to an increase in the prevalence of chronic health conditions [1]. Consequently, health services utilization, including the emergency care system, among older adults is relevant. In 2012, 22% of the two million older people living in the Lombardy region had undergone at least one ED visit [2]. Another study estimated that 14.3% of old age subjects in Italy had at least one preventable ED visit over a one-year follow-up, defined as an ED visit that ended with a direct discharge home [3]. The risk of unplanned revisits after a first discharge in the geriatric population, estimated between 15.8% and 33% at 30 days of follow-up [4–6], further burdens EDs. There is evidence that up to six months after discharge unpredicted re-admission rate can reach 50% [7, 8]. Apart from the economic implications of this pattern of Emergency Services usage, ED admissions carry a high potential for adverse events in the following months, such as increased mortality (estimated at 3% at one month after discharge to 10% at three months) and a reduction in autonomy in the daily activities [9]. All the epidemiological issues we have illustrated justify research aimed to define and deliver interventions for preventing unpredicted ED revisits. A meta-analysis has identified three kinds of interventions potentially useful to ease the ED discharge process for older adults [10]: “referral”, which is an assessment of the patient by a care provider (usually a nurse or social worker) in the ED, followed by recommendations to community-based agencies or referral for follow-up with the regular physician, “program or follow up” intervention, consisting in ongoing support or care for the patient after discharge from the index ED visit, and, finally, “integrated models of care,” defined as those interventions in which a care facilitator was embedded into the patient's care plans. A wide range of follow-up interventions has been proposed [10, 11], from telephone calls after discharge to complex transitional programs. The former were not associated with any reduction in the risk of unplanned returns in a systematic review of two large trials [12], probably because the intervention may not be enough to deal with the complexity of health issues experienced. The latter appear challenging to implement on a large scale in a busy ED, and their effectiveness has never been demonstrated in a clinical trial designed explicitly in an ED [11]. To explore such pressing problems, we proposed a model of care based on a geriatric clinic evaluation following ED discharge. The Geriatric ambulatory service, led by geriatric physicians also operating in the ED, offers a Global Geriatric Evaluation, a psycho-geriatric evaluation if needed, an assessment of the evolution of clinical conditions and of the adherence to the indications given at the ED discharge, and a pharmacologic therapy reconciliation based on validated appropriateness criteria. This study aimed to investigate the efficacy of a referral to a dedicated Geriatric Comprehensive Evaluation after ED discharge in preventing unexpected ED revisit within 30 days. As a secondary outcome, we considered hospital admission within 30 days from ED discharge. Methods This is an observational, retrospective, case–control study in which patients evaluated in a geriatric clinic after an ED visit were compared to subjects for whom specific post-ED assessment was not arranged. The study took place at Ordine Mauriziano Hospital, a 450-bed university-affiliated Hospital located in the urban area of Turin with a mean census of about 50,000 visits/year. The study protocol was approved by the local Ethical Committee (protocol number 159/2021). In this study, cases are patients consecutively sent to the Geriatric Re-evaluation Clinic (GRC) after an ED visit between October 22nd, 2018, and March 3rd, 2020, and between June 22nd, 2020, and October 26th, 2020. The activity of the GRC was suspended during COVID-19 pandemic. The control group was constituted of subjects 65 years or older, discharged from the ED between October 5th, 2018, and March 2nd, 2020, and between June 1st, 2020, and October 26th, 2020, for whom an evaluation to the GRC was not scheduled. The minimal difference in the recruiting periods was adopted to take into account the waiting time between ED discharge and the GRC evaluation. Exclusion criteria for the control group were:Incomplete ED access (left without being seen, leave against medical advice, leave without notification to ED personnel after the medical visit); Age less than 65 years; Hospital admission after ED evaluation; ICD-CM9 diagnosis codes not included among those employed in ED discharge of the cases; GRC evaluation after ED discharge. A synthesis of the selection of the subjects included in the control group is presented in Fig. 1. For each patient in the control group, only the first ED access during the study period was taken into account.Fig. 1 The selection process for the control group The population identified by the inclusion and exclusion criteria underwent a further selection and balancing process using a propensity score matching technique based on age, gender, and ICD-9 discharge diagnosis to create a 1 case: 2 controls ratio. In 2018, a Geriatric Re-evaluation Clinic (GRC) was established at our Institution, led by geriatricians who also work in the ED, specifically aimed to manage patients aged 65 years or older directly discharged from the ED. At the GRC, functional status is assessed employing the Barthel Index [13] and the Instrumental Activities of Daily Living scale [14], while the cognitive function is evaluated using the Short Portable Mental Status Questionnaire [15]. The screening for pharmacologic therapy appropriateness is conducted following the indications of the STOPP&START framework [16] with monitoring of the adherence to ED discharge instructions. In selected cases was offered counseling about social security interventions and home-based assistance. Patients were scheduled for the GRC evaluation after ED discharge at the discretion of the Emergency Physician. Data collection We extracted data from the electronic health records archived in the Hospital Database. After association with a five-digit numeric code, randomly generated data were entered in a spreadsheet using Excel 365 (Microsoft Corp, Redmond, WA). The following data were collected: age, sex, number of drugs taken every day, hospital admissions and number of ED accesses during the year preceding study enrollment, chronic comorbidities, ICD9-CM codes for ED discharge diagnosis, ED revisits, and hospital admission at 30 days follow-up. The comorbidities burden was addressed using Charlson Comorbidity Index [17]. Outcomes The main outcome of the study was unplanned ED revisits within 30 days from ED discharge. Unplanned revisits are returns to the ED which were not previously agreed upon (e.g., to complete diagnostic workup). The secondary outcome was unplanned hospital admission within 30 days from ED discharge. Statistical analysis Descriptive statistics for primary demographic and clinical data were conducted using medians with Interquartile Range (IQR) and proportions with 95% confidence intervals as needed. Proportions were compared using Chi-square test or Fisher Exact Test as indicated. Medians were compared with Wilcoxon Rank SumTest. A difference was considered statistically significant if p < 0.05. We estimated the enrollment ratio (1 case: 2 controls) using the number of cases available and the results of a previously unpublished data collection conducted at our Institution, from which a potential 50% reduction of ED revisits was estimated. We calculated unadjusted odds ratios (OR) to estimate intervention efficacy. Aiming to take into account potential imbalances among study groups, we estimated adjusted OR using logistic regression. The models employed included the variables with statistically significant differences among the two study groups at univariate analysis. Statistical analysis was conducted using R v. 4.0.3 [18]. Results During the study period, 142 patients were evaluated at the GRC. Due to a lack of follow-up data and repeated evaluation to the GRC, 21 subjects were excluded, leaving 121 cases for the study analysis. The process of selection of the 242 subjects included in the control group is reported in Fig. 1. Overall, median age of the subjects enrolled was 85 years. Men and women were equally represented (Table 1).Table 1 Demographic characteristics and medical history of the study population Patients scheduled to the G.R.C. (n = 121) (%, C.I. 95%)* Patients not scheduled for G.R.C. (n = 242) (%, I.C. 95%)* p Age (years), median (IQR.) 85 (81–88) 85 (80–88) 0.27 Age ≥ 85 years 54.6 (45.3–63.8) 50.4 (43.9–56.9) 0.53 Women 57 (47.7–66) 55.8 (49.3–62.1) 0.91 Chronic heart failure 14.9 (8.1–21.6) 26.5 (20.7–32.2) 0.02 Cerebrovascular disease 47.1 (37.8–56.4) 36.4 (30.1–42.6) 0.06 Diabetes 24 (16–32) 15.3 (10.6–20) 0.06 Hypertension 68.6 (59.9–77.3) 67.4 (61.3–73.5) 0.91 Chronic kidney failure 8.3 (2.9–13.6) 9.1 (5.3–12.9) 0.95 Cancer 17.4 (10.2–24.5) 19 (13.9–24.2) 0.81 Dementia 52.9 (43.6–62.2) 21.9 (16.5–27.3) < 0.01 Charlson comorbidity index, median (IQR.) 6 (5–8) 6 (5–7) 0.01 Charlson comorbidity index ≤ 6 52.9 (43.6–62.2) 67.4 (61.3–73.5) 0.01 Number of daily drugs, median (IQR.) 5 (3–7) 4 (2–6) 0.01 Polypharmacotherapy* 45.5 (36.2–54.7) 34.3 (28.1–40.5) 0.05 Hospital admission during the previous year 14.9 (8.1–21.6) 11.6 (7.3–15.8) 0.47 ED visit during the previous year 52.1 (42.8–61.4) 43.8 (37.3–50.3) 0.17 Married 37.2 (28.2–46.2) 59.7 (46–73.3) 0.01 If not otherwise indicated; including both solid and hematological neoplasms; **patients assuming five or more drugs daily. CI confidence interval; GRC Geriatric Re-evaluation Clinic; IQR interquartile range Patients in both study groups had a relevant and comparable burden of comorbidities; a substantial difference could be detected for dementia and chronic heart failure. The clinical complexity of this group of patients was confirmed by the higher number of drugs taken every day, which was higher among cases. Patients referred to the GRC more frequently had a final diagnosis of dementia or were visited in the ED for behavioral issues presumably due to cognitive decline, demonstrating that the Emergency Physicians felt that this category of patients is at high risk of ED frequent use. We found a 9.1% prevalence of 30 days ED unplanned returns in patients scheduled for a Comprehensive Geriatric Assessment compared to 18.6% in a matched cohort of patients for whom no formal follow-up to the clinic was arranged. The unadjusted OR for unplanned return in the ED within 30 days of discharge is 0.44 (CI 95% 0.22–0.89), suggesting a positive impact of the GRC referral in our study population. An unadjusted OR for unplanned admission within 30 days after discharge was 0.56 (CI 95% 0.18–1.74). Study groups were not balanced for every parameter. For this reason, we built two logistic regression models for estimating the adjusted OR. The first included the Charlson Comorbidity Index, the number of Table 2 daily drugs, and mental disorders related discharge diagnosis from the ED. The beneficial effects of the GRC referral remained, with an adjusted OR of 0.44 (CI 0.20–0.89) for 30 days unplanned ED re-admission and 0.52 (CI 0.14–1.56) for 30 days unexpected hospitalization.Table 2 Diagnosis at discharge from the ED using ICD9-CM classification Patients scheduled to the G.R.C. (n = 121) (%, C. I 95%) Patients not scheduled for G.R.C. (n = 242) (%, I.C. 95%) p Symptoms, signs, and ill-defined conditions 33.1 (24.3–41.9) 28.5 (22.6–34.4) 0.44 Mental disorders 24 (16–32) 9.1 (5.3–12.9) < 0.01 Injury and poisoning 12.4 (6.1–18.7) 16.9 (12–21.9) 0.33 Diseases of circulatory system 11.6 (5.5–17.7) 13.6 (9.1–18.2) 0.70 Diseases of the nervous system and sense organs 5 (1–8.9) 6.6 (3.3–10) 0.70 Diseases of musculoskeletal systems and connective tissue 5 (1–8.9) 5.8 (2.6–8.9) 0.94 Endocrine, nutritional, and metabolic diseases, and immunity disorders 2.5 (0–5.3) 3.3 (1–5.6) 0.76 Diseases of the blood and blood-forming organs 0.8 (0–2.5) 7.4 (3.9–11) < 0.01 Other diseases 5.8 (1.2–10.4) 8.7 (4.9–12.4) 0.44 Other diagnoses: diseases of the respiratory system, congenital anomalies, diseases of the digestive system, diseases of the genitourinary system, diseases of the skin and subcutaneous tissue, neoplasms, infectious diseases In the second model, we substituted the Charlson Comorbidity Index with history of dementia and heart failure obtaining similar adjusted OR: 0.43 (CI 0.21–0.92) and 0.55 (CI 0.16–1.87) for 30-days unplanned ED return and 30-days unexpected hospitalization, respectively. Data for marital status were available only for 57 patients (23.6%) in the control group: for this reason, we did not include the parameter in the final model. Finally, also hematological discharge diagnoses were not balanced in the study groups due to issues in control group selection operated by the statistical software. Still, they were not included in the multivariate model because of the marginal prevalence and the absence, to our knowledge, of any evidence of impact on unplanned short-term returns in the ED. Discussion In developed countries, the population is aging. As a major consequence, the prevalence of chronic diseases is increasing, which in turn relates to an ever-growing utilization of emergency health services. Nowadays, compelling evidence demonstrates a negative impact of ED admission and prognosis in older and frail patients [19]. Assuring a specialist follow-up to senior patients after an ED visit may reduce unplanned returns in the hospital, contributing to easing pressure on these frequently over-crowded services. Referral to GRC, which was at the discretion of the discharging emergency physician, was more frequent for patients with cognitive impairment-related issues. Even though the decision to schedule patients to the GRC was mainly based on physician gestalt, the relevance of dementia as a risk factor for repeated attendances to ED has been reported by other authors [20–24]. Unplanned returns to the ED are related to caregiver burden [25], which is particularly high for patients affected by cognitive impairment [26], and the opportunity for a comprehensive evaluation of geriatrics issues may contribute to reducing its intensity [27]. In our opinion, this "emotional" unloading constitutes the major explanation for our findings, along with the chance to re-evaluate the effectiveness of, and the compliance to, the therapeutic instructions given at discharge from the ED. Apart from mental disorders, the other covariates included in the logistic regression model were also validated in the literature as significant predictors of unplanned revisits in the ED [8, 20, 21, 28–30]. For example, Gips and co-authors reported an increased probability of repeated ED visits in patients with a Charlson Comorbidity higher than 4 [20]. Polypharmacy, defined as taking three or more medications daily, was present in 22% of patients with an unplanned return to the ED vs. 17% in the control group in the study by McCusker et al. (p = 0.03) [21]. Other studies investigated the implementation of in-person follow-up. Ballabio et al. published a report of an observational, before-after study, including 222 patients ≥ 75 years old, discharged from the ED. They showed a reduction of ED unplanned re-admission from 20% in the three months before enrollment to 11% in the three months of follow-up. The reported positive effect on cognitive performance and caregiver burden is of great interest [27]. In a quasi-randomized trial in Denmark, patients aged 75 years or more discharged from the ED or a geriatric ward were allocated to an intervention consisting of a follow-up home visit the working day after the discharge. Hazard ratio for re-admission within 30 days was 0.49 (95% C.I. 0.33–0.72). The findings of this study cannot be directly compared to ours because it included patients both discharged from the ED or an acute geriatric ward. Nonetheless, it demonstrates an in-person follow-up evaluation strategy's strong efficacy in preventing unplanned hospital returns after a discharge for an acute health condition [31]. Finally, Runciman and collaborators evaluated the potential impact of an evaluation by a health visitor in a population of patients 75 years or older discharged from the ED in a randomized controlled trial which included 424 patients. The health visitor assessed the needs of the study subjects and arranged a package of appropriate services. Twenty-eight days after ED discharge, 11.6% of patients in the intervention group and 9.3% in the control group were readmitted at the ED, suggesting no advantage from the intervention [32]. The findings can be partly explained by the low overall prevalence of unplanned returns in this study, requiring a much larger sample size. Another reason could be that the health visitor is not a physician, preventing the possibility of implementing essential elements of geriatric follow-up like therapeutic reconciliation. Our study has some limitations. The use of a case–control design to investigate the association between the exposure to an intervention and a subsequent outcome has been used infrequently employed in medical literature [33]; in our case, the adoption was motivated by the minimal number of patients sent to the GRC. Moreover, the case–control design may introduce selection bias. We are aware that our study has an exploratory scope and aims to generate working hypotheses that should be verified with a more robust research strategy. The lack of randomization makes it possible for unaccounted confounders to affect our estimations of efficacy. In particular, the inability to acquire data on clinical frailty and dependence in daily activities for the patients in the control group prevented us from taking these parameters into account in the final efficacy estimates. The monocentric nature of our work and the discretional referral to the GRC affect the generalizability of the results. We only considered unplanned return visits in our ED; for this reason, we cannot exclude an underestimation of outcome occurrence, being possible for the patients to seek care in other hospitals in our city [34]. Due to the study's retrospective design, we cannot estimate the proportion of patients in whom the behavioral alterations leading to the ED admission were caused by delirium. Even though this limitation may have determined an overestimation of dementia prevalence among the subjects included in the intervention group, in our opinion does not affect the finding about the overall usefulness of the GRC in the post-ED discharge management of older patients. In a population of older patients discharged from the ED, the referral to a GRC was associated with a lower rate of unplanned returns. The strengths of our study are the real-world setting in which it has been conducted, the relevance of the association between the intervention and the outcome, and the clear description of a model which could be readily applied to everyday practice. The involvement of a geriatrician in managing the more complex patients through an in-person evaluation may contribute to better identification of all patient's health issues, apart from the acute complaints which drove him to the ED. Notably, an assessment of adherence to treatment and of its effectiveness is important in preventing future decompensations leading to ED admission. In our opinion, our model is relatively less expensive both in terms of human and financial resources required compared to others available in the literature, making it worth consideration even in a low-resource setting. If the effectiveness of the proposed intervention will be confirmed in studies with more robust designs, such as randomized clinical trials, new research perspectives will open aimed at evaluating its beneficial effects. Indeed, the chance of reducing unplanned returns to the ED could impact favorably by lowering the risk of adverse effects and increased disability related to repeated ED visits in the geriatric population [9]. Also, it is reasonable to hypothesize positive consequences on the side of health organizations, bearing in mind the organizational and economic burden [35] related to ED revisits. Funding This study received no funding. Availability of data and material Not applicable. Code availability Not applicable. Declarations Conflict of interest The authors declare that they have no conflict of interest. Ethical approval The study protocol was approved by the Ethical Committee of AOU “Città della Salute e della Scienza” in Turin, AO “Ordine Mauriziano” and ASL “Città di Torino”. Consent to participate Not applicable. Consent for publication Not applicable. informed consent Informed consent was available for the intervention group; for the control group it was waived, considering the retrospective design of the study. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Laires PA Perelman J The current and projected burden of multimorbidity: a cross-sectional study in a Southern Europe population Eur J Ageing 2019 16 181 192 10.1007/s10433-018-0485-0 31139032 2. Franchi C Cartabia M Santalucia P Baviera M Mannucci PM Fortino I Emergency department visits in older people: pattern of use, contributing factors, geographical differences and outcomes Aging Clin Exp Res 2017 29 319 326 10.1007/s40520-016-0550-5 26931325 3. Gasperini B Cherubini A Pierri F Barbadoro P Fedecostante M Prospero E Potentially preventable visits to the emergency department in older adults: Results from a national survey in Italy PLoS ONE 2017 12 e0189925 10.1371/journal.pone.0189925 29267333 4. Moons P De Ridder K Geyskens K Sabbe M Braes T Flamaing J Screening for risk of re-admission of patients aged 65 years and above after discharge from the emergency department: predictive value of four instruments Eur J Emerg Med 2007 14 315 323 10.1097/MEJ.0b013e3282aa3e45 17968196 5. Salvi F Morichi V Lorenzetti B Rossi L Spazzafumo L Luzi R Risk stratification of older patients in the emergency department: comparison between the identification of seniors at risk and triage risk screening tool Rejuvenation Res 2012 15 288 294 10.1089/rej.2011.1239 22730956 6. Suffoletto B Miller T Shah R Callaway C Yealy DM Predicting older adults who return to the hospital or die within 30 days of emergency department care using the ISAR tool: subjective versus objective risk factors Emerg Med J 2016 33 4 9 10.1136/emermed-2014-203936 25987595 7. Sri-on J Tirrell GP Bean JF Lipsitz LA Liu SW Revisit, subsequent hospitalization, recurrent fall, and death within 6 months after a fall among elderly emergency department patients Ann Emerg Med 2017 70 516 521.e2 10.1016/j.annemergmed.2017.05.023 28688769 8. Hastings SN Schmader KE Sloane RJ Weinberger M Goldberg KC Oddone EZ Adverse health outcomes after discharge from the emergency department—incidence and risk factors in a veteran population J Gen Intern Med 2007 22 1527 1531 10.1007/s11606-007-0343-9 17828432 9. Aminzadeh F Dalziel WB Older adults in the emergency department: A systematic review of patterns of use, adverse outcomes, and effectiveness of interventions Ann Emerg Med 2002 39 238 247 10.1067/mem.2002.121523 11867975 10. Karam G Radden Z Berall LE Cheng C Gruneir A Efficacy of emergency department-based interventions designed to reduce repeat visits and other adverse outcomes for older patients after discharge: A systematic review Geriatr Gerontol Int 2015 15 1107 1117 10.1111/ggi.12538 26171554 11. Hughes JM Freiermuth CE Shepherd-Banigan M Ragsdale L Eucker SA Goldstein K Emergency department interventions for older adults: a systematic review J Am Geriatr Soc 2019 67 1516 1525 10.1111/jgs.15854 30875098 12. van Loon-van GM van Winsen B van der Linden MC Gussekloo J van der Mast RC The effect of a telephone follow-up call for older patients, discharged home from the emergency department on health-related outcomes: a systematic review of controlled studies Int J Emerg Med 2021 14 13 10.1186/s12245-021-00336-x 33602115 13. Mahoney FI Barthel DW Functional evaluation: the barthel index Md State Med J 1965 14 61 65 14. Lawton MP Brody EM Assessment of older people: self-maintaining and instrumental activities of daily living Gerontologist 1969 9 179 186 10.1093/geront/9.3_Part_1.179 5349366 15. Pfeiffer E A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients J Am Geriatr Soc 1975 23 433 441 10.1111/j.1532-5415.1975.tb00927.x 1159263 16. O'Mahony D O'Sullivan D Byrne S O'Connor MN Ryan C Gallagher P STOPP/START criteria for potentially inappropriate prescribing in older people: version 2 Age Ageing 2014 44 213 218 10.1093/ageing/afu145 25324330 17. Charlson ME Pompei P Ales KL MacKenzie CR 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 18. R Core Team (2022) R: A Language and Environment for Statistical Computing. 19. Nagurney JM Han L Leo-Summers L Allore HG Gill TM risk factors for disability after emergency department discharge in older adults Acad Emerg Med 2020 27 1270 1278 10.1111/acem.14088 32673434 20. Gips E Spilsbury K Boecker C Rebecca NG Arendts G Do frailty and comorbidity indices improve risk prediction of 28 day ED reattendance? Reanalysis of an ED discharge nomogram for older people Aging Clin Exp Res 2019 31 1401 1406 10.1007/s40520-018-1089-4 30560431 21. McCusker J Cardin S Bellavance F Belzile É Return to the emergency department among elders: patterns and predictors Acad Emerg Med 2000 7 249 259 10.1111/j.1553-2712.2000.tb01070.x 10730832 22. Friedmann PD Jin L Karrison TG Hayley DC Mulliken R Walter J Early revisit, hospitalization, or death among older persons discharged from the ED Am J Emerg Med 2001 19 125 129 10.1053/ajem.2001.21321 11239256 23. Kent T Lesser A Israni J Hwang U Carpenter C Ko KJ 30 Day emergency department revisit rates among older adults with documented dementia J Am Geriatr Soc 2019 67 2254 2259 10.1111/jgs.16114 31403717 24. LaMantia MA Stump TE Messina FC Miller DK Callahan CM Emergency department use among older adults with dementia Alzheimer Dis Assoc Disord 2016 30 35 40 10.1097/WAD.0000000000000118 26523710 25. Bonin Guillaume S Durand A-C Yahi F Curiel-Berruyer M Lacroix O Cretel E Predictive factors for early unplanned rehospitalization of older adults after an ED visit: role of the caregiver burden Aging Clin Exp Res 2015 27 883 891 10.1007/s40520-015-0347-y 25835219 26. Svendsboe EJ Testad I Terum T Jörg A Corbett A Aarsland D Patterns of carer distress over time in mild dementia Int J Geriatr Psychiatry 2018 33 987 993 10.1002/gps.4882 29575109 27. Ballabio C Bergamaschini L Mauri S Baroni E Ferretti M Bilotta C A comprehensive evaluation of elderly people discharged from an Emergency Department Intern Emerg Med 2008 3 245 249 10.1007/s11739-008-0151-1 18421427 28. Deschodt M Devriendt E Sabbe M Knockaert D Deboutte P Boonen S Characteristics of older adults admitted to the emergency department (ED) and their risk factors for ED re-admission based on comprehensive geriatric assessment: a prospective cohort study BMC Geriatr 2015 15 54 10.1186/s12877-015-0055-7 25928799 29. Bari MD Balzi D Roberts AT Barchielli A Fumagalli S Ungar A Prognostic Stratification of older persons based on simple administrative data: development and validation of the "silver code," to be used in emergency department triage J Gerontol A Biol Sci Med Sci 2010 65A 159 164 10.1093/gerona/glp043 30. Graf CE Giannelli SV Herrmann FR Sarasin FP Michel J-P Zekry D identification of older patients at risk of unplanned re-admission after discharge from the emergency department-comparison of two screening tools Swiss Med Wkly 2012 141 w13327 10.4414/smw.2011.13327 31. Pedersen LH Gregersen M Barat I Damsgaard EM Early geriatric follow-up after discharge reduces re-admissions–A quasi-randomised controlled trial Eur Geriatr Med 2016 7 443 448 10.1016/j.eurger.2016.03.009 32. Runciman P Currie CT Nicol M Green L McKay V Discharge of elderly people from an accident and emergency department: evaluation of health visitor follow-up J Adv Nurs 1996 24 711 718 10.1046/j.1365-2648.1996.02479.x 8894888 33. Shapiro ED Using case-control studies to assess the prevention of inflicted traumatic brain injury Am J Prev Med 2008 34 S153 S156 10.1016/j.amepre.2008.01.022 18374267 34. Shy BD Loo GT Lowry T Kim EY Hwang U Richardson LD Bouncing back elsewhere: multilevel analysis of return visits to the same or a different hospital after initial emergency department presentation Ann Emerg Med 2018 71 555 563.e1 10.1016/j.annemergmed.2017.08.023 28967514 35. Duseja R Bardach NS Lin GA Yazdany J Dean ML Clay TH Revisit rates and associated costs after an emergency department encounter: a multistate analysis Ann Intern Med 2015 162 750 756 10.7326/M14-1616 26030633	36471122	PMC9734542	NO-CC CODE	2022-12-14 23:28:29	no	Eur Geriatr Med. 2022 Dec 6;:1-7	utf-8	Eur Geriatr Med	2,022	10.1007/s41999-022-00726-1	oa_other
==== Front Gender Issues Gender Issues Gender Issues 1098-092X 1936-4717 Springer US New York 9307 10.1007/s12147-022-09307-9 Original Article A Multilevel Grounded Theory of Quantitative Job Quality Among Mothers, Fathers and Childless Women and Men in a Gendered, Classed and Aged “Growth-Driven” Organisation http://orcid.org/0000-0002-8758-6678 Turnbull Beth [email protected] 1 Graham Melissa 1 Taket Ann 2 1 grid.1018.8 0000 0001 2342 0938 School of Psychology and Public Health, La Trobe University, Bundoora, VIC Australia 2 grid.1021.2 0000 0001 0526 7079 School of Health and Social Development, Deakin University, Burwood, VIC Australia 6 12 2022 122 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. Poor quality jobs, incorporating job demands, resources and rewards, can impact employees’ health and wellbeing inside and outside work. However, jobs’ changing nature and employees’ increasingly diverse backgrounds mean existing job quality models may not adequately explain individuals’ job quality experiences within their individual, organisational and societal contexts. The paper aimed to understand mothers, fathers and childless women and men’s gendered, classed and aged experiences of quantitative job demands (including work amount, speed, effort, length and timing) and their resources and rewards, within multilevel contexts. We conducted a qualitative case-study of an Australian organisation, employing a critical feminist grounded theory design. We collected and analysed data from open-ended questionnaire responses from 47 employees and iterative in-depth interviews with 10 employees. Participants’ experiences of excessive quantitative demands, whether they could meet such demands, and whether they felt extrinsically or intrinsically resourced and rewarded for doing so, were embedded within ComCo’s masculine-neoliberal-capitalist growth imperative, cultures, policies and practices reinforcing growth, and quantitatively extreme and qualitatively conformant ideal worker discourses, as well as participants’ organisationally and societally-embedded individual, family and community-level contexts; producing nuanced gendered, classed and aged experiences among mothers, fathers and childless women and men. Although confirming well-established objective job quality dimensions, our research suggests individuals’ nuanced and subjective job quality experiences are embedded within individual, family, community, organisational and societal contexts. Keywords Gender Class Age Parent-status Job quality Australia ==== Body pmcIntroduction Although high-quality employment is associated with improved health and wellbeing, poor-quality jobs are those that can have severe impacts on mental and physical health and interfere with employees’ lives inside and outside work [1–4]. Jobs and workers have transformed dramatically in recent decades, with information technology, globalisation and international competition increasing work intensity and spatial and temporal boundlessness, and societal and demographic shifts bringing into workplaces more women, people from different backgrounds, and people with different family structures, including coupled and single parents and childless couples and single people [4, 5]. Accordingly, existing job quality models may not adequately explain employees’ experiences [6, 7]. Existing job quality theories help to explain work contexts that impact employees’ working lives and wellbeing. Many have identified working conditions, including demands, resources and rewards, which constitute job quality dimensions, such as Karasek and Theorell’s demands and control model [1], Siegrist’s [2] effort-reward imbalance model, Demerouti, Bakker, Nachreiner and Schaufeli’s [3] demands and resources model, and Polanyi and Tompa’s [6] qualitative exploration of established and emerging job quality dimensions. Demands can be broadly categorised as quantitative and qualitative. This paper focuses on quantitative demands, and their associated rewards and resources, which previous research has established as job quality dimensions that interact to influence employees’ physical and mental health and wellbeing [1–3]. Quantitative demands consist of interrelated demands relating to work amount (quotas, outputs and workloads); speed and effort (working hard, fast, intensely and under time pressure); length (working hours) and timing (working hours distribution, regularity and predictability) [1–3, 7], as well as conflicts between different quantitative demands, and between quantitative demands and employees’ health, needs, interests, attributes and non-work responsibilities (3, 5–9]. Rewards have traditionally included extrinsic rewards, such as adequate respect and recognition, advancement opportunities, and income and incentives, in return for employees’ efforts [2, 5–8]. However, others have identified intrinsic rewards, including feelings of accomplishment and success [5, 6], which can enhance the personal resource of intrinsic job motivation [9]. Finally, resources include supervisor and colleague [1, 3, 7, 10], leadership [5, 8] and organisational support [10, 11]. Some resources have also been categorised as rewards [9, 12], including job security, flexibility and control [1–3, 6, 7, 10]. Beyond identifying job quality dimensions, some researchers have emphasised interacting multilevel influences on job quality, including organisational, leader and team-level influences within organisations [13]; industry, labour market and societal influences beyond organisations [4, 7, 14]; and, congruent with person-environment fit theories [15], whether job conditions align with individuals’ life-contexts, including personal, family and community circumstances [7], and needs, interests, aspirations, personalities and identities [6, 16]. Accordingly, different job quality dimensions can be experienced as any or all of hindrance demands, challenge demands [17], resources or rewards, depending on the individuals experiencing them and the multilevel contexts in which they are embedded [6, 8, 18]. Importantly, Pocock and Skinner [7] have argued each of these interacting individual, organisational and societal-level influences on job quality, are influenced by employer-employee and gender power relations. Combined, extant job quality theories provide a gestalt understanding of job quality and its interacting multilevel individual, team, manager, organisational and societal influences. Accordingly, this paper responds to calls to redress the dearth of multilevel and qualitative research on job quality [16, 18], by presenting a grounded theory explicating how employees of a large Australian company, “ComCo,” experienced and gave meaning to quantitative job quality (that is, quantitative job demands and their interlinking rewards and resources), within their individual, organisational and societal contexts. Aligning with Pocock and Skinner’s [7] emphasis on power relations’ pervasion of societal, organisational and individual-level influences on job quality, and the demographic shifts bringing women and men with changing family structures into workplaces, this paper illustrates multilevel influences on quantitative job quality by exploring similarities and nuances in gendered, classed and aged experiences among mothers, fathers and childless women and men. Methods The study was conducted in Australia from early 2019 to mid-2020. The research employed critical feminist grounded theory [19–21] to understand how mothers, fathers and childless women and men represented and gave meaning to quantitative job quality in their gendered, classed and aged individual, organisational and societal-level contexts. While extant multilevel power relations theories informed our sensitising concepts, understanding quantitative job quality within ComCo’s unique context required enrichment of theory grounded in data [22], which drew upon job quality literature and theory as analysis and theory development proceeded, to contextualise participants’ experiences and facilitate deeper explanation of the emerging theory [19, 20, 23]. Accordingly, rather than reviewing empirical literature at the outset, it is (excepting the theoretical background) intertwined with the findings to privilege the grounded theory process [24]. Furthermore, we acknowledge researchers construct knowledge, and accordingly remained open to perspectives other than our own (as feminists, one mother and two childfree women) by endeavouring to recruit participants from various backgrounds and inductively analysing data. In 2018, ComCo agreed to its involvement, including business-hours employee participation. However, ComCo consented only to “white collar” employees’ participation (who were paid by annual salary), and not to that of blue collar and casual employees’ (who were paid by hourly wage), due to expense and impracticality. In early 2019, eligible employees were invited to complete a self-administered online questionnaire. This paper incorporates data from forty-seven of the respondents’ answers to open-ended questions asking them to describe any positive or negative experiences of working at ComCo connected with being mothers, fathers or childless women or men. From late-2019 to mid-2020, in-depth interviews were conducted with 10 employees, recruited using questionnaire contact details and snowball sampling (noting this paper focuses on participants’ experiences before COVID-19). Participants were interviewed twice for between 45 and 90 min per interview, employing semi-structured interviews exploring participants’ experiences of working at ComCo, asking participants, for example, to describe a typical working day from when they work up to when they went to bed, and how they experienced their everyday working conditions. Second interviews explored topics not covered in first interviews and issues emerging from first interviews. The audio-recorded interviews were transcribed verbatim. Three interviewees validated their transcripts. Although organisational constraints limited us to the original ten interviewees, combining qualitative questionnaire and interviewee data provided a range of perspectives and enabled saturation of many categories. However, while there were similar numbers of participants from support, commercial and operations departments, women and men, and mothers, fathers and childless women, few childless men participated, and participants were dominated by people who were head office-based, full-time employees, managers, aged 35 years or over, bachelor-qualified, spoke only English at home, Australian-born and heterosexual. Finally, using QSR NVivo 12, data were inductively and iteratively analysed using open, axial and selective coding, facilitated by data immersion, memo writing and constant comparison between data, emergent theory, and pre-existing literature and theory [19–21]. The remainder of this paper presents the multilevel quantitative job quality grounded theory, including quantitative job demands, their multilevel influences, and (de)motivating, (un)rewarding, (un)rewarded and (un)resourced quantitative hindrance/challenge demands, and our discussion and conclusions. Throughout, we use terms such as masculine, feminine, middle/working-classed, patriarchal, neoliberal and capitalist, to refer to attributes, practices or cultures that can be seen through a feminist lens as having been socially, culturally and discursively configured as gendered or classed, or as being produced by, and reproducing, unequal power relations, to emphasise power relations’ influence on participants’ experiences. Finally, to protect ComCo’s confidentiality, we have broadly identified its industry and not specified the number of employees. Additionally, we have carefully edited participant quotations, and attribute quotations with the minimum descriptors required to compare nuanced experiences, similar to strategies for protecting confidentiality in other organisational research [25]. Within-paragraph quotations are presented in double quotation marks. A Multilevel Grounded Theory of Quantitative Job Quality in a Gendered, Classed and Aged “Growth-Driven” Organisation ComCo, a multinational subsidiary with offices around the world, was an incorporated private company beholden to “meeting [and] exceeding shareholder expectations,” which manufactured consumer goods. In Australia, employees worked in what participants described as “middle-classed” and “heterosexual” but gender-diverse national office, “blue collar” and male-dominated factories, state offices or the field; and in male-dominated operations (research, development, quality, manufacturing, logistics, supply chain), support (finance, legal, corporate affairs and female-dominated human resources), and commercial (male-dominated sales, female-dominated marketing) areas, suggesting gendered and classed divisions of labour [25]. As illustrated in Fig. 1 (Multilevel grounded theory of gendered, classed and aged quantitative job quality), the grounded theory suggested participants’ experiences of masculine [26–28] excessive quantitative job demands, whether they could meet such demands, and whether they felt extrinsically and intrinsically resourced and rewarded for doing so, flowed from ComCo’s societally-embedded gendered, classed and aged inequality regimes [25, 29] (which, as discussed in depth elsewhere [30], triangulated organisational documentation and participant narratives about ComCo’s values, cultures, policies, practices and expectations, suggested were produced by ComCo’s overarching masculine-neoliberal-capitalist [27, 31] growth imperative by which ComCo sought “growth on growth on growth on growth,” which cascaded to diversified (including both masculine and feminine) growth mechanisms and quantitatively extreme and qualitatively conformant ideal worker discourses that were represented as contributing to growth); as well as individual, family and community-level contexts themselves embedded within gendered, classed and aged organisational and societal contexts [4, 27, 32, 33]. Although excessive quantitative demands aligned predominantly with ComCo’s masculine quantitatively extreme ideal worker discourses, they were also influenced by some qualitatively conformant ideal worker discourses [30]. However, interactions between multilevel influences produced nuanced and inequitable gendered, classed, aged and tenured experiences of quantitative job quality among mothers, fathers and childless women and men. We discuss the data from which the grounded theory emerged in the following sections. Fig. 1 Notes: 1 Constituted by interacting laws, policies, political and media discourses and ideologies, divisions of labour and parenting and working beliefs, aspirations and performances [32]; 2 Growth imperative, cascading to diversified growth mechanisms and quantitatively extreme and qualitatively conformant ideal workers, discursively constituted by participant narratives and organisational documentation regarding ComCo’s organisational, leadership, team and workplace cultures, values, policies, practices and expectations [30]; 3 Shows only those conformant worker expectations that are relevant to quantitative job quality Quantitative Job Demands Although research has found some differences in the extent of quantitative job demands reported by mothers, fathers and childless women and men [34–36], participants experienced quantitative job demands aligning with ComCo’s idealised masculine extreme worker discourses and established job quality dimensions [1, 3, 5, 37], regardless of sex or parent-status. Many participants experienced and normalised as “doing [their] jobs,” overarching high performance demands, including “constant pressure” to go “above and beyond” to “deliver” “high,” “unrealistic” and “undoable” “targets,” “goals,” “numbers” and “growth.”I see it as just doing my job, [but I’m] constantly getting recognised for going above and beyond … delivering what I need to deliver. [Childless man, non-manager] In turn, most participants experienced interlinking quantitative demands as necessary to achieve high performance and do their jobs, as elsewhere [36–38]. These demands included “ridiculous” and “overwhelming” “work overload” that sometimes “never ended,” whether consistently or in “peaks and troughs.”Last month, almost every week I was thinking, ‘I’m still behind, I don’t know how to get on top of it, I don’t know how to get on top of it.’ Now there’s enough there, but I’m on top of it. [Mother, manager] Additionally, many participants experienced “the daily grind of relentless work,” including working “extremely hard,” “intensely,” “efficiently,” “single-mindedly” and “urgently,” to manage excessive workloads and meet “targets,” “goals” and “tight deadlines,” and not having enough time to “complete tasks” before “rushing to the next task.”It’s not an easy job. It’s an all day, every day, all year, every year, effort. [Childless woman, non-manager] Moreover, many participants needed to work “excessive” or “crazy” hours, whether traditionally or flexibly, to manage “unreasonable” workloads, deadlines and targets, either “constantly” or during “peaks,” including part-time workers with full-time workloads, reflecting ComCo’s gendered extreme, but penalised, part-time worker discourses (Fig. 1).Constantly working excessive hours as the workload is relentless. [Mother, manager] It was, ‘It’s a full-time role. If you wanna do it in four days, that’s fine, but just work your arse off to make it happen, and we’ll pay you 20 percent less. And on the day you’re not working, we expect you to communicate, be on email and approve things.’ [Mother, manager] Finally, many participants needed to be “always” available, connected and flexible for work in which there was no temporal or spatial “clock on, clock off.” Many participants experienced physical availability demands during and outside business hours. Many also needed to be “always connected” during evenings, weekends, carer’s leave, annual leave and part-time workers’ “days off,” in an evolution of entrenched visibility cultures [39] driven by technology and flexibility to include online visibility.You’re always connected electronically … If you have a sick child and you’ve got a 30-minute call, sometimes you can distract your child and do the call. [Mother, manager] Additionally, some participants’ roles’ travel requirements imposed geographic flexibility demands.I was away all the time, most weekends and of course during the week. [Childless woman, non-manager] Although most participants experienced some or all of ComCo’s excessive quantitative demands, some whose life-contexts ran afoul of organisational discourses doubting the capability and commitment of gendered and aged employee categories [30], experienced a greater burden of proving they were meeting quantitative demands. For example, reflecting aged and tenured organisational discourses of “doubted” and not yet “trusted” younger and newer employees, some newer employees felt they needed to “prove [their] worth” by “delivering.” Similarly, emanating from societal and organisational discourses of mothers “not dedicated” to work, mothers who had returned from parental leave or were working flexibly, felt pressured to “deliver,” “perform” and “justify” their commitment, as in Thornton’s study [38].I think they judge me when I race out the door and they’re still working … I’ll do more hours [at night], but I feel like I have to justify it. [Mother, manager] Similarly, while most participants strained to meet some or all of ComCo’s excessive quantitative demands to the extent they were able, research has demonstrated that masculine quantitative demands create barriers to not only meeting and being rewarded for meeting such demands for those unable or unwilling to perform the masculine ideal (overwhelmingly women and mothers given the family-level context of disproportionate caring and domestic burdens); but also engaging in non-working lives for those able or expected to perform the masculine ideal, including childless people and fathers [27, 28, 40, 41], reflecting and reinforcing societal and organisational-level policies, ideologies and discourses constructing caregiving, career-sacrificing mothers, unencumbered breadwinning fathers and citizen-working childless people, which create unequal conditions for complying with nominally gender-neutral economic citizenship and ideal working [27, 28, 32, 42]. Conversely, a childless woman described personal characteristics, such as not being “efficient” or liking “having a lot on,” which limited her capacity to meet her job’s “unreasonable” quantitative demands without “tipping over.” Although, as elsewhere [38, 43], she believed parents were “forced” to develop those skills, she was cognisant such skills did not necessarily translate to being rewarded, which was predicated upon her temporally unconstrained ability to meet quantitative demands.I’m not good at managing stress and thinking on my feet and being efficient … I don’t like having a lot on … If I push too hard on that, I would tip over. I think having children … forces you to develop those skills … [However] it was easier for me to do things that were beneficial professionally … I could keep going, and they couldn’t… and that isn’t necessarily fair. [Childless woman, non-manager] Multilevel Drivers of Quantitative Demands As we explore at length elsewhere [30], many participants ascribed masculine [26–28] performance, workload, hard work, working hours and availability demands to ComCo’s globally and societally embedded growth imperative, growth mechanisms, and extreme and conformant ideal worker discourses (Fig. 1). Briefly, ComCo’s masculine-neoliberal-capitalist growth imperative [27, 31] entailed “nearly undoable expectations [for] growth on growth on growth on growth;” while masculine-capitalist [31] exploitative cost-cutting through redundancies and restructures created “bigger and busier roles” and exacerbated working-hours availability demands because “no-one was available” to “cover” absences. Masculine, neoliberal [26, 27, 44–46] hierarchical, self-interested leaders and managers who wanted to maximise their “substantial” bonuses, “pressured” participants to meet “unreasonable” expectations and targets in “unrealistic” timeframes. Performing feminine-collective [27, 45, 46] collaborative ideal worker practices entailed “back-to-back meetings” and “constant” emails that consumed some participants’ working days to the extent they had no time for “work,” producing a more than full-time workload; and required many participants to “be there” physically to meet local colleagues’ needs and attend meetings scheduled during and after business hours, as well as virtually, to collaborate with global colleagues “at different times of night.” Finally, performing qualitatively conformant ideal worker characteristics and practices (including going over and above core roles for masculine career growth [27, 47], building feminine relationships [27, 46], and masculine-neoliberal self-promotion [27, 45, 46] upon which bonuses and salary increases for high performance could depend) and “organising” and “improving” labour flowing from feminine beneficent growth mechanisms [27, 31] (such as committee membership or organising “fun” activities), necessitated time and availability beyond participants’ core roles, as Ely and Meyerson have observed [27]. Conversely, some organisational contexts aligned with resources provided by managers, teams and colleagues that alleviated or enabled participants to manage quantitative demands [3, 10]. Unlike hierarchical leaders, some participants’ managers personified feminine supportive leadership [27, 31, 46]: they provided instrumental support for managing quantitative demands by “taking on as much as they can,” “restructuring” workloads, “backing” participants to “prioritise,” encouraging participants not to “take on too much,” “readjusting” goals and respecting participants’ work-life “boundaries.” Conversely, a childless man felt “inconsiderately” drawn into being available outside working hours by flexibly working managers, despite their supportive “intent.”On the bottom of their email they’ll say, “I work odd hours, please don’t feel pressured to respond after hours or on weekends.” … Regardless of your intent, the impact is different … I don’t feel pressured to respond, but I can’t stop myself from reading it … then you can’t stop it from affecting you and making you think about work on the weekend. [Childless man, non-manager] In this respect, a flexibly working manager mother who felt negatively judged by colleagues, performed after-hours “online” visibility to “justify” her commitment. Thus, despite ComCo’s mainstreamed flexibility initiative (Fig. 1), lingering societal and organisational stigmatisation of “uncommitted” mothers and flexible workers may have driven some flexibly working managers’ visibly outside-hours emails, which in turn “bred anxiety” in subordinates. Similarly, ComCo’s discourses idealising feminine collaboration and teamwork [27, 45, 46] flowed to reciprocal instrumental support sometimes exacerbating and sometimes alleviating quantitative demands. For example, many participants’ and their teams assumed greater workload or availability demands to alleviate each other’s.We all try to help each other out, to smooth out those peaks. [Father, manager] [Manager] works some mornings from home … that generally will become the whole day, because I’ll say, ‘I’m happy to cover that meeting, there’s no need for you to come in.’ [Childless man, non-manager] However, other participants experienced imposed, rather than voluntarily, increased workloads to support colleagues.Having to do more work than your peers because they can’t manage. [Childless man, non-manager] Additionally, although performing ComCo’s feminine [27, 45, 46] relationship-building ideal worker practices could exacerbate workloads, once established, relationships and networks provided resources enabling many participants to meet or alleviate quantitative demands. For example, many participants felt networks facilitated “getting work done” in interconnected jobs or knowing who to influence about what to work on.I don’t think it’s ever going to get down to a relaxing amount of work. But you can … shift what’s important and sell that into people. You get better at it the longer you’re there. You know how to do it and who to do it to. [Father, manager] Finally, some participants described “excessive” quantitative demands as personal “choices” resulting from personally “high standards of performance,” wanting to “say yes” to and “do everything,” being “driven” and “self-motivated,” and “loving the work;” reflecting ComCo’s masculine [47] qualitatively committed ideal worker discourses, and internalising societal and organisational neoliberal individually responsible worker discourses absolving ComCo of responsibility for excessive demands (Fig. 1) [48].I’m probably my worst enemy in terms of how much I take on. [Mother, manager] Similarly, some participants attributed working intensely to not only workloads and urgency, but also life-contexts such as children, relationships, or maintaining physical and mental health, that constrained them to limiting their working hours. For example, expanding on research with mothers [38, 49], some fathers and a childless man worked more “efficiently,” or “focused” on work rather than building relationships, in order to minimise office hours and increase time for life. Participants’ attributions of quantitative demands to personal contexts supported arguments job demands can be exacerbated by personal demands [5, 8]. However, individual responsibility discourses can underestimate organisational and societal contexts (such as masculine [26–28] quantitatively extreme and masculine-neoliberal [47] “driven,” “committed,” “passionate” self-actualised ideal worker discourses) which constrain and manipulate individuals’ choices and identities in favour of those which serve organisational and societal imperatives [33, 50]. (De)Motivating, (Un)Rewarding, (Un)Rewarded And (Un)Resourced Quantitative Hindrance/Challenge Demands Supporting Polanyi and Tompa’s findings [6], participants described nuanced experiences of quantitative demands and their attendant rewards and resources. While many participants experienced some quantitative demands as demotivating, unrewarding and unrewarded hindrance demands, others experienced them as intrinsically rewarding and motivating, extrinsically rewarded, and providing resources for managing demands. Although these experiences aligned with job quality theories incorporating job contexts’ alignment with individuals’ needs, aspirations, preferences, attributes and circumstances [6, 8, 16], job and individual contexts were embedded within and interacted with organisational and societal contexts [7]. Intrinsically (De)motivating and (Un)rewarding Hindrance or Challenge Demands. Quantitative demands created challenge and hindrance demands for different participants. Many participants felt “engaged” by the “challenge” of the high-performance demand of “getting more growth,” or felt intrinsically “rewarded” and experienced “pride” or “achievement” when “delivering well” or achieving “tangible results,” sometimes despite “working really hard” or receiving inadequate extrinsic rewards [6, 9].There was a big push to ... achieve [target]. With a lot of hard work and time I managed to [meet target] and felt a sense of achievement. Not that I received much from the business. [Mother, non-manager] As in other research [51], feeling intrinsically rewarded by meeting performance demands was particularly important to some mothers, to counterbalance “unrewarding” aspects of mothering; mirroring societal-level discourses rhetorically valuing mothering, but constituting paid employment as integral to self-actualisation and citizenship (Fig. 1).I love being a mum, but sometimes it’s really unrewarding … At work if I hit a target, it’s an achievement. [Mother, manager] However, few participants felt intrinsically motivated or rewarded by other quantitative demands. Some geographically untethered [4] childless participants felt rewarded by meeting masculine [4, 27] geographic flexibility demands: a man “loved it” because it kept him “fresh,” while a woman experienced working internationally as “personally and professionally” rewarding. Other participants felt motivated by “busyness.”I thrive on that busyness sometimes. Not all the time, and not for a prolonged period, but it can make you productive and give you a sense of urgency. [Childless woman, manager] Conversely, many participants experienced “unreasonable” performance demands and their prerequisite quantitative demands, as intrinsically demotivating, demoralising and unrewarding hindrances, which reduced their work enjoyment, engagement, morale and functioning, and conflicted with or crowded out other job quality dimensions [7]. Many participants experienced excessive quantitative demands as “unachievable” hindrances preventing them from meeting both masculine quantitative demands and personal “high standards” of “doing everything” and “performing their best.”Doing your role with 70 percent satisfaction does wear you down. [Mother, manager] Moreover, quantitative demands deprived many participants of adequate time to perform the practices of ComCo’s qualitatively conformant ideal workers (Fig. 1) aligning with qualitative job quality dimensions [1, 6, 7], such as feminine socialising and relationships [27, 45, 46] or masculine personal development [27, 47], which simultaneously deprived them of enjoyable, valued or rewarding aspects of their working lives.Volume of work’s so high, if I take an hour lunch break, I have to make that up in my own time … You’re weighing up the benefit of catching up with people, developing relationships, working and enjoying it versus … trying to get work done. [Father, manager] Similarly, regardless of sex or parent-status, many participants experienced “overwhelming” quantitative demands as impacting other job quality dimensions by conflicting with their mental and physical health and non-working lives [3, 5, 7], by making them feel “ill,” “stressed,” “anxious,” “miserable,” “burnt out” and “exhausted,” and reducing their time and energy for non-working lives and relationships.Before the restructure I could complete my job without impacting my mental and physical health. Now I can’t. [Father, non-manager] Interestingly, some childless participants and parents with older children, who described less time-intensive non-work responsibilities, perceived quantitative job demands’ impact on their health and non-working lives as less severe than many parents with younger children, despite similar descriptions of “never comfortable,” “tiring” and “stressful” demands. However, a childless woman’s efforts to meet her role’s consistently excessive demands “broke her health,” suggesting her lack of temporal limits enabled strain that surpassed her physical and mental limits. Such experiences support arguments [6, 7, 16] that job quality, including quantitative demands and their impact on non-working lives, is subjective and contextual, and can be influenced by feminine-encumbered-fallible (as opposed to masculine-unencumbered-infallible) time, effort and energy-consuming life-contexts such as caregiving responsibilities, relationships, personal interests and health. Many such life-contexts are immersed within patriarchal-neoliberal-capitalist societal structures and discourses [14], overwhelmingly reproduced in family-level structures [32, 52], of caregiving mothers, breadwinning fathers and citizen-working childless people (Fig. 1). Extrinsically Rewarded and Resourced Demands. Emanating from ComCo’s diversified growth mechanisms, such as masculine individual and feminine collective rewards [27, 31], and feminine beneficent flexible working arrangements [27, 39, 42] (Fig. 1), which sought to create resources and rewards enabling and motivating all employees to meet ComCo’s quantitative demands and “accelerate” masculine-neoliberal-capitalist company growth [32], many participants experienced individual, team and company high performance as extrinsically recognised and rewarded, congruent with job quality models [2, 6–8]. Conversely, meeting workload, working hours, intensity and availability demands, though necessary for doing jobs, was rarely sufficient to attract extrinsic rewards without also achieving high performance. Many participants felt “acknowledged,” “appreciated,” valued,” “respected,” “proud” and “motivated” by recognition of their performance and (to a lesser extent) underlying hard work, through informal “day-to-day” peer and manager recognition of doing “a good job” or their “efforts;” formal performance ratings as “exceeding expectations” (which influenced awards, promotions, salary increases and bonuses), formal individual and team performance awards, and company celebrations.Team won [award] … a lot of hard work, hitting targets … going beyond that … doing that time and time again. I was absolutely stoked. [Mother, manager] As elsewhere [38], some participants felt “entitled to” and “deserving” of more substantive non-material rewards for meeting “performance” and “output” demands, whereby managers “trusted” them to “get the job done” and accordingly gave them “control” over how, when and where they did their work, and supported flexible working arrangements.I’ve got KPIs to reach, but they let me do it whichever way I want to … I really appreciate that trust. [Mother, manager] Reflecting organisational discourses of flexibility for work and life (Fig. 1), many such participants valued the job quality dimensions of control and flexibility [6, 18] as resources enabling them to, for example, manage masculine excessive workloads [26–28] by being more “efficient” and “productive” when working remotely, as well as feminine, encumbered life demands [27, 31, 47, 53], such caregiving responsibilities, long commutes or maintaining mental and physical health.I need to take my dog to the vet … so I’ll work from home … Because you don’t have the distractions … you get so much work done. [Childless woman, manager] However, as we discuss in detail elsewhere [30], participant narratives suggested that, despite ComCo’s mainstreamed flexibility initiative, these resources were not automatically bestowed, but rewards contingent upon masculine “output” and “high performance,” excluding newer employees who had not yet “earned their stripes.” Additionally, the kind of flexibility “rewarded” participants could use to manage work and life demands, was influenced by sex and parent-status, amplifying societal discourses of caregiving mothers, breadwinning fathers and childless citizen workers: mothers had the most, and fathers and childless people the least, access to flexible hours to manage life; while remote working for managing work as well as life, was more widely “embraced” and available. Finally, “blue-collar” and “male-dominated” operations and factory workers, who “needed to be available” and had “command and control” leaders, were excluded from being rewarded with flexibility and control. Similarly, aligning with job security as a job quality dimension constituting both a resource enabling employees to cope with demands [3] and an extrinsic reward [2], some participants felt secure in their roles despite cost-cutting restructures (Fig. 1) because their managers and ComCo “valued” them, their work added “value,” or there was “so much workload” ComCo could not “move it around;” revealing job security as a reward for masculine [26–28] high performance and excessive workloads.I look at our team and the value it brings and impact it makes … I don’t feel worried at all. [Childless man, non-manager] However, other participants were “realistic” job insecurity was “just what happens these days” in a “business that keeps restructuring and saving money,” internalising neoliberal-capitalist organisational discourses of employees as expendable growth mechanisms (Fig. 1).There’s always this unsettling level of, ‘I could be made redundant at any point.’ You just go into that thinking it’s a financial transaction … and find something else. [Mother, manager] Although different participants from support and commercial areas felt either “secure” or “uncertain” about their jobs, some participants observed that gendered and classed restructures and redundancies disproportionately “affected [blue collar, male-dominated] waged and manufacturing areas” and “operations.” Confirming another job quality dimension [2, 8], some participants experienced career development opportunities or promotions as “tangible” performance rewards.Being promoted through the business … based on my performance and track record … makes me feel valued. [Childless man, non-manager] However, career rewards were nuanced and gendered [6, 8]. Many participants felt intrinsically motivated and rewarded by “more difficult challenges” and “responsibility” accompanying promotions. However, some men who described themselves as “ambitious,” but only one woman, also got “value out of a title” or “more profiled” roles, and experienced “progressing” as demonstrating to themselves and others they were “achieving” and “excelling;” internalising societal and organisational discourses valorising masculine ambition and career progression [27, 47]. Finally, reflecting financial rewards as job quality dimensions [2, 7], many participants normalised an obligation to meet masculine excessive quantitative demands such as working “hard” and “long hours,” in return for ComCo’s “good” base salaries. Accordingly, many participants felt paid “well” and “fairly,” despite awareness that long hours meant hourly rates “wouldn’t be very good.”I’ve got a good salary … higher than similar jobs at different organisations … There’s an expectation you work really hard for that. [Father, manager] Additionally, ComCo rewarded high individual performance with annual salary increases, and combined individual, team and company performance with individual bonuses [30]. However, ComCo’s seniority-based employee class-structure and participants’ life contexts produced nuanced experiences of ComCo’s masculine-neoliberal financial rewards [27]. Some mothers and fathers in more senior roles, for whom bonuses were a “significant part” of their remuneration that could be “maximised” by “performing well,” described bonuses as a “factor” in “wanting to perform.” Indeed, the lure of bonuses undermined feminine [27, 31, 46] manager support alleviating workloads, for some participants cognisant that “doing less” meant “not getting a high-performance rating …. then [getting] a smaller bonus.” Conversely, many lower-level managers and non-managers, who received relatively insubstantial bonuses, were motivated more by wanting to “give their best,” as elsewhere [6].I’ve never worked a minute harder to get a bonus. You do your job to the best of your ability at all times. [Father, manager] Moreover, as in research in which few employees were motivated by money in itself [6], many participants felt motivated to “work hard” or “perform highly” for ComCo’s “generous” salaries and bonuses, because societally-embedded life-contexts made money “important.” For mothers, fathers and childless people, these included relying on sole, main or dual incomes to meet current and future practical financial obligations, such as “exorbitant” childcare, private school and university fees and mortgages; and aspirations for their own or children’s “lifestyles” and “financial security,” reflecting the neoliberal expansion of breadwinning and economic citizenship to women (Fig. 1).Work means money … providing a lifestyle I want to live. But it’s also a gateway towards future prospects … You read about how expensive house prices are … all these areas affect my generation. [Childless man, non-manager] Additionally, many women (but no men) were motivated by financial independence, echoing societal neoliberal-feminist discourses of independent women and mothers [28].I’ve always wanted to be independent and not rely on someone else. I want that for my daughter as well. [Mother, manager] Inadequate Extrinsic Rewards and Penalties. As in studies linking inadequate extrinsic rewards with job dissatisfaction and psychological strain [2, 5, 37], some participants experienced inadequate respect, recognition and rewards from colleagues, managers and “the business” for “working hard” and “exceeding” expectations, which reduced their “engagement.”The daily grind of relentless work, with little recognition or thanks. [Mother, manager] In this respect, women and other employees can experience societally and organisationally embedded barriers to being adequately rewarded [5, 54, 55], elucidated by participants’ gendered, classed, aged and tenured experiences of ComCo’s growth culture and mechanisms, and extreme and conformant ideal worker discourses (Fig. 1), discussed at length elsewhere [30]. Briefly, illustrating barriers to being rewarded with career opportunities and concomitant salary increases disproportionately affecting women and mothers in this and other research [4, 27], workload and availability-exacerbating “above and beyond” career growth prerequisites, and perceptions of “enormous” working hours and after-hours commitments “the further you move up the chain” that were possible for parents only with “full-time” at-home support, created barriers for participants with feminine [27, 31, 47, 53] encumbered time-intensive non-working responsibilities and priorities; requisite relocation to other sites or state, national or international offices for promotions imposed barriers for participants experiencing feminine [4, 27] geographic constraints requiring them to work nearby extended families and schools; and feminine [28] part-time employees were perceived as unable to meet high performance expectations and “blocked” from development and promotion opportunities. Exemplifying employee growth’s subordination to neoliberal-capitalist company growth, and aged, tenured and seniority-based class relationships in ComCo’s hierarchical organisational structure (Fig. 1), a highly performing younger employee was denied “promised” career development by a leader who wanted to use his expertise for “business ends;” while others felt expected to endlessly “broaden” and “get paid the same” because senior roles, salaries and bonuses were “monopolised” by a “35 to 45” year-old “age club” of employees with substantial tenure, who “do everything they can to keep that revenue in” to pay for “mortgages and kids,” mirroring the financial motivations of parents in more senior roles. Another younger participant had to “contest” his underpaid bonus, which he understood as a seniority-based “fight for resources.”It’s a fight for resources within the company … when you’re further down the chain, you’ve gotta fight even harder. [Childless man, non-manager] Furthermore, some experiences of inadequate recognition suggested conforming to or deviating from gendered, classed and aged qualitatively conformant worker discourses (Fig. 1) overrode or undermined high performance. For example, some participants’ achievements were “ignored” by masculine [46, 47, 56] authoritarian managers who praised and promoted feminine-submissive [47, 57] “favourites” who “toed the line,” “said yes” and “didn’t challenge,” and denigrated reputations of employees who “said no,” resulting in participants who were “doing a good job” and “exceeding” performance expectations feeling like they “were never going to progress.”One person [who was praised and promoted] never challenged the manager and that was what [manager] liked. [Mother, manager] Some participants’ descriptions of senior leaders’ “subjective” influence on bonuses and promotions in meetings “like a popularity test” where they “reviewed performance gradings of their subordinates,” “rated teams on … the next steps in their career” and developed career plans for “everyone … years in advance,” further explicated demands to “toe the line,” “not challenge” and “self-promote” with managers and senior leaders in order be adequately rewarded (Fig. 1). As in Casey’s research [33], two manager mothers experienced the necessity of masculine [27, 45, 46] “self-promotion” and “selling achievements” to achieve high performance ratings upon which salary increases, bonuses and promotions were based; requiring both women to undertake workload-exacerbating activities such as “tracking,” “recording” and “sharing” achievements, and one to compromise her feminine [27, 47, 53] “quietly achieving” nature.I thought, ‘I’m going to make sure people know what I achieve … and I’m going to track and share our progress” … because I felt like my rating was based on self-promotion, versus what I achieved. [Mother, manager] As elsewhere [37, 49], other participants described inadequate incomes in return for meeting quantitative demands, suggesting a lack of effort-reward fairness [2] in service of exploitative masculine-capitalist [31] cost-cutting and cost-saving. Some participants were given substantial additional work “from other people’s roles” or resulting from redundancies “with no increase in pay;” a part-time working mother was paid “20 per cent less” for a “full-time” workload; and a manager’s mostly female casually employed team members received inadequate wages despite their results.My team is not so good … we are one of the lowest [paid] in the market, I’m pretty sure … and they did their all. [Mother, manager] Given restructures and redundancies predominantly affected male and working-class-dominated operations and factory-based roles, and part-time and casual roles were female-dominated, these experiences suggested gendered and classed distribution of adequate financial rewards in ComCo. Finally, flowing from ComCo’s masculine [47, 58] growth mechanism of suppressing and penalising failure (Fig. 1), some participants described unfair extrinsic penalties for failing to meet “undoable” and “unachievable” masculine performance expectations. These included “negative” performance ratings, which impacted promotions, salary increases and bonuses; senior leaders “questioning” whether a previously highly performing employee “was right for the role” after a performance “dip;” and “redundancies” rather than “help and support.”My manager and colleague were made redundant. I didn’t receive a promotion or change in pay grade when I took on the majority of their workload. I then received negative reviews at end of year because I was struggling to cope with the significant change in workload. [Childless woman, non-manager] Discussion and Conclusions The multilevel quantitative job quality grounded theory integrates arguments job quality dimensions are ambiguous and overlapping [18] depending upon the individuals experiencing them and their interacting personal [6, 16], family and community [7], team, leader and organisational [13] and societal contexts [7]. That is, it suggests employees’ experiences of quantitative job demands’ existence, whether such demands are hindrances, challenges, intrinsically or extrinsically resourced, rewarding and rewarded, and barriers and facilitators to meeting such demands, are influenced by intertwining gendered, classed and aged (and likely raced and abled) global, societal, organisational, leadership, manager, team, community, family and individual-level contexts, which produce nuanced and inequitable job quality experiences among mothers, fathers and childless women and men. Many such experiences reflect and reinforce societal and organisational discourses of intensive, career-compromising mothers, and unencumbered, ideal working breadwinning fathers and childless citizen-workers excluded from non-working lives [28, 32]. Participants’ experiences of masculine [26–28] excessive performance, workload, working hours, speed and intensity, and availability demands align with established job quality dimensions [1–3], which our research illustrates can undermine other job quality dimensions, including feminine [27, 45, 46] socialising and capacity to maintain health and engage in life, and masculine [27, 47] personal development [7]. Our research also confirms leader and manager behaviours (including feminine [27, 31, 46] supporting, empowering and rewarding or masculine [46, 47, 56] obstructing, directing and ignoring) can profoundly influence quantitative job quality [13], including the existence and extent of quantitative demands, and whether employees are adequately extrinsically resourced and rewarded with respect and recognition, control, flexibility and job security, career opportunities, and bonuses and pay increases. Similarly, our research illustrates Karasek and Theorell’s [1] contention that teams and colleagues can provide instrumental support to alleviate and meet quantitative demands, exacerbate quantitative demands by expecting physical or online availability for meetings, emails and collaboration during and outside working hours, and contribute to extrinsic rewards by recognising colleagues’ efforts. However, experiences of resources and rewards also support less established job quality dimensions: whether employees feel adequately rewarded is influenced not only by the extrinsic rewards they receive, but also whether employees feel motivated by such rewards, and whether they feel intrinsically rewarded and motivated by meeting quantitative demands [5, 6]; experiences which are embedded within gendered individual, family, community, organisational and societal contexts [4, 27, 33]. Additionally, dimensions traditionally identified as resources, such as job security, flexibility and control, can also be experienced as rewards [9, 12]; creating a cycle in which crucial resources for meeting quantitative demands are not automatically bestowed, but rewards contingent upon meeting demands, creating barriers for unrewarded employees who have not yet met demands, to receiving resources that would support them to do so. Finally, our research facilitates a deeper understanding of quantitative demands and their non-material, career and financial resources and rewards, including leader, manager, team and colleague influences, by locating them within societally gendered, classed and aged organisational imperatives and their concomitant organisational, leadership, team and workplace policies, cultures, practices and ideal worker discourses. This research’s strengths lie in its demonstration that exploring multilevel contexts can enhance researchers’ and organisations’ understanding of job quality, by providing nuanced understandings of job quality dimensions, drivers, experiences, barriers and facilitators. Importantly, such understandings enable targeted strategies for improving poor job quality by addressing its underlying causes. However, the research was limited by the relatively small and homogenous sample, which consisted of employees who were exclusively white-collar and predominantly full-time employed, managerial, head office-based, aged 35 years or over, university educated, heterosexual, Australian born and spoke English at home. Nevertheless, integrating interviewees’ deep descriptions, qualitative questionnaire data and previous research, strengthened the grounded theory. Although it cannot be generalised beyond ComCo and the perspectives of the study’s relatively homogenous participants, the grounded theory may be relevant to other profit-driven incorporated companies [59]. Given the utility of understanding multilevel contexts, we call for more such research, and for other organisations to follow ComCo’s example by participating in such research, in order to identify and address multilevel contexts contributing to job quality experiences. Acknowledgements This work was supported by an Australian Government Research Training Program Scholarship. Author Contributions Conceptualization, B.T., M.G., A.T.; methodology, B.T., M.G., A.T.; formal analysis, B.T., M.G., A.T.; investigation, B.T.; writing—original draft preparation, B.T.; writing—review and editing, B.T., M.G., A.T.; visualization, B.T. Funding No funding was received for conducting this study. Declarations Competing Interests The authors have no relevant financial or non-financial interests to declare. Ethics Approval The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by La Trobe University’s Human Research Ethics Committee (HEC19285, 15 July 2019). Consent to Participate Informed consent was obtained from all individual participants included in the study. Consent to Publish Participants provided informed consent for publication of their data. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References 1. Karasek, R., & Theorell, T. (1990). Healthy work: stress, productivity and the reconstruction of working life. Basic Books. 2. Siegrist J Adverse health effects of high-effort/low-reward conditions Journal of Occupational Health Psychology 1996 1 1 27 41 10.1037/1076-8998.1.1.27 9547031 3. Demerouti E Bakker AB Nachreiner F Schaufeli WB The job demands-resources model of burnout Journal of Applied Psychology 2001 86 3 499 512 10.1037/0021-9010.86.3.499 11419809 4. Pocock B Charlesworth S Multilevel work–family interventions: creating good-quality employment over the life course Work and Occupations 2017 44 1 23 46 10.1177/0730888415619218 5. Cooper, C. L., Dewe, P., & O’Driscoll, M. P. (2001). Organizational stress: a review and critique of theory, research, and applications. SAGE Publications. 6. Polanyi M Tompa E Rethinking work-health models for the new global economy: a qualitative analysis of emerging dimensions of work Work (Reading, Mass.) 2004 23 1 3 18 15328458 7. Pocock, B., & Skinner, N. (2012). Good jobs, bad jobs and the australian experience. In C. Warhurst, F. Carre, P. Findlay, & C. Tilly (Eds.), Are bad jobs inevitable (pp. 61–77). Palgrave MacMillan. 8. Bakker AB Demerouti E Job demands–resources theory: taking stock and looking forward Journal of Occupational Health Psychology 2017 22 3 273 285 10.1037/ocp0000056 27732008 9. Victor, J. A., & Hoole, C. (2021). Rejuvenating the rewards typology: qualitative insights into reward preferences. SA Journal of Industrial Psychology, 47, 10.4102/sajip.v47i0.1832. 10. Cousins R Mackay CJ Clarke SD Kelly C Kelly PJ McCaig RH Management standards and work related stress in the UK Work and Stress 2004 18 2 113 136 10.1080/02678370410001734322 11. D’aleo N Stebbins P Lowe R Lees D Ham D Managing workplace stress: psychosocial hazard risk profiles in public and private sector Australia The Australian Journal of Rehabilitation Counselling 2007 13 2 68 87 10.1375/jrc.13.2.68 12. Aletraris L How satisfied are they and why? A study of job satisfaction, job rewards, gender and temporary agency workers in Australia Human Relations 2010 63 8 1129 1155 10.1177/0018726709354131 13. Bakker, A. B., & Demerouti, E. (2018). Multiple levels in job demands-resources theory: implications for employee well-being and performance. In E. Diener, S. Oishi, & L. Tay (Eds.), Handbook of well-being. DEF Publishers. https://doi.org/nobascholar.com. 14. Pocock B Williams P Skinner N Conceptualizing work, family and community: a socio-ecological systems model, taking account of power, time, space and life stage British Journal of Industrial Relations 2012 50 3 391 411 10.1111/j.1467-8543.2011.00852.x 15. van Vianen AE Person–environment fit: a review of its basic tenets Annual Review of Organizational Psychology and Organizational Behavior 2018 5 75 101 10.1146/annurev-orgpsych-032117-104702 16. Findlay P Kalleberg AL Warhurst C The challenge of job quality Human Relations 2013 66 4 441 451 10.1177/0018726713481070 17. LePine JA Podsakoff NP LePine MA A meta-analytic test of the challenge stressor–hindrance stressor framework: an explanation for inconsistent relationships among stressors and performance Academy of Management Journal 2005 48 5 764 775 10.5465/amj.2005.18803921 18. Cooke GB Donaghey J Zeytinoglu IU The nuanced nature of work quality: evidence from rural Newfoundland and Ireland Human Relations 2013 66 4 503 527 10.1177/0018726712464802 19. Corbin, J., & Strauss, A. (2008). Basics of qualitative research: techniques and procedures for developing grounded theory (3rd ed.). SAGE Publications. 20. Fassinger RE Paradigms, praxis, problems, and promise: grounded theory in counseling psychology research Journal of Counseling Psychology 2005 52 2 156 166 10.1037/0022-0167.52.2.156 21. Layder, D. (1993). New strategies in social research. Polity Press. 22. Mishra P Gupta R Bhatnagar J Grounded theory research: exploring work-family enrichment in an emerging economy Qualitative Research Journal 2014 14 3 289 306 10.1108/QRJ-01-2013-0001 23. Wuest, J. (1995). February 1, 1995). Feminist Grounded Theory: An Exploration of the Congruency and Tensions between Two Traditions in Knowledge Discovery. Qualitative Health Research, 5(1), 125–137. 10.1177/104973239500500109 24. Dunne C The place of the literature review in grounded theory research International Journal of Social Research Methodology 2011 14 2 111 124 10.1080/13645579.2010.494930 25. Connell R Glass ceilings or gendered institutions? Mapping the gender regimes of public sector worksites Public Administration Review 2006 66 6 837 849 10.1111/j.1540-6210.2006.00652.x 26. Acker J Gendered organizations and intersectionality: problems and possibilities Equality Diversity and Inclusion: An International Journal 2012 31 3 214 224 10.1108/02610151211209072 27. Ely RJ Meyerson DE Theories of gender in organizations: a new approach to organizational analysis and change Research in Organizational Behavior 2000 22 103 151 10.1016/S0191-3085(00)22004-2 28. Pocock, B., & Pocock, B. (2003). The work/life collision: what work is doing to Australians and what to do about it. Federation Press. 29. Acker J Inequality regimes gender, class, and race in organizations Gender & Society 2006 20 4 441 464 10.1177/0891243206289499 30. Turnbull, B., Graham, M., & Taket, A. (2022). Diversified inequality regimes and ideal workers in a “growth-driven,” “diverse,” “flexible” organization: a multilevel grounded theory. Social Sciences, 11(8), 325. 10.3390/socsci11080325. 31. Maier, M. (1999). On the gendered substructure of organization: dimensions and dilemmas of corporate masculinity. In G. N. Powell (Ed.), Handbook of gender and work (pp. 69–93). SAGE Publications. 32. Turnbull, B., Graham, M., & Taket, A. (2020) Hierarchical femininities and masculinities in Australia based on parenting and employment: a multidimensional, multilevel, relational and intersectional perspective. Journal of Research in Gender Studies, 10(2), 9-62. 10.22381/JRGS10220201 33. Casey, C. (1995). Work, self and society: after industrialism. Taylor & Francis Group. 34. Chapman J Skinner N Pocock B Work–life interaction in the twenty-first century australian workforce: five years of the australian work and life index Labour & Industry: A Journal of the Social and Economic Relations of Work 2014 24 2 87 102 10.1080/10301763.2014.915786 35. Drago R Tseng YP Wooden M Usual and preferred working hours in couple households Journal of Family Studies 2005 11 1 46 61 10.5172/jfs.327.11.1.46 36. Pocock, B., Hutchinson, C., & Skinner, N. (2012). The big squeeze: work, home and care in 2012. University of South Australia. 37. Trudel L Vonarx N Simard C Freeman A Vézina M Brisson C Vinet A Bourbonnais R Dugas N The adverse effects of psychosocial constraints at work: a participatory study to orient prevention to mitigate psychological distress Work (Reading, Mass.) 2009 34 3 345 357 10.3233/WOR-2009-0933 20037250 38. Thornton M Work/life or work/work? Corporate legal practice in the twenty-first century International Journal of the Legal Profession 2016 23 1 13 39 10.1080/09695958.2015.1093939 39. Kornberger M Carter C Ross-Smith A Changing gender domination in a big four accounting firm: flexibility, performance and client service in practice Accounting Organizations and Society 2010 35 8 775 791 10.1016/j.aos.2010.09.005 40. Burnett SB Gatrell CJ Cooper CL Sparrow P Fathers at work: a ghost in the organizational machine Gender Work & Organization 2013 20 6 632 646 10.1111/gwao.12000 41. Doyle, J., Pooley, J. A., & Breen, L. (2013). March 1, 2013). A phenomenological exploration of the childfree choice in a sample of Australian women. Journal of Health Psychology, 18(3), 397–407. 10.1177/1359105312444647 42. Swanberg JE Illuminating gendered organization assumptions: an important step in creating a family-friendly organization: a case study Community Work & Family 2004 7 1 3 28 10.1080/1366880042000200271 43. Kugelberg C Constructing the deviant other: Mothering and fathering at the workplace Gender Work & Organization 2006 13 2 152 173 10.1111/j.1468-0432.2006.00301.x 44. Matos K O’Neill O Lei X Toxic leadership and the masculinity contest culture: how “win or die” cultures breed abusive leadership Journal of Social Issues 2018 74 3 500 528 10.1111/josi.12284 45. Cheryan S Markus HR Masculine defaults: identifying and mitigating hidden cultural biases Psychological Review 2020 127 6 1022 1052 10.1037/rev0000209 32804526 46. Van Vianen AE Fischer AH Illuminating the glass ceiling: the role of organizational culture preferences Journal of Occupational and Organizational Psychology 2002 75 3 315 337 10.1348/096317902320369730 47. Berdahl JL Cooper M Glick P Livingston RW Williams JC Work as a masculinity contest Journal of Social Issues 2018 74 3 422 448 10.1111/josi.12289 48. Teasdale N Fragmented sisters? The implications of flexible working policies for professional women’s workplace relationships Gender Work & Organization 2013 20 4 397 412 10.1111/j.1468-0432.2012.00590.x 49. Cahusac E Kanji S Giving up: how gendered organizational cultures push mothers out Gender Work & Organization 2014 21 1 57 70 10.1111/gwao.12011 50. Lewis S The integration of paid work and the rest of life. Is post-industrial work the new leisure? Leisure Studies 2003 22 4 343 345 10.1080/02614360310001594131 51. Diamond C Baird M Whitehouse G Maternity leave and return to work in Australia-Accessibility and Use in a state utility Australian Bulletin of Labour 2007 33 2 134 157 52. Hilbrecht M Shaw SM Johnson LC Andrey J ‘I’m home for the kids’: contradictory implications for work–life balance of teleworking mothers Gender Work & Organization 2008 15 5 454 476 10.1111/j.1468-0432.2008.00413.x 53. Lewis S Humbert AL Discourse or reality?: “Work-life balance”, flexible working policies and the gendered organization Equality Diversity and Inclusion: An International Journal 2010 29 3 239 254 10.1108/02610151011028840 54. Watson I Decomposing the gender pay gap in the australian managerial labour market Australian Journal of Labour Economics 2010 13 1 49 79 55. Petersen, T., Penner, A., & Hogsnes, G. (2007). From motherhood penalties to fatherhood premia: the new challenge for family policy. Institute for Research on Labor and Employment. 56. Wajcman, J. (1998). Managing like a man: women and men in corporate management. Pennsylvania State University Press. 57. Williams JC Blair-Loy M Berdahl JL Cultural schemas, social class, and the flexibility stigma Journal of Social Issues 2013 69 2 209 234 10.1111/josi.12012 58. Reid EM O’Neill OA Blair-Loy M Masculinity in male‐dominated occupations: how teams, time, and tasks shape masculinity contests Journal of Social Issues 2018 74 3 579 606 10.1111/josi.12285 59. Lewis, J., & Ritchie, J. (2013). Generalising from qualitative research. In J. Ritchie, J. Lewis, C. M. Nicholls, & R. Ormston (Eds.), Qualitative research practice: a guide for social science students and researchers (pp. 263–286). SAGE Publications.	0	PMC9734544	NO-CC CODE	2022-12-14 23:28:29	no	Gender Issues. 2022 Dec 6;:1-22	utf-8	Gender Issues	2,022	10.1007/s12147-022-09307-9	oa_other
==== Front China Int Strategy Rev. China International Strategy Review 2524-5627 2524-5635 Springer Nature Singapore Singapore 123 10.1007/s42533-022-00123-0 Original Paper Beyond AUKUS: the emerging grand maritime alliance http://orcid.org/0000-0001-9970-6094 Shi Xiaoqin [email protected] grid.13402.34 0000 0004 1759 700X School of Public Affairs, Zhejiang University, West Quad, 866, Yuhangtang Rd, Hangzhou, China 4 12 2022 120 17 10 2022 11 11 2022 © The Institute of International and Strategic Studies (IISS), Peking 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. The AUKUS agreement is the first in history to allow a non-nuclear country to have nuclear-powered submarines, but not nuclear weapons. It pushes the Australia–UK–US alliance to the level of a closer military, scientific, and industrial community and outlines a prototype for a new “maritime alliance.” The closer the internal relationship of this alliance, the more it will estrange China. The alliance seeks to create a “de-Sinicized” defense industry chain and has thus become the forerunner for the restructuring of international relations. The “integrated deterrence” strategy that the alliance is building will change the strategic deterrence structure in the Indo-Pacific region and some essential features of the US alliance system’s deterrence strategy against China. China should pay great attention to the political will, strategic structure, deterrence concept, and military tactics displayed by this major agreement. Keywords AUKUS agreement Maritime alliance Military Scientific and industrial community Integrated deterrence http://dx.doi.org/10.13039/501100012456 National Social Science Fund of China 20VHQ003 Shi Xiaoqin ==== Body pmcIntroduction On September 15, 2021, US President Joe Biden, then Australian Prime Minister Scott Morrison, and then British Prime Minister Boris Johnson jointly announced the creation of the Australia–UK–US Trilateral Security Partnership, or AUKUS (White House 2021a). They declared that the first action of this security partnership would be the construction of a fleet of eight nuclear-powered submarines for Australia, the details of which would be released over the following 18 months. This was the first time since 1958 that the US had shared with other countries’ nuclear submarine technology. In April 2022, the three countries announced a second action: cooperation on the development of hypersonic weapons (White House 2022a). From a global perspective, the AUKUS agreement sends three major signals: first, a new “grand maritime alliance” is about to emerge; second, the US military alliance system is becoming a military, scientific, and industrial community; third, the military, scientific, and industrial community of the US and its allies will take integrated strategic capabilities and integrated deterrence as its future development goals, marking the beginning of the division of the world into blocs. A new “grand maritime alliance” is about to emerge The AUKUS agreement proclaims that Australia, the UK, and the US will share naval nuclear propulsion information, including “the design, arrangement, development, manufacture, testing, operation, administration, training, maintenance, and repair of the propulsion plants of naval nuclear-powered ships and prototypes, including the associated shipboard and shore-based nuclear support facilities.” The governments of the three countries have established working groups to work on issues such as safety, design, construction, operation, maintenance, disposal, regulations, training, and environmental protection, as well as full management of installations, infrastructure, bases, manpower, force structure, etc. (House of Lords Library 2022). In addition to naval nuclear power, AUKUS covers a wide range of fields, including military, science, technology, industry, and training. Even for military alliances, sharing high-end technical information and jointly building industrial capabilities are displays of deep trust and cooperation. UK National Security Advisor Stephen Lovegrove calls AUKUS “the most significant capability collaboration anywhere in the world in the past six decades” (Cabinet Office and Sir Stephen Lovegrove 2021). AUKUS was built on the bones of an Australian–French submarine deal. After Australia turned its back on France and joined the US and the UK, the new trilateral agreement became more than just a submarine construction project. Due to various factors, it has become a new way for the US to connect with its allies and a new model for the US alliance system. It also serves as a bridge between two strategic regions: the Atlantic Ocean and the Indo-Pacific Ocean. This was made clear in a speech delivered by US President Biden on September 15, 2021, when he announced the trilateral agreement. He said, “This is about investing in our greatest source of strength—our alliances—and updating them to better meet the threats of today and tomorrow. It’s about connecting America’s existing allies and partners in new ways and amplifying our ability to collaborate, recognizing that there is no regional divide separating the interests of our Atlantic and Pacific partners” (White House 2021b). As a result of this, the three countries’ position offshore the Eurasian Continent has become prominent. AUKUS connects the US’s Euro-Atlantic and Indo-Pacific theaters AUKUS’s negotiation process was extremely confidential. However, after the agreement came out, its potential global strategic significance was soon discovered. At a press conference following the announcement of AUKUS, Biden said that the agreement concerns “the future of each of our nations—and indeed the world” (White House 2021b). On April 1, 2022, the US Congress formed the AUKUS Caucus, which is dedicated to advancing the trilateral alliance. The Caucus is chaired by Chairman of the Seapower and Projection Forces Subcommittee Joe Courtney, Republican senator Mike Gallagher, Democratic senator Derek Kilmer, and Republican representative Blake Moore. According to Gallagher, “AUKUS is a critical new partnership that should be at the forefront of our security architecture in the Indo-Pacific.” Kilmer labels the UK and Australia as “critical allies.” Moore views the AUKUS alliance as a global one to contain China (Courtney 2022). Taken together, according to this Caucus, AUKUS apparently is a global maritime alliance and maritime community which aims to contain China. The strategic document, Advantage at Sea: Prevailing with Integrated All-Domain Naval Power (Department of the Navy, United States Marine Corps, and United States Coast Guard 2020), which was released by the US in late 2020, does not discuss global cooperation that includes China. Instead, it stresses securing the US’s maritime advantage in every domain. But the strategic document leaves out two major questions. First, how will the US confront both China and Russia at the same time? Second, how should the US distribute its strength between the Atlantic theater and the Indo-Pacific theater? In other words, the challenge that confronts the US is this: does the US have a coherent global strategy or only two independent regional strategies? This gap is now filled by AUKUS. AUKUS seems to incoporate Indo-Pacific strategy into the global strategy of the US. Since the birth of AUKUS, the US has been greatly pushing forward integration between the Indo-Pacific theater and the Euro-Atlantic theater. The Indo-Pacific Strategy of the United States, which was released in February 2022, says that the US has noticed that the EU and NATO “are increasingly committing new attention to the Indo-Pacific” and that the US “will harness this opportunity to align our approaches and will implement our initiatives in coordination to multiply our effectiveness” (White House 2022b). In his China policy speech on May 26, 2022, US Secretary of State Antony Blinken said that even though the Russia–Ukraine war has not yet ended, the US “will remain focused on the most serious long-term challenge to the international order—and that’s posed by the People’s Republic of China.” To cope with the China challenge, the US will be “building bridges among our Indo-Pacific and European partners, including by inviting Asian allies to the NATO summit in Madrid next month” (Blinken 2022). Later, Australia, New Zealand, Japan, and South Korea participated in the NATO summit in Madrid at the end of June 2022. Australia established its pivotal position in new global geopolitics by signing up for AUKUS Australia once hoped it could be a bridge between China and the US. In 2014, China and Australia established a “comprehensive strategic partnership.” In the same year, China, the US, and Australia conducted their first joint military exercise in Australia. In October 2015, the China–Australia Free Trade Agreement came into effect. In the same year, despite US opposition, Australia joined the Asian Infrastructure Investment Bank (AIIB) advocated by China. The government of Australia’s Northern Territory also leased a terminal in Darwin Port to China’s Landbridge Group for 99 years. The level of closeness and trust between China and Australia reached a historical peak. Australia believed it did not have to choose sides between the US and China back then (Schreer 2015; Thomson 2013). However, in 2015, the Landbridge Group’s Darwin Port lease deal unexpectedly came to the attention of the US, which made Australia aware that its relationship with China could jeopardize its relationship with its American ally (Perlez 2016). This event became the turning point in the bilateral relationship between China and Australia. Australia’s concerns about security outweighed its economic interests. In 2016, the Australian government blocked the sale of Ausgrid to Chinese holding companies. In 2017, then Australian Prime Minister Malcolm Turnbull accused foreign countries (China) of interfering in the internal affairs of Australia and even dramatically claimed that Australian people are standing up just like the Chinese people did in 1949 (Gribbin 2017). On June 28, 2018, the Australian parliament passed the Foreign Influence Transparency Scheme. In August of the same year, the Australian government banned China’s Huawei and ZTE from participating in Australia's national 5G network construction under the excuse of national security. In early 2019, Australia even lobbied the Five Eyes alliance to ban Huawei technology. In April 2020, Australia took the lead in demanding an international probe into the origins of the COVID-19 pandemic. On July 29, 2020, Australia followed the US in submitting a report to the United Nations denying the legal basis of China’s claims in the South China Sea. The move was hailed by the US ambassador to Australia as exercising “leadership in the region” (Galloway and Bagshaw 2020). In July 2020, Australia released its Defence Strategic Update, in which it cites a sharp deterioration in its security environment and blames the rapid growth of certain forces in the Indo-Pacific region. The Australian prime minister believes the region is in the midst of an unprecedented and far-reaching strategic restructuring (Morrison 2020). In just a few years, Australia has reassessed its regional security environment, repositioned its position between China and the US, and prepared to play a greater role in the new US strategic framework (Jennings 2019). At the press conference after the announcement of the AUKUS agreement, then Australian Prime Minister Scott Morrison called AUKUS a “forever partnership” 13 times. Australia has been “all-in” on the alliance with the US and the UK (Babones 2022). Whether from east to west or from south to north, Australia happens to be at the geographic hub of America’s new Indo-Pacific strategy. It is located in the center of the southern part of the Earth, with the Indian Ocean and the Pacific Ocean as its two wings. Australia also is fairly distant from China compared to the US allies of South Korea and Japan. It can deploy long-range weapons to deter China, and it is still at the outer edge of China's long-range strike capability (Dibb 2020). Therefore, for the US, Australia can still be regarded as a relatively secure forward base with ability to conduct long-distance deterrence against China. The US and Australia are both former colonies of the UK. The US inherited global maritime hegemony from the UK, while Australia first closely followed the UK and then went all out to the US for protection. Moreover, in the Indo-Pacific region, compared with other US allies, Australia and the US are the closest and most intimate in terms of history, culture, psychology, economy, technology, and values. If the US wants to focus on a country in the Indo-Pacific theater to cultivate as a special ally, or if we assess which country has the most potential to become a US special ally, the top candidate is undoubtedly Australia (US Department of Defense n.d.). The special relationship between Britain and the US after World War II needs no further discussion. The three countries of Australia, the US, and the UK are the most homogeneous countries in the international community in terms of their views on the world, the nature of their power, and their history and culture. In international power struggle, the three countries have the same way of thinking—that is, to maintain maritime superiority in the global maritime domain and to pursue limited intervention on land. The US and Britain have provided a large amount of military assistance to Ukraine, but they have not sent troops to intervene on land. In addition, their maritime forces have maintained vigilance in the Baltic Sea and the Mediterranean Sea. This is a continuation of the traditional British and American way of using sea power. It is not difficult to predict that the US, Australia, and the UK will work together to maintain maritime superiority over China, a land–sea power with long-range strike capability and nuclear weapons. The UK is pursuing a “Global Britain” strategy that emphasizes AUKUS and the Indo-Pacific It has been more than half a century since the British withdrew from the east of the Suez Canal. With Brexit and the rise of the Asia–Pacific, the UK has now adopted a “Global Britain” strategy that emphasizes the country’s global nature (British Cabinet Office 2021a). In 2018, the UK reopened the Bahrain military base it evacuated from in 1947, which is now the UK’s largest overseas naval operation base. In September 2020, the British Ministry of Defense announced the expansion of the Port of Duqm in Oman to facilitate the deployment of the British Navy to the Indian Ocean (Binnie 2020; Ministry of Defence and The Rt Hon Ben Wallace MP 2020). In March 2021, the UK launched the Integrated Review, which sets out a blueprint for its “Global Britain” goals. In this strategy document, the UK mentions the Indo-Pacific region more than 30 times and states that it will be deeply engaged in the Indo-Pacific as the European partner with the broadest and most integrated presence (British Cabinet Office 2021a). Clearly, Britain is reviving its past as a great sea power. In 2020, then British Prime Minister Boris Johnson promised to make the UK Europe’s “foremost naval power.” The UK is the second largest NATO country in terms of military spending after the US (Ewing 2020). According to the Integrated Review, the new era is the most severe era in the international environment since the end of the Cold War. It is an era of maritime competition between China and the US, and the Indo-Pacific region is the center stage of global maritime strategic competition. In this new era, the UK will be “free to go its own way” and will be open to a global network of allies and partners with whom it will forge new and deeper relationships (British Cabinet Office 2021b). In July 2021, the UK sent a fleet of unprecedented scale on a global deployment. Before the British fleet set sail, then Defence Secretary Ben Wallace said, “When our Carrier Strike Group sets sail next month, it will be flying the flag for Global Britain—projecting our influence, signaling our power, engaging with our friends and reaffirming our commitment to addressing the security challenges of today and tomorrow” (Sky News 2021). On July 6, the British “Elizabeth” aircraft carrier battle group passed through the Suez Canal, crossed the Indian Ocean, entered the South China Sea, visited South Korea and Japan, and later conducted joint exercises with the Japan Maritime Self-Defense Force and the US Navy. It also held joint exercises with the navies of Australia, New Zealand, Malaysia, and Singapore to celebrate the fiftieth anniversary of the signing of the Five Power Defence Arrangements (Ministry of Defence and The Rt Hon Ben Wallace MP 2021). The global deployment demonstrates the degree of US–UK integration, with US destroyers providing air and anti-submarine protection to the British fleet and ten US Marine Corps F-35B fighter jets providing the British fleet with additional air defense capabilities. The AUKUS agreement announced on September 15, 2021 has naturally become a new signpost on this path, as well as a gateway for the UK to intervene in the Asia–Pacific. Following its signing, former UK Prime Minister Boris Johnson said that the UK, the US, and Australia are natural allies and that AUKUS is a concrete step toward implementing the UK’s Integrated Review, a big step toward global security, a partnership between three like-minded allies, and a new partnership for sharing technology (British Prime Minister’s Office 2021). The AUKUS agreement has undoubtedly consolidated the UK’s presence in the Indo-Pacific region; promoted the building of integrated defense capabilities among Australia, the UK, and the US; and implemented the UK’s strategic intention of leaning toward the Indo-Pacific and returning to the global ocean (Patalano 2021). Australia, the UK, and the US are forming a division of labor to deter China through their grand maritime alliance “An important feature of US sea power is that it regards allies as an important part of US sea power” (Shi 2012, 120). The Biden administration has repeatedly affirmed the value of allies and partners in its maintenance of its global order. Among its many allies, Australia is the gateway to the Pacific Ocean and the Indian Ocean, and the UK is the chokepoint connecting the European continent to the Atlantic Ocean. From the two flanks of the Atlantic and the Pacific to the junction of the Pacific and the Indian Ocean, the UK, the US, and Australia have built a huge strategic arc which spans the outer seas of the Eurasian continent. This strategic arc constitutes a line of deterrence against China and Russia. The AUKUS agreement clearly outlines this strategic arc. From the perspective of task division, the UK can guard against Russia in the Atlantic region, and Australia can guard against China in the Indo-Pacific region. The close relationship between the US, the UK, and Australia can integrate the two strategic theaters into a single comprehensive theater. Moreover, the new “grand maritime alliance” may expand. AUKUS has sparked Japan’s imagination. Former Japanese Prime Minister Shinzo Abe and Japanese Ambassador to Australia Shingo Yamagami have both said that Japan is willing to participate in AI and cybersecurity cooperation under the AUKUS framework. Although limited by its Three Non-Nuclear Principles, Japan still expects to cooperate with AUKUS through some kind of “AUKUS + Forum.” On January 6, 2022, the Japan-Australia Reciprocal Access Agreement came into force (Ministry of Foreign Affairs of Japan 2022), making Australia the first country after the US to sign a visiting force status agreement with Japan. This is of great significance for the enhancement of interoperability and coordination capabilities between Japanese and Australian forces. Both Japan and Australia are big purchasers of US arms, and the importance of the improvement of their synergistic capability to the overall strength of the US alliance system is self-evident. After developments over the past few years, Japan and Australia have become each other’s closest military partner aside from the US (Wilkins 2022). This provides reason to be optimistic for the prospect of closer Japan–Australia–US trilateral security cooperation. If Japan joins AUKUS in some way, then the “grand maritime alliance” will expand to Northeast Asia. At the same time, the Russian–Ukrainian war has resulted in a new European security posture in favor of the US: the UK has shown an unexpected toughness, and the newly appointed British Prime Minister Liz Truss (who was then foreign secretary) has called for the “reboot, recast and remodel” of the Western approach to deterrence (Cameron-Chileshe 2022). The UK’s role as “offshore balancer” is back. Germany, which used to be very cautious and hesitant in the use of force overseas, has also made changes and will be more actively involved in European defense (Economist 2022), and it is starting to provide Ukraine with heavy weapons. As of present, all former Warsaw Pact member states, aside from Ukraine and Belarus, have joined NATO. Sweden and Finland, which did not join NATO during the whole Cold War era, will soon join NATO (Reuters 2022). Finland and Estonia plan to jointly guard the Gulf of Finland, which could cut off Russia’s access to the Baltic Sea in times of war (Häggblom 2022). The present Russia–Ukraine war “has altered the political, strategic and economic framework conditions for European regional cooperation and transatlantic relations” (Reuters 2022). At the NATO summit held in Madrid, Spain from June 29 to 30 this year, members decided that NATO will expand its troops and prepare for war and that NATO troops will be deployed on the territory of former Warsaw Pact countries. NATO has become stronger and more united (Heads of State and Government at the NATO Summit in Madrid 2022). Europe and NATO have invested more and become more united and active on defense issues. Further, Russia is in a defensive position because of its failure on the battlefield. The balance of power between NATO and Russia has dispelled some people’s initial fear of the Atlantic and Indo-Pacific competing for US resources. On the contrary, with the support of AUKUS, the intent of the US and its allies to divide labor to contain China in the Indo-Pacific and Russia in the Euro-Atlantic relatively is becoming a reality. Closer and deeper than a military alliance: toward a military–science–industrial community On September 15, 2021, at the video conference that announced the AUKUS trilateral security partnership agreement, then Australian Prime Minister Scott Morrison said that the trilateral security agreement is not only a combination of defense forces, but also a combination of technical and scientific cooperation, industrial capabilities, and supply chain integration. It will be a combination of Australian, British, and American technology, scientists, industry, and defense forces (Prime Minister of Australia 2021). It can be seen that the ambition of AUKUS is not limited to an arms sales contract. In explaining why Australia did not build a nuclear-powered submarine fleet from the start in 2016 (the year Australia signed a submarine construction contract with France), Morrison revealed, “That wasn’t on the table. It wasn’t on the table for a range of reasons.” But a nuclear-powered submarine fleet is now “a feasible option for Australia,” and “We now have the support and expertise of the United States and the United Kingdom” (ABC News 2021). Obviously, the decisive factor was the US agreeing to supply nuclear technology to Australia. This shows that the US plays the role of an enabler in the alliance system, a change from the traditional simple cooperation model in which the US provides and its allies receive capabilities and its allies receive capabilities. The US alliance system is closer and more integrated. The “weapon system community” is an important means for the US alliance system to consolidate and enhance the overall power of the alliance in international relations after World War II Since World War II, the US has always regarded international equipment cooperation as a symbol of alliance unity as well as an important means of coordinating alliance military strategy and foreign policy goals. The US has never forgotten the international strategic significance of the planning and layout of the defense industry (Hartley 2006). The military and industrial cooperation between the US and its allies has gone through roughly five stages: the sale of US military equipment through intergovernmental procurement agreements; the issuance of licenses to allow allies to produce or co-produce; allies’ imitation of the production of US-made weapons; the US and its allies’ joint development of the next generation of existing weapon systems; and now, the execution of joint research, development, and investment. The F-35 fighter jet program is a typical example of this most recent stage. This project includes eight founding countries, including Australia, Canada, Denmark, Italy, the Netherlands, Norway, the UK, and the US. The total amount spent on this project is in the trillions, and a full life estimate of this spending is about $1.45 trillion. This is the largest weapons project in modern history and is known as the “project of the century” (Shalal-Esa 2012). Such joint R&D, investment, production, and sales are motivated both strategically and economically. Economically, unit costs can be reduced through co-production, hedging R&D spending, and increasing exports. But in such cooperation, there is also competition. This competition is largely centered on technology sharing. Such cooperation is very asymmetric and the US holds the source code. Thus, the US’s allies try their best to gain some control. The UK once demanded special treatment on the grounds of its special relationship with the US and even threatened to withdraw from the project to obtain source code and other key technologies (Vucetic 2013). In a sense, through the F-35 program, the US and its allies have formed an “F-35 community.” Such a community greatly improves the common training and combat capability of its members. At present, Australia, the UK, and Japan are the largest buyers of the US’s F-35, with orders of 100, 138, and 147, respectively (Hoehn 2022). NATO is to have four hundred F-35s by 2030 (Lockheed Martin n.d.). F-35 fighter jets of all countries can take off and land on the aircraft carriers of other allies (Shelbourne 2021). In the ongoing Russian–Ukrainian war, we clearly see the importance of interoperability in weaponry and training in warfare. To rapidly facilitate Ukraine’s combat power, the US and NATO countries have provided weapon systems to former Eastern Bloc countries, replacing their former Soviet weapon systems in order to provide them to Ukraine (Ismay and Schmitt 2022). At the global political and strategic level, the “weapon system community” represented by the F-35 program also has flagship significance. Energizing global deterrence through interoperability and synergy among global allies was one of the original motivations for the project (Lockheed Martin n.d.). At the same time, because of the high cost, the opportunity cost is also high. Once committed, it is very difficult for politicians to withdraw from the program. This strengthens the alliance and makes it more durable. In addition, due to the substantial increase in the opportunities for and depth of interaction among all parties, the relevant departments within each country—from scientific research to technology departments and from industry to personnel departments, and even departments that provide support and security in related areas—will deepen their mutual familiarity and sense of intimacy. Cohesion among allies is strengthening from government to society. Such psychological and cultural familiarity and connection also has positive implications for the overall power of alliances in international relations. Despite occasional tense relations among participants during the development process, these “weapon systems communities” ultimately tighten interdependence among the US and its allies. Taking a high-tech weapon system such as the F-35 as an example, the key to improving combat effectiveness lies in the continuous upgrade of software systems. However, the US firmly keeps these key technologies in its own hands, leading to dissatisfaction among allies (Vucetic 2013). To assuage its allies’ dissatisfaction with US control of core technology, the US has agreed to waive export licenses from Australia and the UK on some weapons projects. In 2007, the US signed Defense Trade Cooperation Treaties with Australia and the UK, allowing the two countries to provide certain weapons and equipment for several activities, including “agreed combined military or counterterrorism operations; cooperative security and defense research, development, production, and support programs; security and defense projects where the Government of Australia is the end user; and for US government end-use” (Australian Department of Defence n.d.). The US alliance system is developing from a “weapon system community” to a “military science, and industry community” There is a mutual push within the US alliance for cooperation on a deeper level. In 2017, the US Congress expanded its National Technology and Industry Base (NTIB) Act to include Australia and Japan as bases, after Canada in 1994. The US’s goal is to establish some kind of “defense free trade area” to achieve further integration among member states in R&D, production, and operations. AUKUS is a new advancement from the established “weapon system community” and the “defense free trade zone.” It is an integrated military, technology, industry, experiment, training, and operating community. The 2021 statement of the annual Australia–US 2 + 2 ministerial talks listed cooperation on an extremely wide range of issues: Australia’s procurement of nuclear-powered submarines; enhanced force posture cooperation and alliance integration; strategic capabilities cooperation; and industry, technology, innovation, and other security issues. Australia will receive all types of US Air Force fighter aircraft, which will be deployed on its land; increase integrated cooperation between the US army and the Australian army; and improve its capabilities to maintain, repair, overhaul, and upgrade US military platforms and components on Australian land. The two sides also signed a secret agreement to improve their integration in science and technology as well as in their strategic capabilities and defense industrial base, strengthen Australia’s industrial capabilities. The US allows Australia to assume the role of a US industrial base, and ensure that Australia can sustain the Indo-Pacific defense supply chain (Australian Department of Foreign Affairs and Trade 2021). Obviously, in addition to the international R&D cooperation between Australia, the UK, and the US on high-tech weapon systems, Australia has increasingly become a key ally for the US to extend its capabilities, providing not only cooperation in the usual sense, but also bases, industrial capabilities, and key supplies. On April 5, 2022, the three countries of Australia, the UK, and the US issued another statement announcing their intent to cooperate on the development of hypersonic, anti-hypersonic, and electronic warfare capabilities and to expand cooperation in information sharing and defense innovation. This is a new collaboration that follows cooperation on cyberspace capabilities, artificial intelligence, quantum technology, and deep-sea technologies. In the statement, the three countries also state that they will seek to engage with other allies and partners (White House 2022a). In addition to cooperation on the nuclear-powered submarine project and hypersonic weapons, there is another area of cooperation between Australia, the UK, and the US that deserves more attention: cyberspace cooperation. In the information age, allies use cloud collaboration to carry out encrypted communication, rapid information transfer, and precise targeting to facilitate alliance operations. But in cyberspace, alliance capabilities may be determined not by the strongest members, but by the weakest. This is different from the land, sea, and air domains. Node states on cloud platforms are critical. The US attaches great importance to improving the “capacity package”—i.e., the ability to integrate weapons systems, components, logistics, training, professional military education, and exercises—of its allies in this area. These capabilities reside in computer projects, databases, and algorithms (Hooper 2021). The US is exploring how to share or co-build such capabilities with Australia. In March 2022, Admiral Aquilino, the commander of US Indo-Pacific Command, visited the Darwin and Amberley bases in Australia to implement the agenda of the 2021 2 + 2 joint statement. The two sides have agreed to cooperate and share technologies in all domains, including cyber and space, and to deepen military alliances, training, and deployment to advance collective security and integrated deterrence in the Indo-Pacific. The Australian Department of Defense press release also revealed that interoperability exercises and training between Australia and the US include complex US B-2 Spirit strategic bombers operating in concert with the Australian Air Force (Australian Department of Defence 2022). The construction of this military science, and industry community is a way for the US to neutralize China’s influence and rebuild its defense industry chain In September 2018, amid the US-initiated trade war against China, then US President Donald Trump signed Executive Order 13806 to establish an interagency task force to assess and suggest improvements for the US’s manufacturing, defense industrial base and supply chain resilience (Office of the Under Secretary of Defense for Acquisition and Sustainment and Office of the Deputy Assistant Secretary of Defense for Industrial Policy 2018). Because the US has listed China as a strategic competitor (US Department of Defense 2018), it has begun to pay more attention to the security of its industrial chain and tried to remove China from the US defense supply chain. On February 24, 2022, in accordance with US Presidential Executive Order 14017, the US Department of Defense released a roadmap for addressing vulnerabilities in the US’s defense industrial base supply chain. In this document, titled Securing Defense-Critical Supply Chains, the Department of Defense incorporates opinions from the National Security Council and the National Economic Council and emphasizes that defense supply chain resilience should first focus on key areas critical to readiness. The document lays out China’s position in the US defense industrial chain. It holds that in the energy storage technology and battery supply chain industry, which is the second most important industry to US security, the US’s dependence on China poses a threat to US national security. In order to exclude China from the US defense supply chain, DoD makes the following recommendations: first, build domestic production capacity; second, engage with partners and allies to improve supply chain resilience; and third, mitigate foreign ownership, control, or influence (FOCI) and safeguard markets. To improve supply chain resilience, DoD and its partners, “through Reciprocal Defense Procurement Agreements, bi-lateral engagements, and continued dialogue…have identified opportunities to collaborate and share information” and over the next year will pursue “strengthening channels of information sharing through bilateral agreements and establishing working groups to pursue joint actions.” In addition, the Department will conduct a front-end assessment of defense programs and mitigate FOCI concerns through early identification (US Office of the Deputy Secretary of Defense 2022). Australia, the UK, and the US have moved toward the establishment of a military, scientific, and industrial community, a defense supply chain that excludes China, and an intragroup supply chain. This has made the US alliance system more closed and its relationship with China more hostile. Further, the alliance system has transcended wartime cooperation and mutual assistance and sought close interdependence in peacetime, turning international relations into a confrontation between blocks and casting a shadow of global division. From “extended nuclear deterrence” to “integrated deterrence” Providing extended nuclear protection for some allies has always been an important pillar of the US alliance system. But with the US’s expectations for its allies increasing, the US is planning to implement a new strategy of US ally “integrated deterrence.” The new strategy no longer emphasizes the traditional one-way relationship in which the US provides protection to allies. Rather, it requires all allies, strategic assets, and postures to be organized around the central goal of “deterring opponents from acting rashly.” This is the idea of “total deterrence.” AUKUS is part of the US’s new extended strategic deterrence. It is “new” because nuclear-powered submarines do not have nuclear strike capability, as they are not equipped with nuclear weapons. However, these submarines do have strategic projection capability. Their nuclear power allows them to quietly travel long distances and approach the enemy’s coast to deliver strikes, and thus contributes to strategic deterrence. This is a kind of non-nuclear but nuclear-powered strategic deterrence. Its crossover nature to some extent has blurred the distinction between nuclear deterrence and conventional deterrence, which is exactly in line with the current strategic thinking of the US, which integrates nuclear deterrence, conventional deterrence, allies, and global capabilities. The significance of AUKUS is that it allows the US to extend a form of nuclear deterrence protection to Australia. As a result, the extended strategic deterrence protection of the US will extend from Northeast Asia to the southern Pacific Ocean. Unlike the US’s assurances to Japan and NATO, Washington has never explicitly and publicly offered Australia “extended nuclear deterrence.” Before the AUKUS deal, the US’s security guarantee to Australia was only that the two would assist each other in times of crisis. In the Australia, New Zealand and United States Security Treaty, or ANZUS Treaty, which was signed in 1951, Article IV states that “each Party recognizes that an armed attack in the Pacific Area on any of the Parties would be dangerous to its own peace and safety and declares that it would act to meet the common danger in accordance with its constitutional processes.” In Australia’s Strategic Policy, which was issued in 1997, Australia interpreted its alliance with the US as an obligation to come to one another’s aid in a crisis—an obligation to “act to meet the common danger,” as Article IV of the ANZUS Treaty puts it. “These undertakings do not amount to a guarantee by the US of Australia’s security. Indeed, the Treaty specifically requires each party to attend to its own capabilities. Nor does it amount to a promise to send armed forces in a crisis. But it provides a sound basis for us to plan on the expectation of substantial and vital non-combat support from the United States in a crisis.” Moreover, “we do not assume that such help would be provided. Indeed, such an assumption would be inconsistent with our self-reliant posture and our alliance obligations” (Australian Department of Defence 1997). The ANZUS Treaty does not contain an obligation to provide extended nuclear deterrence, and there is currently no public information that can prove the US in fact provides extended nuclear protection to Australia (Hood and Cormier 2021). But Australia has long had such expectations, and its defense policy has made clear its reliance on US extended nuclear deterrence. According to Australia’s Strategic Policy, “In one specific respect the alliance does provide a clearer expectation of US support—that is, defence against nuclear attack. While the risk of nuclear attack on Australia remains very low, the possibility cannot entirely be ruled out. In those circumstances we would rely on the extended deterrence provided by the United States to deter such an attack” (Australian Department of Defence 1997). Australia: Defence 2000 reiterates that the Australia–US alliance should not be a relationship of dependency, but rather one of mutual help. But there is one important exception to this principle of self-reliance: Australia relies on the extended deterrence provided by US nuclear forces to deter the remote possibility of any nuclear attack on Australia (Commonwealth of Australia 2000, 36). However, even though the US said it will continue to strengthen extended deterrence with South Korea and Japan in the Indo-Pacific Strategy of the United States, which was issued in February 2022, it still did not mention Australia (White House 2022b). Australia’s demand for extended nuclear protection from the US has never been openly satisfied. In Australia’s past judgments, the probability of a nuclear attack on it was low, so Canberra was not concerned too much about the prospect of not having the US’s nuclear protection. But with China’s rise in recent years, Australia believes the strategic environment of the region which it is located in has significantly changed: there is no mature arms control regime in the Indo-Pacific, and in a period of rapid technological development, the boundary between conventional and nuclear deterrence has become blurred. The Australian strategic circle has explored three possible ways to acquire credible nuclear deterrence: first, the deployment of US nuclear weapons on Australian territory; second, the agreement of a nuclear sharing arrangement with the UK; and third, the development of its own nuclear capabilities (Ogilvie-White 2020). The AUKUS deal tactfully combines US strategic considerations with Australia’s strategic demands. The US does not need to directly deploy or transfer nuclear weapons to Australia, which would constitute evading its obligations under the nuclear non-proliferation regime. At the same time, by signing a nuclear power technology cooperation agreement, Australia’s long-range strike capability has been greatly enhanced. Nuclear-powered submarines can travel much longer, quieter, and faster than conventional ones. They are more difficult to detect, have stronger firepower, are stealthier, and have a larger payload capacity, which provides Australia with non-nuclear-weapon strategic deterrence. Australia’s acquisition of strategic deterrence capability will also help the US, Britain, and Australia redistribute their work at sea. From another perspective, Australia will possess nuclear-powered ships but not nuclear weapons, leading to a new situation: while the US is transferring extended strategic deterrence capability to Australia, the US cannot call it so in the traditional sense. With this new extended strategic deterrence capability, the US and Australia are moving toward an “integrated deterrence” strategy that combines nuclear and conventional deterrence. On April 30, 2021, at a US Indo-Pacific Command change of command ceremony, US Secretary of Defense Lloyd Austin said in his speech that “the cornerstone of America’s defense is still deterrence.” Lloyd also pointed out that the US would seek to utilize new and existing capabilities, hand in hand with US allies, to extend deterrence across multiple realms (Austin 2021). In June 2021, when the AUKUS deal was still being negotiated, the Under Secretary of Defense for Policy Colin Kahl gave an explanation for “integrated deterrence,” saying that integrated deterrence is a cornerstone of the US’s new national defense strategy. He said, “deterrence is at its heart an active dissuasion,” adding, “The classic in the nuclear domain is deterrence by punishment… But you also need to think carefully about deterrence by denial. That is, how can you deny the benefits of aggression through some mix of defense and resilience, or casting doubt on the effectiveness of the other side’s attacks” (Kahl 2021). According to Kahl (2021),…there’s also what some people call defense by entanglement, which is creating a normative and rule-based order around which the benefits of maintaining the status quo are elevated, and if actors step out and commit aggression, they can be assured that they will meet an international community which imposes diplomatic costs, economic costs, and military costs on them. So when the Secretary of Defense talks about integrated deterrence, he’s talking about deterrence integrated across a number of different categories. Integrated across domains, so deterrence that is integrated across nuclear, conventional, space, cyber, informational. Deterrence across the spectrum of conflict. So everything from high-end nuclear and conventional conflict scenarios on one end, to hybrid and gray zone competition on the other end. He means integrated across the instruments of national power. Since many of the things the US needs to be doing to deter don’t necessarily fall in the military domain, it may be elements of US diplomacy or economic state craft or intelligence and information operations. And then lastly, integrated across US’ allies and partners. Kahl also said that the Biden administration is “inclined to look for a way to reduce the role of nuclear weapons in its national security strategy…because there are a huge number of threats that nuclear weapons do not address.” Compared to its predecessor, the Biden administration has paid special attention to allies in Europe and Asia, and it wants to ensure that its deterrence-related commitments to them are seen as credible (Kahl 2021). Several months later, Kahl stated again that the whole alliance system is crucial to integrated deterrence. “We have to work alongside our allies and partners so that our adversaries know that they’re not just taking on the United States, they’re taking on a coalition of countries who are committed to upholding a rules-based international order,” he said (US Department of Defense 2021a). On September 16, 2021, the day after the announcement of the AUKUS deal, Austin talked about the concept of “integrated deterrence” again. He said it refers to “the ability for the United States military to work more effectively with our allies and partners” (US Department of Defense 2021b). In February 2022, the US released the Indo-Pacific Strategy of the United States. The strategy document states that the US will continue its military presence and deterrence in the Indo-Pacific region with an “integrated deterrence” approach: the US “will more tightly integrate our efforts across warfighting domains and the spectrum of conflict to ensure that the United States, alongside our allies and partners, can dissuade or defeat aggression in any form or domain.” Deepening joint strategic, diplomatic, and military planning between the US and Australia is critical to a collective deterrence strategy. The alliance should make plans for high-end deterrence scenarios and expand Australia’s ability to operate forward in the Indo-Pacific (White House 2022b). It also states that the US network of security alliances and partnerships is the greatest asymmetric strength of the US. On March 8, 2022, the US Department of Defense submitted a classified version of the new National Defense Strategy to Congress, with the declassified version expected to soon follow. At the core of the US’s national defense strategy will be “integrated deterrence,” which Deputy Under Secretary of Defense for Policy Sasha Baker says “is a framework for working across warfighting domains, theaters and the spectrum of conflict, in collaboration with all instruments of national power, as well as with US allies and our partners” (US Department of Defense 2022a). For the first time, the Department conducted its strategic reviews in a fully integrated way, incorporating the Nuclear Posture Review and Missile Defense Review into the National Defense Strategy. The new National Defense Strategy lists four priorities: “defending the homeland, paced to the growing multi-domain threat posed by the PRC; deterring strategic attacks against the United States, Allies, and partners; deterring aggression, while being prepared to prevail in conflict when necessary, prioritizing the PRC challenge in the Indo-Pacific, then the Russia challenge in Europe; [and] building a resilient Joint Force and defense ecosystem.” The National Defense Strategy says the Department of Defense “will incorporate ally and partner perspectives, competencies, and advantages at every stage of defense planning” and advance the US’s goals through integrated deterrence. According to the Department, “Integrated deterrence entails developing and combining US’s strengths to maximum effect, by working seamlessly across warfighting domains, theaters, the spectrum of conflict, other instruments of US national power, and its unmatched network of Alliances and partnerships. Integrated deterrence is enabled by combat-credible forces, backstopped by a safe, secure, and effective nuclear deterrent” (US Department of Defense 2022b). Judging from various statements of the US, “integrated deterrence” contains everything. In terms of guiding ideology, “integrated deterrence” is like some kind of philosophy that clearly delineates a camp in which all countries, assets, and capabilities exist in a state of total deterrence. It sounds like a “total cold war.” The key feature of “integrated deterrence” is its network of assets, capabilities, and allies. As far as AUKUS itself is concerned, it has all military, diplomatic, scientific, as well as economic elements of total deterrence. Australia’s nuclear-powered submarine fleet can be seen as an experiment of “integrated deterrence.” Conclusion The AUKUS deal is the first agreement in history to allow a non-nuclear-weapon state to have nuclear-powered submarines but not nuclear weapons. It will push the Australia–UK–US alliance to the level of a closer military, scientific and industrial community, lay out the beginning of a new grand alliance on the ocean, and change both the strategic deterrence layout in the Indo-Pacific region and some features of the US alliance system’s deterrence strategy against China. China should pay great attention to the political will, strategic layout, and deterrence concept displayed by this major development. At the global strategic level, the Russia–Ukraine conflict that started in February 2022 and the Taiwan Strait crisis in August 2022 pushed the international landscape in one direction: toward the convergence and interaction of the US’s Atlantic and Indo-Pacific strategies. A new NATO strategic document for the first time listed “intensifying cooperation between China and Russia” as a focus (Heads of State and Government at the NATO Summit in Madrid 2022). This is a wake-up call. At present, the new global strategic structure of the US alliance system has taken shape. A “maritime grand alliance” now targets Russia and China. Its strategic goal is to “offshore balance” in Europe and maintain “integrated deterrence” against China in the Asia-Pacific. Its long-term objective is to stabilize Europe and engage in a stable competition with China. Funding Funding was provided by National Social Science Fund of China, 20VHQ003. Declarations Conflict of interest The author declares that there is no conflict of interest regarding the publication of this paper. ==== Refs References ABC News. 2021. In full: PM Scott Morrison speaks after AUKUS submarine announcement. https://www.youtube.com/watch?v=o8Gknmk6FDo. Accessed 21 Aug 2021. Austin, Lloyd J. III. 2021. Secretary of Defense remarks for the US INDOPACOM change of Command ceremony. US Department of Defense. https://www.defense.gov/News/Speeches/Speech/Article/2592093/secretary-of-defense-remarks-for-the-us-indopacom-change-of-command/. Accessed 21 Aug 2021. Australian Department of Defence. 1997. Australia’s strategic policy. Canberra. http://repository.jeffmalone.org/files/defence/SR97.pdf. Accessed 21 Aug 2021. Australian Department of Defence. n.d. Australia–US defence trade cooperation treaty. https://www.defence.gov.au/business-industry/export/controls/us-trade-treaty. Accessed 5 Apr 2022. Australian Department of Defence. 2022. Visit to Australia by Commander Indo-Pacific Command, Admiral John C Aquilino. https://news.defence.gov.au/media/media-releases/visit-australia-commander-indo-pacific-command-admiral-john-c-aquilino. Accessed 21 Aug 2021. Australian Department of Foreign Affairs and Trade. 2021. Joint statement Australia–US ministerial consultations (AUSMIN) 2021. https://www.dfat.gov.au/geo/united-states-of-america/ausmin/joint-statement-australia-us-ministerial-consultations-ausmin-2021. Accessed 21 Aug 2021. Babones, Salvatore. 2022. Understanding the “forever” alliance: What AUKUS means for Australia and the world. Valdai Discussion Club. https://valdaiclub.com/a/highlights/understanding-the-forever-alliance-what-aukus/. Accessed 21 Aug 2021. Binnie, Jeremy. 2020. UK to expand base in Oman. Janes. https://www.janes.com/defence-news/news-detail/uk-to-expand-base-in-oman. Accessed 21 Aug 2021. Blinken, Antony J. 2022. The administration’s approach to the People’s Republic of China. https://www.state.gov/the-administrations-approach-to-the-peoples-republic-of-china/. Accessed 21 Aug 2021. British Cabinet Office. 2021a. Global Britain in a competitive age: The integrated review of security, defence, development and foreign policy. https://www.gov.uk/government/publications/global-britain-in-a-competitive-age-the-integrated-review-of-security-defence-development-and-foreign-policy. Accessed 21 Aug 2021. British Cabinet Office. 2021b. Global Britain in a competitive age: The integrated review of security, defence, development and foreign policy. https://www.gov.uk/government/publications/global-britain-in-a-competitive-age-the-integrated-review-of-security-defence-development-and-foreign-policy/global-britain-in-a-competitive-age-the-integrated-review-of-security-defence-development-and-foreign-policy. Accessed 21 Aug 2021. British Prime Minister’s Office. 2021. UK, US and Australia launch new security partnership. https://www.gov.uk/government/news/uk-us-and-australia-launch-new-security-partnership. Accessed 21 Aug 2021. Cabinet Office and Sir Stephen Lovegrove. 2021. Sir Stephen Lovegrove speech at the Council on Geostrategy. https://www.gov.uk/government/speeches/sir-stephen-lovegrove-speech-at-the-council-on-geostrategy. Accessed 21 Aug 2021. Cameron-Chileshe, Jasmine. 2022. Liz Truss to warn of “misery across Europe” if Russia wins in Ukraine. Financial Times. https://www.ft.com/content/8fb923ad-a0f5-4fe8-ac3c-b76e0710402f. Accessed 21 Aug 2021. Commonwealth of Australia. 2000. Defence 2000: Our future defence force. Canberra. https://www.files.ethz.ch/isn/155594/Australia-2000.pdf. Accessed 21 Aug 2021. Courtney, Joe. 2022. Reps. Courtney, Gallagher, Kilmer, and Moore announce formation of the new bipartisan AUKUS working group, in support of the undersea alliance between the US, UK, and Australia. https://courtney.house.gov/media-center/press-releases/reps-courtney-gallagher-kilmer-and-moore-announce-formation-new. Accessed 21 Aug 2021. Department of the Navy, United States Marine Corps, and United States Coast Guard. 2020. Advantage at sea: Prevailing with integrated all-domain naval power. Washington, DC. https://media.defense.gov/2020/Dec/16/2002553074/-1/-1/0/TRISERVICESTRATEGY.PDF. Accessed 21 Aug 2021. Dibb, Paul. 2020. How Australia can deter China. The Strategist. Australia Strategic Policy Institute. https://www.aspistrategist.org.au/how-australia-can-deter-china/. Accessed 21 Aug 2021. Economist. 2022. “We might see a very different sort of Germany from the one that we become very familiar with”—a new posture. https://www.economist.com/podcasts/2022/03/18/we-might-see-a-very-different-sort-of-germany-from-the-one-that-we-become-very-familiar-with-a-new-posture. Accessed 21 Aug 2021. Ewing, Philip. 2020. UK Prime Minister Boris Johnson pledges $32b military build-up. USNI News. https://news.usni.org/2020/11/19/u-k-prime-minister-boris-johnson-pledges-32b-military-build-up. Accessed 21 Aug 2021. Galloway, Anthony, and Eryk Bagshaw. 2020. Australia–US looking to ramp up joint military exercises in South China Sea. Sydney Morning Herald. https://www.smh.com.au/politics/federal/australia-us-looking-to-ramp-up-joint-military-exercises-in-south-china-sea-20200727-p55fvn.html. Accessed 21 Aug 2021. Gribbin, Caitlyn. 2017. Malcolm Turnbull declares he will “stand up” for Australia in response to China’s criticism. ABC Net News. https://www.abc.net.au/news/2017-12-09/malcolm-turnbull-says-he-will-stand-up-for-australia/9243274. Accessed 21 Aug 2021. Häggblom, Robin. 2022. Finnish-Estonian cooperation turning Baltic Sea into a NATO-lake. Naval News. https://www.navalnews.com/author/robin/. Accessed 21 Aug 2021. Hartley Keith Defence industrial policy in a military alliance Journal of Peace Research 2006 43 4 473 489 10.1177/0022343306064976 Heads of State and Government at the NATO Summit in Madrid. 2022. NATO 2022 strategic concept. https://www.nato.int/nato_static_fl2014/assets/pdf/2022/6/pdf/290622-strategic-concept.pdf. Accessed 21 Aug 2021. Hoehn, John R. 2022. F-35 joint strike fighter (JSF) Program. Congressional Research Service: RL30563. US Library of Congress. https://sgp.fas.org/crs/weapons/RL30563.pdf. Accessed 21 Aug 2021. Hood, Anna, and Monique Cormier. 2021. The role of ANZUS in Australia’s reliance on US extended nuclear deterrence. Australian Outlook. Australian Institute of International Affairs. https://www.internationalaffairs.org.au/australianoutlook/the-role-of-anzus-in-australias-reliance-on-us-extended-nuclear-deterrence/. Accessed 21 Aug 2021. Hooper, Charles W. 2021. US national defense strategy and the future of foreign military sales. Atlantic Council. https://www.atlanticcouncil.org/blogs/new-atlanticist/us-national-defense-strategy-and-the-future-of-foreign-military-sales/. Accessed 21 Aug 2021. House of Lords Library. 2022. AUKUS agreement to exchange naval nuclear propulsion information. https://lordslibrary.parliament.uk/aukus-agreement-to-exchange-naval-nuclear-propulsion-information/. Accessed 21 Aug 2021. Ismay, John, and Eric Schmitt. 2022. Seeking arms for Ukraine, Pentagon buyers scour eastern European factories. New York Times. https://www.nytimes.com/2022/04/25/us/ukraine-pentagon-arms-buyers.html. Accessed 21 Aug 2021. Jennings, Peter. 2019. America’s new Asia strategy opens doors for Australia. The Strategist. Australia Strategic Policy Institute. https://www.aspistrategist.org.au/americas-new-asia-strategy-opens-doors-for-australia/. Accessed 21 Aug 2021. Kahl, Colin. 2021. Keynote address: Colin Kahl, 2021 Carnegie international nuclear policy conference. Carnegie Endowment for International Peace. https://ceipfiles.s3.amazonaws.com/pdf/Colin+Kahl+Keynote_Transcript.pdf. Accessed 21 Aug 2021. Lockheed Martin. n.d. Essential to allied airpower: F-35 Lightning II. https://www.f35.com/f35/global-enterprise.html. Accessed 21 Aug 2021. Ministry of Defence and The Rt Hon Ben Wallace MP. 2020. Defence Secretary announces investment in strategic Omani port. https://www.gov.uk/government/news/defence-secretary-announces-investment-in-strategic-omani-port. Accessed 21 Aug 2021. Ministry of Defence and The Rt Hon Ben Wallace MP. 2021. Record size and scope of carrier strike group deployment announced. https://www.gov.uk/government/news/record-size-and-scope-of-carrier-strike-group-deployment-announced. Accessed 21 Aug 2021. Ministry of Foreign Affairs of Japan. 2022. Japan–Australia Reciprocal Access Agreement. https://www.mofa.go.jp/a_o/ocn/au/page4e_001195.html. Accessed 21 Aug 2021. Morrison, Scott. 2020. Launch of the 2020 defence strategic update—address transcript. Australian Journal of Defence and Strategic Studies 2(2): 173–188. https://defence.gov.au/ADC/Publications/AJDSS/documents/volume2-number2/launch-2020-strategic-update.pdf. Accessed 21 Aug 2021. NATO. 2022. Madrid Summit ends with far-reaching decisions to transform NATO. https://www.nato.int/cps/en/natohq/news_197574.htm?selectedLocale=en. Accessed 21 Aug 2021. Office of the Under Secretary of Defense for Acquisition and Sustainment and Office of the Deputy Assistant Secretary of Defense for Industrial Policy. 2018. Assessing and strengthening the manufacturing and defense industrial base and supply chain resiliency of the United States: Report to President Donald J. Trump by the interagency task force in fulfillment of executive order 13806. Washington, DC. https://s3.amazonaws.com/static.militarytimes.com/assets/eo-13806-report-final.pdf. Accessed 21 Aug 2021. Ogilvie-White, Tanya. 2020. Australia and extended nuclear assurance. Perspectives on nuclear deterrence in the 21st century. Chatham House. https://www.chathamhouse.org/2020/04/perspectives-nuclear-deterrence-21st-century-0/australia-and-extended-nuclear-assurance. Accessed 21 Aug 2021. Patalano Alessio The Indo-Pacifc “tilt” and the return of British maritime strategy 2021 London King’s College London Perlez, Jane. 2016. US casts wary eye on Australian port leased by Chinese. New York Times. https://www.nytimes.com/2016/03/21/world/australia/china-darwin-port-landbridge.html. Accessed 21 Aug 2021. Prime Minister of Australia. 2021. Joint Leaders’ Statement on AUKUS. https://www.pm.gov.au/media/joint-leaders-statement-aukus. Accessed 21 Aug 2021. Reuters. 2022. Finland and Sweden could join NATO quickly, Stoltenberg says. https://www.reuters.com/world/europe/finland-sweden-could-join-nato-quickly-stoltenberg-says-2022-04-28/. Accessed 21 Aug 2021. Schreer, Ben. 2015. Australia doesn’t have to choose between the US and China. World Politics Review. https://www.worldpoliticsreview.com/australia-doesn-t-have-to-choose-between-the-u-s-and-china/. Accessed 21 Aug 2021. Shalal-Esa, Andrea. 2012. Exclusive: U.S. sees lifetime cost of F-35 fighter at $1.45 trillion. Reuters. https://www.reuters.com/article/us-lockheed-fighter-idUSBRE82S03L20120329. Accessed 21 Aug 2021. Shelbourne, Mallory. 2021. U.K. welcomes chance to put British F-35Bs on American warships. USNI News, July 14, 2021, https://news.usni.org/2021/07/14/u-k-welcomes-chance-to-put-british-f-35bs-on-american-warships. Accessed 21 Aug 2021. Shi, Xiaoqin[师小芹]. 2012. On seapower and the US–China relationship (论海权与中美关系). Beijing: Junshi kexue chubanshe. Sky News. 2021. Deployment of HMS Queen Elizabeth and carrier strike group will “fly flag for global Britain”. https://news.sky.com/story/deployment-of-hms-queen-elizabeth-and-carrier-strike-group-will-fly-flag-for-global-britain-12287281. Accessed 21 Aug 2021. Thomson, Mark. 2013. We don’t have to choose between the US and China. The Strategist. Australia Strategic Policy Institute. https://www.aspistrategist.org.au/we-dont-have-to-choose-between-the-us-and-china/. Accessed 21 Aug 2021. US Department of Defense. n.d. US relations with Australia. https://www.state.gov/u-s-relations-with-australia/. Accessed 21 Aug 2021. US Department of Defense. 2018. Summary of the National Defense Strategy: Sharpening the American military’s competitive edge. Washington, DC. https://dod.defense.gov/Portals/1/Documents/pubs/2018-National-Defense-Strategy-Summary.pdf. Accessed 21 Aug 2021. US Department of Defense. 2021a. Concept of integrated deterrence will be key to national defense strategy, DOD official say. https://www.defense.gov/News/News-Stories/Article/Article/2866963/concept-of-integrated-deterrence-will-be-key-to-national-defense-strategy-dod-o/. Accessed 21 Aug 2021. U.S. Department of Defense. 2021b. Australia, US alliance is stronger, deeper than ever, officials say. https://www.defense.gov/News/News-Stories/Article/Article/2777798/australia-us-alliance-is-stronger-deeper-than-ever-officials-say/. Accessed 21 Aug 2021. US Department of Defense. 2022a. Integrated deterrence at center of upcoming national defense strategy. https://www.defense.gov/News/News-Stories/Article/Article/2954945/integrated-deterrence-at-center-of-upcoming-national-defense-strategy/. Accessed 21 Aug 2021. US Department of Defense. 2022b. Fact sheet: 2022 national defense strategy. https://media.defense.gov/2022b/Mar/28/2002964702/-1/-1/1/NDS-FACT-SHEET.PDF. Accessed 21 Aug 2021. US Office of the Deputy Secretary of Defense. 2022. Securing defense-critical supply chains: an action plan developed in response to President Biden's executive order 14017. Washington, DC. https://media.defense.gov/2022/Feb/24/2002944158/-1/-1/1/DOD-EO-14017-REPORT-SECURING-DEFENSE-CRITICAL-SUPPLY-CHAINS.PDF. Accessed 21 Aug 2021. Vucetic, Srdjan. 2013. Before the cut: the global politics of the F-35 joint strike fighter. Paper presented at the CIPSS/CEPSI Workshop on International Cooperation, McGill University. https://ciaotest.cc.columbia.edu/wps/ceri/0029416/f_0029416_23866.pdf. Accessed 21 Aug 2021. White House. 2021a. Joint leaders statement on AUKUS. https://www.whitehouse.gov/briefing-room/statements-releases/2021a/09/15/joint-leaders-statement-on-aukus/. Accessed 21 Aug 2021. White House. 2021b. Remarks by President Biden, Prime Minister Morrison of Australia, and Prime Minister Johnson of the United Kingdom announcing the creation of AUKUS. https://www.whitehouse.gov/briefing-room/speeches-remarks/2021b/09/15/remarks-by-president-biden-prime-minister-morrison-of-australia-and-prime-minister-johnson-of-the-united-kingdom-announcing-the-creation-of-aukus/. Accessed 21 Aug 2021. White House. 2022a. AUKUS leaders’ level statement. https://www.whitehouse.gov/briefing-room/statements-releases/2022a/04/05/aukus-leaders-level-statement/. Accessed 21 Aug 2021. White House. 2022b. Indo-Pacific Strategy of the United States. https://www.whitehouse.gov/wp-content/uploads/2022b/02/U.S.-Indo-Pacific-Strategy.pdf. Accessed 21 Aug 2021. Wilkins, Thomas. 2022. Another piece in the jigsaw: Australia and Japan sign long-awaited Reciprocal Access Agreement. Australian Outlook. Australian Institute of International Affairs. https://www.internationalaffairs.org.au/australianoutlook/another-piece-jigsaw-australia-japan-sign-long-awaited-reciprocal-access-agreement/. Accessed 21 Aug 2021.	0	PMC9734547	NO-CC CODE	2022-12-14 23:28:29	no	China Int Strategy Rev.. 2022 Dec 4;:1-20	utf-8	null	null	null	oa_other
==== Front SN Bus Econ SN Bus Econ Sn Business & Economics 2662-9399 Springer International Publishing Cham 331 10.1007/s43546-022-00331-1 Original Article An income equalisation policy and social welfare: beyond inequality http://orcid.org/0000-0002-4511-7806 Teramoto Hiroaki [email protected] grid.443705.1 0000 0001 0741 057X Master of Economics, Hiroshima Shudo University, Hiroshima Shudo Daigaku, Hiroshima, Japan 7 12 2022 2023 3 1 413 6 2021 31 8 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 inequality of income distribution is one of the most serious problems to our society. Traditional economics proposes the tax increase on wealthy classes and transfers its revenue to low income group. This is the most appropriate one. But, the total of this and donation will unfortunately be insufficient. For procuring funds to ease inequality, I propose to erase the government bonds which are already purchased and owned by the central bank. The government, then, issues new government bonds (the total sum of which should be less than erased one) to expend for equalization. The amount of new government bonds is determined from the maximisation of social welfare function concerning to income distribution. By this policy, first, the socially optimal equalization will be attained; second, through the productivity increase caused by the expenditure for low income group, the national economy will grow; third, because of the erasure process, the government bonds outstanding will not increase. In these days of pandemic, much more government expenditures are needed and thereby government deficits are piled up. I think the valid policy is not the heavy tax increase, but the execution of erasure of government bonds outstanding and to equalize income distribution. Keywords Inequality of income distribution Maximisation of social welfare Erasure of government bonds Government expenditure for equalization Pandemic Market Economy JEL Classification D31 E58 H53 issue-copyright-statement© Springer Nature Switzerland AG 2023 ==== Body pmcIntroduction In the history of humankind, political and economic inequalities have always coexisted. At the end of the eighteenth century, the citizen’s revolution alleviated political inequality, which gradually spread across the globe. After the end of absolutism and advent of the industrial revolution, people expected economic equality and the prosperity of ordinary workers, and considerable effort was spent to achieve it. However, economic inequality still exists. Economic society can resolve the problem of inequality of income and properties. Economics is an effective discipline having a wide range. The study on the allocation and distribution of scarce resources is an important field. The studies on resource allocation have obtained excellent results, but the studies on distribution have not achieved the desired results despite the strenuous efforts of economists. To rectify income and property inequality, economics mainly proposes tax increases or the increase of the rate of progressive tax. This is the most appropriate and authentic measure that has successfully reduced inequality. However, inequality still persists and the policy of tax increase cannot solely resolve the ongoing inequality. As the tax increase policy faces the strong political and economic oppositions and, moreover, has an upper limit, the inequality problem will exist semi-permanently. In this study, a new policy is proposed that can resolve the problem of economic inequality. This policy suggests discontinuing the existing government bonds with the central bank, and then issuing new government bonds to expend for income equalisation. The amount of new issue will be less than the discontinued sum and will be determined from the standpoint of social welfare maximisation. The process of erasure and expense is devised to repeat at a declining rate. By this policy, the income distribution approaches the socially optimal point. The expenditure for income equalisation will increase people’s income which in turn will increase the tax revenue, thereby reducing the government deficit. This study proposes an alternative to the existing policy of progressive taxation for reducing income inequality. I believe that this study makes contribution to the literature because it discusses in detail the erasure policy that can accomplish absolute income equality which the traditional tax policy has not realized. As this erasure policy faces the problems of inflation and lax financial policy, close attention and deliberate countermeasures are indispensable. The causes of present inequality The inequality pressure from outside The prevalence of the market economy in most parts of the world led to the acceptance of direct investment and the newest technology by developing countries, which enabled these countries to produce high-quality goods with low costs along with low wage rates, low rent, and diligence. They export these goods to industrialised countries. Thus, both sets of countries benefit from international trade. However, the domestic firms of industrialised countries must compete with the imports that are of high quality and low price; otherwise, employment and the general wage rate will decrease. As a result of globalization and economic liberalization, the people of developing countries move to high-wage countries. This is advantageous to firms of industrialised countries but disadvantageous to workers who have been employed on high wages. The income of middle class workers who stably supported society has considerably decreased, resulting in a dwindling middle class. The inequality pressure from inside Currently, many start-ups have been founded in the fields of Internet of Things (IoT), artificial intelligence, biotechnology, and others. Some of them have grown into worldwide enterprises which have de facto standards. These enterprises supply goods and services which enhance the productivity of economy tremendously and therefore become the indispensable items for society to progress. These high-quality goods are improved continuously, and the real price of these have a tendency to decline. This also contributes towards the economy. The founders, investors, and entrepreneurs of these companies gain enormous income and wealth as a reward. These innovative enterprises usher changes to the traditional old-fashioned companies and in some cases induce withdrawal or bankruptcy. This leads to the increase in company efficiency; however, it also increases the number of unemployed people and decreases wages. Most of the rising innovative enterprises are fabless and outsource production to foreign manufacturers, leading to a decrease in domestic employment. Thus, a rise in innovative enterprises causes income disparity as a few successful persons become extremely rich; however, workers’ average income decreases gradually, and at worst, causes a rise in unemployment. The COVID-19 pandemic The recent COVID-19 pandemic has aggravated the unequal distribution of income. To prevent the spread of the pandemic, many countries declared a state of emergency, affecting international as well as domestic economies. Numerous workers lost their jobs. The benefits provided by the government to the workers have not been enough. Low- or middle-income patients suffering from the long-term symptoms of the pandemic often lose the ability to work. The disparities between the wealthy and unemployed or sick people are increasing, leading to an increase in the government deficit. Many economists insist that tax increases are inevitable to restore the soundness of the government budget. Literatures on inequality Market economy and democracy have historically overcome many obstacles. There are various distinguished contributions to the analysis of inequality. In the era, the bourgeoisie was on the rise, Adam Smith (1776) formulated the significant concept that through the working of the market mechanism (and the adoption of the division of labour system), the capitalist class, the land-owing class, and the working class can live harmoniously. This concept conveys that the market economy does not bring about severe inequality. But reality contradicts this. Workers’ long-lasting struggles to fight inequality still continue. Pigou (1920) founded the welfare economics and in it he insisted that equalisation of national dividend (or national income) will, ceteris paribus, increase the economic welfare. After his analysis, welfare analysis on income distribution developed.1 Seeing the unprecedented massive unemployment caused by the Great Depression (started from 1929), Keynes (1936) had a doubt about the effectiveness of neoclassical economics2 to solve it. Massive unemployment is one of the most serious threats to the society and to the income inequality.3 He founded the macroeconomics. In it he affirmatively evaluated the activities of entrepreneurs and workers, and placed low value on the activities of the people who does not contribute to production. A representative example of the latter is the rentier. Keynes considered that the rentiers did not contribute to the recovery of Britain’s economy at that time. Kuznets (1955) examined economic growth and income distribution. The ‘Kuznets Curve’ has been one of the bases of inequality analysis. Although Friedman (1962) believed in the efficient working of the market, he did not fail to notice the inequality of income and property. He advanced the system of negative income tax and mentioned that the determination of the minimum income level of this system depends on the sum the society can subsidize. He was also sensitive towards the inequality brought about by the market. Sen (1973; 1982) insisted that both utility level and the level of basic goods are not appropriate for measuring welfare. Instead, he proposed the concept “basic capability4” and used it to revise inequality. Atkinson’s empirical studies (1975; 1987; 20115) elucidate and tackle the inequality problem. He also published a book (2015) that proposed practical policies to rectify inequality. Deaton’s approach (2013) involves the relationship between health and wealth, and turns attention towards people of developing countries. Piketty’s studies (20036; 2014) cover a long span and criticize Kuznets’ findings, stirring debates and reinforcing people’s interest. Regarding the policy prescription for income equalisation, Friedman’s concept (1969) of ‘helicopter money’ can be applicable to the problems of inequality as well as deflation. Modern monetary theory (MMT) proposed by Wray (2015) and Kelton (2020) among others, presents a new way of understanding public finance and monetary theory. This theory observed the behaviour of the Bank of Japan (BOJ; central bank) and financial market. A government’s budget deficit is not the reason to revise unequal income distribution. Ravallion (2016) made an intensive study on the economic inequality. The study is structured as follows: the contents of the policy are explained in (“Economic growth and social welfare”, “Erasure of government bonds”, “The pandemic and income equalisation”, “Apprehension regarding inflation” and “The effectiveness of income equalizing expenditure R”). The next section elucidates (“Traditional measures of equalisation”), and the subsequent section describes the (“Multiplier analysis”). (“Concluding remarks”) concludes the paper. Economic growth and social welfare Social welfare function concerning income distribution This study presents a normative analysis of the relationship between economic growth and income inequality. The variables of this model are those of macroeconomics. The social welfare function developed by Bergson (1938) and Samuelson (1947) is applied to consider the problem. From the standpoint of economic policy, this function is maximised, and the method of analysis is that of microeconomics. Therefore, this model is developed by the combination of macroeconomics and microeconomics. The national income of a country is denoted as Y, and is assumed to be composed of labour income wL (where w: wage rate, L: labour), capital income rK (r: interest rate, K: capital), and entrepreneurs’ income or profit, Π, i.e. Y=wL+rK+Π. The national income Y is divided into two parts: labour income wL denoted as YL; and the sum of capital income rK and entrepreneurs’ income (profit) Π, termed as property income and denoted as YP. That is:1 YNationalincome==wLLabourincome++rKCapitalincome++ΠEntrepreneurs'incomeprofitYNationalincome==YLLabourincome++YPPropertyincome This is illustrated in Fig. 1, where the labour income YL is measured along the vertical axis and property income YP along the horizontal axis. The national income Y is depicted as lines, Y1Y1,Y2Y2, Y3Y3,⋯, according to the income level Y1,Y2,Y3,⋯. The line YiYi(i=1,2,3,⋯) is called the ‘national income line’. The slope of this line is − 45°, and it shifts above to the right as the economy (national income) grows.Fig. 1 Socially equal distribution The workers’ utility from labour income YL is denoted as uLYL, and capital owners’ and entrepreneurs’ utility from property income YP is denoted as uP(YP). Then, the social welfare function W concerning income distribution is formulated as:2 W=WuLYL,uPYP The social (welfare) indifference curves concerning income distribution are derived from (2), and depicted as W1,W2,W3 in Fig. 1. As to the relative height of the utility, it is assumed that inequality W1<W2<W3 holds. The slope of these curves is shown as:3 dYLdYP=-∂W/∂uP∂uP/∂YP∂W/∂uL∂uL/∂YL. If social welfare (2) is maximised subject to the income distribution constraint (1), the optimal conditions are derived as:4 ∂W∂uL∂uL∂YL=-λ,∂W∂uP∂uP∂YP=-λ,YL+YP=Y The conditions (4) are the points of contact of the national income lines with social indifference curves and are shown as E1,E2,E3. As national income Y increases, or as the economy grows, the point of contact moves above to the right, and the path is depicted as the optimal growth distribution (OGD). The locus of this curve is shown as:5 dYLdY=WLPuP′uL′-WPPuP′2+WPuP″ΔdYPdY=WPLuP′uL′-(WLL(uL′)2+WLuL″)Δ where WL=∂W/∂uL,WP=∂W/∂uP,WLP=∂2W/∂uP∂uL,WPL=∂2W/∂uL∂uP, WLL=∂2W/∂uL2,WPP=∂2W/∂uP2,uL′=∂uL/∂YL,uP′=∂uP/∂YP,uL″=∂2uL/∂YL2,uP″=∂2uP/∂YP2 andΔ=WLPuL′uP′+WPLuP′uL′-WLLuL′2+WLuL″+WPPuP′2+WPuP″>0.7 Equation (5) shows the social optimal economic growth-income distribution curve OGD, which represents the locus of socially optimal income distribution when the economy grows. On both sides of this OGD curve, the socially permissible border lines regarding income inequality are depicted. The border line of the minimum permissible labour income, from the standpoint of social welfare, is depicted as LL in Fig. 1, and the property income is depicted as LP. Regarding the slope of social welfare indifference curves (3), the sign of terms ∂uL/∂YL and ∂uP/∂YP is assumed to be positive. The sign of terms ∂W/∂uL and ∂W/∂uP must be examined. Three regions are classified:6 RegionI: regionbetweentheLLlineandLPline∂W/∂uL>0,∂W/∂uP>0RegionII: regionrighttothelineLL∂W/∂uL>0,∂W/∂uP<0RegionIII: regionlefttothelineLP∂W/∂uL<0,∂W/∂uP>0 Expression (6) shows that in Region I, income is distributed equally or is permissibly equal. Therefore, the increase in workers’ utility uL due to increase in income YL is considered the increase in social welfare, ∂W/∂uL>0. The same is applied to the increase of utility of capitalists and entrepreneurs uP, i.e. ∂W/∂uP>0. In Region II, income is distributed unequally (i.e. too little to the workers and too much to capitalists and entrepreneurs). Therefore, the increase in workers’ utility is considered as the increase in social welfare, ∂W/∂uL>0, and that of capitalists and entrepreneurs is considered as negative, ∂W/∂uP<0. In Region III, income is distributed contrary to the Region II. Then, the relationships ∂W/∂uL<0 and ∂W/∂uP>0 are supposed to hold. From expression (6), the slope of the social welfare indifference curves (3) can be shown as:7 RegionI: theslopeisnegative,i.e.dYL/dYP<0RegionIIandIIItheslopeispositive,i.edYL/dYP>0. Expression (7) in Fig. 1 can be explained as follows: in Region I, the income distribution is equal or permissibly equal, and so, from the social welfare perspective, the distributed incomes YL and YP are goods, and the slope of the indifference curves is negative. In Region II, the income distribution to YL is too little and that to YP is too much. So, YL is goods and YP is bads. In this situation, if YP increases, to keep the level of social welfare constant, YL must be increased. Therefore, the slope of the indifference curves is positive. In Region III, YL is bads and YP is goods. So, the slope of indifference curves is positive. The curvature of the social welfare indifference curves is assumed as smooth and convex to the origin 0. The actual growth distribution curve and socially equal economy The OGD curve which is socially optimum does not always coincide with the actual one. The actual growth distribution curve is denoted as AGD and depicted in Fig. 1. When the national income level is Yi(i=1,2,3,⋯), the actual income distribution between YL and YP is plotted as pi on the national income line YiYi. The locus of these points p1,p2,p3 constitutes the AGD curve. As this curve passes through the Region I which is between the two permissible borders LL and LP, the income distribution of this economy is said to be socially (more or less) equal. Socially impermissible extremely unequal economy An extremely unequal economy is assumed next and explained in Fig. 2. Here, only two social groups, α and β, are analysed. The rest, whose income is stable, is assumed to be constant and omitted from the analysis. Group α constitutes the people who formerly belonged to the middle class, but owing to economic changes and the immigration policies of the economy are facing challenges. Group β constitutes successful innovative entrepreneurs and well-off property owners. In Fig. 2, Group α’s income is measured along the vertical axis and that of Group β’s in the horizontal axis. The social welfare indifference curves are depicted as w1,w2, and w3. The total income line of Groups α and β is depicted as y1y1,y2y2,y2′y2′, and y3y3. (This total income excludes the stable incomes of the rest of the people). The social OGD curve is depicted as ogd, and socially permissible borders are depicted as lα and lβ. The AGD curve is depicted as a sharply bent curve, agd. This curve means that in the process of increase of total income from y1 to y2′, the people in Groups α and β gained considerable equal shares of income. This corresponds to the point p1,p2, and p2′ on the agd curve. When the total income increases from y2′ to y3, the points which represent income share move from p2′ to p3. This movement shows that Group α’s income share decreased disadvantageously.Fig. 2 Socially impermissible extreme distribution We consider the relationship between the increase in income level and social welfare of Group α. When the economy grows and total income increases from y1 to y2′, Group α’s income also increases and its income distribution is within the socially permissible borders lα and lβ. In this process, the social welfare increases and approaches the highest attainable level w2. The situation, however, changes remarkably when the total income exceeds y2′. If the total income increases to y3, the social welfare decreases from w2 to w1. The income share of this point is represented by p3, and this point is located much farther to the right of the permissible border line lα. This means that, from the social welfare perspective, the income shares of Group α is extremely low compared to Group β. If the total income increases further, the situation becomes more complex. Group α may resent the economic growth in a market economy where they are not rewarded. Traditional measures for equalisation Several policies have been presented for the equalisation of income distribution. Two of these policies are discussed here and thereafter the new policy is explained. The royal road Progressive taxation of high-income classes and transfer of its revenue to low-income classes is a well-known policy for the equalisation of income distribution (Fig. 3). As the economy grows, the national income Y on the AGD curve is divided between YL and YP. The actual income share is shown as points P0,P1,P2, and P3 on the AGD curve.Fig. 3 Taxation and transfer The point P3 is set as the starting point of analysis. The point represents the income distribution between YL and YP when the national income is Y3. Because this point P3 is considerably to the right of the socially permissible border LL, from the social welfare perspective, the redistribution of income is desirable. At the income level of Y3, the social welfare is maximised at the point E3. To reach this point, it is necessary to tax as much as P3A on the property income and transfer the same amount of tax revenue AE3 to labour income. This thoroughgoing measure realizes the maximum social welfare. The economic value of this progressive taxation and its transfer is inferred as follows: the starting point P3 and attained point E3 are both on the same national income line Y3. The point P3 is on the social welfare indifference curve W1 and its level of social welfare is the same as E1 on the OGD curve. So, although point P3 is on the Y3 line, its economic value from the standpoint of social welfare is equivalent to the income value of Y1. Contrary to the point P3, the attained point E3 is on the social welfare indifference curve W3 and is on the OGD curve. Therefore, the economic value of the progressive taxation and its transfer has the economic value shown by the distance E1E3 or the income difference Y3Y1. This policy is, indeed, extremely efficient to attain maximum social welfare and may be called as ‘the royal road’ to income equalisation. The royal road policy is ideal for the economy. The rich capitalists (or propertied) class as well as successful entrepreneurs will, however, strongly oppose this policy. One of the most persuasive reasons for opposition is that unlimited progressive taxation will destroy the social order, or extraordinary heavy tax will break their incentive to innovate. If these oppositions are strong, a compromised, downgraded policy is implemented. In this situation, the extent of P3a is taxed and its revenue of ab(=P3a) is transferred. The inequality of income distribution is, then, slightly revised. The economic value of this compromised policy is inferred as the distance E1E2 or the income difference Y2Y1. These down-sized policies are adopted in almost every industrialised economy. But, in many cases, it may not be enough to cope with the ongoing inequality. Donation (or contribution) In a market economy, entrepreneurs plan the production and investment scheme to maximise profit, while consumers plan the expenditure and time-allocation scheme to maximise utility. These behaviours come from their self-interest or egoism and are the driving forces of a market economy.8 Altruism (or philanthropy) At the opposite extreme, there may be altruism or philanthropy. The development of the market economy till now had caused many harmful effects. Nevertheless, it is well known that capitalists, entrepreneurs, and wealthy persons have donated a considerable part of their earnings and properties. The motives for their donation can be classified into three types. First is the pure or genuine altruism motive, such as human welfare and religious sentiments. Second is that of ordinary altruism such as donations or the establishment of foundations with their names. Finally, there is impure or non-genuine altruism such as vanity or pomposity. Currently, due to remarkable innovations, many multi-billionaire entrepreneurs donate their vast wealth to society or establish foundations across the world. These foundations have far-reaching effects. The advancement of art, culture and sciences, support to education and medical activities, and eradication of poverty are well-known examples. Donations based on the motive of altruism is considered in Fig. 4. The starting point of analysis is set at point c where the labour income is YLc and property income is YPc. Here, the existence of property owners’ (i.e. the capitalists’ and entrepreneurs’) utility function U=U(YP,YL) is supposed, and their income constraint is shown as YP+YL=YPc (YPc: constant). Then, at the point c, the property owners’ utility maximisation problem is formulated as:8 maxU=UYP,YL,subject\,toYP+YL=YPcYPc:constant Fig. 4 Altruistic (or Philanthropic) donation If the property owners have no concern for the income level of labour YL, or consider the existing level of labour income YLc to be enough, the solution of (8) is a corner solution and attained at point c. In this case, the non-altruistic property owners do not donate, and the income distribution is unchanged. Conversely, if the altruistic property owners consider that the existing income distribution is unequal and disadvantageous for labour, they will donate their income to labour. Then, the solution of (8) is shown as:9 ∂U/∂YL/∂U/∂YP=1,YP+YL=YPc In this situation, the altruistic property owners choose, for example, the point d where the property owners’ indifference curve (derived from their utility function U) Ua is tangential to the national income line Y3Y3. The altruistic property owners donate their income cf(=YPcYPd), and labour receives this. So, the labour income increases to fd(=YLdYLc). By virtue of the property owners’ donation, the point of income distribution changes from c to d. Thus, on one hand, the level of social welfare increases from W1 to W2, and, on the other hand, the altruistic property owners’ utility also increases from U¯a to Ua. The increase in the utility of property owners is shown in terms of income. The property owners’ indifference curve U¯a is tangential to the income line YaYa at point g, which is parallel to Y3Y3. Then, the distance gd represents the increase of property owners’ income and is equivalent to YaY3. This income increase may be called as ‘the altruistic donator’s benefit’. The increase in social welfare is also measured in terms of income. As the social welfare increases from W1 to W2, the corresponding equivalent value increases from E1 to E2, meaning that social welfare in terms of income increases to Y1Y2. In Fig. 4, the point d after donation, however, is located to the right of the labour’s socially permissible border line LL. This indicates that although property owners’ donations are valuable, it is insufficient to attain socially optimal income distribution.9 Egoism (or self-protection) Over time, if the income inequality reaches an extreme, people who become desperately poor or have long been in dire straits will feel strong resentment towards the society, market economy, capitalists, entrepreneurs, and the propertied people. This will lead to social unrest across the country. In such situations, some of the wealthiest people may individually or collectively try to prevent the occurrence of the worst case scenario (such as riots or revolutions) by donating a part of their wealth to the people in need or destitute workers (hereafter, these people are denoted as the destitute). The behaviour of the wealthiest people is analysed in Fig. 5. The destitute is classified as Group α, and the wealthiest is classified as Group β. In this analysis, only two Groups α and β of the society are examined and the rest, whose income is stable, is assumed to be constant and omitted from the analysis.Fig. 5 Egoistic (or Self-protective) donation; social unrest effect, utility effect, and income effect The starting point in the Figure is set at point h (on the agd curve) where the destitute α’s income Yαh is extremely low and that of the wealthiest group β is Yβh and is extremely high. First, the existence of the utility function of the wealthiest U=U(Yα,Yβ,I) is supposed, where Yi(i=α,β) is the income of group i, and I is the indicator of social unrest brought about by income inequality. This indicator reflects problems between the destitute and wealthiest, demonstrations against income inequality, a riot to achieve economic justice, and so on. Although the indicator I is a function of income inequality, to simplify the analysis, it is assumed as a parameter. In the figure, the indifference curve UA of the wealthiest group is depicted. At point H, which is very close to the neighbourhood of h,10 the utility of the wealthiest is maximised. At point H, the indifference curve of the wealthiest UA is tangential to the income line yByB. This maximisation problem is formulated as:10 maxU=U(Yα,Yβ,I),subjecttoYα+Yβ=Yβh(Yβh:constant), The solution is shown as:11 ∂U/∂Yi=-λ,Yα+Yβ=YβhYβh:constant, where λ is a Lagrange multiplier. The wealthiest β at first satisfies the situation shown by point H. This point, however, represents the spread of social unrest and the threat felt by the wealthiest β. If the indicator I which reflects social unrest increases, their utility distinctly decreases (i.e., ∂U/∂I<0), and the shape and location of the utility function U of the wealthiest change and thereby their indifference curves also change. The indifference curve UA changes to uA. After the increase of the social unrest indicator I, the point H is not the optimal point for the wealthiest.11 They seek the point which maximizes their utility and will find that j, where the new indifference curve uB is tangential to the income line yByB, is the new optimal point. Accordingly, the wealthiest people donate their income by hk(=YβhYβj), and the destitute receive the same amount as kj(=YαhYαj). This movement of optimal point from H to j is called ‘the social unrest effect’. It is disintegrated into two parts, i.e. from H to m, and from m to j. The first movement is that between the points at the same utility level. This movement is called ‘the utility effect’ (of the increase of the social unrest indicator I). The second is ‘the income effect’ (of the indicator I). The total and net loss incurred by the wealthiest due to the extreme concentration of wealth which caused social unrest can be considered in the same figure. The total loss is shown as the movement from m to j and j to n. This loss is expressed in income terms as yAyC. Because the wealthiest donated their wealth to Hk to mitigate social unrest, their utility level increased from uC to uB. This leads to the decrease in total loss as yByC. Then, the net loss caused by the social unrest is shown as yAyC-yByC or yAyB, and this net loss is equal to ‘the income effect’ of the increase in social unrest I. The relationships among ‘the social unrest effect’, ‘the utility effect’, and ‘the income effect’ are shown mathematically as follows: The social unrest effect is derived from the optimal conditions (11):12 dYα/dIYβh:const.=UαI-UβI/ΔIdYβ/dIYβh:const.=-UαI-UβI/ΔI where Uαβ=∂2U/∂Yβ∂Yα,Uβα=∂2U/∂Yα∂Yβ,Uαα=∂2U/∂Yα2,Uββ=∂2U/∂Yβ2,UαI=∂2U/∂I∂Yα, and UβI=∂2U/∂I∂Yβ. The term ΔI=Uαβ+Uβα-Uαα-Uββ is positive from the condition of utility maximisation of the wealthiest. The utility effect is shown as:13 ΔYαΔIU:const.=UαI-UβIΔI-UI/UYβUαβ-UββΔIΔYβΔIU:const.=-UαI-UβIΔI-UI/UYβUβα-UααΔI The income effect is shown as:14 ΔYαΔYβhI:const.=Uαβ-UββΔIΔYβΔYβhI:const.=Uβα-UααΔI Therefore, by considering the Eqs. (12) ~ (14), the social unrest effect or the total effect is rewritten as:15 dYα/dIYβh:const.=dYα/dIU:const.+UI/UYβdYα/dYβhI:const.dYβ/dIYβh:const.=dYβ/dIU:const.+UI/UYβdYβ/dYβhI:const. where UI=∂U/∂I<0 and UYβ=∂U/∂Yβ. As to the signs of Eqs. (12–15), no confirmed inference can be obtained without additional assumptions. The concrete determination of signs is, however, obtained from the situation depicted in Fig. 5. In this specific situation, the following conclusions are derived: As to the utility effect,13’ dYα/dIU:const.>0,dYβ/dIU:const.<0 As to the income effect,14’ dYα/dYβhI:const.<0,dYβ/dYβhI:const.<0 As to the social unrest effect or total effect,15’ dYα/dIYβh:const.>0,dYβ/dIYβh:const.<0 The change in social welfare is next considered briefly in Fig. 6. The optimal point moved from H to j due to the egoistic donation of the wealthiest. According to this movement, the level of social welfare increases from w 1 to w 2, and the corresponding equivalent value increases from e1 to e2. This implies that even if the motive of donation is egoistic, the donation contributes to the increase in the level of social welfare. However, the point j after the donation is located to the right of the socially permissible line of the destitute lα. This indicates that the egoistic donation of the wealthiest group is insufficient to resolve the difficult situation.Fig. 6 Egoistic (or Self-Protective) donation and its effect on social welfare Erasure of government bonds Although ‘helicopter money’ and ‘MMT’ policies have a wide range of purposes, they are effective in managing income inequality. It is reasonable to implement these policy measures if the price level is stable. Japan as an example: the amendment of law and erasure After evaluating the above policies, this study proposes a new policy to mitigate or overcome the income distribution inequality. The policy is that, by amendment of law, the government erases the government bonds (owned by the central bank) for income equalisation. This study considers Japan as an example. At the end of September 2021, the amount of outstanding government bonds is approximately 9.5 trillion dollars, which is a little less than twice as much as Japan’s GDP.12 The ratio, government bonds outstanding/GDP, is the highest (or worst) in the industrialised countries, leading to fears of bankruptcy of the Japanese government. BOJ owns approximately 4.6 trillion dollars. By subtracting 4.6 trillion dollars from 9.5 trillion dollars, 4.9 trillion dollars remain. If this amount is divided by the GDP, the quotient is about 1.0. So, the ratio of the debt burden decreases drastically,13 which is the key issue. If 4.6 trillion dollars of government bonds owned by BOJ remain intact and 9.5 trillion dollars of total government bonds remain, problems will occur because of the following reasons: first, because the government bonds outstanding is 1.9 times as large as Japan’s GDP, the exogenous shocks such as war, massive earthquakes, endogenous great depressions, and pandemics can cause unexpected scenarios. Second, if massive amounts of government bonds remain intact, consumers will anticipate a future tax increase and feel uneasy about their social security, medical care, education, and so on, and refrain from consumption as much as possible. Moreover, entrepreneurs will expect an increase in corporate tax and after surveying the attitude of consumers, will hesitate to invest. When massive and enduring government bonds (or debts) exist, the economy has the tendency towards nearly zero rate of growth or stagnation. Therefore, people who slid down from the middle class to the low-income class and low-income group people are forced to endure this situation. Therefore, the government bonds which the central bank possesses should be erased. This measure will minimise the fear of tax increase, and consumers’ and entrepreneurs’ psychological pressure will be significantly reduced. From the macroeconomic perspective, both the consumption function and investment function will shift favourably. Regarding the side effects of the erasure policy, heavy confusion regarding the bonds market or the economy may be anticipated. However, the side effects are limited. As the amount of government bonds decreases drastically, and the government bonds held by private sectors and foreign countries are not erased and guaranteed firmly, its scarcity value will increase. The possibility of the default of Japan’s government bonds will vanish. This will increase the demand for Japan’s government bonds which in turn will increase its price leading to a decrease in the interest rate. New issue of government bonds for equalisation After the erasure of the existing government bonds, new government bonds need to be issued to equalise income distribution. The generated revenue is spent on low-income people. This increases their income level and improves their health condition, provides opportunity for higher education, human capital formation to increase their productivity, and so on. Therefore, their future income will continuously increase. These strong measures will improve low-income situations and lead the whole economy towards steady growth. The summated issue depends on the sum needed to equalise income distribution, but it should not exceed the erased sum. This ensures that debts do not accumulate. In Japan’s case, the amount of outstanding government bonds is now approximately 9.5 trillion dollars and, of these, BOJ owns about 4.6 trillion dollars. After the amendment of the law, the government of Japan erases these 4.5 trillion dollars of government bonds held by BOJ. The remaining government bonds (5.0 trillion dollars) are still owned by domestic and foreign investors.14 The government must firmly guarantee the obligation. As the possibility of default completely disappears by the erasure of government bonds (held by BOJ), the scarcity value of the 5 trillion dollars of government bonds will increase, thereby increasing the demand for Japanese government bonds. Specifically, there will be an excess demand for Japanese government bonds. To realize the optimal income equalisation, I consider how much the government should expend for the low-income people. In this model, to simplify the analysis, the labour income is considered to be that of low-income people. Given the existing property income level, the total socially optimal sum R required for income distribution is defined as:16 R=thelabour'ssociallyoptimalincomelevel- thelabour'sexistingincomelevel where the labour’s socially optimal income level satisfies the optimal conditions (4). I consider the above using an abstract example and then a numerical example which may also have an empirical reality. The initially erased sum is denoted as A. The total optimal sum required for income equalisation is assumed to be R trillion dollars. To avoid rapid changes which will be caused by the lump-sum expenditure of R, the process of further erasure and expenditure should be divided into a geometric series. These divisions will be useful to smoothen the change and constrain the occurrence of inflation. If, during these processes, inflation occurs, the erasure and expenditure policy should be suspended and resumed after the resolution of inflation. The upper limit of the inflation is about 2%. The first step is to newly issue some ratio of the erased sum. This ratio is called ‘newly issued rate’, denoted by ϕ, i.e.17 newlyissuedrateϕ=newlyissuedgovernmentbondsinitiallyerasedgovernmentbondsA,(1>ϕ>0) The newly issued government bonds are denoted as ϕA. This sum ϕA is expended for the low-income group, which involves social security including reinforcement of the pension system, medical care, education for human capital formation to enhance their future earning capacities, and assistance for undeveloped countries (the income redistribution for foreign low-income group). These will increase the welfare of the low-income group and consequently increase their economic productivity. Then, because the government bonds have increased to ϕA, the government erases ϕ2A of the government bonds. The residual non-erased portion 1-ϕϕA is still held by domestic and foreign investors and firmly guaranteed by the government. The above process is shown as follows: The second step is to issue new government bonds by as much as ϕ3A and expend the same amount for the low-income group. Then, of these increased bonds, the government erases ϕ4A and the residual non-erased portion 1-ϕϕ3A held by domestic and foreign investors is firmly guaranteed by the government. This process is shown as follows: The third step is similarly shown as: To summarize these steps PI,PII,PIII,⋯, the total optimal expenditure R for the low-income group is equal to:18 R=ϕA+ϕ3A+ϕ5A+⋯=Aϕ/(1-ϕ2) From this, the value of ϕ is calculated as:19 ϕ=-A+A2+4R2/2R When the government’s optimal expenditure R for the low-income group is implemented gradually, the multiplier effects work at each step. If the scale of its effects is m, the total created income V is shown as:20 V=mϕA+mϕ3A+mϕ5A+⋯=mAϕ/(1-ϕ2) The total residual non-erased sum S which is held by domestic and foreign investors is calculated as:21 S=1-ϕϕA+1-ϕϕ3A+1-ϕϕ5A+⋯=Aϕ/(1+ϕ) By such increases in created income, tax revenues also increase. If the average and marginal tax rate is denoted as t, the total increases of tax revenue T is:22 T=tmϕA+tmϕ3A+tmϕ5A+⋯=tmAϕ/(1-ϕ2) If this tax revenue T is used to redeem the government bonds, the net increase of government bonds N(=S-T) is:23 N=S-T=Aϕ1-ϕ-tm/(1-ϕ2) If the government bonds that were outstanding before the erasure policy is denoted as Z, the total sum of government bonds held by domestic and foreign (including government) investors Σ is:24 Σ=Z-A+N=Z-A+Aϕ1-ϕ-tm/(1-ϕ2) The above expressions (18–24) are elucidated using numerical examples. First, the expressions (18–20) are explained diagrammatically. Because government’s first erased sum A is 4.50 trillion dollars, if the total optimal sum R required for income redistribution is, for example, 3 trillion dollars, the government’s new issue rate ϕ can be obtained by calculating the expression (18), i.e.25 3.00=4.50ϕ/(1-ϕ2) From this, the value of ϕ is determined as 0.50. This implies that if the government has a heavy burden of outstanding government bonds of 9.50 trillion dollars (out of which BOJ holds 4.60 trillion dollars) and needs 3 trillion dollars to expend for optimal income redistribution, the government should erase 4.50 trillion dollars of government bonds held by BOJ and, in the first period, newly issue the 0.50×4.50triliondollars=2.25 trillion dollars of government bonds for income redistribution and, in the second period, newly issue the 0.503×4.50=0.56 trillion dollars of government bonds for income redistribution and, in the third period, 0.505×4.50=0.14 trillion dollars of government bonds, ⋯. Then, from Eq. (18), the following relationship is obtained:18’ 3.00≒Requiredsum≒2.25+firstperiod+0.56+secondperiod+0.14thirdperiod++…… Here, to implement the income equalisation, expenditure should be made over the years to make the transfer payment and construction of various systems (such as social security, medical care, education, foreign assistances, etc.) smooth. Beside these, gradual steps of expenditure are indispensable to avoid inflation. If the multiplier effect over all period is 2.00, the total created income V(=2.00R) is shown as:V=2.25×2.00+0.56×2.00+0.14×2.00+0.04×2.00+⋯≒6.00trilliondollars The effect of optimal redistribution money R on income creation V and social welfare W is explained in Fig. 7. The starting point is P4 where the erasure of government bonds held by central bank has been completed. The total optimal sum R required for income equalisation is shown as the distance P4g where the point g is on the OGD curve and therefore this point g satisfies the social optimal conditions (4).Fig. 7 The effect of expenditure on income equalization At point P4, the property income is shown as Oe. From the social welfare point of view, the optimal level of labour income is eg. The existing level of labour income is eP4. So, the total optimal sum R required for income equalisation is:26 R=eg-eP4=P4g It is assumed that R is equal to 3.00 trillion dollars, or P4g=3.00trillion dollars. To finance this sum and to not accumulate too much debt, I proposed to erase nearly all of the 4.50 trillion government bonds held by BOJ and to issue new government bonds which amounts to half of the total erased sum. Through this, the government obtains 2.25 trillion dollars. The process starts and continues as follows: First, 2.25 trillion dollars are divided, for example, by 5 and, each 0.45(= 2.25/5) trillion dollars are expended over five years. The first year’s expenditure is shown as P4ρ1 and taking into its multiplier effect 2.00, the expenditure creates additional income shown as ρ1ρ2. The next year’s expenditure is shown as ρ2ρ3 and by the multiplier effect, this expenditure creates the additional income ρ3ρ4. This process continues to the point ρ10, where the redistributed money is 2.25 trillion dollars and the total additional income is 4.50(= 2.25 × 2.00) trillion dollars. Second, 0.56 trillion dollars are divided by 5 and each 0.11(= 0.56/5) trillion dollars are expended over five years. This year’s expenditure is shown as ρ10ρ11, and it creates additional income ρ11ρ12,⋯. These processes go on till point P5. The sum of redistributed money R is 3.00 trillion dollars and the total created income V is 6.00 trillion dollars. The total increased income V is shown as Y4Y5 (or P4m, or P4q) and of this income V, the increased labour income is shown as P4l(=nq) and increased property income is lm(=P4n). It is shown that by the introduction of optimal redistribution of money R, the level of social welfare increases from W0 to W5, and the economic value increases from E0 to E5. The result of this policy expressed in terms of income is Y4Y5. Next, the expressions (21–24) are explained. The sum of government bonds newly issued (but not erased) and purchased by domestic and foreign investors S (trillion dollars), i.e. the total residual non-erased sum is calculated as:S=1.13+0.28+0.07+0.02+⋯≒1.50(trilliondollars) The total increase in tax revenue brought about by the created income is calculated. If the average and marginal income tax rate is 0.20, the total increase in tax revenue T is shown as:T=0.20V=2.25×2.00×0.20+0.56×2.00×0.20+0.14×2.00×0.20+0.04×2.00×0.20+⋯≒1.20trilliondollars Then, if this tax revenue is used to redeem the government bonds, the net increase in government bonds can be calculated. The sum N(=S-T) is shown as:N=S-T≒1.50-1.20=0.30trilliondollars) Hence, the total sum of government bonds held by domestic and foreign investors Σ (trillion dollars) is shown as:Σ=5.00+0.30=5.30trilliondollars The above numerical example is summarized as follows: when the total optimal expenditure R is needed by 3.00 trillion dollars, the erasure rule demands that newly issued rate ϕ should be 0.50. If the multiplier effects m on redistribution expenditure R is 2.00, the total created income V becomes 2R or 6.00 trillion dollars. The total residual non-erased sum S is 1.50 trillion dollars. If the average and marginal tax rate t is 0.20, the total increase of tax revenue T is 0.20V or 1.20 trillion dollars. The net increase in government bonds N(=S-T) is therefore 1.50-1.20=0.30 trillion dollars. As a result, the total sum of government bonds Σ=5.00+0.30 is equal to 5.30 trillion dollars. Other values of total optimal expenditure for redistribution R are also examined. These are shown in Table 1. The table shows the relationship between R and ϕ,V,S,T,N, and Σ (when the value of multiplier effect m is 2.00, and tax rate is 0.20).Table 1 The Relationship between R and ϕ,V,S,T,N, and Σ Total optimal sum for redistribution R 1.00 1.67 2.00 3.00 4.22 5.00 … 10.00 … 17.21 … 21.32 … 25.00 Newly issue rate ϕ 0.21 0.33 0.38 0.50 0.60 0.65 0.80 0.88 0.90 0.91 Total created income V=2R 2.00 3.34 4.00 6.00 8.44 10.00 20.00 34.42 42.64 50.00 Total residual non erased sum S 0.78 1.12 1.24 1.50 1.69 1.77 2.00 2.11 2.13 2.14 Total increase of tax revenue T=0.4R 0.40 0.67 0.80 1.20 1.69 2.00 4.00 6.88 8.53 10.00 Net increase of government bonds N=S-T 0.38 0.45 0.44 0.30 0.00 − 0.23 − 2.00 − 4.77 − 6.40 − 7.86 Total sum of Government bonds outstanding Σ= 5.00+N 5.38 5.45 5.44 5.30 5.00 4.77 3.00 0.23 − 1.40 − 2.86 R=ϕA1-ϕ2,ϕ=newlyissuedgovernmentbonds/erasedgovernmentbonds, A=4.50trilliondollars,Z=9.50trilliondollars,V=mR,S=ϕA1+ϕ, N=ϕA1-ϕ-mt1-ϕ2,Σ=Z-A+ϕA1-ϕ-mt1-ϕ2,m=2.00,t=0.20 From Table 1, two relationships are established and considered. One is the relationship between R and ϕ, depicted as the curve R in Fig. 8. The slope of this curve R is positive and increasing, i.e. ∂R/∂ϕ>0 and ∂2R/∂ϕ2>0.15 The curve R shows that, given the value of the initially erased sum A, when the total optimal sum R is determined, the newly issue rate ϕ is correspondingly calculated. This figure is intended to be read from the vertical axis to the horizontal axis, or from R to ϕ. Therefore, if the total optimal sum R required for income equalisation is 1.67 trillion dollars, the newly issued rate ϕ is determined as 0.33, and if R is 4.22, ϕ is 0.60.Fig. 8 The relationship between the total optimal sum R and newly issued rate ϕ Another relationship is between ϕ and S,T,N(=S-T) and Σ(=5.00+N), depicted in Fig. 9. The relationship between the total residual non-erased sum S and ϕ is shown as S curve. The slope of this curve S is positive and decreasing.16 The relationship between the total increase in tax revenue T and ϕ is shown as the T curve, and the slope of this T curve is positive and increasing.17 The net increase in government bonds N is equal to S-T. The curvature of the curve N is convex and sloping upwards18 and reaches a maximum point when ϕ is equal to 0.33.19 By connecting these two figures, the relationship between the total optimal sum for redistribution R and net increase in government bonds N, and the total sum of government bonds outstanding Σ(=5.00+N) is shown. This study explains three characteristic points:When the optimal sum for redistribution R is 1.67 trillion dollars, the newly issued rate ϕ is 0.33. At this rate, as explained above, the net increase in government bonds N reaches the peak, and the total sum of government bonds outstanding Σ reaches the peak. In the interval of 0<ϕ<0.33, N and Σ increase, and in the interval of 0.33<ϕ<1, N and Σ decrease. When R is 4.22 trillion dollars, ϕ is 0.60. At this rate, the total residual non-erased sum S is equal to the total increase in tax revenue, and therefore N becomes zero. This means that total sum of government bonds outstanding Σ is 5 trillion dollars which is equivalent to the sum before the new issue of government bonds. When R is 17.21 trillion dollars, ϕ=0.88. At this rate, S is 2.11 trillion dollars and T is 6.88 trillion dollars. Thus, the net increase in government bonds N becomes -4.77 trillion dollars, and the total sum of government bonds outstanding Σ becomes 0.23 trillion dollars, or nearly zero. This means that the outstanding government bonds almost disappear.20 Fig. 9 The relationship between ϕ and total created income V, total residual non-erased sum S, total increase in tax revenue T, net increase in government bonds N(=S-T), and total sum of outstanding government bonds Σ It must be noted that because the optimal value of R is determined from the (constrained) social welfare maximisation conditions (4), the level of total sum of government bonds outstanding Σ is not the object of some sort of maximisation, but the result of the determination of R. In other words, social welfare maximisation conditions determine the values of R,ϕ,V,S,T,N, and Σ. It is natural that the relatively small sum of R is easily carried out. But, using the method of erasing government bonds, the relatively large sum of R is carried out without accumulating government bonds or with decreasing government bonds. The government can revise income inequality if it wishes to do so. The pandemic and income equalisation History reveals that humankind has suffered from many disasters. One of the most serious disasters is the COVID-19 pandemic. The prevalence of the plague, smallpox, cholera, new influenza, and new coronavirus are the representative epidemics. From the income inequality perspective, high-income groups seem to have a relatively advantageous position in avoiding infectious diseases. This is because the high-income groups live in a favourable environment (good sanitary conditions, housing, access to medical facilities, education regarding hygiene, etc.). Conversely, low-income groups lack these facilities which leaves them vulnerable to serious infectious diseases. Thus, income distribution inequality causes the unequal susceptibility to epidemics and diseases. If the pandemic spreads worldwide, its victims will mostly be a part of the low-income group than the high-income group. During the pandemic, governments take drastic measures such as closing the borders and restricting the movement of people and commodities. If the situation persists, the total demand and supply of the economy substantially decreases which has a negative impact on economic activities. The high-income group also suffers tremendous losses, but, owing to their accumulated wealth, many of them can deal with it. In contrast, if low-income groups lose their jobs, they get into a predicament. Thus, the pandemic affects the low-income group more than the high-income group. Governments need to take prompt and adequate measures to tackle the pandemic. One suitable measure is the income equalisation or payment of adequate benefits to low-income groups. Figure 10 explains this. The point P on the AGD curve shows the income level and its distribution before the outbreak of pandemic. As the pandemic becomes severe, the economic situation worsens and the point on the AGD curve moves from P to P~. This movement shows the high disadvantage for the low-income group. Both the absolute income level and relative income share of low-income group decreases. To resolve this grave situation, the government needs to revise the inequality. The following numerical example explains this. At the point P which represents the situation before pandemic, the total optimal sum required for income equalisation R is shown as the distance PG and its amount is assumed to be 3.00 trillion dollars. The economic situation moves down to P~ due to the outbreak and spread of the pandemic. This indicates that, to revise the inequality completely, R must be the distance P~G and this amount is assumed to be 5.00 trillion dollars.Fig. 10 Pandemic and the effect of expenditure on income equalization To revise this severe inequality, 5.00 trillion dollars should be expended for low-income groups. The process and results are shown in Table. If R is equal to 5.00 (trillion dollars), the newly issued rate ϕ is set at 0.65 (percent). If the multiplier effect of R is 2.00, the total created income V is 10.00 (million dollars). The total residual non-erased sum S is 1.77 (trillion dollars). If the average and marginal tax rate t is 0.20, the total increase in tax revenue T is 2.00 (trillion dollars). Then, the net increase in government bonds decreases. As a result, the total sum of government bonds outstanding Σ is 4.77 (trillion dollars), which is below the initial level of 5.00. Through this process, the economy grows from Y~ to Y~~, and the increment of national income ΔY is equal to 2R(=10trilliondollars). The low-income group who was facing challenges are now better off. Their income level increases to P~L. The level of social welfare has increased from W∗ to W∗∗, and its economic value is expressed as the increase in income level from Y∗ to Y∗∗. This policy is summarized as follows: the 5.00 trillion dollars of income equalisation expenditure creates 10.00 trillion dollars of national income, and by the policy of erasure of government bonds and multiplier effect, the net increase in government bonds is approximately zero. The income level of the low-income group and level of social welfare improves significantly. Apprehension regarding inflation This section examines inflation. This study considers that, compared to the previous century, the probability of inflation is not so high currently. Two factors may be stated: First, as is widely supported, after the end of the Cold War, countries, such as China, India, Russia, and many others entered the world market economy. These countries provided high-quality goods with reasonable prices. The industrialised countries imported these goods which reduced the domestic prices of the goods. These tendencies are pervasive and will persist unless irregular exogenous shocks hit the economy. Second, especially after the public access to the Internet, the world economy is now in a severe competition to innovate in the IoT field. Conventional goods and services are substituted by those made by high-technology companies. Such goods are not always cheap. However, as the efficiency of IoT industries is remarkably high, the hedonic prices of these goods are showing a downward trend due to technological progress induced by tough competition. Third, the price of energy will not surge for a long time. Alternative energy is now being developed. An example is shale oil and shale gas. Shale oil and gas reserves are abundant.21 In addition to this, the worldwide prevalence of the development and utilization of renewable energy will resist any energy price increase. The above-mentioned reasons will suppress the possibility of future inflation. However, the possibility of the occurrence of inflation cannot be dismissed.22 If the expenditure for income equalisation R affects the inflation rate i, this relationship is shown as:27 i=i(R,F¯) where F¯ is the other factors which affect the inflation rate i, and to simplify the analysis, assumed to be constant. Regarding the shape of the function i, the following relationships are assumed: di/dR≧0 and d2i/dR2≧0. This function i is shown in Fig. 11. This hypothetical curve i represents that when the value of R is 3.00 trillion dollars, the inflation rate i is 1.60%, when R is 4.00 trillion dollars, i is 2.00%, and when R is 5.00 trillion dollars, i becomes 2.70%.Fig. 11 The relationship between the total optimal sum R and inflation rate i Thus, when the society or government opposes the inflation rate which exceeds 2%, the scale of the expenditure for income equalisation R is constrained by this inflation rate. In this situation, if, for example, 3.00 trillion dollars of R is planned, this may be permitted from the standpoint of controlling inflation. If, however, 5.00 trillion of R is planned, this may not be permitted.23 As inflation occurs also due to the other factors, to avoid an undesirable rate of inflation, these factors must be taken into account. If by any chance,24 an undesirably high inflation rate occurs, the countermeasure is to absorb the money in the market. The government should issue government bonds. If the amount of this issue is too much, the price of government bonds will fall and the interest rate will rise. If this happens in a short time, many of the domestic and foreign investors will suffer huge losses. The government does not suffer any loss because, by erasure, government has little government bonds outstanding. It must, however, be stressed that inflation negatively affects the low-income group, but, by this policy, as the low-income group upgrades to the middle income group, from the income distribution perspective, the inflation does not exert too much negative impact. As long as the inflation rate is low, the majority does not suffer. If a high inflation rate occurs, the policy of erasure and expenditure for the low-income group should be discontinued for some time and resumed after the resolution of inflation. Multiplier analysis This section considers the above policy measure, i.e. income equalizing expenditure R, and then compares the effectiveness between it and public investment for countercyclical purpose G. First, the consumption function C is expressed as:28 C=CY-T+R,Yf(R,G) where T is the tax and is assumed to be proportional to national income Y, i.e. denoting t as an average and marginal tax rate, T=tY. R is the government expenditure for income equalisation.25 Then, Y-T+R represents the disposable income in a broad sense. Yf is the future income which is assumed to depend on government expenditure for income equalisation R and public investment for countercyclical measures G, i.e. Yf=Yf(R,G). The study assumes the following relationships: 1>∂C/∂(Y-T+R)>0,1>∂C/∂Yf>0,∂Yf/∂R>0, and ∂Yf/∂G<0.26 Second, the investment function I is expressed as:29 I=IrR,G,E(R,aG,b(G) where r is the market interest rate which is assumed to be influenced by the government expenditure for income equalisation R and public investment G,27 i.e. r=r(R,G). When the parameters R and G increase, r may increase a little. Thus, the following relationships are assumed: ∂r/∂R≧0,∂r/∂G≧0, and as to the relationship between r and I, it is assumed to be negative, i.e. ∂I/∂r<0. E is a factor which is related to the efficiency of investment I and depends on R and G. I call E ‘an efficient factor in investment’. When R increases, the income of low-income people also increases. Through this, people can be better off and have the opportunity to learn more, and are able to improve their health conditions. In this situation, the efficiency of investment I will increase. Therefore, the following relationships are assumed: ∂I/∂E>0 and ∂E/∂R>0. As to the effects of public investment for countercyclical purpose G on investmentI, two effects are supposed. One is the improvement in infrastructure and through this, investment efficiency will increase. This effect is called the ‘infrastructure effect’ and denoted as a and calculated by: a=aG,da/dG>0,dE/da>0, and dI/dE>0. Another is investing in pork-barrel public projects, which is often done so that politicians can secure votes, which have negative effects on the investment efficiency I. This effect is called the ‘futility effect’ and denoted as b and calculated by: b=bG,db/dG>0,dE/db<0, and dI/dE>0. The equation for the determination of national income is:30 Y=CY-T+R,Yf(R,G)+IrR,G,E(R,aG,b(G)+R+G The effect of distribution equalizing expenditure R on the national income Y is calculated as follows:31 dYdR=1+∂C∂(Y-T+R)+∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R/1-∂C∂Y-T+R(1-t) From the assumptions above, the denominator of (31) is positive. Among the terms of the numerator, the term ∂I/∂r∂r/∂R is nearly zero or zero.28 However, a negative value of this term will not overwhelm the total positive values of the other terms. So, the numerator will be positive. Then, as to the sign of (31), the following relationship may hold:32 dY/dR>0 The effects of the government’s distribution equalizing expenditure R on the labour income YL and property income YP can be analysed. In Fig. 7, the tangent line of the actual growth distribution curve AGD at point P4 is expressed as33 YL=αYP+βα=αR,β:constant where it is assumed that α depends on the parameter R, i.e. α=α(R) and α(R) is positive and dα/dR is also positive.29 This tangential line means that the change in the parameter R affects the relationship between YL and YP; i.e. if R changes, the following relationship holds: dYL=α′YPdR+αdYP, where α′=dα/dR. As Y=YL+YP, the Eq. (30) is revised as:34 YL+YP=C1-tLYL+1-tPYP+R,YfR,G+IrR,G,E(R,aG,bG+R+G where tL and tP are the income tax rate on YL and YP, respectively, and tL<tP. From (34), the following relationships are obtained:35 dYLdR=α′YP1-C′·1-tP+αC′+1+α∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R1-C′·1-tP+α1-C′·1-tL 36 dYPdR=-α′YP1-C′·1-tL+C′+1+∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R1-C′·1-tP+α1-C′·1-tL where C′=∂C/∂1-tLYL+1-tPYP+R. As to the denominators of (35) and (36), the terms 1-C′·(1-tP) and 1-C′·1-tL are positive and as α is positive, the denominators are positive. As to the numerator of (35), from the assumptions α and α′=dα/dR are positive, so the sum of the first and second terms α′YP1-C′·1-tP+αC′+1 is positive. The element ∂C/∂Yf∂Yf/∂R of the third item is positive because when R increases, future disposable income Yf also increases, and accordingly consumption C increases. The element ∂I/∂r∂r/∂R may be negative because when R increases, interest rate r may increase slightly or unchanged, then the investment I decreases slightly or unchanged. The element ∂I/∂E∂E/∂R is positive. When R increases, the income of low-income class increases, and this will have the effect of improving the health condition, productivity, and the education opportunities of the low-income class. Then, it will heighten the efficiency factor of investment E and will lead to increase investment I. Therefore, if the negative value of the element ∂I/∂r∂r/∂R does not overwhelm30 the positive values of the first and second terms and the other elements of the third term, the numerator of (35) is positive. Hence, the following relationship is established:37 dYL/dR>0. As to the numerator of (36), the first term is negative and the element ∂I/∂r∂r/∂R is also negative. So, the sign of (36) is not clear. If the income tax of YL and YP are the same and equal to t, Eq. (31) is obtained by summing (35) and (36). The effectiveness of income equalising expenditure R When the government’s income equalizing expenditure R is implemented, its path is depicted as P4P5 (Fig. 7). This path now forms the actual growth distribution curve. From (35) and (36), the slope of the path dYL/dYP is shown as:38 dYLdYP=α′YP1-C′·1-tP+αC′+1+α∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R-α′YP1-C′·1-tL+C′+1+∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R The condition that this expenditure for income equalisation R is effective, or whether this increases the social welfare depends on the condition that the slope of the P4P5 must be steeper than that of the social welfare indifference curve W0. So, to be effective, the following relationship must hold at the point P4, i.e.:39 α′YP1-C′·1-tP+αC′+1+α∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R-α′YP1-C′·1-tL+C′+1+∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R>-∂W/∂uP∂uP/∂YP∂W/∂uL∂uL/∂YLthe slope of income equalizing expenditureRthe slope of social welfare indifference curve If this condition is satisfied, the income equalizing expenditure R is justified. If, by any chance, this condition is not satisfied, R should not be implemented. The ineffectiveness of public investment for countercyclical purpose G The effect of public investment for countercyclical purpose G on national income Y is calculated from (30), i.e.:40 dYdG=1+∂C∂Yf∂Yf∂G+∂I∂r∂r∂G+∂I∂E∂E∂a∂a∂G+∂I∂E∂E∂b∂b∂G/1-∂C∂Y-T+R1-t The denominator of (40) is positive. As to the numerator, the second term ∂C/∂Yf∂Yf/∂G is assumed to be negative. This effect is called the ‘negative future effect on consumption’.31 In the presence of the ‘crowding out effect’, the third term ∂I/∂r∂r/∂G is negative; i.e. if the public investment G increases, the interest rate will increase, and therefore ∂r/∂G is positive, and if the interest increases, the investment I decreases, and therefore ∂I/∂r is negative. The fourth term ∂I/∂E∂E/∂a∂a/∂G will be positive if public investment contributes to the development of infrastructure a; i.e. ∂a/∂G is positive, and if this improve the efficiency E of investment, ∂E/∂a is positive, and if it increases the investment I, ∂I/∂E is positive. In sum, the fourth term is positive. The fifth term ∂I/∂E∂E/∂b ∂b/∂G will be negative if public investment is done mainly for securing elections and if it is favouring pork-barrel public works projects. In this situation, the waste or futility b increases, i.e. ∂b/∂G>0, and this will decrease the efficiency E of investment, i.e. ∂E/∂b<0, and this will decrease the investment I, i.e. ∂I/∂E>0. So, the effect of the fifth term is called the ‘allocation disturbing effect’. Overall, if the extent of ‘negative future effect on consumption’, ‘crowding out effect’, and ‘allocation disturbing effect’ do not exceed those of other positive terms, the following relationship holds:41 dY/dG>0 This relationship means that the public investment for the countercyclical purpose G increases national income. However, if, due to three negative effects, the multiplier effect shown by (41) is not as large as expected, then, its value may be less than 2 and this type of public investment G increases the income only from point P4 to j, or from Y4 to Y~ in Fig. 7. The multiplier, in this case, is in approximate unity. Moreover, if the situation is such that nearly full employment is already attained and resource allocation is almost desirable, then the government’s public investment intended to win the national election will have unanticipated effects. In this situation, the extent of the ‘negative future effect on consumption’, ‘crowding out effect’, and ‘allocation disturbing effect’ may be so strong that they overwhelm other terms. Then, the following situation will hold:42 dY/dG<0. This indicates that the futile public investment G has the possibility to demolish a sound economy and decrease national income from point P4 to s, or from Y4 to Y~~ in Fig. 7 and retain the financial deficit. The effects of public investment G on the labour income YL and property income YP are also analysed. The change in G does not directly affect the ratio of YL and YP. Comparison of the effects of R and G on national income Y It may seem that the income equalizing expenditure R has a weaker effect on national income Y than the public investment for the countercyclical purpose G. This, however, may not true. As explained above, in some situations, especially in developed countries, the effect of R is not always weaker than that of G. The amount of R and G is set to be equal, and by subtracting (40) from (31), the following relationship is obtained:43 dYdR-dYdGR=G=∂C∂Y-T+R⏞first⏟++∂C∂Yf∂Yf∂R⏟+⏞second-∂C∂Yf∂Yf∂G⏟+⏞third+∂I∂r∂r∂R⏟-⏞fourth-∂I∂r∂r∂G⏟+⏞fifth+∂I∂E∂E∂R⏟+⏞sixth-∂I∂E∂E∂a∂a∂G⏟-⏞seventh-∂I∂E∂E∂b∂b∂G⏟+⏞eighth/1-∂C∂Y-T+R1-t As to the right-hand side numerator of (43),32 the first term ∂C/∂(Y-T+R) represents the marginal propensity to consume and is positive. The second term ∂C/∂Yf∂Yf/∂R refers to the effect of R on (present) consumption C through the increase in future disposable income Yf and is assumed to be positive. The third term -∂C/∂Yf∂Yf/∂G refers to the effect of G on consumption C through Yf and is assumed to be positive.33 The fourth term ∂I/∂r∂r/∂R refers to the effect of R on investment I through the increase of interest r and is assumed to be slightly negative or zero. The fifth term (-)∂I/∂r ∂r/∂G refers to the effect of G on investment I through r, and is assumed to be slightly positive or zero. The sixth term ∂I/∂E∂E/∂R refers to the effect of R on I through the efficiency factor (in investment) E, and is assumed to be positive. The seventh term (-)∂I/∂E∂E/∂a ∂a/∂G refers to the effect of G on I through the infrastructure effect a and efficiency factor E, and is assumed to be negative. The eighth term (-)∂I/∂E ∂E/∂b∂b/∂G refers to the effect of G on I through the futility effect b and efficiency factor E, and is assumed to be positive. In general, the sign of the relationship (43) is indeterminate. However, the sign may be determined by making certain assumptions. First, with a zero or negative interest rate, the crowding-out effect may be zero or negligible. Then, the fourth and fifth terms may be considered as zero. Second, the seventh term is the infrastructure effect and this may be strong in developing countries.34 In developed countries, however, this effect may not be stronger than that in developing countries because in developed countries many fundamental infrastructures are already constructed. So, the marginal efficiency of public investment G on infrastructure is low. If some new firms succeed in making innovative products, the infrastructure needed to make or use innovative products may be in many cases, constructed by these firms. The government’s role in these situations may not be to construct new infrastructure but to provide new legislation to manage the new situation. This provision of new legislation is not considered in this study. We can conclude that the infrastructure effect in developed countries is not very strong. So, if the extent of the infrastructure effect, i.e. the seventh term does not overwhelm the other terms, the effect of R on Y surpasses the effect of G on Y, i.e.:44 dY/dR>dY/dG,R=G This tendency is strengthened when the eighth term is strong, or when public investment G is focused on pork-barrel public projects. Concluding remarks Inequality seems to be almost inevitable in the process of economic growth, especially when considering the income distribution inequality. From the end of twentieth century, globalization and tremendous breakthroughs in information and communication technology have had profound effects on the economy as well as society (divided into the extremely rich, poor, and dwindling middle class). To deal with this issue, economics proposes an increase in taxes. Tax increases, especially, the increase in progressive tax on high income and property is essential for reducing inequality. Tax increases, however, must confront political opposition from the wealthy classes. This situation is compared to the case of land reclamation. To make a low land higher, the earth and sand of higher places are dug and conveyed. By this, the difference of altitude will decrease. If in this case the owners of higher places do not oppose much, the project will succeed. If, however, they strongly oppose, it will face difficulties. In general, if the scale of the project becomes larger, the difficulty of gathering much earth and sand will increase more. In reality, almost every people does not like tax increase. If heavy taxation oppresses entrepreneurship, it will hinder economic growth. Furthermore, tax increases have an upper limit. To offset the enormous accumulated financial deficit, I think the policy of tax increase is inadequate and a thorough new policy is required. This situation is compared to the case of large scale urban development. If this causes considerable volume of surplus soil from the construction site, that soil may be conveyed and used to cover over the lower land. The difference of altitude between the lower land and higher places decreases. In this case, as the higher places are not dug, the owners of it do not oppose. If the enormous accumulated sum of government bonds is intact, the continued large-scale government expenditure on social security, supply of public services, and countermeasures to pandemics and so on, may cause serious problems in the future. If a world war, civil war, terrorist event, or financial crisis happen, along with the behaviour of speculators, the national bonds market as well as the whole market may become turbulent. This will cause a decrease in consumption and investment and will lead to a long and severe depression. Government expenditure may be unable to cover this decrease in total demand because of the turbulent government bonds market. The price of government bonds may fall considerably. To avoid these scenarios and heavy tax increase caused by the enormous accumulated government bonds, this study proposes to erase the government bonds which are owned by the central bank. The erasure of government bonds means the disappearance of the possibility of a government default and also a decrease in the supply of government bonds, leading to a price increase of it and decrease in the interest rate. The features of erasure policy are summarized. First, the erasure policy will maximise the social welfare function which involves both altruism and egoism. The desirability of each income distribution state is judged according to this welfare function. The main purpose of this erasure policy is to accomplish widespread income equality which traditional tax policies have not realized. Second, as this policy devotedly expends for the low income groups, it enables low-income groups to obtain the chance to heighten their overall human capital or their basic capability and thereby increase future earning capacity and enhance their quality of life. Then, the productivity of the economy increase and the national economy will grow. Third, as this policy continually erases the government bonds which the central bank owns, and expends some ratio ϕ(1>ϕ>0) of the sum, the remaining government bonds decrease. This tendency is strengthened by the increase in tax revenue caused by the multiplier effect of the expenditure. Fourth, different from tax increase policy, this policy does not inflict any burden to the people and firms. Therefore, this policy neither reduces the incentive of anybody nor that of entrepreneurs to make efforts to achieve more. In this sense, this policy is feasible. Fifth, this policy is intended to implement gradually. The process of further erasure and expenditure is divided into a decreasing geometric series. By this, the changes are made smooth. The economic structure is easy to correspond to the change. Moreover, the recipient of government expenditure can act in a carefully planned way. Sixth, the grounds for concern must be noted. One is the apprehension regarding inflation. As this erasure policy increases the money supply, the demand side pressure created by it always exists. If the inflationary pressures are low, the erasure policy should be implemented. If the inflationary pressure is moderate, the policy should be implemented gradually. If the inflationary pressure is high, the policy should be suspended until the pressure becomes moderate. As this policy is implemented gradually, the government is easy to respond to the inflation. I think the supply side effect created by the productivity increase of the recipients of government expenditure will contribute to hold back inflation. Seventh, another limitation is that the erasure policy may have the possibility to effectuate a lax financial policy. As mentioned earlier, the expenditure financed by the erasure policy should be strictly applied only for income equalisation. If it is applied to pork-barrel public works projects, it worsens income distribution and, to make matters worse, the efficiency of the economy will decline and the economy itself will become stagnant. Acknowledgements I appreciate the kind and precise advices of the reviewers and the editor of this journal. Author contributions I am a single author and no other author contributed to this manuscript. Funding I have not obtained any funds at all. Data availability I cited the fundamental data on GDP and Government Bonds from the home pages of Ministry of Finance, JAPAN, and Bank of Japan. Code availability Microsoft Word. Declarations Conflict of interest I have no conflicts of interest/competing interest concerning to my manuscript and my research. Ethical approval This manuscript is submitted only to SN Business & Economics. This manuscript is original and have never been published elsewhere in any form or language. No data, text, or theories by others are presented as if they were the author’s own. I swear that I followed all the ethical responsibilities of an author. Consent for publication I consent to publish my manuscript. 1 As Pigou’s proposition entailed the interpersonal comparisons of utility, the criticism and reinforcement on his theory are presented much. See, for example, Robbins (1932), Bergson (1938), Kaldor (1939), Scitovsky (1941), Samuelson (1947) and Harsanyi (1955). 2 Neoclassical economics believes in the market mechanism, or working of flexible price system to solve many economic problems. 3 With regard to the unemployment and inequality, Keynes wrote as follows: “The outstanding faults of the economic society in which we live are its failure to provide for full employment and its arbitrary and inequitable distribution of wealth and income. ….” See, Keynes (1936), Chapter 24, Page 372. 4 Basic capability may be interpreted as the human’s ability to attain what the human desires. 5 A joint paper with Thomas Piketty and Emmanuel Saez. 6 A joint paper with Emmanuel Saez. 7 From the social welfare maximisation conditions, the sign of Δ is positive. 8 Adam Smith (1776) indicated that self-interest is the driving force of the development of market economy and, before this, he also had pointed out that people behave conscious of the other’s sentiment or judgment (1759). In the analysis of human behavior, Smith esteemed the sympathy as the most influential factor. The concept of sympathy has the affinity with some sort of altruism. 9 It must be stressed that altruistic donation between the destitute or members of low income group plays more valuable roles than those of the wealthiest to mitigate the inequality. This kind of donation enhances the essential values of their lives. 10 As it is assumed that the point H is very close to the beginning point h, hereafter these two points are considered as indifferent. 11 At the point H, the indifference curve uC the level of which is lower than uB intersects the income line yByB. 12 In 2021, GDP of Japan is about 4.9 trillion dollars. (This is calculated at the rate of 112 yen to the US dollars.). 13 At that time, the amount of government bond and T-bill outstanding is approximately 10.9 trillion dollars. This is 2.2 times as large as GDP. 14 To put it accurately, this 5 trillion dollars involves 0.1 trillion dollars of government bonds which is not erased and owned still by the central bank. To simplify explanation, this sum is ignored. 15 ∂R/∂ϕ=A(1+ϕ2)/(1-ϕ2)2>0 and ∂2R/∂ϕ2=2Aϕ(1-ϕ2)2+2(1-ϕ4/(1-ϕ2)4>0. 16 These are shown as, ∂S/∂ϕ=A/(1+ϕ)2>0 and ∂2S/∂ϕ2=-2A(1+ϕ)/(1+ϕ)4<0. 17 As the slope of curve R in Fig. 8 is positive and increasing, the slope of curve T=mtR in the Fig. 9 is also positive and increasing. 18 From the definition N=S-T, the following expression is obtained, i.e.∂2N/∂ϕ2=∂2S/∂ϕ2-∂2T/∂ϕ2 where, ∂2S/∂ϕ2=-2A(1+ϕ)/(1+ϕ)4<0,∂2T/∂ϕ2=2tmAϕ(1-ϕ2)(3+ϕ2)/(1-ϕ2)2>0. then, ∂2N/∂ϕ2<0. 19 In the case∂N/∂ϕ=Aϕ21-tm-2ϕ+(1-tm)/(1-ϕ2)2=0, only the value ϕ=0.33 satisfies this first-order condition (where, m=2.00 and t=0.20 are assumed). 20 When R is more than 17.21 trillion dollars, Σ becomes negative. This negative value represents the financial surplus. 21 To mine these, however, close attention to the environment is needed. 22 I recognize that as of March 2022, the worldwide inflation is prevailing. 23 If the pandemic or disaster is severe, I think the inflation rate a little more than 2.00%, that is, around 3.00% may be permitted. 24 The causes of recent inflation are reduction of the supply of crude oil induced by the global movement of carbon neutral, outbreak of pandemic and the war of aggression. 25 It must be noted that because the amount of R may not always be sufficient to attain socially desirable level of equality, or, cannot reach within the region of socially permissible income distribution, the word “income equalization” means, in these cases, the direction to reach the goal. 26 I assume that public investment for counter cyclical measures G may have positive effect on current national income Y, but will increase the government debt and thereby bring about future tax increase. So, G decrease the future income Yf. 27 Interest rate r is influenced also by the money supply of the central bank and many market factors. To simplify the analysis, these are assumed to be given, but, instead, the parameter R plays the central role in my model. 28 Because R is expended after the large scale erasure of government bonds, or, the price of bonds are high and interest rate is forced to be the lowest, the value of fraction ∂r/∂R may be nearly zero or zero. The term ∂I/∂r is negative. In sum, this term is slightly negative, or zero. 29 So, when the value of R increases, the slope of tangent line becomes steep. 30 As mentioned earlier, the part of this element ∂r/∂R may be nearly zero or zero. 31 It is known that almost every public investment for countercyclical purpose leaves financial deficit and decrease future income. 32 The denominator of right side (43) is positive. 33 It must be noted that the fraction ∂C/∂Yf∂Yf/∂G is assumed to be negative. 34 In developing countries, the amount of infrastructure may not be enough. So, in this situation, the government investment for building the infrastructure may be effective. ==== Refs References Atkinson AB The Economics of Inequality 1975 Oxford University Press Atkinson AB On the Measurement of Poverty Econometrica 1987 55 749 764 10.2307/1911028 Atkinson AB Thomas piketty, and Emmanuel Saez, “Top incomes in the long run of history” J Econ Lit 2011 49 1 3 71 10.1257/jel.49.1.3 Atkinson AB Inequality: what can be done? 2015 Harvard University Press Bergson A A reformulation of certain aspects of welfare economics Quart J Econ 1938 52 310 334 10.2307/1881737 Deaton A The Great Escape: Health, and the Origins of Inequality 2013 Princeton, NJ Princeton University Press Friedman M Capitalism and Freedom 1962 The University of Chicago Press Friedman M The optimum quantity of money and other essays 1969 Aldine Publishing Company Harsanyi JC Cardinal welfare, individualistic ethics, and interpersonal comparisons of utility J Polit Econ 1955 63 309 321 10.1086/257678 Kaldor N Welfare propositions of economics and interpersonal comparisons of utility Econ J 1939 49 549 552 10.2307/2224835 Kelton S (2020) The Deficit Myth Modern Monetary Theory and the Birth of the People’s Economy, Public Affairs, New York Keynes JM The general theory of employment, interest, and money 1936 London Macmillan Kuznets S Economic growth and income inequality Am Econ Rev 1955 45 1 28 Pigou AC (1920) (1971) The Economics of Welfare, 4th ed., London Macmillan Piketty T Saez E Income inequality in the United States, 1913–1988 Quart J Econ 2003 118 1 39 10.1162/00335530360535135 Piketty T Capital in the twenty-first century 2014 Cambridge, MA Harvard University Press Ravallion M The economics of poverty, history, measurement, and policy 2016 Oxford University Press Robbins L The nature and significance of economic science 1932 London Macmillan Samuelson PA Foundations of economic analysis 1947 Cambridge, Massachusetts Harvard University Press; Enlarged Second Edition 1983 Scitovsky T A note on welfare propositions in economics Rev Econ Stud 1941 9 77 88 10.2307/2967640 Sen AK On economic inequality 1973 Oxford Clarendon Press and New York, Norton Sen AK Choice, Welfare and Measurement 1982 Basil Blackwell Smith A (1759) The theory of moral sentiments, A. Millar, Strand; and A. Kincaid and J. Bell, Edinburgh Smith A (1776) An Inquiry into the Nature and Causes of the Wealth of Nations, Edinburgh, "new edition" Glasgow (1805) Wray LR (2015) Modern money theory: the primer on macroeconomics for sovereign monetary systems, second edition	0	PMC9734554	NO-CC CODE	2022-12-14 23:28:29	no	SN Bus Econ. 2023 Dec 7; 3(1):4	utf-8	SN Bus Econ	2,022	10.1007/s43546-022-00331-1	oa_other
==== Front SN Bus Econ SN Bus Econ Sn Business & Economics 2662-9399 Springer International Publishing Cham 331 10.1007/s43546-022-00331-1 Original Article An income equalisation policy and social welfare: beyond inequality http://orcid.org/0000-0002-4511-7806 Teramoto Hiroaki [email protected] grid.443705.1 0000 0001 0741 057X Master of Economics, Hiroshima Shudo University, Hiroshima Shudo Daigaku, Hiroshima, Japan 7 12 2022 2023 3 1 413 6 2021 31 8 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 inequality of income distribution is one of the most serious problems to our society. Traditional economics proposes the tax increase on wealthy classes and transfers its revenue to low income group. This is the most appropriate one. But, the total of this and donation will unfortunately be insufficient. For procuring funds to ease inequality, I propose to erase the government bonds which are already purchased and owned by the central bank. The government, then, issues new government bonds (the total sum of which should be less than erased one) to expend for equalization. The amount of new government bonds is determined from the maximisation of social welfare function concerning to income distribution. By this policy, first, the socially optimal equalization will be attained; second, through the productivity increase caused by the expenditure for low income group, the national economy will grow; third, because of the erasure process, the government bonds outstanding will not increase. In these days of pandemic, much more government expenditures are needed and thereby government deficits are piled up. I think the valid policy is not the heavy tax increase, but the execution of erasure of government bonds outstanding and to equalize income distribution. Keywords Inequality of income distribution Maximisation of social welfare Erasure of government bonds Government expenditure for equalization Pandemic Market Economy JEL Classification D31 E58 H53 issue-copyright-statement© Springer Nature Switzerland AG 2023 ==== Body pmcIntroduction In the history of humankind, political and economic inequalities have always coexisted. At the end of the eighteenth century, the citizen’s revolution alleviated political inequality, which gradually spread across the globe. After the end of absolutism and advent of the industrial revolution, people expected economic equality and the prosperity of ordinary workers, and considerable effort was spent to achieve it. However, economic inequality still exists. Economic society can resolve the problem of inequality of income and properties. Economics is an effective discipline having a wide range. The study on the allocation and distribution of scarce resources is an important field. The studies on resource allocation have obtained excellent results, but the studies on distribution have not achieved the desired results despite the strenuous efforts of economists. To rectify income and property inequality, economics mainly proposes tax increases or the increase of the rate of progressive tax. This is the most appropriate and authentic measure that has successfully reduced inequality. However, inequality still persists and the policy of tax increase cannot solely resolve the ongoing inequality. As the tax increase policy faces the strong political and economic oppositions and, moreover, has an upper limit, the inequality problem will exist semi-permanently. In this study, a new policy is proposed that can resolve the problem of economic inequality. This policy suggests discontinuing the existing government bonds with the central bank, and then issuing new government bonds to expend for income equalisation. The amount of new issue will be less than the discontinued sum and will be determined from the standpoint of social welfare maximisation. The process of erasure and expense is devised to repeat at a declining rate. By this policy, the income distribution approaches the socially optimal point. The expenditure for income equalisation will increase people’s income which in turn will increase the tax revenue, thereby reducing the government deficit. This study proposes an alternative to the existing policy of progressive taxation for reducing income inequality. I believe that this study makes contribution to the literature because it discusses in detail the erasure policy that can accomplish absolute income equality which the traditional tax policy has not realized. As this erasure policy faces the problems of inflation and lax financial policy, close attention and deliberate countermeasures are indispensable. The causes of present inequality The inequality pressure from outside The prevalence of the market economy in most parts of the world led to the acceptance of direct investment and the newest technology by developing countries, which enabled these countries to produce high-quality goods with low costs along with low wage rates, low rent, and diligence. They export these goods to industrialised countries. Thus, both sets of countries benefit from international trade. However, the domestic firms of industrialised countries must compete with the imports that are of high quality and low price; otherwise, employment and the general wage rate will decrease. As a result of globalization and economic liberalization, the people of developing countries move to high-wage countries. This is advantageous to firms of industrialised countries but disadvantageous to workers who have been employed on high wages. The income of middle class workers who stably supported society has considerably decreased, resulting in a dwindling middle class. The inequality pressure from inside Currently, many start-ups have been founded in the fields of Internet of Things (IoT), artificial intelligence, biotechnology, and others. Some of them have grown into worldwide enterprises which have de facto standards. These enterprises supply goods and services which enhance the productivity of economy tremendously and therefore become the indispensable items for society to progress. These high-quality goods are improved continuously, and the real price of these have a tendency to decline. This also contributes towards the economy. The founders, investors, and entrepreneurs of these companies gain enormous income and wealth as a reward. These innovative enterprises usher changes to the traditional old-fashioned companies and in some cases induce withdrawal or bankruptcy. This leads to the increase in company efficiency; however, it also increases the number of unemployed people and decreases wages. Most of the rising innovative enterprises are fabless and outsource production to foreign manufacturers, leading to a decrease in domestic employment. Thus, a rise in innovative enterprises causes income disparity as a few successful persons become extremely rich; however, workers’ average income decreases gradually, and at worst, causes a rise in unemployment. The COVID-19 pandemic The recent COVID-19 pandemic has aggravated the unequal distribution of income. To prevent the spread of the pandemic, many countries declared a state of emergency, affecting international as well as domestic economies. Numerous workers lost their jobs. The benefits provided by the government to the workers have not been enough. Low- or middle-income patients suffering from the long-term symptoms of the pandemic often lose the ability to work. The disparities between the wealthy and unemployed or sick people are increasing, leading to an increase in the government deficit. Many economists insist that tax increases are inevitable to restore the soundness of the government budget. Literatures on inequality Market economy and democracy have historically overcome many obstacles. There are various distinguished contributions to the analysis of inequality. In the era, the bourgeoisie was on the rise, Adam Smith (1776) formulated the significant concept that through the working of the market mechanism (and the adoption of the division of labour system), the capitalist class, the land-owing class, and the working class can live harmoniously. This concept conveys that the market economy does not bring about severe inequality. But reality contradicts this. Workers’ long-lasting struggles to fight inequality still continue. Pigou (1920) founded the welfare economics and in it he insisted that equalisation of national dividend (or national income) will, ceteris paribus, increase the economic welfare. After his analysis, welfare analysis on income distribution developed.1 Seeing the unprecedented massive unemployment caused by the Great Depression (started from 1929), Keynes (1936) had a doubt about the effectiveness of neoclassical economics2 to solve it. Massive unemployment is one of the most serious threats to the society and to the income inequality.3 He founded the macroeconomics. In it he affirmatively evaluated the activities of entrepreneurs and workers, and placed low value on the activities of the people who does not contribute to production. A representative example of the latter is the rentier. Keynes considered that the rentiers did not contribute to the recovery of Britain’s economy at that time. Kuznets (1955) examined economic growth and income distribution. The ‘Kuznets Curve’ has been one of the bases of inequality analysis. Although Friedman (1962) believed in the efficient working of the market, he did not fail to notice the inequality of income and property. He advanced the system of negative income tax and mentioned that the determination of the minimum income level of this system depends on the sum the society can subsidize. He was also sensitive towards the inequality brought about by the market. Sen (1973; 1982) insisted that both utility level and the level of basic goods are not appropriate for measuring welfare. Instead, he proposed the concept “basic capability4” and used it to revise inequality. Atkinson’s empirical studies (1975; 1987; 20115) elucidate and tackle the inequality problem. He also published a book (2015) that proposed practical policies to rectify inequality. Deaton’s approach (2013) involves the relationship between health and wealth, and turns attention towards people of developing countries. Piketty’s studies (20036; 2014) cover a long span and criticize Kuznets’ findings, stirring debates and reinforcing people’s interest. Regarding the policy prescription for income equalisation, Friedman’s concept (1969) of ‘helicopter money’ can be applicable to the problems of inequality as well as deflation. Modern monetary theory (MMT) proposed by Wray (2015) and Kelton (2020) among others, presents a new way of understanding public finance and monetary theory. This theory observed the behaviour of the Bank of Japan (BOJ; central bank) and financial market. A government’s budget deficit is not the reason to revise unequal income distribution. Ravallion (2016) made an intensive study on the economic inequality. The study is structured as follows: the contents of the policy are explained in (“Economic growth and social welfare”, “Erasure of government bonds”, “The pandemic and income equalisation”, “Apprehension regarding inflation” and “The effectiveness of income equalizing expenditure R”). The next section elucidates (“Traditional measures of equalisation”), and the subsequent section describes the (“Multiplier analysis”). (“Concluding remarks”) concludes the paper. Economic growth and social welfare Social welfare function concerning income distribution This study presents a normative analysis of the relationship between economic growth and income inequality. The variables of this model are those of macroeconomics. The social welfare function developed by Bergson (1938) and Samuelson (1947) is applied to consider the problem. From the standpoint of economic policy, this function is maximised, and the method of analysis is that of microeconomics. Therefore, this model is developed by the combination of macroeconomics and microeconomics. The national income of a country is denoted as Y, and is assumed to be composed of labour income wL (where w: wage rate, L: labour), capital income rK (r: interest rate, K: capital), and entrepreneurs’ income or profit, Π, i.e. Y=wL+rK+Π. The national income Y is divided into two parts: labour income wL denoted as YL; and the sum of capital income rK and entrepreneurs’ income (profit) Π, termed as property income and denoted as YP. That is:1 YNationalincome==wLLabourincome++rKCapitalincome++ΠEntrepreneurs'incomeprofitYNationalincome==YLLabourincome++YPPropertyincome This is illustrated in Fig. 1, where the labour income YL is measured along the vertical axis and property income YP along the horizontal axis. The national income Y is depicted as lines, Y1Y1,Y2Y2, Y3Y3,⋯, according to the income level Y1,Y2,Y3,⋯. The line YiYi(i=1,2,3,⋯) is called the ‘national income line’. The slope of this line is − 45°, and it shifts above to the right as the economy (national income) grows.Fig. 1 Socially equal distribution The workers’ utility from labour income YL is denoted as uLYL, and capital owners’ and entrepreneurs’ utility from property income YP is denoted as uP(YP). Then, the social welfare function W concerning income distribution is formulated as:2 W=WuLYL,uPYP The social (welfare) indifference curves concerning income distribution are derived from (2), and depicted as W1,W2,W3 in Fig. 1. As to the relative height of the utility, it is assumed that inequality W1<W2<W3 holds. The slope of these curves is shown as:3 dYLdYP=-∂W/∂uP∂uP/∂YP∂W/∂uL∂uL/∂YL. If social welfare (2) is maximised subject to the income distribution constraint (1), the optimal conditions are derived as:4 ∂W∂uL∂uL∂YL=-λ,∂W∂uP∂uP∂YP=-λ,YL+YP=Y The conditions (4) are the points of contact of the national income lines with social indifference curves and are shown as E1,E2,E3. As national income Y increases, or as the economy grows, the point of contact moves above to the right, and the path is depicted as the optimal growth distribution (OGD). The locus of this curve is shown as:5 dYLdY=WLPuP′uL′-WPPuP′2+WPuP″ΔdYPdY=WPLuP′uL′-(WLL(uL′)2+WLuL″)Δ where WL=∂W/∂uL,WP=∂W/∂uP,WLP=∂2W/∂uP∂uL,WPL=∂2W/∂uL∂uP, WLL=∂2W/∂uL2,WPP=∂2W/∂uP2,uL′=∂uL/∂YL,uP′=∂uP/∂YP,uL″=∂2uL/∂YL2,uP″=∂2uP/∂YP2 andΔ=WLPuL′uP′+WPLuP′uL′-WLLuL′2+WLuL″+WPPuP′2+WPuP″>0.7 Equation (5) shows the social optimal economic growth-income distribution curve OGD, which represents the locus of socially optimal income distribution when the economy grows. On both sides of this OGD curve, the socially permissible border lines regarding income inequality are depicted. The border line of the minimum permissible labour income, from the standpoint of social welfare, is depicted as LL in Fig. 1, and the property income is depicted as LP. Regarding the slope of social welfare indifference curves (3), the sign of terms ∂uL/∂YL and ∂uP/∂YP is assumed to be positive. The sign of terms ∂W/∂uL and ∂W/∂uP must be examined. Three regions are classified:6 RegionI: regionbetweentheLLlineandLPline∂W/∂uL>0,∂W/∂uP>0RegionII: regionrighttothelineLL∂W/∂uL>0,∂W/∂uP<0RegionIII: regionlefttothelineLP∂W/∂uL<0,∂W/∂uP>0 Expression (6) shows that in Region I, income is distributed equally or is permissibly equal. Therefore, the increase in workers’ utility uL due to increase in income YL is considered the increase in social welfare, ∂W/∂uL>0. The same is applied to the increase of utility of capitalists and entrepreneurs uP, i.e. ∂W/∂uP>0. In Region II, income is distributed unequally (i.e. too little to the workers and too much to capitalists and entrepreneurs). Therefore, the increase in workers’ utility is considered as the increase in social welfare, ∂W/∂uL>0, and that of capitalists and entrepreneurs is considered as negative, ∂W/∂uP<0. In Region III, income is distributed contrary to the Region II. Then, the relationships ∂W/∂uL<0 and ∂W/∂uP>0 are supposed to hold. From expression (6), the slope of the social welfare indifference curves (3) can be shown as:7 RegionI: theslopeisnegative,i.e.dYL/dYP<0RegionIIandIIItheslopeispositive,i.edYL/dYP>0. Expression (7) in Fig. 1 can be explained as follows: in Region I, the income distribution is equal or permissibly equal, and so, from the social welfare perspective, the distributed incomes YL and YP are goods, and the slope of the indifference curves is negative. In Region II, the income distribution to YL is too little and that to YP is too much. So, YL is goods and YP is bads. In this situation, if YP increases, to keep the level of social welfare constant, YL must be increased. Therefore, the slope of the indifference curves is positive. In Region III, YL is bads and YP is goods. So, the slope of indifference curves is positive. The curvature of the social welfare indifference curves is assumed as smooth and convex to the origin 0. The actual growth distribution curve and socially equal economy The OGD curve which is socially optimum does not always coincide with the actual one. The actual growth distribution curve is denoted as AGD and depicted in Fig. 1. When the national income level is Yi(i=1,2,3,⋯), the actual income distribution between YL and YP is plotted as pi on the national income line YiYi. The locus of these points p1,p2,p3 constitutes the AGD curve. As this curve passes through the Region I which is between the two permissible borders LL and LP, the income distribution of this economy is said to be socially (more or less) equal. Socially impermissible extremely unequal economy An extremely unequal economy is assumed next and explained in Fig. 2. Here, only two social groups, α and β, are analysed. The rest, whose income is stable, is assumed to be constant and omitted from the analysis. Group α constitutes the people who formerly belonged to the middle class, but owing to economic changes and the immigration policies of the economy are facing challenges. Group β constitutes successful innovative entrepreneurs and well-off property owners. In Fig. 2, Group α’s income is measured along the vertical axis and that of Group β’s in the horizontal axis. The social welfare indifference curves are depicted as w1,w2, and w3. The total income line of Groups α and β is depicted as y1y1,y2y2,y2′y2′, and y3y3. (This total income excludes the stable incomes of the rest of the people). The social OGD curve is depicted as ogd, and socially permissible borders are depicted as lα and lβ. The AGD curve is depicted as a sharply bent curve, agd. This curve means that in the process of increase of total income from y1 to y2′, the people in Groups α and β gained considerable equal shares of income. This corresponds to the point p1,p2, and p2′ on the agd curve. When the total income increases from y2′ to y3, the points which represent income share move from p2′ to p3. This movement shows that Group α’s income share decreased disadvantageously.Fig. 2 Socially impermissible extreme distribution We consider the relationship between the increase in income level and social welfare of Group α. When the economy grows and total income increases from y1 to y2′, Group α’s income also increases and its income distribution is within the socially permissible borders lα and lβ. In this process, the social welfare increases and approaches the highest attainable level w2. The situation, however, changes remarkably when the total income exceeds y2′. If the total income increases to y3, the social welfare decreases from w2 to w1. The income share of this point is represented by p3, and this point is located much farther to the right of the permissible border line lα. This means that, from the social welfare perspective, the income shares of Group α is extremely low compared to Group β. If the total income increases further, the situation becomes more complex. Group α may resent the economic growth in a market economy where they are not rewarded. Traditional measures for equalisation Several policies have been presented for the equalisation of income distribution. Two of these policies are discussed here and thereafter the new policy is explained. The royal road Progressive taxation of high-income classes and transfer of its revenue to low-income classes is a well-known policy for the equalisation of income distribution (Fig. 3). As the economy grows, the national income Y on the AGD curve is divided between YL and YP. The actual income share is shown as points P0,P1,P2, and P3 on the AGD curve.Fig. 3 Taxation and transfer The point P3 is set as the starting point of analysis. The point represents the income distribution between YL and YP when the national income is Y3. Because this point P3 is considerably to the right of the socially permissible border LL, from the social welfare perspective, the redistribution of income is desirable. At the income level of Y3, the social welfare is maximised at the point E3. To reach this point, it is necessary to tax as much as P3A on the property income and transfer the same amount of tax revenue AE3 to labour income. This thoroughgoing measure realizes the maximum social welfare. The economic value of this progressive taxation and its transfer is inferred as follows: the starting point P3 and attained point E3 are both on the same national income line Y3. The point P3 is on the social welfare indifference curve W1 and its level of social welfare is the same as E1 on the OGD curve. So, although point P3 is on the Y3 line, its economic value from the standpoint of social welfare is equivalent to the income value of Y1. Contrary to the point P3, the attained point E3 is on the social welfare indifference curve W3 and is on the OGD curve. Therefore, the economic value of the progressive taxation and its transfer has the economic value shown by the distance E1E3 or the income difference Y3Y1. This policy is, indeed, extremely efficient to attain maximum social welfare and may be called as ‘the royal road’ to income equalisation. The royal road policy is ideal for the economy. The rich capitalists (or propertied) class as well as successful entrepreneurs will, however, strongly oppose this policy. One of the most persuasive reasons for opposition is that unlimited progressive taxation will destroy the social order, or extraordinary heavy tax will break their incentive to innovate. If these oppositions are strong, a compromised, downgraded policy is implemented. In this situation, the extent of P3a is taxed and its revenue of ab(=P3a) is transferred. The inequality of income distribution is, then, slightly revised. The economic value of this compromised policy is inferred as the distance E1E2 or the income difference Y2Y1. These down-sized policies are adopted in almost every industrialised economy. But, in many cases, it may not be enough to cope with the ongoing inequality. Donation (or contribution) In a market economy, entrepreneurs plan the production and investment scheme to maximise profit, while consumers plan the expenditure and time-allocation scheme to maximise utility. These behaviours come from their self-interest or egoism and are the driving forces of a market economy.8 Altruism (or philanthropy) At the opposite extreme, there may be altruism or philanthropy. The development of the market economy till now had caused many harmful effects. Nevertheless, it is well known that capitalists, entrepreneurs, and wealthy persons have donated a considerable part of their earnings and properties. The motives for their donation can be classified into three types. First is the pure or genuine altruism motive, such as human welfare and religious sentiments. Second is that of ordinary altruism such as donations or the establishment of foundations with their names. Finally, there is impure or non-genuine altruism such as vanity or pomposity. Currently, due to remarkable innovations, many multi-billionaire entrepreneurs donate their vast wealth to society or establish foundations across the world. These foundations have far-reaching effects. The advancement of art, culture and sciences, support to education and medical activities, and eradication of poverty are well-known examples. Donations based on the motive of altruism is considered in Fig. 4. The starting point of analysis is set at point c where the labour income is YLc and property income is YPc. Here, the existence of property owners’ (i.e. the capitalists’ and entrepreneurs’) utility function U=U(YP,YL) is supposed, and their income constraint is shown as YP+YL=YPc (YPc: constant). Then, at the point c, the property owners’ utility maximisation problem is formulated as:8 maxU=UYP,YL,subject\,toYP+YL=YPcYPc:constant Fig. 4 Altruistic (or Philanthropic) donation If the property owners have no concern for the income level of labour YL, or consider the existing level of labour income YLc to be enough, the solution of (8) is a corner solution and attained at point c. In this case, the non-altruistic property owners do not donate, and the income distribution is unchanged. Conversely, if the altruistic property owners consider that the existing income distribution is unequal and disadvantageous for labour, they will donate their income to labour. Then, the solution of (8) is shown as:9 ∂U/∂YL/∂U/∂YP=1,YP+YL=YPc In this situation, the altruistic property owners choose, for example, the point d where the property owners’ indifference curve (derived from their utility function U) Ua is tangential to the national income line Y3Y3. The altruistic property owners donate their income cf(=YPcYPd), and labour receives this. So, the labour income increases to fd(=YLdYLc). By virtue of the property owners’ donation, the point of income distribution changes from c to d. Thus, on one hand, the level of social welfare increases from W1 to W2, and, on the other hand, the altruistic property owners’ utility also increases from U¯a to Ua. The increase in the utility of property owners is shown in terms of income. The property owners’ indifference curve U¯a is tangential to the income line YaYa at point g, which is parallel to Y3Y3. Then, the distance gd represents the increase of property owners’ income and is equivalent to YaY3. This income increase may be called as ‘the altruistic donator’s benefit’. The increase in social welfare is also measured in terms of income. As the social welfare increases from W1 to W2, the corresponding equivalent value increases from E1 to E2, meaning that social welfare in terms of income increases to Y1Y2. In Fig. 4, the point d after donation, however, is located to the right of the labour’s socially permissible border line LL. This indicates that although property owners’ donations are valuable, it is insufficient to attain socially optimal income distribution.9 Egoism (or self-protection) Over time, if the income inequality reaches an extreme, people who become desperately poor or have long been in dire straits will feel strong resentment towards the society, market economy, capitalists, entrepreneurs, and the propertied people. This will lead to social unrest across the country. In such situations, some of the wealthiest people may individually or collectively try to prevent the occurrence of the worst case scenario (such as riots or revolutions) by donating a part of their wealth to the people in need or destitute workers (hereafter, these people are denoted as the destitute). The behaviour of the wealthiest people is analysed in Fig. 5. The destitute is classified as Group α, and the wealthiest is classified as Group β. In this analysis, only two Groups α and β of the society are examined and the rest, whose income is stable, is assumed to be constant and omitted from the analysis.Fig. 5 Egoistic (or Self-protective) donation; social unrest effect, utility effect, and income effect The starting point in the Figure is set at point h (on the agd curve) where the destitute α’s income Yαh is extremely low and that of the wealthiest group β is Yβh and is extremely high. First, the existence of the utility function of the wealthiest U=U(Yα,Yβ,I) is supposed, where Yi(i=α,β) is the income of group i, and I is the indicator of social unrest brought about by income inequality. This indicator reflects problems between the destitute and wealthiest, demonstrations against income inequality, a riot to achieve economic justice, and so on. Although the indicator I is a function of income inequality, to simplify the analysis, it is assumed as a parameter. In the figure, the indifference curve UA of the wealthiest group is depicted. At point H, which is very close to the neighbourhood of h,10 the utility of the wealthiest is maximised. At point H, the indifference curve of the wealthiest UA is tangential to the income line yByB. This maximisation problem is formulated as:10 maxU=U(Yα,Yβ,I),subjecttoYα+Yβ=Yβh(Yβh:constant), The solution is shown as:11 ∂U/∂Yi=-λ,Yα+Yβ=YβhYβh:constant, where λ is a Lagrange multiplier. The wealthiest β at first satisfies the situation shown by point H. This point, however, represents the spread of social unrest and the threat felt by the wealthiest β. If the indicator I which reflects social unrest increases, their utility distinctly decreases (i.e., ∂U/∂I<0), and the shape and location of the utility function U of the wealthiest change and thereby their indifference curves also change. The indifference curve UA changes to uA. After the increase of the social unrest indicator I, the point H is not the optimal point for the wealthiest.11 They seek the point which maximizes their utility and will find that j, where the new indifference curve uB is tangential to the income line yByB, is the new optimal point. Accordingly, the wealthiest people donate their income by hk(=YβhYβj), and the destitute receive the same amount as kj(=YαhYαj). This movement of optimal point from H to j is called ‘the social unrest effect’. It is disintegrated into two parts, i.e. from H to m, and from m to j. The first movement is that between the points at the same utility level. This movement is called ‘the utility effect’ (of the increase of the social unrest indicator I). The second is ‘the income effect’ (of the indicator I). The total and net loss incurred by the wealthiest due to the extreme concentration of wealth which caused social unrest can be considered in the same figure. The total loss is shown as the movement from m to j and j to n. This loss is expressed in income terms as yAyC. Because the wealthiest donated their wealth to Hk to mitigate social unrest, their utility level increased from uC to uB. This leads to the decrease in total loss as yByC. Then, the net loss caused by the social unrest is shown as yAyC-yByC or yAyB, and this net loss is equal to ‘the income effect’ of the increase in social unrest I. The relationships among ‘the social unrest effect’, ‘the utility effect’, and ‘the income effect’ are shown mathematically as follows: The social unrest effect is derived from the optimal conditions (11):12 dYα/dIYβh:const.=UαI-UβI/ΔIdYβ/dIYβh:const.=-UαI-UβI/ΔI where Uαβ=∂2U/∂Yβ∂Yα,Uβα=∂2U/∂Yα∂Yβ,Uαα=∂2U/∂Yα2,Uββ=∂2U/∂Yβ2,UαI=∂2U/∂I∂Yα, and UβI=∂2U/∂I∂Yβ. The term ΔI=Uαβ+Uβα-Uαα-Uββ is positive from the condition of utility maximisation of the wealthiest. The utility effect is shown as:13 ΔYαΔIU:const.=UαI-UβIΔI-UI/UYβUαβ-UββΔIΔYβΔIU:const.=-UαI-UβIΔI-UI/UYβUβα-UααΔI The income effect is shown as:14 ΔYαΔYβhI:const.=Uαβ-UββΔIΔYβΔYβhI:const.=Uβα-UααΔI Therefore, by considering the Eqs. (12) ~ (14), the social unrest effect or the total effect is rewritten as:15 dYα/dIYβh:const.=dYα/dIU:const.+UI/UYβdYα/dYβhI:const.dYβ/dIYβh:const.=dYβ/dIU:const.+UI/UYβdYβ/dYβhI:const. where UI=∂U/∂I<0 and UYβ=∂U/∂Yβ. As to the signs of Eqs. (12–15), no confirmed inference can be obtained without additional assumptions. The concrete determination of signs is, however, obtained from the situation depicted in Fig. 5. In this specific situation, the following conclusions are derived: As to the utility effect,13’ dYα/dIU:const.>0,dYβ/dIU:const.<0 As to the income effect,14’ dYα/dYβhI:const.<0,dYβ/dYβhI:const.<0 As to the social unrest effect or total effect,15’ dYα/dIYβh:const.>0,dYβ/dIYβh:const.<0 The change in social welfare is next considered briefly in Fig. 6. The optimal point moved from H to j due to the egoistic donation of the wealthiest. According to this movement, the level of social welfare increases from w 1 to w 2, and the corresponding equivalent value increases from e1 to e2. This implies that even if the motive of donation is egoistic, the donation contributes to the increase in the level of social welfare. However, the point j after the donation is located to the right of the socially permissible line of the destitute lα. This indicates that the egoistic donation of the wealthiest group is insufficient to resolve the difficult situation.Fig. 6 Egoistic (or Self-Protective) donation and its effect on social welfare Erasure of government bonds Although ‘helicopter money’ and ‘MMT’ policies have a wide range of purposes, they are effective in managing income inequality. It is reasonable to implement these policy measures if the price level is stable. Japan as an example: the amendment of law and erasure After evaluating the above policies, this study proposes a new policy to mitigate or overcome the income distribution inequality. The policy is that, by amendment of law, the government erases the government bonds (owned by the central bank) for income equalisation. This study considers Japan as an example. At the end of September 2021, the amount of outstanding government bonds is approximately 9.5 trillion dollars, which is a little less than twice as much as Japan’s GDP.12 The ratio, government bonds outstanding/GDP, is the highest (or worst) in the industrialised countries, leading to fears of bankruptcy of the Japanese government. BOJ owns approximately 4.6 trillion dollars. By subtracting 4.6 trillion dollars from 9.5 trillion dollars, 4.9 trillion dollars remain. If this amount is divided by the GDP, the quotient is about 1.0. So, the ratio of the debt burden decreases drastically,13 which is the key issue. If 4.6 trillion dollars of government bonds owned by BOJ remain intact and 9.5 trillion dollars of total government bonds remain, problems will occur because of the following reasons: first, because the government bonds outstanding is 1.9 times as large as Japan’s GDP, the exogenous shocks such as war, massive earthquakes, endogenous great depressions, and pandemics can cause unexpected scenarios. Second, if massive amounts of government bonds remain intact, consumers will anticipate a future tax increase and feel uneasy about their social security, medical care, education, and so on, and refrain from consumption as much as possible. Moreover, entrepreneurs will expect an increase in corporate tax and after surveying the attitude of consumers, will hesitate to invest. When massive and enduring government bonds (or debts) exist, the economy has the tendency towards nearly zero rate of growth or stagnation. Therefore, people who slid down from the middle class to the low-income class and low-income group people are forced to endure this situation. Therefore, the government bonds which the central bank possesses should be erased. This measure will minimise the fear of tax increase, and consumers’ and entrepreneurs’ psychological pressure will be significantly reduced. From the macroeconomic perspective, both the consumption function and investment function will shift favourably. Regarding the side effects of the erasure policy, heavy confusion regarding the bonds market or the economy may be anticipated. However, the side effects are limited. As the amount of government bonds decreases drastically, and the government bonds held by private sectors and foreign countries are not erased and guaranteed firmly, its scarcity value will increase. The possibility of the default of Japan’s government bonds will vanish. This will increase the demand for Japan’s government bonds which in turn will increase its price leading to a decrease in the interest rate. New issue of government bonds for equalisation After the erasure of the existing government bonds, new government bonds need to be issued to equalise income distribution. The generated revenue is spent on low-income people. This increases their income level and improves their health condition, provides opportunity for higher education, human capital formation to increase their productivity, and so on. Therefore, their future income will continuously increase. These strong measures will improve low-income situations and lead the whole economy towards steady growth. The summated issue depends on the sum needed to equalise income distribution, but it should not exceed the erased sum. This ensures that debts do not accumulate. In Japan’s case, the amount of outstanding government bonds is now approximately 9.5 trillion dollars and, of these, BOJ owns about 4.6 trillion dollars. After the amendment of the law, the government of Japan erases these 4.5 trillion dollars of government bonds held by BOJ. The remaining government bonds (5.0 trillion dollars) are still owned by domestic and foreign investors.14 The government must firmly guarantee the obligation. As the possibility of default completely disappears by the erasure of government bonds (held by BOJ), the scarcity value of the 5 trillion dollars of government bonds will increase, thereby increasing the demand for Japanese government bonds. Specifically, there will be an excess demand for Japanese government bonds. To realize the optimal income equalisation, I consider how much the government should expend for the low-income people. In this model, to simplify the analysis, the labour income is considered to be that of low-income people. Given the existing property income level, the total socially optimal sum R required for income distribution is defined as:16 R=thelabour'ssociallyoptimalincomelevel- thelabour'sexistingincomelevel where the labour’s socially optimal income level satisfies the optimal conditions (4). I consider the above using an abstract example and then a numerical example which may also have an empirical reality. The initially erased sum is denoted as A. The total optimal sum required for income equalisation is assumed to be R trillion dollars. To avoid rapid changes which will be caused by the lump-sum expenditure of R, the process of further erasure and expenditure should be divided into a geometric series. These divisions will be useful to smoothen the change and constrain the occurrence of inflation. If, during these processes, inflation occurs, the erasure and expenditure policy should be suspended and resumed after the resolution of inflation. The upper limit of the inflation is about 2%. The first step is to newly issue some ratio of the erased sum. This ratio is called ‘newly issued rate’, denoted by ϕ, i.e.17 newlyissuedrateϕ=newlyissuedgovernmentbondsinitiallyerasedgovernmentbondsA,(1>ϕ>0) The newly issued government bonds are denoted as ϕA. This sum ϕA is expended for the low-income group, which involves social security including reinforcement of the pension system, medical care, education for human capital formation to enhance their future earning capacities, and assistance for undeveloped countries (the income redistribution for foreign low-income group). These will increase the welfare of the low-income group and consequently increase their economic productivity. Then, because the government bonds have increased to ϕA, the government erases ϕ2A of the government bonds. The residual non-erased portion 1-ϕϕA is still held by domestic and foreign investors and firmly guaranteed by the government. The above process is shown as follows: The second step is to issue new government bonds by as much as ϕ3A and expend the same amount for the low-income group. Then, of these increased bonds, the government erases ϕ4A and the residual non-erased portion 1-ϕϕ3A held by domestic and foreign investors is firmly guaranteed by the government. This process is shown as follows: The third step is similarly shown as: To summarize these steps PI,PII,PIII,⋯, the total optimal expenditure R for the low-income group is equal to:18 R=ϕA+ϕ3A+ϕ5A+⋯=Aϕ/(1-ϕ2) From this, the value of ϕ is calculated as:19 ϕ=-A+A2+4R2/2R When the government’s optimal expenditure R for the low-income group is implemented gradually, the multiplier effects work at each step. If the scale of its effects is m, the total created income V is shown as:20 V=mϕA+mϕ3A+mϕ5A+⋯=mAϕ/(1-ϕ2) The total residual non-erased sum S which is held by domestic and foreign investors is calculated as:21 S=1-ϕϕA+1-ϕϕ3A+1-ϕϕ5A+⋯=Aϕ/(1+ϕ) By such increases in created income, tax revenues also increase. If the average and marginal tax rate is denoted as t, the total increases of tax revenue T is:22 T=tmϕA+tmϕ3A+tmϕ5A+⋯=tmAϕ/(1-ϕ2) If this tax revenue T is used to redeem the government bonds, the net increase of government bonds N(=S-T) is:23 N=S-T=Aϕ1-ϕ-tm/(1-ϕ2) If the government bonds that were outstanding before the erasure policy is denoted as Z, the total sum of government bonds held by domestic and foreign (including government) investors Σ is:24 Σ=Z-A+N=Z-A+Aϕ1-ϕ-tm/(1-ϕ2) The above expressions (18–24) are elucidated using numerical examples. First, the expressions (18–20) are explained diagrammatically. Because government’s first erased sum A is 4.50 trillion dollars, if the total optimal sum R required for income redistribution is, for example, 3 trillion dollars, the government’s new issue rate ϕ can be obtained by calculating the expression (18), i.e.25 3.00=4.50ϕ/(1-ϕ2) From this, the value of ϕ is determined as 0.50. This implies that if the government has a heavy burden of outstanding government bonds of 9.50 trillion dollars (out of which BOJ holds 4.60 trillion dollars) and needs 3 trillion dollars to expend for optimal income redistribution, the government should erase 4.50 trillion dollars of government bonds held by BOJ and, in the first period, newly issue the 0.50×4.50triliondollars=2.25 trillion dollars of government bonds for income redistribution and, in the second period, newly issue the 0.503×4.50=0.56 trillion dollars of government bonds for income redistribution and, in the third period, 0.505×4.50=0.14 trillion dollars of government bonds, ⋯. Then, from Eq. (18), the following relationship is obtained:18’ 3.00≒Requiredsum≒2.25+firstperiod+0.56+secondperiod+0.14thirdperiod++…… Here, to implement the income equalisation, expenditure should be made over the years to make the transfer payment and construction of various systems (such as social security, medical care, education, foreign assistances, etc.) smooth. Beside these, gradual steps of expenditure are indispensable to avoid inflation. If the multiplier effect over all period is 2.00, the total created income V(=2.00R) is shown as:V=2.25×2.00+0.56×2.00+0.14×2.00+0.04×2.00+⋯≒6.00trilliondollars The effect of optimal redistribution money R on income creation V and social welfare W is explained in Fig. 7. The starting point is P4 where the erasure of government bonds held by central bank has been completed. The total optimal sum R required for income equalisation is shown as the distance P4g where the point g is on the OGD curve and therefore this point g satisfies the social optimal conditions (4).Fig. 7 The effect of expenditure on income equalization At point P4, the property income is shown as Oe. From the social welfare point of view, the optimal level of labour income is eg. The existing level of labour income is eP4. So, the total optimal sum R required for income equalisation is:26 R=eg-eP4=P4g It is assumed that R is equal to 3.00 trillion dollars, or P4g=3.00trillion dollars. To finance this sum and to not accumulate too much debt, I proposed to erase nearly all of the 4.50 trillion government bonds held by BOJ and to issue new government bonds which amounts to half of the total erased sum. Through this, the government obtains 2.25 trillion dollars. The process starts and continues as follows: First, 2.25 trillion dollars are divided, for example, by 5 and, each 0.45(= 2.25/5) trillion dollars are expended over five years. The first year’s expenditure is shown as P4ρ1 and taking into its multiplier effect 2.00, the expenditure creates additional income shown as ρ1ρ2. The next year’s expenditure is shown as ρ2ρ3 and by the multiplier effect, this expenditure creates the additional income ρ3ρ4. This process continues to the point ρ10, where the redistributed money is 2.25 trillion dollars and the total additional income is 4.50(= 2.25 × 2.00) trillion dollars. Second, 0.56 trillion dollars are divided by 5 and each 0.11(= 0.56/5) trillion dollars are expended over five years. This year’s expenditure is shown as ρ10ρ11, and it creates additional income ρ11ρ12,⋯. These processes go on till point P5. The sum of redistributed money R is 3.00 trillion dollars and the total created income V is 6.00 trillion dollars. The total increased income V is shown as Y4Y5 (or P4m, or P4q) and of this income V, the increased labour income is shown as P4l(=nq) and increased property income is lm(=P4n). It is shown that by the introduction of optimal redistribution of money R, the level of social welfare increases from W0 to W5, and the economic value increases from E0 to E5. The result of this policy expressed in terms of income is Y4Y5. Next, the expressions (21–24) are explained. The sum of government bonds newly issued (but not erased) and purchased by domestic and foreign investors S (trillion dollars), i.e. the total residual non-erased sum is calculated as:S=1.13+0.28+0.07+0.02+⋯≒1.50(trilliondollars) The total increase in tax revenue brought about by the created income is calculated. If the average and marginal income tax rate is 0.20, the total increase in tax revenue T is shown as:T=0.20V=2.25×2.00×0.20+0.56×2.00×0.20+0.14×2.00×0.20+0.04×2.00×0.20+⋯≒1.20trilliondollars Then, if this tax revenue is used to redeem the government bonds, the net increase in government bonds can be calculated. The sum N(=S-T) is shown as:N=S-T≒1.50-1.20=0.30trilliondollars) Hence, the total sum of government bonds held by domestic and foreign investors Σ (trillion dollars) is shown as:Σ=5.00+0.30=5.30trilliondollars The above numerical example is summarized as follows: when the total optimal expenditure R is needed by 3.00 trillion dollars, the erasure rule demands that newly issued rate ϕ should be 0.50. If the multiplier effects m on redistribution expenditure R is 2.00, the total created income V becomes 2R or 6.00 trillion dollars. The total residual non-erased sum S is 1.50 trillion dollars. If the average and marginal tax rate t is 0.20, the total increase of tax revenue T is 0.20V or 1.20 trillion dollars. The net increase in government bonds N(=S-T) is therefore 1.50-1.20=0.30 trillion dollars. As a result, the total sum of government bonds Σ=5.00+0.30 is equal to 5.30 trillion dollars. Other values of total optimal expenditure for redistribution R are also examined. These are shown in Table 1. The table shows the relationship between R and ϕ,V,S,T,N, and Σ (when the value of multiplier effect m is 2.00, and tax rate is 0.20).Table 1 The Relationship between R and ϕ,V,S,T,N, and Σ Total optimal sum for redistribution R 1.00 1.67 2.00 3.00 4.22 5.00 … 10.00 … 17.21 … 21.32 … 25.00 Newly issue rate ϕ 0.21 0.33 0.38 0.50 0.60 0.65 0.80 0.88 0.90 0.91 Total created income V=2R 2.00 3.34 4.00 6.00 8.44 10.00 20.00 34.42 42.64 50.00 Total residual non erased sum S 0.78 1.12 1.24 1.50 1.69 1.77 2.00 2.11 2.13 2.14 Total increase of tax revenue T=0.4R 0.40 0.67 0.80 1.20 1.69 2.00 4.00 6.88 8.53 10.00 Net increase of government bonds N=S-T 0.38 0.45 0.44 0.30 0.00 − 0.23 − 2.00 − 4.77 − 6.40 − 7.86 Total sum of Government bonds outstanding Σ= 5.00+N 5.38 5.45 5.44 5.30 5.00 4.77 3.00 0.23 − 1.40 − 2.86 R=ϕA1-ϕ2,ϕ=newlyissuedgovernmentbonds/erasedgovernmentbonds, A=4.50trilliondollars,Z=9.50trilliondollars,V=mR,S=ϕA1+ϕ, N=ϕA1-ϕ-mt1-ϕ2,Σ=Z-A+ϕA1-ϕ-mt1-ϕ2,m=2.00,t=0.20 From Table 1, two relationships are established and considered. One is the relationship between R and ϕ, depicted as the curve R in Fig. 8. The slope of this curve R is positive and increasing, i.e. ∂R/∂ϕ>0 and ∂2R/∂ϕ2>0.15 The curve R shows that, given the value of the initially erased sum A, when the total optimal sum R is determined, the newly issue rate ϕ is correspondingly calculated. This figure is intended to be read from the vertical axis to the horizontal axis, or from R to ϕ. Therefore, if the total optimal sum R required for income equalisation is 1.67 trillion dollars, the newly issued rate ϕ is determined as 0.33, and if R is 4.22, ϕ is 0.60.Fig. 8 The relationship between the total optimal sum R and newly issued rate ϕ Another relationship is between ϕ and S,T,N(=S-T) and Σ(=5.00+N), depicted in Fig. 9. The relationship between the total residual non-erased sum S and ϕ is shown as S curve. The slope of this curve S is positive and decreasing.16 The relationship between the total increase in tax revenue T and ϕ is shown as the T curve, and the slope of this T curve is positive and increasing.17 The net increase in government bonds N is equal to S-T. The curvature of the curve N is convex and sloping upwards18 and reaches a maximum point when ϕ is equal to 0.33.19 By connecting these two figures, the relationship between the total optimal sum for redistribution R and net increase in government bonds N, and the total sum of government bonds outstanding Σ(=5.00+N) is shown. This study explains three characteristic points:When the optimal sum for redistribution R is 1.67 trillion dollars, the newly issued rate ϕ is 0.33. At this rate, as explained above, the net increase in government bonds N reaches the peak, and the total sum of government bonds outstanding Σ reaches the peak. In the interval of 0<ϕ<0.33, N and Σ increase, and in the interval of 0.33<ϕ<1, N and Σ decrease. When R is 4.22 trillion dollars, ϕ is 0.60. At this rate, the total residual non-erased sum S is equal to the total increase in tax revenue, and therefore N becomes zero. This means that total sum of government bonds outstanding Σ is 5 trillion dollars which is equivalent to the sum before the new issue of government bonds. When R is 17.21 trillion dollars, ϕ=0.88. At this rate, S is 2.11 trillion dollars and T is 6.88 trillion dollars. Thus, the net increase in government bonds N becomes -4.77 trillion dollars, and the total sum of government bonds outstanding Σ becomes 0.23 trillion dollars, or nearly zero. This means that the outstanding government bonds almost disappear.20 Fig. 9 The relationship between ϕ and total created income V, total residual non-erased sum S, total increase in tax revenue T, net increase in government bonds N(=S-T), and total sum of outstanding government bonds Σ It must be noted that because the optimal value of R is determined from the (constrained) social welfare maximisation conditions (4), the level of total sum of government bonds outstanding Σ is not the object of some sort of maximisation, but the result of the determination of R. In other words, social welfare maximisation conditions determine the values of R,ϕ,V,S,T,N, and Σ. It is natural that the relatively small sum of R is easily carried out. But, using the method of erasing government bonds, the relatively large sum of R is carried out without accumulating government bonds or with decreasing government bonds. The government can revise income inequality if it wishes to do so. The pandemic and income equalisation History reveals that humankind has suffered from many disasters. One of the most serious disasters is the COVID-19 pandemic. The prevalence of the plague, smallpox, cholera, new influenza, and new coronavirus are the representative epidemics. From the income inequality perspective, high-income groups seem to have a relatively advantageous position in avoiding infectious diseases. This is because the high-income groups live in a favourable environment (good sanitary conditions, housing, access to medical facilities, education regarding hygiene, etc.). Conversely, low-income groups lack these facilities which leaves them vulnerable to serious infectious diseases. Thus, income distribution inequality causes the unequal susceptibility to epidemics and diseases. If the pandemic spreads worldwide, its victims will mostly be a part of the low-income group than the high-income group. During the pandemic, governments take drastic measures such as closing the borders and restricting the movement of people and commodities. If the situation persists, the total demand and supply of the economy substantially decreases which has a negative impact on economic activities. The high-income group also suffers tremendous losses, but, owing to their accumulated wealth, many of them can deal with it. In contrast, if low-income groups lose their jobs, they get into a predicament. Thus, the pandemic affects the low-income group more than the high-income group. Governments need to take prompt and adequate measures to tackle the pandemic. One suitable measure is the income equalisation or payment of adequate benefits to low-income groups. Figure 10 explains this. The point P on the AGD curve shows the income level and its distribution before the outbreak of pandemic. As the pandemic becomes severe, the economic situation worsens and the point on the AGD curve moves from P to P~. This movement shows the high disadvantage for the low-income group. Both the absolute income level and relative income share of low-income group decreases. To resolve this grave situation, the government needs to revise the inequality. The following numerical example explains this. At the point P which represents the situation before pandemic, the total optimal sum required for income equalisation R is shown as the distance PG and its amount is assumed to be 3.00 trillion dollars. The economic situation moves down to P~ due to the outbreak and spread of the pandemic. This indicates that, to revise the inequality completely, R must be the distance P~G and this amount is assumed to be 5.00 trillion dollars.Fig. 10 Pandemic and the effect of expenditure on income equalization To revise this severe inequality, 5.00 trillion dollars should be expended for low-income groups. The process and results are shown in Table. If R is equal to 5.00 (trillion dollars), the newly issued rate ϕ is set at 0.65 (percent). If the multiplier effect of R is 2.00, the total created income V is 10.00 (million dollars). The total residual non-erased sum S is 1.77 (trillion dollars). If the average and marginal tax rate t is 0.20, the total increase in tax revenue T is 2.00 (trillion dollars). Then, the net increase in government bonds decreases. As a result, the total sum of government bonds outstanding Σ is 4.77 (trillion dollars), which is below the initial level of 5.00. Through this process, the economy grows from Y~ to Y~~, and the increment of national income ΔY is equal to 2R(=10trilliondollars). The low-income group who was facing challenges are now better off. Their income level increases to P~L. The level of social welfare has increased from W∗ to W∗∗, and its economic value is expressed as the increase in income level from Y∗ to Y∗∗. This policy is summarized as follows: the 5.00 trillion dollars of income equalisation expenditure creates 10.00 trillion dollars of national income, and by the policy of erasure of government bonds and multiplier effect, the net increase in government bonds is approximately zero. The income level of the low-income group and level of social welfare improves significantly. Apprehension regarding inflation This section examines inflation. This study considers that, compared to the previous century, the probability of inflation is not so high currently. Two factors may be stated: First, as is widely supported, after the end of the Cold War, countries, such as China, India, Russia, and many others entered the world market economy. These countries provided high-quality goods with reasonable prices. The industrialised countries imported these goods which reduced the domestic prices of the goods. These tendencies are pervasive and will persist unless irregular exogenous shocks hit the economy. Second, especially after the public access to the Internet, the world economy is now in a severe competition to innovate in the IoT field. Conventional goods and services are substituted by those made by high-technology companies. Such goods are not always cheap. However, as the efficiency of IoT industries is remarkably high, the hedonic prices of these goods are showing a downward trend due to technological progress induced by tough competition. Third, the price of energy will not surge for a long time. Alternative energy is now being developed. An example is shale oil and shale gas. Shale oil and gas reserves are abundant.21 In addition to this, the worldwide prevalence of the development and utilization of renewable energy will resist any energy price increase. The above-mentioned reasons will suppress the possibility of future inflation. However, the possibility of the occurrence of inflation cannot be dismissed.22 If the expenditure for income equalisation R affects the inflation rate i, this relationship is shown as:27 i=i(R,F¯) where F¯ is the other factors which affect the inflation rate i, and to simplify the analysis, assumed to be constant. Regarding the shape of the function i, the following relationships are assumed: di/dR≧0 and d2i/dR2≧0. This function i is shown in Fig. 11. This hypothetical curve i represents that when the value of R is 3.00 trillion dollars, the inflation rate i is 1.60%, when R is 4.00 trillion dollars, i is 2.00%, and when R is 5.00 trillion dollars, i becomes 2.70%.Fig. 11 The relationship between the total optimal sum R and inflation rate i Thus, when the society or government opposes the inflation rate which exceeds 2%, the scale of the expenditure for income equalisation R is constrained by this inflation rate. In this situation, if, for example, 3.00 trillion dollars of R is planned, this may be permitted from the standpoint of controlling inflation. If, however, 5.00 trillion of R is planned, this may not be permitted.23 As inflation occurs also due to the other factors, to avoid an undesirable rate of inflation, these factors must be taken into account. If by any chance,24 an undesirably high inflation rate occurs, the countermeasure is to absorb the money in the market. The government should issue government bonds. If the amount of this issue is too much, the price of government bonds will fall and the interest rate will rise. If this happens in a short time, many of the domestic and foreign investors will suffer huge losses. The government does not suffer any loss because, by erasure, government has little government bonds outstanding. It must, however, be stressed that inflation negatively affects the low-income group, but, by this policy, as the low-income group upgrades to the middle income group, from the income distribution perspective, the inflation does not exert too much negative impact. As long as the inflation rate is low, the majority does not suffer. If a high inflation rate occurs, the policy of erasure and expenditure for the low-income group should be discontinued for some time and resumed after the resolution of inflation. Multiplier analysis This section considers the above policy measure, i.e. income equalizing expenditure R, and then compares the effectiveness between it and public investment for countercyclical purpose G. First, the consumption function C is expressed as:28 C=CY-T+R,Yf(R,G) where T is the tax and is assumed to be proportional to national income Y, i.e. denoting t as an average and marginal tax rate, T=tY. R is the government expenditure for income equalisation.25 Then, Y-T+R represents the disposable income in a broad sense. Yf is the future income which is assumed to depend on government expenditure for income equalisation R and public investment for countercyclical measures G, i.e. Yf=Yf(R,G). The study assumes the following relationships: 1>∂C/∂(Y-T+R)>0,1>∂C/∂Yf>0,∂Yf/∂R>0, and ∂Yf/∂G<0.26 Second, the investment function I is expressed as:29 I=IrR,G,E(R,aG,b(G) where r is the market interest rate which is assumed to be influenced by the government expenditure for income equalisation R and public investment G,27 i.e. r=r(R,G). When the parameters R and G increase, r may increase a little. Thus, the following relationships are assumed: ∂r/∂R≧0,∂r/∂G≧0, and as to the relationship between r and I, it is assumed to be negative, i.e. ∂I/∂r<0. E is a factor which is related to the efficiency of investment I and depends on R and G. I call E ‘an efficient factor in investment’. When R increases, the income of low-income people also increases. Through this, people can be better off and have the opportunity to learn more, and are able to improve their health conditions. In this situation, the efficiency of investment I will increase. Therefore, the following relationships are assumed: ∂I/∂E>0 and ∂E/∂R>0. As to the effects of public investment for countercyclical purpose G on investmentI, two effects are supposed. One is the improvement in infrastructure and through this, investment efficiency will increase. This effect is called the ‘infrastructure effect’ and denoted as a and calculated by: a=aG,da/dG>0,dE/da>0, and dI/dE>0. Another is investing in pork-barrel public projects, which is often done so that politicians can secure votes, which have negative effects on the investment efficiency I. This effect is called the ‘futility effect’ and denoted as b and calculated by: b=bG,db/dG>0,dE/db<0, and dI/dE>0. The equation for the determination of national income is:30 Y=CY-T+R,Yf(R,G)+IrR,G,E(R,aG,b(G)+R+G The effect of distribution equalizing expenditure R on the national income Y is calculated as follows:31 dYdR=1+∂C∂(Y-T+R)+∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R/1-∂C∂Y-T+R(1-t) From the assumptions above, the denominator of (31) is positive. Among the terms of the numerator, the term ∂I/∂r∂r/∂R is nearly zero or zero.28 However, a negative value of this term will not overwhelm the total positive values of the other terms. So, the numerator will be positive. Then, as to the sign of (31), the following relationship may hold:32 dY/dR>0 The effects of the government’s distribution equalizing expenditure R on the labour income YL and property income YP can be analysed. In Fig. 7, the tangent line of the actual growth distribution curve AGD at point P4 is expressed as33 YL=αYP+βα=αR,β:constant where it is assumed that α depends on the parameter R, i.e. α=α(R) and α(R) is positive and dα/dR is also positive.29 This tangential line means that the change in the parameter R affects the relationship between YL and YP; i.e. if R changes, the following relationship holds: dYL=α′YPdR+αdYP, where α′=dα/dR. As Y=YL+YP, the Eq. (30) is revised as:34 YL+YP=C1-tLYL+1-tPYP+R,YfR,G+IrR,G,E(R,aG,bG+R+G where tL and tP are the income tax rate on YL and YP, respectively, and tL<tP. From (34), the following relationships are obtained:35 dYLdR=α′YP1-C′·1-tP+αC′+1+α∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R1-C′·1-tP+α1-C′·1-tL 36 dYPdR=-α′YP1-C′·1-tL+C′+1+∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R1-C′·1-tP+α1-C′·1-tL where C′=∂C/∂1-tLYL+1-tPYP+R. As to the denominators of (35) and (36), the terms 1-C′·(1-tP) and 1-C′·1-tL are positive and as α is positive, the denominators are positive. As to the numerator of (35), from the assumptions α and α′=dα/dR are positive, so the sum of the first and second terms α′YP1-C′·1-tP+αC′+1 is positive. The element ∂C/∂Yf∂Yf/∂R of the third item is positive because when R increases, future disposable income Yf also increases, and accordingly consumption C increases. The element ∂I/∂r∂r/∂R may be negative because when R increases, interest rate r may increase slightly or unchanged, then the investment I decreases slightly or unchanged. The element ∂I/∂E∂E/∂R is positive. When R increases, the income of low-income class increases, and this will have the effect of improving the health condition, productivity, and the education opportunities of the low-income class. Then, it will heighten the efficiency factor of investment E and will lead to increase investment I. Therefore, if the negative value of the element ∂I/∂r∂r/∂R does not overwhelm30 the positive values of the first and second terms and the other elements of the third term, the numerator of (35) is positive. Hence, the following relationship is established:37 dYL/dR>0. As to the numerator of (36), the first term is negative and the element ∂I/∂r∂r/∂R is also negative. So, the sign of (36) is not clear. If the income tax of YL and YP are the same and equal to t, Eq. (31) is obtained by summing (35) and (36). The effectiveness of income equalising expenditure R When the government’s income equalizing expenditure R is implemented, its path is depicted as P4P5 (Fig. 7). This path now forms the actual growth distribution curve. From (35) and (36), the slope of the path dYL/dYP is shown as:38 dYLdYP=α′YP1-C′·1-tP+αC′+1+α∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R-α′YP1-C′·1-tL+C′+1+∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R The condition that this expenditure for income equalisation R is effective, or whether this increases the social welfare depends on the condition that the slope of the P4P5 must be steeper than that of the social welfare indifference curve W0. So, to be effective, the following relationship must hold at the point P4, i.e.:39 α′YP1-C′·1-tP+αC′+1+α∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R-α′YP1-C′·1-tL+C′+1+∂C∂Yf∂Yf∂R+∂I∂r∂r∂R+∂I∂E∂E∂R>-∂W/∂uP∂uP/∂YP∂W/∂uL∂uL/∂YLthe slope of income equalizing expenditureRthe slope of social welfare indifference curve If this condition is satisfied, the income equalizing expenditure R is justified. If, by any chance, this condition is not satisfied, R should not be implemented. The ineffectiveness of public investment for countercyclical purpose G The effect of public investment for countercyclical purpose G on national income Y is calculated from (30), i.e.:40 dYdG=1+∂C∂Yf∂Yf∂G+∂I∂r∂r∂G+∂I∂E∂E∂a∂a∂G+∂I∂E∂E∂b∂b∂G/1-∂C∂Y-T+R1-t The denominator of (40) is positive. As to the numerator, the second term ∂C/∂Yf∂Yf/∂G is assumed to be negative. This effect is called the ‘negative future effect on consumption’.31 In the presence of the ‘crowding out effect’, the third term ∂I/∂r∂r/∂G is negative; i.e. if the public investment G increases, the interest rate will increase, and therefore ∂r/∂G is positive, and if the interest increases, the investment I decreases, and therefore ∂I/∂r is negative. The fourth term ∂I/∂E∂E/∂a∂a/∂G will be positive if public investment contributes to the development of infrastructure a; i.e. ∂a/∂G is positive, and if this improve the efficiency E of investment, ∂E/∂a is positive, and if it increases the investment I, ∂I/∂E is positive. In sum, the fourth term is positive. The fifth term ∂I/∂E∂E/∂b ∂b/∂G will be negative if public investment is done mainly for securing elections and if it is favouring pork-barrel public works projects. In this situation, the waste or futility b increases, i.e. ∂b/∂G>0, and this will decrease the efficiency E of investment, i.e. ∂E/∂b<0, and this will decrease the investment I, i.e. ∂I/∂E>0. So, the effect of the fifth term is called the ‘allocation disturbing effect’. Overall, if the extent of ‘negative future effect on consumption’, ‘crowding out effect’, and ‘allocation disturbing effect’ do not exceed those of other positive terms, the following relationship holds:41 dY/dG>0 This relationship means that the public investment for the countercyclical purpose G increases national income. However, if, due to three negative effects, the multiplier effect shown by (41) is not as large as expected, then, its value may be less than 2 and this type of public investment G increases the income only from point P4 to j, or from Y4 to Y~ in Fig. 7. The multiplier, in this case, is in approximate unity. Moreover, if the situation is such that nearly full employment is already attained and resource allocation is almost desirable, then the government’s public investment intended to win the national election will have unanticipated effects. In this situation, the extent of the ‘negative future effect on consumption’, ‘crowding out effect’, and ‘allocation disturbing effect’ may be so strong that they overwhelm other terms. Then, the following situation will hold:42 dY/dG<0. This indicates that the futile public investment G has the possibility to demolish a sound economy and decrease national income from point P4 to s, or from Y4 to Y~~ in Fig. 7 and retain the financial deficit. The effects of public investment G on the labour income YL and property income YP are also analysed. The change in G does not directly affect the ratio of YL and YP. Comparison of the effects of R and G on national income Y It may seem that the income equalizing expenditure R has a weaker effect on national income Y than the public investment for the countercyclical purpose G. This, however, may not true. As explained above, in some situations, especially in developed countries, the effect of R is not always weaker than that of G. The amount of R and G is set to be equal, and by subtracting (40) from (31), the following relationship is obtained:43 dYdR-dYdGR=G=∂C∂Y-T+R⏞first⏟++∂C∂Yf∂Yf∂R⏟+⏞second-∂C∂Yf∂Yf∂G⏟+⏞third+∂I∂r∂r∂R⏟-⏞fourth-∂I∂r∂r∂G⏟+⏞fifth+∂I∂E∂E∂R⏟+⏞sixth-∂I∂E∂E∂a∂a∂G⏟-⏞seventh-∂I∂E∂E∂b∂b∂G⏟+⏞eighth/1-∂C∂Y-T+R1-t As to the right-hand side numerator of (43),32 the first term ∂C/∂(Y-T+R) represents the marginal propensity to consume and is positive. The second term ∂C/∂Yf∂Yf/∂R refers to the effect of R on (present) consumption C through the increase in future disposable income Yf and is assumed to be positive. The third term -∂C/∂Yf∂Yf/∂G refers to the effect of G on consumption C through Yf and is assumed to be positive.33 The fourth term ∂I/∂r∂r/∂R refers to the effect of R on investment I through the increase of interest r and is assumed to be slightly negative or zero. The fifth term (-)∂I/∂r ∂r/∂G refers to the effect of G on investment I through r, and is assumed to be slightly positive or zero. The sixth term ∂I/∂E∂E/∂R refers to the effect of R on I through the efficiency factor (in investment) E, and is assumed to be positive. The seventh term (-)∂I/∂E∂E/∂a ∂a/∂G refers to the effect of G on I through the infrastructure effect a and efficiency factor E, and is assumed to be negative. The eighth term (-)∂I/∂E ∂E/∂b∂b/∂G refers to the effect of G on I through the futility effect b and efficiency factor E, and is assumed to be positive. In general, the sign of the relationship (43) is indeterminate. However, the sign may be determined by making certain assumptions. First, with a zero or negative interest rate, the crowding-out effect may be zero or negligible. Then, the fourth and fifth terms may be considered as zero. Second, the seventh term is the infrastructure effect and this may be strong in developing countries.34 In developed countries, however, this effect may not be stronger than that in developing countries because in developed countries many fundamental infrastructures are already constructed. So, the marginal efficiency of public investment G on infrastructure is low. If some new firms succeed in making innovative products, the infrastructure needed to make or use innovative products may be in many cases, constructed by these firms. The government’s role in these situations may not be to construct new infrastructure but to provide new legislation to manage the new situation. This provision of new legislation is not considered in this study. We can conclude that the infrastructure effect in developed countries is not very strong. So, if the extent of the infrastructure effect, i.e. the seventh term does not overwhelm the other terms, the effect of R on Y surpasses the effect of G on Y, i.e.:44 dY/dR>dY/dG,R=G This tendency is strengthened when the eighth term is strong, or when public investment G is focused on pork-barrel public projects. Concluding remarks Inequality seems to be almost inevitable in the process of economic growth, especially when considering the income distribution inequality. From the end of twentieth century, globalization and tremendous breakthroughs in information and communication technology have had profound effects on the economy as well as society (divided into the extremely rich, poor, and dwindling middle class). To deal with this issue, economics proposes an increase in taxes. Tax increases, especially, the increase in progressive tax on high income and property is essential for reducing inequality. Tax increases, however, must confront political opposition from the wealthy classes. This situation is compared to the case of land reclamation. To make a low land higher, the earth and sand of higher places are dug and conveyed. By this, the difference of altitude will decrease. If in this case the owners of higher places do not oppose much, the project will succeed. If, however, they strongly oppose, it will face difficulties. In general, if the scale of the project becomes larger, the difficulty of gathering much earth and sand will increase more. In reality, almost every people does not like tax increase. If heavy taxation oppresses entrepreneurship, it will hinder economic growth. Furthermore, tax increases have an upper limit. To offset the enormous accumulated financial deficit, I think the policy of tax increase is inadequate and a thorough new policy is required. This situation is compared to the case of large scale urban development. If this causes considerable volume of surplus soil from the construction site, that soil may be conveyed and used to cover over the lower land. The difference of altitude between the lower land and higher places decreases. In this case, as the higher places are not dug, the owners of it do not oppose. If the enormous accumulated sum of government bonds is intact, the continued large-scale government expenditure on social security, supply of public services, and countermeasures to pandemics and so on, may cause serious problems in the future. If a world war, civil war, terrorist event, or financial crisis happen, along with the behaviour of speculators, the national bonds market as well as the whole market may become turbulent. This will cause a decrease in consumption and investment and will lead to a long and severe depression. Government expenditure may be unable to cover this decrease in total demand because of the turbulent government bonds market. The price of government bonds may fall considerably. To avoid these scenarios and heavy tax increase caused by the enormous accumulated government bonds, this study proposes to erase the government bonds which are owned by the central bank. The erasure of government bonds means the disappearance of the possibility of a government default and also a decrease in the supply of government bonds, leading to a price increase of it and decrease in the interest rate. The features of erasure policy are summarized. First, the erasure policy will maximise the social welfare function which involves both altruism and egoism. The desirability of each income distribution state is judged according to this welfare function. The main purpose of this erasure policy is to accomplish widespread income equality which traditional tax policies have not realized. Second, as this policy devotedly expends for the low income groups, it enables low-income groups to obtain the chance to heighten their overall human capital or their basic capability and thereby increase future earning capacity and enhance their quality of life. Then, the productivity of the economy increase and the national economy will grow. Third, as this policy continually erases the government bonds which the central bank owns, and expends some ratio ϕ(1>ϕ>0) of the sum, the remaining government bonds decrease. This tendency is strengthened by the increase in tax revenue caused by the multiplier effect of the expenditure. Fourth, different from tax increase policy, this policy does not inflict any burden to the people and firms. Therefore, this policy neither reduces the incentive of anybody nor that of entrepreneurs to make efforts to achieve more. In this sense, this policy is feasible. Fifth, this policy is intended to implement gradually. The process of further erasure and expenditure is divided into a decreasing geometric series. By this, the changes are made smooth. The economic structure is easy to correspond to the change. Moreover, the recipient of government expenditure can act in a carefully planned way. Sixth, the grounds for concern must be noted. One is the apprehension regarding inflation. As this erasure policy increases the money supply, the demand side pressure created by it always exists. If the inflationary pressures are low, the erasure policy should be implemented. If the inflationary pressure is moderate, the policy should be implemented gradually. If the inflationary pressure is high, the policy should be suspended until the pressure becomes moderate. As this policy is implemented gradually, the government is easy to respond to the inflation. I think the supply side effect created by the productivity increase of the recipients of government expenditure will contribute to hold back inflation. Seventh, another limitation is that the erasure policy may have the possibility to effectuate a lax financial policy. As mentioned earlier, the expenditure financed by the erasure policy should be strictly applied only for income equalisation. If it is applied to pork-barrel public works projects, it worsens income distribution and, to make matters worse, the efficiency of the economy will decline and the economy itself will become stagnant. Acknowledgements I appreciate the kind and precise advices of the reviewers and the editor of this journal. Author contributions I am a single author and no other author contributed to this manuscript. Funding I have not obtained any funds at all. Data availability I cited the fundamental data on GDP and Government Bonds from the home pages of Ministry of Finance, JAPAN, and Bank of Japan. Code availability Microsoft Word. Declarations Conflict of interest I have no conflicts of interest/competing interest concerning to my manuscript and my research. Ethical approval This manuscript is submitted only to SN Business & Economics. This manuscript is original and have never been published elsewhere in any form or language. No data, text, or theories by others are presented as if they were the author’s own. I swear that I followed all the ethical responsibilities of an author. Consent for publication I consent to publish my manuscript. 1 As Pigou’s proposition entailed the interpersonal comparisons of utility, the criticism and reinforcement on his theory are presented much. See, for example, Robbins (1932), Bergson (1938), Kaldor (1939), Scitovsky (1941), Samuelson (1947) and Harsanyi (1955). 2 Neoclassical economics believes in the market mechanism, or working of flexible price system to solve many economic problems. 3 With regard to the unemployment and inequality, Keynes wrote as follows: “The outstanding faults of the economic society in which we live are its failure to provide for full employment and its arbitrary and inequitable distribution of wealth and income. ….” See, Keynes (1936), Chapter 24, Page 372. 4 Basic capability may be interpreted as the human’s ability to attain what the human desires. 5 A joint paper with Thomas Piketty and Emmanuel Saez. 6 A joint paper with Emmanuel Saez. 7 From the social welfare maximisation conditions, the sign of Δ is positive. 8 Adam Smith (1776) indicated that self-interest is the driving force of the development of market economy and, before this, he also had pointed out that people behave conscious of the other’s sentiment or judgment (1759). In the analysis of human behavior, Smith esteemed the sympathy as the most influential factor. The concept of sympathy has the affinity with some sort of altruism. 9 It must be stressed that altruistic donation between the destitute or members of low income group plays more valuable roles than those of the wealthiest to mitigate the inequality. This kind of donation enhances the essential values of their lives. 10 As it is assumed that the point H is very close to the beginning point h, hereafter these two points are considered as indifferent. 11 At the point H, the indifference curve uC the level of which is lower than uB intersects the income line yByB. 12 In 2021, GDP of Japan is about 4.9 trillion dollars. (This is calculated at the rate of 112 yen to the US dollars.). 13 At that time, the amount of government bond and T-bill outstanding is approximately 10.9 trillion dollars. This is 2.2 times as large as GDP. 14 To put it accurately, this 5 trillion dollars involves 0.1 trillion dollars of government bonds which is not erased and owned still by the central bank. To simplify explanation, this sum is ignored. 15 ∂R/∂ϕ=A(1+ϕ2)/(1-ϕ2)2>0 and ∂2R/∂ϕ2=2Aϕ(1-ϕ2)2+2(1-ϕ4/(1-ϕ2)4>0. 16 These are shown as, ∂S/∂ϕ=A/(1+ϕ)2>0 and ∂2S/∂ϕ2=-2A(1+ϕ)/(1+ϕ)4<0. 17 As the slope of curve R in Fig. 8 is positive and increasing, the slope of curve T=mtR in the Fig. 9 is also positive and increasing. 18 From the definition N=S-T, the following expression is obtained, i.e.∂2N/∂ϕ2=∂2S/∂ϕ2-∂2T/∂ϕ2 where, ∂2S/∂ϕ2=-2A(1+ϕ)/(1+ϕ)4<0,∂2T/∂ϕ2=2tmAϕ(1-ϕ2)(3+ϕ2)/(1-ϕ2)2>0. then, ∂2N/∂ϕ2<0. 19 In the case∂N/∂ϕ=Aϕ21-tm-2ϕ+(1-tm)/(1-ϕ2)2=0, only the value ϕ=0.33 satisfies this first-order condition (where, m=2.00 and t=0.20 are assumed). 20 When R is more than 17.21 trillion dollars, Σ becomes negative. This negative value represents the financial surplus. 21 To mine these, however, close attention to the environment is needed. 22 I recognize that as of March 2022, the worldwide inflation is prevailing. 23 If the pandemic or disaster is severe, I think the inflation rate a little more than 2.00%, that is, around 3.00% may be permitted. 24 The causes of recent inflation are reduction of the supply of crude oil induced by the global movement of carbon neutral, outbreak of pandemic and the war of aggression. 25 It must be noted that because the amount of R may not always be sufficient to attain socially desirable level of equality, or, cannot reach within the region of socially permissible income distribution, the word “income equalization” means, in these cases, the direction to reach the goal. 26 I assume that public investment for counter cyclical measures G may have positive effect on current national income Y, but will increase the government debt and thereby bring about future tax increase. So, G decrease the future income Yf. 27 Interest rate r is influenced also by the money supply of the central bank and many market factors. To simplify the analysis, these are assumed to be given, but, instead, the parameter R plays the central role in my model. 28 Because R is expended after the large scale erasure of government bonds, or, the price of bonds are high and interest rate is forced to be the lowest, the value of fraction ∂r/∂R may be nearly zero or zero. The term ∂I/∂r is negative. In sum, this term is slightly negative, or zero. 29 So, when the value of R increases, the slope of tangent line becomes steep. 30 As mentioned earlier, the part of this element ∂r/∂R may be nearly zero or zero. 31 It is known that almost every public investment for countercyclical purpose leaves financial deficit and decrease future income. 32 The denominator of right side (43) is positive. 33 It must be noted that the fraction ∂C/∂Yf∂Yf/∂G is assumed to be negative. 34 In developing countries, the amount of infrastructure may not be enough. So, in this situation, the government investment for building the infrastructure may be effective. ==== Refs References Atkinson AB The Economics of Inequality 1975 Oxford University Press Atkinson AB On the Measurement of Poverty Econometrica 1987 55 749 764 10.2307/1911028 Atkinson AB Thomas piketty, and Emmanuel Saez, “Top incomes in the long run of history” J Econ Lit 2011 49 1 3 71 10.1257/jel.49.1.3 Atkinson AB Inequality: what can be done? 2015 Harvard University Press Bergson A A reformulation of certain aspects of welfare economics Quart J Econ 1938 52 310 334 10.2307/1881737 Deaton A The Great Escape: Health, and the Origins of Inequality 2013 Princeton, NJ Princeton University Press Friedman M Capitalism and Freedom 1962 The University of Chicago Press Friedman M The optimum quantity of money and other essays 1969 Aldine Publishing Company Harsanyi JC Cardinal welfare, individualistic ethics, and interpersonal comparisons of utility J Polit Econ 1955 63 309 321 10.1086/257678 Kaldor N Welfare propositions of economics and interpersonal comparisons of utility Econ J 1939 49 549 552 10.2307/2224835 Kelton S (2020) The Deficit Myth Modern Monetary Theory and the Birth of the People’s Economy, Public Affairs, New York Keynes JM The general theory of employment, interest, and money 1936 London Macmillan Kuznets S Economic growth and income inequality Am Econ Rev 1955 45 1 28 Pigou AC (1920) (1971) The Economics of Welfare, 4th ed., London Macmillan Piketty T Saez E Income inequality in the United States, 1913–1988 Quart J Econ 2003 118 1 39 10.1162/00335530360535135 Piketty T Capital in the twenty-first century 2014 Cambridge, MA Harvard University Press Ravallion M The economics of poverty, history, measurement, and policy 2016 Oxford University Press Robbins L The nature and significance of economic science 1932 London Macmillan Samuelson PA Foundations of economic analysis 1947 Cambridge, Massachusetts Harvard University Press; Enlarged Second Edition 1983 Scitovsky T A note on welfare propositions in economics Rev Econ Stud 1941 9 77 88 10.2307/2967640 Sen AK On economic inequality 1973 Oxford Clarendon Press and New York, Norton Sen AK Choice, Welfare and Measurement 1982 Basil Blackwell Smith A (1759) The theory of moral sentiments, A. Millar, Strand; and A. Kincaid and J. Bell, Edinburgh Smith A (1776) An Inquiry into the Nature and Causes of the Wealth of Nations, Edinburgh, "new edition" Glasgow (1805) Wray LR (2015) Modern money theory: the primer on macroeconomics for sovereign monetary systems, second edition	36481921	PMC9734564	NO-CC CODE	2022-12-14 23:28:29	no	Nat Rev Urol. 2022 Dec 8;:1	latin-1	Nat Rev Urol	2,022	10.1038/s41585-022-00700-3	oa_other
==== Front Affect Sci Affect Sci Affective Science 2662-2041 2662-205X Springer International Publishing Cham 168 10.1007/s42761-022-00168-9 Commentary / Opinions Introduction to Special Issue on Affective Science in Animals: Toward a Greater Understanding of Affective Processes in Non-Human Animals http://orcid.org/0000-0001-7521-6904 Rogers Forrest D. 12 http://orcid.org/0000-0001-5826-2095 Bales Karen L. [email protected] 345 1 grid.16750.35 0000 0001 2097 5006 Princeton Neuroscience Institute, Princeton, NJ USA 2 grid.16750.35 0000 0001 2097 5006 Department of Molecular Biology, Princeton University, Princeton, NJ USA 3 grid.27860.3b 0000 0004 1936 9684 Department of Psychology, University of California, Davis, CA USA 4 grid.27860.3b 0000 0004 1936 9684 Department of Neurobiology, Physiology, and Behavior, University of California, Davis, CA 95616 USA 5 grid.27860.3b 0000 0004 1936 9684 California National Primate Research Center, Davis, CA USA Handling editor: Wendy Berry Mendes 3 12 2022 12 2022 3 4 697702 20 10 2022 21 11 2022 © The Society for Affective Science 2022 How should we characterize the affective lives of non-human animals? There is a large body of work studying affective processes in non-human animals, yet this work is frequently overlooked. Ideas about the affective lives of animals have varied across culture and time and are reflected in literature, theology, and philosophy. Our contemporary ideas about animal affect are philosophically important within the discipline of affective science, and these ideas have consequences in several domains, including animal husbandry, conservation, and human and veterinary medicine. The articles contained within this special volume cover several levels of analysis and broad representation of species, from the non-mammalian, to rodents, to primates; but together, these articles are collectively concerned with the topic of affective processes in non-human animals. Keywords Animals Animal welfare Emotion Affect Behavior Neuroscience issue-copyright-statement© The Society for Affective Science 2022 ==== Body pmc ‘Well, then,’ the Cat went on, ‘you see, a dog growls when it’s angry, and wags its tail when it’s pleased. Now I growl when I’m pleased, and wag my tail when I’m angry.’ (Carroll, 1865) The affective states of non-human animals (hereafter, animals) have long been the target of presumption, speculation, denial, and disregard. Animal affect has been reflected upon by authors, philosophers, and theologians; and most laypeople are likely to have some perspective on the topic. It makes sense that we should think about the affective lives of animals. Our own personal affective experiences are deeply rooted in our social lives, and we often think deeply about the affective lives of our social partners. We regularly engage in social exchanges with nonhuman animals, and at times we cultivate relationships with them, from passing interactions with wild animals to more consistent relationships with livestock or beloved pets. We may even come to regard the latter as members of our extended families (or even describe and often treat them like our own children, e.g., fur babies). Through these interactions, many of us seek to attribute affective states to these heterospecific social partners. For some, this attribution of often human-like affective experience to animals is an anthropomorphic overapplication, yet to others, the rejection of animal affect is viewed as anthropocentric. How should we, then, characterize the affective lives of animals? Historical Perspectives The extent to which one attributes affect to animals, or whether one attributes affect to animals at all, is perhaps based in one’s respective views on the nature of affect. These views have been simultaneously cultivated in distinct cultures across several historical periods. For the Western perspective, one might consider the lineage from the ancient Greek philosophical tradition and throughout the development of Christian theology. Aristotle (384–322 BCE, Greece) vacillated in his view of animal affect. On one hand, he believed cognition was a prerequisite for the experience of affect; so when he denied cognitive ability wholesale to animals, he therefore denied affect to animals (Fortenbaugh, 1971). Yet on the other hand, he at times attributed fear, pity, and other affective states to nonhuman animals (e.g., “… the eagle expels the nestlings because of jealousy”) (Fortenbaugh, 1971; Sihvola, 1996). Aristotle may have drawn a distinction between affect that is rational versus those that are irrational; and accordingly he attributed the former exclusively to adult humans while applying the latter (e.g., θυμός [thumos], or “non-rational, spirited desire”) to animals (and human children) (Sihvola, 1996). This distinction between the rational human and irrational animal was similarly applied by the Stoics (Passmore, 1975; Sihvola, 1996) and later echoed by some Christian theologians (e.g., Augustine of Hippo). This theme continues in the philosophy of Thomas Aquinas who proposed that human rationality gives us the right to govern other, irrational animals. Aquinas taught that it is good to treat animals humanely and with compassion because it promotes such feelings between humans, not because animals experience affective suffering—neither do they, in this view, experience affect more generally nor do they have a capacity for morality (Passmore, 1975). Eastern philosophical traditions developed in parallel to Greek and Christian philosophy, with Confucianism having been established in China at least 100 years prior to the birth of Aristotle and persisting with extensive regional influence over the next two thousand years. By the late Ming Dynasty (1368–1644), Chinese philosophers and writers in the Confucian tradition had cultivated the affective concept of qing (情)1 (roughly translated as feeling or passion), which was at times contrasted with the concept of xing (性), or inborn [human] nature (Huang, 1998; Sung, 2016). Animals are characterized as possessing qing, but not xing. For example, the poet Yang Shen (杨慎) wrote “What will happen if one promotes xing but neglects qing? He will become dead ashes. What will happen if one is moved by qing but forgets about xing? He will become an animal” (Huang, 1998, p. 156—157). Charles Darwin took great interest in the topic of non-human affect, extensively detailing his perspectives on animal affective expression (Darwin, 1872). He attributed to a variety of non-human animals the experiences of both positive (e.g., joy, affection) and negative (e.g., anger, terror) affective states and the expression thereof to communicate intention, for example:… The appearance of a dog approaching another dog with hostile intentions, namely, with erected ears, eyes intently directed forwards, hairs on the neck and back bristling, gait remarkably stiff, with the tail upright and rigid. So familiar is this appearance to us, that an angry man is sometimes said “to have his back “up”. (Darwin, 1872, p. 116) Paul Ekman would later describe Darwin’s work on expression as “the first pioneering study of emotion” and perhaps even foundational to psychology itself (Ekman, 2009). Darwin, with his evolutionary perspective, registered his objection to the idea that affect is solely experienced and expressed by humans (Darwin, 1872; Ekman, 2009). Bridging the Gap Bridging the gap between our experience of human affect, and our attempt to infer specific states in non-human animals, requires particular philosophical assumptions (Barrett, 2012). That is, because animals cannot describe their affective experiences with language, researchers must instead use various behavioral and physiological indicators of presumed affective states. For example, researchers may register freezing behavior as a proxy for “fear”; or they might alternatively measure neural activity from circuits known a priori to be active during particular human affective experiences or otherwise connected to a particular affectively associated behavior (Barrett, 2012). One perspective is that philosophical assumptions are not necessary if researchers stop attempting to measure specific emotions in animals (in general) or trying to identify hardwired signatures of said emotions in animals; rather, they should study the ingredients of emotion (e.g., affect, conceptual knowledge, language, and social context) per the theories of constructed emotion (Bliss-Moreau, 2017). In this special issue, Mendl and colleagues provide an alternative perspective in which they consider how to handle these philosophical inferences (Mendl et al., 2022). In this commentary, they argue that it is possible to systematically measure animal affective states with a framework of three pathways of inference through which one might consider how to translate emotion concepts, emotion indicators and emotion-generating contexts. Respectively, Mendl and colleagues consider a series of three questions: What types of emotions are likely to exist in other species? How can one assess animal emotional states? Can one establish an animal’s “ground truth” emotional state at any one time (and use that to identify indicators of animal emotion)? They then consider what might be necessary to determine if such affective states are consciously experienced as conscious emotional feelings by animals. Of Voles and Men The successful generation of mental health treatments has relied and continues to rely heavily on the translation of animal research, particularly that in rodent models (Milton & Holmes, 2018). While there is not a perfect rodent model for any particular human affective disorder, the establishment of assays for behavioral and physiological indicators of negative affect has been foundational to the subsequent generation of pharmacological interventions for affective disorders (e.g., anti-depressants; Robinson, 2018). Advances in genetic, molecular, and other neuroscientific tools have made the generation of improved rodent models and identification of biological markers of psychiatric disorders increasingly promising (Canetta & Kellendonk, 2018). The vast majority of this research is conducted in mice and rats; yet mice and rats are not always the best model for human affective experiences, which are notably rooted in social relationships. One alternative to more common murine models is the prairie vole (Microtus ochrogaster), which is central to two papers included in this special issue. Early work by Getz and Carter established that prairie voles have relatively rich social lives, with wild prairie voles often integrated into large social networks of extended families, life-long mating, and biparental care, all indicative of their socially monogamous and cooperatively breeding social strategy (Carter & Getz, 1993; Getz & Carter, 1996; Getz et al., 1981; Roberts et al., 1998; Williams et al., 1992). Some intriguing similarities exist between prairie voles and humans, some of which may be less expected. For instance, the same sort of misattribution of arousal observed in human men by Dutton and Aron in their “shaky bridge” experiment (Dutton & Aron, 1974) is also observed in gonadally male prairie voles (DeVries et al., 1996), such that anxiogenic experiences preceding interaction with an sexually attractive stimulus are found to yield heightened sexual arousal and social approach in both species.2 Yet, beyond their social behavior alone, prairie voles have proven a useful model for understanding the neurobiology of pair bonding and paternal care (Rogers & Bales, 2019; Seelke et al., 2018; Young & Wang, 2004). The advent of the COVID-19 pandemic and associated stay-at-home orders fueled increased interest in the negative affective consequences of social isolation (Bland et al., 2022; Palgi et al., 2020), building on previous seminal work on loneliness (Ernst & Cacioppo, 1999). Here, Akinbo and colleagues (Akinbo et al., 2022) apply several levels of analysis at the behavioral, physiological, and neurobiological levels to illustrate the detrimental affective consequences of prolonged social isolation in prairie voles, as well as the potential for environmental enrichment to partially ameliorate these deleterious effects. They demonstrated that four weeks of social isolation effectively increases depression- and anxiety-like behavior in the forced swim task and elevated plus maze as well as physiological markers of long-term stress (e.g., adrenal weight) in prairie voles. Moreover, socially isolated prairie voles showed altered dendritic morphology in the basolateral amygdala (BLA), a region of the brain associated with the processing of negative affect. Akinbo et al. further demonstrate that an intervention of environmental enrichment substantially ameliorated the aforementioned negative behavioral and physiological outcomes of social stress, yet it did not reverse the associated outcomes in dendritic morphology in the BLA. For species like humans and prairie voles, our social attachments are deeply important to our mental and physical health. Recent advancements in molecular and genetic tools have allowed researchers to more deeply probe the neural mechanisms of attachment in rodent models of attachment, principally prairie voles. In a review included in this special issue, Berendzen and Manoli (2022) guide us through the relevant developmental and neurobiological concepts necessary to understand the genetic and neuroendocrine factors (e.g., the oxytocinergic system) that subserve social attachment in animals and humans. They synthesize historical findings with new perspectives and elegantly juxtapose the orthodox model of oxytocinergic action on processes of attachment with new, updated models which intriguingly redefine the role of the oxytocinergic system. Our Closest Relatives We often attribute a similarity of affective experience between non-human primates and humans, due to our phylogenetic closeness (Darwin, 1872; Preuschoft & van Hooff, 1995). This issue contains several articles which explore the affective processes of monkeys and apes. Kim and colleagues (In Press) explored the social context of the bared-teeth display in chimpanzees (Pan troglodytes). The bared-teeth display is a facial expression which bears a resemblance to the human smile. Dominants directed these displays towards subordinates during affiliation, suggesting a reassurance function, while the same display was directed by subordinates towards dominance in aggressive contexts, suggesting an appeasement function. The authors thus found multicontextual use for this communicative signal, as well as slight variations in the display that may have different communicatory meanings. Comparative research that includes physiological data measurable in both humans and animals, as well as non-verbal tasks that both humans and animals can perform, can be especially enlightening regarding animal affect. For instance, differing pupil sizes can be an indicator of state of arousal or other social information. In addition, the dot-probe task is utilizable in similar ways in humans and animals (van Rooijen et al., 2017). Zijlstra and colleagues (2022) utilized a dot-probe task to investigate attentional bias towards an affective cue (i.e., differing pupil size), collecting similar data in both humans and bonobos (Pan paniscus). In this task, two stimuli are presented simultaneously, followed by a dot where one of the stimuli was displayed. If attention was consistently on one of the stimuli, the delay to touch the dot when it appears on the same side should be shorter than to touch the dot when it appears on the other side which is receiving less attention. Humans displayed a significant bias towards individuals with larger pupils, while bonobos did not display a bias. While the bonobo results were based on a relatively small sample size, if replicable they will indicate an interesting evolutionary difference in the role of pupil size in social interactions. Laméris and colleagues (2022) also use the dot-probe methodology in orangutans (Pongo pygmaeus) to examine attentional bias towards affective scenes. Using Bayesian analysis, they support the likelihood that orangutans lack this bias which has been found in some other primate species, including rhesus monkeys (Lacreuse et al., 2013) and bonobos (Kret et al., 2016), but significantly not in chimpanzees (Wilson & Tomonaga, 2018). The need for testing of additional subjects, species, and affective contexts provides a promising area for future studies. Debracque and colleague (2022) utilize a different technology, functional near infrared spectroscopy, to explore cerebral activity in baboons (Papio anubis) while hearing aggressive calls from either their own species or a different species (chimpanzee calls). The subjects tested had quite heterogeneous results; however, the technology is promising as a form of non-invasive technology which can be utilized to examine neural reactions to a wide range of affective stimuli. Back to our Lizard Brains On the opposite end of the phylogenetic continuum, it is often more difficult for us to imagine that non-mammals (fish, reptiles, amphibians, invertebrates) experience affect, and the literature in this area is relatively sparse (Braithwaite et al., 2013; Lambert et al., 2019). Paul MacLean suggested that the mammalian limbic system gave mammals superior affective processing and linked this to the mammalian care of their offspring—even calling parental care “the Big News” in mammalian evolution (MacLean, 1977). Maciejewski and Bell (2022) now take the example of parenting behavior to provide a review of this behavior in non-mammalian vertebrates, as well as evidence that there is a conserved similarity in the neurobiological mechanisms for this affect-linked behavior. They also point out that the relative frequency of male and biparental care in non-mammals provides special opportunities to expand our understanding of this behavior, which is rare in mammals (Kleiman, 1977; Kleiman & Malcolm, 1981). Technological Frontiers in the Study of Animal Affect As technology improves, so does our ability to interrogate affect in animals—for instance, via the dot-probe task used in several of the papers in this issue. Positron emission tomography (PET) scans, as well as other modes of imaging and the fNIRS method used by Debracque and colleagues (2022), give us the ability to non-invasively study affect by putting animals in evocative situations during the radiotracer uptake period (Zablocki-Thomas et al., 2022). Eye-tracking can also be used non-invasively in animals to give measures of attention to videos or photos with affective content (Ryan et al., 2019; Yorzinski et al., 2013). Autonomic tools also give us a non-invasive window into the internal affective lives of animals (Murphy et al., 2019). Conclusions and Practical Implications of the Study of Animal Affect The basic philosophical and scientific consideration of animal affect is on its own important, with significant implications for how we consider both the ultimate (evolutionary) and proximate mechanisms of our own human affect. Yet outside of these academic considerations, the study of animal affect has tangible consequences for practices of animal welfare, husbandry, and conservation, as well as human and animal medicine. For example, meat consumption and standards for animal welfare are influenced by how we conceptualize animals as beings with the ability to experience, think, and feel (Braithwaite et al., 2013; Loughnan et al., 2014; Morris et al., 2012; Wilkins et al., 2015). The translation of studies on animal affect into pharmacological treatments to improve human mental health implies a certain amount of shared affective reality between humans and animals (Milton & Holmes, 2018). The papers published in this special issue span the phylogenetic tree from fish to humans but leave room for a great deal of additional exploration in many new species that have not yet been studied, and more in depth questions in those that have. We particularly hope that this special issue highlights the strong role that animal research should play in affective science, as well as the openness of this journal to articles on animal affect. A century and a half ago, Darwin argued that the study of animal emotion deserved further attention: “…we may conclude that the philosophy of our subject has well deserved the attention which it has already received from several excellent observers, and that it deserves still further attention, especially from any able physiologist.” (Darwin, 1872). The collection of articles in this special issue suggests a multitude of new directions in which to take the study of animal affect. Additional Information Funding Not applicable. Conflicts of Interest The authors declare no competing interests. Data Availability Not applicable. Code Availability Not applicable. Authors’ Contribution Both authors contributed to the original draft, editing, and approving the final version. 1 The concept of qing has changed over time and the description here is appropriate for the Ming Dynasty, but not necessarily appropriate in earlier texts, for example, those conveyed by Mencius. Interestingly, an early Confucian philosopher Xunzi (荀況,circa 310 – 238 BCE) defined qing as five specific emotions of hao wu xi nu ai le (好恶喜怒哀乐), or preferences, happiness, anger, grief, and enjoyment (Bruya, 2001). 2 In the study by Dutton and Aron (1974), male gendered (assumed heterosexual) human participants crossed a subjectively dangerous bridge, an anxiogenic experience, followed by an interaction with a subjectively attractive female interviewer. Compared to controls who crossed a subjectively safer bridge or interacted with a male interviewer, these participants demonstrated outcomes consistent with heighted sexual arousal. In the study of gonadally male prairie voles (DeVries et al., 1996), formation of a pair bond with a prospective female mate was facilitated by prior anxiogenic experiences of either the forced swim task or having been injected with corticosterone. In both cases, anxiogenic experiences promoted sexual arousal and social approach. ==== Refs References Akinbo OI McNeal N Hylin M Hite N Dagner A Grippo AJ The influence of environmental enrichment on affective and neural consequences of social isolation across development Affective Science 2022 10.1007/s42761-022-00131-8 Barrett LF Emotions are real Emotion 2012 12 3 413 429 10.1037/a0027555 22642358 Berendzen, K.M., & Manoli, D.S. (2022). Rethinking the Architecture of Attachment: New Insights into the Role for Oxytocin Signaling. Affective Science. 10.1007/s42761-022-00142-5 Bland AR Roiser JP Mehta MA Sahakian BJ Robbins TW Elliott R The impact of COVID-19 social isolation on aspects of emotional and social cognition Cognition and Emotion 2022 36 1 49 58 10.1080/02699931.2021.1892593 33632068 Bliss-Moreau E Constructing nonhuman animal emotion Current Opinion in Psychology 2017 17 184 188 10.1016/j.copsyc.2017.07.011 28950967 Braithwaite VA Huntingford F van den Bos R Variation in emotion and cognition among fishes Journal of Agricultural and Environmental Ethics 2013 26 1 7 23 10.1007/s10806-011-9355-x Bruya, B. (2001). Quing 情 and emotion in early chinese thought, Ming Qing Yanjiu, 10(1), 151–176. 10.1163/24684791-90000410 Canetta S Kellendonk C Can we use mice to study schizophrenia? Philosophical Transactions of the Royal Society b: Biological Sciences 2018 373 1742 20170032 10.1098/rstb.2017.0032 Carroll L Alice’s adventures in wonderland 1865 MacMillan & Company Carter CS Getz LL Monogamy and the prairie vole Scientific American 1993 268 6 100 106 10.1038/scientificamerican0693-100 8516669 Darwin C The expression of the emotions in man and animals 1872 John Murray Debracque, C., Gruber, T., Lacoste, R., Meguerditchian, A. & Grandjean, D. (2022). Cerebral Activity in Female Baboons (Papio anubis) During the Perception of Conspecific and Heterospecific Agonistic Vocalizations: a Functional Near Infrared Spectroscopy Study. Affective Science. 10.1007/s42761-022-00164-z DeVries AC DeVries MB Taymans SE Carter CS The effects of stress on social preferences are sexually dimorphic in prairie voles Proceedings of the National Academy of Sciences 1996 93 21 11980 11984 10.1073/pnas.93.21.11980 Dutton DG Aron AP Some evidence for heightened sexual attraction under conditions of high anxiety Journal of Personality and Social Psychology 1974 30 4 510 517 10.1037/h0037031 4455773 Ekman P Darwin’s contributions to our understanding of emotional expressions Philosophical Transactions of the Royal Society b: Biological Sciences 2009 364 1535 3449 3451 10.1098/rstb.2009.0189 Ernst JM Cacioppo JT Lonely hearts: Psychological perspectives on loneliness Applied and Preventive Psychology 1999 8 1 1 22 10.1016/S0962-1849(99)80008-0 Fortenbaugh WW Aristotle: Animals, emotion, and moral virtue Arethusa 1971 4 2 137 165 Getz LL Carter CS Prairie-vole partnerships American Scientist 1996 84 1 56 62 Getz, L. L., Carter, C. S., & Gavish, L. (1981). Field and Laboratory Evidence for Pair-Bonding. Behavioral Ecology and Sociobiology, 8, 189–194. Huang MW Sentiments of desire: Thoughts on the cult of Qing in Ming-Qing literature Chinese Literature: Essays, Articles, Reviews (CLEAR). 1998 20 153 10.2307/495268 Kleiman DG Monogamy in mammals The Quarterly Review of Biology 1977 52 1 39 69 10.1086/409721 857268 Kleiman DG Malcolm JR Gubernick DJ Klopfer PH The evolution of male parental investment in mammals Parental Care in Mammals 1981 Springer 347 387 Kret ME Jaasma L Bionda T Wijnen JG Bonobos ( Pan paniscus ) show an attentional bias toward conspecifics’ emotions Proceedings of the National Academy of Sciences 2016 113 14 3761 3766 10.1073/pnas.1522060113 Lacreuse A Schatz K Strazzullo S King HM Ready R Attentional biases and memory for emotional stimuli in men and male rhesus monkeys Animal Cognition 2013 16 6 861 871 10.1007/s10071-013-0618-y 23463380 Lambert Carder D’Cruze Given the cold shoulder: A review of the scientific literature for evidence of reptile sentience Animals 2019 9 10 821 10.3390/ani9100821 31627409 Laméris, D.W., van Berlo, E., Roth, T.S., & Kret, M.E. (2022). No Evidence for Biased Attention Towards Emotional Scenes in Bornean Orangutans (Pongo pygmaeus). Affective Science. 10.1007/s42761-022-00158-x Loughnan S Bastian B Haslam N The psychology of eating animals Current Directions in Psychological Science 2014 23 2 104 108 10.1177/0963721414525781 Maciejewski, M.F., & Bell, A.M. (2022). Insights into Parental Care from Studies on Non-mammalian Vertebrates. Affective Science. 10.1007/s42761-022-00127-4 MacLean PD The triune brain in conflict Psychotherapy and Psychosomatics 1977 28 14 207 220 10.1159/000287065 416456 Mendl M Neville V Paul ES Bridging the gap: Human emotions and animal emotions Affective Science 2022 10.1007/s42761-022-00125-6 Milton AL Holmes EA Of mice and mental health: Facilitating dialogue and seeing further Philosophical Transactions of the Royal Society b: Biological Sciences 2018 373 1742 20170022 10.1098/rstb.2017.0022 Morris P Knight S Lesley S Belief in animal mind: Does familiarity with animals influence beliefs about animal emotions? Society & Animals 2012 20 3 211 224 10.1163/15685306-12341234 Murphy, A. M., Ross, C. N., & Bliss‐Moreau, E. (2019). Noninvasive cardiac psychophysiology as a tool for translational science with marmosets. American Journal of Primatology, 81(9). 10.1002/ajp.23037 Palgi Y Shrira A Ring L Bodner E Avidor S Bergman Y Cohen-Fridel S Keisari S Hoffman Y The loneliness pandemic: Loneliness and other concomitants of depression, anxiety and their comorbidity during the COVID-19 outbreak Journal of Affective Disorders 2020 275 109 111 10.1016/j.jad.2020.06.036 32658811 Passmore J The treatment of animals Journal of the History of Ideas 1975 36 2 195 10.2307/2708924 11610245 Preuschoft S van Hooff JARAM Homologizing primate facial displays: A critical review of methods Folia Primatologica 1995 65 3 121 137 10.1159/000156878 Roberts RL Williams JR Wang AK Carter CS Cooperative breeding and monogamy in prairie voles: Influence of the sire and geographical variation Animal Behaviour 1998 55 5 1131 1140 10.1006/anbe.1997.0659 9632499 Robinson ESJ Translational new approaches for investigating mood disorders in rodents and what they may reveal about the underlying neurobiology of major depressive disorder Philosophical Transactions of the Royal Society b: Biological Sciences 2018 373 1742 20170036 10.1098/rstb.2017.0036 Rogers FD Bales KL Mothers, fathers, and others: Neural substrates of parental care Trends in Neurosciences 2019 42 8 552 562 10.1016/j.tins.2019.05.008 31255381 Ryan AM Freeman SM Murai T Lau AR Palumbo MC Hogrefe CE Bales KL Bauman MD Non-invasive eye tracking methods for new world and old world monkeys Frontiers in Behavioral Neuroscience 2019 13 39 10.3389/fnbeh.2019.00039 30890923 Seelke AMH Bond JM Simmons TC Joshi N Settles ML Stolzenberg D Rhemtulla M Bales KL Fatherhood alters gene expression within the MPOA Environmental Epigenetics 2018 4 4 dvy026 10.1093/eep/dvy026 30568805 Sihvola, J. (1996). Emotional animals: Do Aristotelian emotions require beliefs? Apeiron, 29(2). 10.1515/APEIRON.1996.29.2.105 Sung W Mencius and Xunzi on Xing (Human Nature): Mencius and Xunzi on Xing (Human Nature) Philosophy Compass 2016 11 11 632 641 10.1111/phc3.12363 van Rooijen R Ploeger A Kret ME The dot-probe task to measure emotional attention: A suitable measure in comparative studies? Psychonomic Bulletin & Review 2017 24 6 1686 1717 10.3758/s13423-016-1224-1 28092078 Wilkins AM McCrae LS McBride EA Factors affecting the human attribution of emotions toward animals Anthrozoös 2015 28 3 357 369 10.1080/08927936.2015.1052270 Williams JR Catania KC Carter CS Development of partner preferences in female prairie voles (Microtus ochrogaster): The role of social and sexual experience Hormones and Behavior 1992 26 3 339 349 10.1016/0018-506X(92)90004-F 1398553 Wilson DA Tomonaga M Exploring attentional bias towards threatening faces in chimpanzees using the dot probe task PLoS ONE 2018 13 11 e0207378 10.1371/journal.pone.0207378 30485317 Yorzinski JL Patricelli GL Babcock JS Pearson JM Platt ML Through their eyes: Selective attention in peahens during courtship Journal of Experimental Biology 2013 216 16 3035 3046 10.1242/jeb.087338 23885088 Young LJ Wang Z The neurobiology of pair bonding Nature Neuroscience 2004 7 1048 1054 10.1038/nn1327 15452576 Zablocki-Thomas PB Rogers FD Bales KL Neuroimaging of human and non-human animal emotion and affect in the context of social relationships Frontiers in Behavioral Neuroscience 2022 10.3389/fnbeh.2022.994504 Zijlstra, T.W., van Berlo, E., & Kret, M.E. (2022). Attention Towards Pupil Size in Humans and Bonobos (Pan paniscus). Affective Science. 10.1007/s42761-022-00146-1	36514490	PMC9734565	NO-CC CODE	2022-12-14 23:43:10	no	Affect Sci. 2022 Dec 3; 3(4):697-702	utf-8	Affect Sci	2,022	10.1007/s42761-022-00168-9	oa_other
==== Front Am J Psychoanal Am J Psychoanal American Journal of Psychoanalysis 0002-9548 1573-6741 Palgrave Macmillan UK London 36470990 9383 10.1057/s11231-022-09383-6 Article Self-constitution and “Infrastructural” Change: An Interdisciplinary Account of Psychoanalytic Action Brakel Linda A. W. [email protected] Linda A. W. Brakel, M.D. is Associate Professor (adj) in Psychiatry and Research Faculty in Philosophy at the University of Michigan and a Faculty Member at The Michigan Psychoanalytic Institute. 525 Third Street, Ann Arbor, MI 48103 USA 5 12 2022 113 © Association for the Advancement of Psychoanalysis 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. Beyond revealing unconscious pathological identifications and traits—including their past usefulness but current toxicity—what techniques in our psychoanalytic practice can lead to change? Radically different from mainstream philosophical views advocating that such undesirable self-aspects should not be endorsed as Self, psychoanalysts hold that these negative traits must instead be understood as part of one’s Self. But then what? Investigating concepts from classical conditioning, neuroscience, the philosophy of mind and action, and psychoanalytic practice itself, this article will suggest a preliminary account of the mechanism of action of psychoanalytic work after insight. Keywords de-identification mind-body physicalism classic conditioning technical issues of working-through akrasia emotional muscle technical issues after insight influence of life experiences of analyst ==== Body pmcIntroduction What leads to change in our psychoanalytic work? How can knowing and understanding unconscious pathological and undesirable elements of ourselves—elements we didn’t know and certainly didn’t understand—help us to be different, better? Considerable effort is required even to learn of these troubling aspects; more still to understand them. This is the work taking place routinely within the normal process of psychoanalysis, where many transferences develop and then are analyzed. So far, so good; but this is indeed only so far. There is an unanswered further question. After the recognition and understanding of problematic self-aspects, after developing insight, we need to ask, Then what? Freud often wrote as if his patients, after understanding the meaningful origins of their unconsciously motivated irrational neurotic behavior, would thereafter just behave more rationally, less neurotically. Here is what he says for instance about Little Hans (Freud, 1909a):But I must now enquire what harm was done to Hans by dragging to light in him complexes that are not only repressed by children but dreaded by their parents? ... On the contrary, the only results of the analysis were that Hans recovered, that he ceased to be afraid of horses, and that he got on to rather familiar terms with his father (Freud, 1909a, p. 145). Even more telling are Freud’s final comments in the Rat Man case (Freud, 1909b). It is is clear here that Freud attributed the curative result to the extensive and detailed uncovering of the Rat Man’s unconscious connections—connections believed by Freud and his patient to have underlain and caused the Rat Man’s former symptomatic and highly undesirable Self-notions.We should not be justified in expecting such severe obsessional ideas as were present in this case to be cleared up in any simpler matter or by any other means. When we reached the solution that has been described above, the patient’s rat delirium disappeared (Freud, 1909b, p. 220). But because this clear and happy outcome is not the experience of most analysts and most patients–in fact it was not the consistent experience of Freud either (see the story of the Wolf Man, [Freud, 1918]2 or the story of Freud’s other long term patient, Frau Elfriede Hirschfeld, [Falzeder, 1994]3)–we need to investigate possible additional mechanisms that allow or promote change toward successful outcomes, when persons “own” their role in their difficulties and identify with more constructive ways of being and living. INITIAL WORK TOWARD DE-IDENTIFYING: WHAT WE ALREADY KNOW AND WHAT WE ALREADY DO Radically different from the account of Self-Constitution widely regarded as mainstream in philosophy of action circles, a philosopher colleague and I presented a new view of Self-Constitution very much influenced by psychoanalytic principles. The mainstream philosophical picture can be summarized thus: “Identify the undesirable elements and then subdue and extrude them…Do not allow [these] unlicensed parts of your psychology to become your will [your Self]” (Fileva & Brakel, forthcoming). This view’s most prominent proponent is Harry Frankfurt. Although Frankfurt later refined his view somewhat (see especially, Frankfurt, 1991), his early distinction between the willing and unwilling addict makes clear the enduring essentials of this sort of extruding account (Frankfurt, 1977). Frankfurt’s recommendations (which follow below) are admittedly outside the therapeutic matrix, without clinical considerations, and from a particular philosophical perspective. They are nonetheless important, targeted not to the willing addict—one who accepts, even embraces the addiction—but to the unwilling addict—the addict who does not endorse the actions and desires associated with his/her pathological addiction:In rejecting the desire…the person withdraws himself from it. He places the rejected desire outside the scope of his preferences…Although he may continue to experience the rejected desire as occurring…the person brings it about…that the occurrence is an external one. The desire is then no longer to be attributed strictly to him… (Frankfurt, 1977, p. 67). Our account, which we call “understanding first” (Fileva & Brakel, forthcoming) and which derives almost entirely from psychoanalytic theory, is obviously quite different. Briefly we aver that toxic identifications, undesirable traits and desires, should not (cannot) be extruded, and thus quite prematurely (wrongly) held as “not-me.” To do so is mere wishful thinking. Instead, such elements need to be recognized as part of one’s Self. This usually takes place in an earlier stage in the usual psychoanalytic work, discovering negative self-aspects and identifications as they emerge in the transference. Next, the patient and analyst can explore the meaningful, perhaps even useful (at least perceived as useful), early role for these toxic identifications and self-aspects when first taken on. Along with this, and quite importantly, the patient can comprehend his/her agential role in employing these toxic elements. Then in the next phase, the patient can readily appreciate a fundamental contrast: The deleterious, undesirable trait/desire/identification, while still a part of the Self, no longer serves any useful role. With this in mind, the patient can agentially try to de-identify, etc. Note, it is quite deliberately that I write “try to de-identify” because, not only is it hard, but more significantly, it is not clear how a person acquires de-identification. The section below is devoted to investigating theories pertaining to how one can de-identify internal aspects, specifically identifications, which are no longer viable and frankly harmful. How to De-identify Why is it so hard? First, let us explore why it is so hard to de-identify. There are several possibilities, with some more relevant for particular persons than others. Thus, for some people an identification that is no longer desirable, and is in fact toxic, comprises a lot of one’s Self. Take for example, Mr. A., a patient who has closely identified with his bullying father—this a rather classic case of identification with the aggressor.4 Mr. A. bullied routinely, with the notion that the world consists only of those who bully and those who get bullied. His father (Dr. A. Sr.) bullied his work associates and patients, and even more painfully, the A. family children, as well as his own wife, Mrs. A. Sr., my patient’s mother. In the treatment Mr. A. was made aware that this sort of either/or choice is typical for young children, but that he had had, and continued to have, some agency in identifying with his bullying father. Actually, other important objects (persons) with whom to identify, neither bullied nor bullying, were available in his childhood, and young adulthood, and continue to be available now in his middle adult years. Mr. A.’s case demonstrates another reason that de-identifying is so hard, namely, that continued identification allows the identify-er to maintain continued “good relations” with the original object, even if this original object is dead. Mr. A.’s father saw the world in terms of Winners and Losers (reminiscent of Trump’s father), and if Mr. A. so much as mitigated his bullying ways, he was sure his father would see him as a Loser. Let me turn to a different patient to illustrate another very potent reason why de-identifying is so hard. The patient, my very first and most enduring psychoanalytic patient, is me, Linda A.W. Brakel. This example describes the influence of an aspect of the analyst’s life experiences. And the issue concerns pleasure. To the extent there is unconscious (or even conscious) pleasure along with the pain of a toxic identification, that negative identification will be hard to modify. As a 3-year-old child I underwent a major surgical procedure, with all its antecedent and post-surgical goings-on. Although it went very well, it was obvious, even to me, that my parents both felt helpless, and that my mother in particular believed from that point forward that I was going to die. I have identified with this view of my mother’s, both to keep in good relations with her, as above; and for the following formerly unconscious pleasures. There is a relief factor, which occurs whenever I undergo an acute attack of “I’m going to die” but soon learn that the medical findings are not so bad. The pleasure here is best understood by reviewing a famous and seminal experiment, whose finding were reported in a Psychological Science journal article with the revealing title: “When More Pain is Preferred to Less: Adding a Better Ending” (Kahneman et al., 1993). Kahneman more recently (2011, p. 382) recounted the basics: People preferred increasing the duration of moderate unpleasure—keeping a hand in very cold water longer; this in order to experience a significant lessening of unpleasure at the end—as water temperature was raised to a comfortable level. Relatedly in my case: the good medical news that the bad medical news is not so bad, is often delivered by doctors, parental transference figures who are quite the opposite of my helpless, overwhelmed parents. A final example of one of the difficulties in de-identification comes from a patient who is an internal medicine doctor. She talked of a long-standing harmful self-view, picturing herself as alone, abandoned, having to take care of herself, even when she is but minimally able, being so young and small. Beginning in early childhood, she felt that for different reasons neither parent was able or willing to attend to her when exigencies would arise. Her mother, successful in the world but also domineering, and often frankly mean, was too self-involved, while her father was too engaged in placating her mother. As an adult professional she finds this self-view highly problematic, especially insofar as it also entails an identification with her arrogant, aggressive mother. This can interfere with relations both at home and at work, as she misjudges others’ intent, feeling put-upon to take on everything, even as she feels incompetent or over-taxed. But my patient suggests that this self-view, as problematic as it is, might be hard to modulate to the extent that it keeps something else at bay—something that is both more realistic, but also more distressing. Note, I am writing this in the Summer of 2021. Given that my patient is a practicing internal medicine doctor who has continued to work during the ongoing deadly COVID-19 pandemic, it is not hard to conjure just what sort of thing, both more realistic and distressing, one would want to defensively “keep at bay.” Techniques for De-identifying Granted that it is a difficult task, what can one do after first recognizing and then wanting to change a pathological identification? First, there are the regular extensions of psychoanalysis. Again, patients can begin to appreciate, likely in the transference, that the patient him/herself has had agency in continuing the harmful identification past its original usefulness. Also, patients can discover that there are (and perhaps were) other objects with whom to identify. Moreover, patients wanting to de-identify can allow and promote understanding of the real person with whom the patient has identified. Why was Mr. A.’s father such a bully? Why did his mother put up with it? What accounted for my patient’s mother’s overarching self-concern? Why was my mother convinced that I would die young?5 Further psychoanalytic techniques are less well understood, but clearly relevant to de-identification processes. I refer here to “working-through” and relatedly grieving. In both, the basic notion is that reviewing/re-working the problematic or painful contents from different angles helps to effect actual change. The working through process is often not smooth, but full of enactments: “When two personalities meet, if they make sufficient contact to be aware of each other, they create an emotional storm… One does not immediately know what the emotional storm is, but the problem is how to make the best of it” (Bion, 1979, p. 247). Michael Feldman (2009) writes that “'making the best of it’ refers to the struggle to become aware of this fact, to tolerate the experience and to begin to examine the nature of the disturbance” (Feldman, 2009, p. 161). The “emotional storm” is Bion’s characterization of resistance/transference as mediated by countertransference and only imperfectly communicated interpretive comments. Although admittedly speculative, I have posited a different perspective (Brakel, 2013, Chapter 3), a brain-based view of mental content—one that could give specific substance to this sort of working-through mechanism. My account advances a particular physicalist (as opposed to dualist) solution to the mind/body problem, “Diachronic Conjunctive Token Physicalism” (DiCoToP). In this view every singular mental event (including mental contents) exists as a brain event consisting of an assembly of neurons, along with whatever neurochemical processes facilitate their connection. That there is nothing over and above these brain goings-on makes this account a physicalist view.6 Each singular instance of the same event/same mental content (each token)—take for example, “I miss my dog”—is populated by a slightly different neuronal network, insofar as it occurs at a different time and likely a different place. The sum of all of these instances of this content, i.e., all of their slightly variable neuronal assemblies (the conjunction), over time (diachronic), comprises the mental content. Let’s use once again the following mental content as a simple example, “I miss my dog.” According to the DiCoToP account of mental events: Dealing with the grief that that particular mental content represents requires a great number of neuronal assembly re-alignments, summed over time (i.e., necessary experiential re-workings of that particular mental content in myriad contexts). Working-through neurotic contents, including harmful negative identifications, would proceed in a similar fashion. In a further (and similarly speculative) attempt to explain working-through, I have held that there is an important parallel between aspects of classic psychoanalytic technique and the de-conditioning (if not complete extinction) of unconditioned aversive responses to various conditioned triggers—triggers that thereby activate pathological identifications and/or other undesirable self-traits. (Brakel, 2013, Chapter 2). The parallel operates as follows: From the side of classic conditioning, researchers (See Mystkowski & Mineka, 2007, p. 218: Mystkowski, Craske, & Echiverri, 2002, p. 414) have found that in order for extinction to be successful, the conditioned stimulus—the event, item, or situation conditioned to predictably precede the unconditioned aversive stimulus and therefore the automatic toxic response—must now be presented in many trials, over time, and in multiple different settings without the unconditioned stimuli to follow. These are termed “Safe-here” and “Safe now” trials. From the side of psychoanalysis, the patient can experience, just such “Safe-here” and “Safe now” episodes of conditioned triggers to which the analyst responds kindly/benignly—in other words always without the unconditioned stimuli (e.g., censure, ridicule) that had usually followed. Without the unconditioned stimuli, the toxic aversive automatic response-sequelae are avoided. And as for the requirement that many and myriad contexts for “Safe-here” “Safe-now” trails are needed—this requirement is met within the multiple different transferences developed over time in the psychoanalytic setting. In addition to working-through, there is another potential avenue to travel in order to change one’s negative identifications after recognizing and then owning them. This avenue can be found in an area best characterized as belonging to two domains: (a) developmental psychoanalysis and (b) the philosophy of action. Child and adult psychoanalysts Kerry and Jack Novick (2001, 2003, 2006, 2010, 2011), in a series of publications advanced a concept they call, “emotional muscle” which they explain “…grew first for us out of our clinical work, as we grappled with the limitations of insight alone to effect lasting change” (2011, p. 137). With improving one’s emotional muscle, a metaphor readily analogous to the familiar task of building physical muscle, the Novicks suggest that a person can increasingly participate in a list of qualities and capacities that are almost universally valued and best described as “…virtues, strengths, will, character, grit…[with] one common thread…that they all imply effort, resolve, and strength” (pp. 135–136). In the same spirit, on the philosophy of action account too, changing harmful but familiar patterns—patterns consequent to primitive but ingrained harmful identifications—requires strength of will. When this strength is insufficient, and a person thereby acts contrary to his/her own rational judgment of his/her best interest, that act is termed an “akratic” act (Davidson, 1970). The word akrasia is Greek and literally means weakness of will. So how can understanding the descriptive characteristics of akrasia help in the process of de-identification? Returning to Mr. A. above, suppose he’s in a situation in which the desire to bully some underling is almost irresistible. “Almost” is the operative word here because he’s recognized not only the origins of this identification, but also its highly problematic nature for his own sense of well-being. So, what can he do? I have proposed (Brakel, 2009) the following psychological reconfiguring in this type of situation: Suppose instead of Mr. A. thinking, “I will just bully this guy now; it is just one more time, and that will be my last bullying act” he could say, “In my new Self-identification as a non-bully, I will forgo bullying this guy now, and resist future opportunities.” My model here was a smoker, desiring henceforth to be a non-smoker, with the pull of one cigarette now at time, t –i.e., synchronic; weighing against the non-smoker designation over time, t, t+1, t+2, …t+n —a new diachronic self-label. Let me offer one final technique for de-identifying, a technique that is at once most easily accessible and also the one entailing the most erudition. I will also give it two names: (a) The Whistle a Happy Tune Method, and (b) Changes in Behavior Effect Brain Synaptic Changes. Noting that (b) is quite consistent with the DiCoToP mind/body view offered (earlier), let’s take up the (a) version first. Consider the following lyrics (shortened a bit) from “Whistle a Happy Tune,” (Rodgers & Hammerstein, 1951) one of the popular songs from “The King and I:” (Lang & Bracket, 1956).Whenever I feel afraid, I…whistle a happy tune, So no one will suspect I'm afraid,… The result of this deception, Is very strange to tell, For…I fool myself as well,… I whistle a happy tune, And every single time, The happiness in the tune, Convinces me that I'm not afraid, Make believe you're brave… [And] You may be as brave, As you make believe you are. First performed in 1951, the science behind the content of the song—that behavioral changes can produce synaptic brain changes [i.e., Method (b) above]—was just beginning. A Polish neurophysiologist, Konorski (1948), first proposed that neuronal connections underwent morphological changes consequent to learning. Shortly after that, Hebb (1949) postulated neural plasticity as the underlying mechanism of learning. “Hebbian plasticity” animates the 70 plus years of research, still being undertaken, and largely bears this postulate out. For example, it has been fairly well established that at least two different types of neuronal changes occur consequent to behavioral tasks, including learning. In a recent summary article in Frontiers in Cellular Neuroscience regarding research in this area, the authors (Mateos-Aparicio & Rodriquez-Moreno, 2019, p. 2) state: “In parallel with activity-dependent changes in synaptic strength and efficiency of synaptic transmission, structural modification of axonal, dendritic branches, and spine morphology occurs, a phenomenon called structural synaptic plasticity.” Moreover, these authors conclude (p. 3): “Indeed the ability to manipulate specific neuronal pathways and synapses has important implications for therapeutic and clinical interventions that will improve our health.” Various promising therapies, they conclude (p.3), behavioral as well as neuropharmacological and neurophysiological (deep brain stimulation) “…are all based on our current understanding of brain plasticity.” Indeed, what follows from the brain’s plasticity as outlined, both here and in the DiCoToP mind/body account, are neuronal synaptic changes leading to neuronal network changes, which when reinforced by continued behavioral inputs, effect behavioral changes. Thus, as you continue to act bravely, you may become brave, and ultimately be brave. CAN THESE TECHNIQUES WORK FOR CHANGING TOXIC POLITICAL IDENTIFICATIONS? This section must unfortunately be rather brief. The simple answer to this section’s title question, is likely “No.” These techniques cannot work insofar as those whose political self-identification are widely regarded as toxic, do not themselves categorize these identifications as toxic. In fact, these very identifications are often worn as proud badges of tribal belonging (see Brakel & Foxall, 2022). Those with these toxic identifications therefore lack any desire to de-identify. They are in this way like Frankfurt’s willing addicts (described earlier)—persons who willfully (with agency) embrace their addiction as part of their Self. There can, however, be exceptions. In the arduous course of analyzing a whole person, even firmly held and frankly toxic political views might be modified—this as a concomitant, unexpected, and happy outcome to the shared analytic work on aspects of the patient’s Self that he/she has deemed painful and in need of change.7 Notes Linda A. W. Brakel, MD, is an Associate Professor (adjunct) in Psychiatry and a Faculty Research Associate (adjunct) in Philosophy, both at the University of Michigan. She is also on the faculty of the Michigan Psychoanalytic Institute. Brakel has had around 40 years of experience as a clinical psychoanalyst, and has spent some decades in empirical investigations of researchable aspects of psychoanalytic theory, mostly involving primary process. Her most recent work is interdisciplinary—philosophy of mind, action, and experimental philosophy; and evolutionary and cognitive psychology. Brakel has co-authored 3 books and authored 4 solo volumes. These last include: Philosophy, Psychology, and the A-Rational Mind (Oxford University Press, 2009), Unconscious Knowing (Oxford University Press, 2010), The Ontology of Psychology (Routledge, 2013); Investigations into the Trans Self and Moore’s Paradox (Palgrave-Macmillan, 2021). “Sergei Pankejeff, a wealthy Russian aristocrat, sought treatment with Freud in 1910, a few years after both his only sibling (the older sister) and his father committed suicide. He suffered from serious depression, spent time in various sanatoriums, in treatment with several doctors, but none of them was able to cure him. He became the famous Wolf Man case of Freud (1918), who conceptualized the inner world of his patient but was unable to deal with the immediacy of the transference of his complex case (Meltzer, 1978). At age 79, Mr. Pankejeff was interviewed and he lamented that all his life he was in and out of analysis, with his condition worsening, and told the interviewer in despair that ‘the whole thing looks like a catastrophe. I am in the same state as when I came to Freud, and Freud is no more’ (Obholzer, 1982)” (as cited in Kupermann, 2017, p. 263). [Eds.] As pointed out earlier, Freud’s early view was that interpretations would lead naturally to timely change (Freud, 1909a, 1909b). Since then, in the widening scope of psychoanalysis (Stone, 1954), we have learned that many patients require a much longer period of enactment, of “working through, possibly full of emotional storms” as Bion (1979) explained. Some of Freud’s long-term patients, for example, the very well-known case of the Wolf Man (Freud, 1918) or the less familiar Frau Elfriede Hirschfeld, often challenged the patience of Freud. Freud treated Frau Hirschfeld for many years, and described her case in at least six papers under different pseudonyms (Falzeder, 1994, p. 325). Frau Hirschfeld chronically frustrated Freud with her refusal or inability to accept his interpretations. He called her “analytically of no use for anybody” (p. 309). Freud struggled with his countertransference toward Frau Hirschfeld, was torn between his empathic devotion and a wish to withdraw from a deep understanding and to treat her strictly. This case led to a turning point of Freud’s more critical and pessimistic assessment of the curative powers of psychoanalysis (p. 318). [Eds.] Frankel (2017) asks: ‘“How do children react to gross emotional abandonment and exploitation by their parents? These children feel compelled to ‘‘subordinate themselves like automata to the will of the aggressor to divine each one of his desires and to gratify these; completely oblivious of themselves they identify themselves with the aggressor’’ (Ferenczi, 1933, p. 162). The other person’s desires replace the child’s own wishes and perceptions. [We note that this use of the term identification with the aggressor differs from, and precedes, Anna Freud’s (Freud, 1936) usage, which is about ridding oneself of the helplessness of victimization by becoming an aggressor toward someone else.] Why does the child identify and comply? To manage the out-of-control parent in order to survive, first of all; and to feel approved of, valued, important, loved, wanted by the parent—to feel they belong in the family, which, for children, is key to survival. The child has no choice. But no one can wholly give herself over to helplessness, or forsake her psychic existence completely. The child will try to salvage part of herself, albeit unconsciously; she will take illusory control through fantasy, and insist that she is loved by her abusive—but loved and needed—parent; the abuse may even be seen as love. Ferenczi wrote that in the abused child’s mind, even as she gives up her self and submits, the aggressor “disappears as part of the external reality, and becomes intra- instead of extra- psychic; the intra-psychic is then subjected, in a dream-like state as is the traumatic trance, to the primary process, i.e. according to the pleasure principle it can be modified or changed by the use of positive or negative hallucinations. In any case, the attack as a rigid external reality ceases to exist and in the traumatic trance the child succeeds in maintaining the previous situation of tenderness”’ (Ferenczi, 1933, p. 162). (Frankel, 2017, p. 219.) [Eds.] This one I can answer at least partially. When my mother was a child, her own mother got a diagnosis from the Cleveland Clinic [no less] that she, my grandmother, would only live into her 30s. This was wrong—my grandmother died at 86—but this prognostic diagnosis led to profound fear, deleteriously affecting my mother’s childhood, and therefore mine. Diachronic Conjunctive Token Physicalism (DiCoToP) is technically speaking, a reductive physicalist account. For a basic level understanding of the differences between reductive and non-reductive physicalism see Brakel, 2013, Chapter 3. One can wonder about other techniques for changing identifications. Take a recent experiment brought to my attention thanks to the editor, Dr. Galdi: Two political scientists (Broockman & Kalla, 2022) conducted a field study with over 700 self-identified strongly conservative Fox-news-watching participants. Forty percent of these persons were randomly assigned to the treatment group. This entailed incentivizing them by giving them $15 an hour to watch 7 hours of CNN news per week for one month. Three days after the viewing portion of the study ended, the researchers found that compared with the control participants, the treated group demonstrated (p.3) “…changes in evaluations of Donald Trump and Republican candidates and elected officials.” These changes were significant and uniformly in the direction of a less positive view of Trump and Republicans. However, the results of a follow-up assessment several weeks later—given the thrust of the current psychoanalytic article—are perhaps more salient. There, Broockman and Kalla (p.3): “…found [that] these impacts largely receded as treated participants primarily returned to their prior viewership habits…[Moreover] that participants’ attitudes meaningfully shifted at first away from…[but] then back towards their partisan side along with changes in their viewership behavior…” Clearly these are not the sort of longstanding therapeutic changes one aspires to and works for in psychoanalysis. The editors agreed with my conclusion, but emphasized that it would be unrealistic to expect lasting changes after a ONE MONTH experiment. The fact that those small, albeit fleeting, changes took place at all, underlines the thesis that a less rigidly strident environment can be potentially helpful for people with strong and inflexibly held identifications. Acknowledgement I would like to thank Dr. Karen E. Klein for many helpful discussions on the matters at issue. Also, I thank the editors, who have enriched this work with useful suggestions. Address correspondence to Linda A. W. Brakel, MD, 525 Third Street, Ann Arbor, Michigan 48103, USA. Email: [email protected] Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Bion WR Bion F Making the best of a bad job Clinical seminars and other works 1979 Routledge 321 332 Brakel LAW Philosophy, psychoanalysis and the a-rational mind 2009 Oxford University Press Brakel, L. A. W. (2013). The ontology of psychology: Questioning foundations in the philosophy of mind. London & New York: Routledge Press. [This was through Routledge philosophy division] Brakel LAW Foxall GF Vaccine refusal—A preliminary interdisciplinary investigation Frontiers in Public Health, section Health Economics 2022 10 917929 10.3389/fpubh.2022.917929 Broockman, D. & Kalla, J. (2022). The manifold effects of partisan media on viewers’ beliefs and attitudes: A field experiment with Fox News viewers. Retrieved April 28, 2022, from 10.31219/osf.io/jrw26 Davidson, D. (1970). How is weakness of the will possible? In J. Feinberg (Ed.), Moral concepts. Oxford Reading in Philosophy (pp. 93–113). Oxford University Press. Republished in D. Davidson (Ed.), Essays on action and events (pp. 21–42). Oxford: Clarendon Press. 1980. Falzeder E My grand-patient, my chief tormentor: A hitherto unnoticed case of Freud’s and the consequences Psychoanalytic Quarterly 1994 63 297 331 10.1080/21674086.1994.11927416 8047634 Feldman M Joseph B Doubt, conviction and analytic process Selected papers of Michael Feldman 2009 Routledge Ferenczi, S. (1933). Confusion of tongues between adults and the child. The language of tenderness and of passion. In Final contribution to the problems and methods of psychoanalysis. (pp. 156–167). London: Karnac Books. 1994. Also in International Journal of Psychoanalysis, 30, 225–230. 1949. Fileva, I. & Brakel, L.A.W. (forthcoming). A new view of Self-constitution. Philosophy, Psychiatry, & Psychology. Frankel J Ferenczi’s evolving conception of narcissistic pathology and its basis in trauma American Journal of Psychoanalysis 2017 77 213 222 10.1057/s11231-017-9097-2 28779109 Frankfurt, H. G. (1977). Identification and externality. In A. O. Rorty (Ed.), The identities of persons. (pp. 239-252). Los Angeles, CA: University of California Press. Republished in D. Frankfurt (Ed.). The importance of what we care about: Philosophical essays. (pp. 58–68). New York: Cambridge University Press, 1988. Frankfurt, H. (1991). The faintest passion. Proceedings and Addresses of the American Philosophical Association, 66/3, 5–16. Freud A The Ego and the mechanisms of defense 1936 International Unversities Press Freud, S. (1909a). Analysis of a phobia in a five-year-old boy. Standard Edition, Vol. 10, (pp. 145–156). London: Hogarth. Freud, S. (1909b). Notes upon a case of obsessional neurosis. Standard Edition, Vol. 10, (pp. 155–318). London: Hogarth. Freud, S. (1918). From the history of an infantile neurosis. Standard Edition, Vol. 17 (pp. 1–122). London: Hogarth. Hebb DO The organization of behavior: A neuropsychological theory 1949 Wiley Kahneman D Thinking fast and slow 2011 Farrar, Straus, and Giroux Kahneman D Frederickson B Schreiber C Redelmeier D When more pain is preferred to less: Adding a better ending Psychological Science. 1993 4 6 401 405 10.1111/j.1467-9280.1993.tb00589.x Konorski J Conditioned reflexes and neuron organization 1948 Cambridge University Press Kupermann D Searching for a sensitive way of working through American Journal of Psychoanalysis 2017 77 255 264 10.1057/s11231-017-9101-x 28740195 Lang, W. (Director) & Bracket, C. (Producer). The King and I. [Motion Picture]. (1956). Los Angeles: 20th Century Fox/Walt Disney. USA. [Based on the novel, Anna and the King of Siam, by Margaret Landon. New York: Harper Collins. 1944]. Mateous-Aparicio, P. & Rodriguez-Moreno, A. (2019). The impact of studying brain plasticity. Frontiers in Cellular Neuroscience. Vol. 13, Article 66, pp. 1–5. 10.3389/fncel.2019.00066 Meltzer D The Klenian development 1978 Oxford Clunie Press Mystkowski J Craske M Echiverri A Treatment context and return of fear in spider phobia Behavior Therapy. 2002 33 399 416 10.1016/S0005-7894(02)80035-1 Mystkowski, J. & Mineka, S. (2007). Behavior therapy for fears and phobias: Context specificity for fear extinction. In T. A. Treat, R. R. Bootzin & T. B. Baker (Eds.), Psychological Clinical Science: Papers in Honor of Richard M. McFall (pp. 197–222). New York: Psychology Press. Novick J Novick K Two systems of self-regulation. In Psychoanalytic approaches to the treatment of children and adolescents Psychoanalytic Social Work 2001 8 95 122 10.1300/J032v08n03_06 Novick J Novick KK Good goodbyes 2006 Jason Aronson Novick KK Novick J Two systems of self-regulation and the differential application of psychoanalytic technique American Journal of Psychoanalysis 2003 63 1 20 10.1023/A:1022323003802 12656197 Novick KK Novick J Emotional muscle: Strong parents, strong children 2010 Bloomington, IN Xlibris Novick KK Novick J Building emotional muscle in children and parents Psychoanalytic Study of the Child 2011 65 131 151 10.1080/00797308.2011.11800835 26027142 Obholzer K Wolf Man: Conversations with Freud’s patient sixty years later 1982 Continuum International Publishing Group Rodgers R Hammerstein O I whistle a happy tune [Musical Score] 1951 New York Williamson Music Stone, The widening scope of indications for psychoanalysis Journal of the American Psychoanalytic Association 1954 2 567 594 10.1177/000306515400200402 13211432	36470990	PMC9734568	NO-CC CODE	2022-12-14 23:28:29	no	Am J Psychoanal. 2022 Dec 5;:1-13	utf-8	Am J Psychoanal	2,022	10.1057/s11231-022-09383-6	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 400 10.1007/s13721-022-00400-3 Original Article Policy analysis and data mining tools for controlling COVID-19 policies http://orcid.org/0000-0002-1826-742X Takefuji Yoshiyasu [email protected] grid.411867.d 0000 0001 0356 8417 Faculty of Data Science, Musashino University, 3-3-3 Ariake Koto-Ku, Tokyo, 135-8181 Japan 5 12 2022 2023 12 1 415 9 2022 20 11 2022 21 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. Much research has been done on the efficacy of vaccines against the COVID-19 pandemic, but the claims have not yet been realized in the real world. This paper proposes three COVID-19 policy outcome analysis tools such as jpscore for scoring and revealing the best prefecture policy in Japan, scorecovid for scoring and revealing the best country policy in the world, and finally hiscovid for visualizing and identifying when policymakers made mistakes in time-series scores. Poorly scored countries or prefectures can learn good strategies from the best country or prefecture with excellent scores. Three tools are based on a single metric dividing the number of COVID-19 deaths by the population in millions. Three tools suggest us that the sustainable mandatory test-isolation strategy should be adopted in the world for mitigating the pandemic. This paper also addresses what is lacking in Japan for scientific evidence-based research for mitigating the pandemic. Visualization tools and sorted and time-series scores of policy outcomes help policymakers make the right decisions. Keywords Scoring COVID-19 policies Evidence-based data COVID-19 policy PyPI tool issue-copyright-statement© Springer-Verlag GmbH Austria, part of Springer Nature 2023 ==== Body pmcIntroduction Data mining on continuous open data or time series datasets plays an important role in discovering scientific facts. Based on the scientific facts uncovered in this study, policymakers need to make the right decision on how to deal with the COVID-19 pandemic, whether to update, strengthen, or mitigate policies. Ristea et al. studied a database that tracked the impact of the COVID-19 pandemic on local communities (Ristea et al. 2022). Policymakers sometimes speak as if science has superhero powers; when it comes to COVID-19, they often speak as if they expect vaccines to return life to the way it used to be but it does not (Shah 2021). Pickersgill et al. addressed the similar research gap (Pickersgill and Smith 2021). Policymakers ignore the past lesson on polio and HIV (Spinney 2022). Long COVID is the latest reminder that epidemics have long tails—biologically, as well as psychologically, economically and socially. Since the persistent effects of COVID-19 were recognized 6 months into the pandemic, up to 200 symptoms have been reported in 10 organ systems, including the skin, brain, heart, and gut (Spinney 2022). Although many papers have emphasized the efficacy of the vaccine against the COVID-19 pandemic, their claim has not yet been realized in real society (Abbasi 2022; Yakusheva et al. 2022; Miller 2021; Sante 2021; Litvak et al. 2021; Ntoumi et al. 2022). This is because vaccine efficacy in the laboratory is different from vaccine effectiveness in the real world. The COVID-19 environment changes with human behavior and new COVID-19 variants. The purpose of this paper is to propose policy outcome analysis tools that score individual policies against COVID-19 and sort the list of scores to help policymakers navigate the pandemic problem. In other words, regardless of the vaccine effectiveness adopted by many countries, the proposed tools will be able to discover the best COVID-19 policy country in the world or prefecture in Japan based on the policy outcomes. Policy outcomes and results can be calculated by the number of COVID-19 deaths. This is because the number of cases is always proportional to the number of COVID-19 deaths. The more COVID-19 deaths the more cases. In other words, the better the policy, the fewer deaths there should be. The contribution of this paper is to proposed policy outcome analysis tools for policymakers to reveal the best policy among countries in the world or prefectures in Japan and they can learn good strategies from excellent scored countries or prefectures. A time-series policy outcome analysis tool allows policymakers to identify and quantify when they made mistakes. Past mistakes cannot be corrected, but mistakes in the future can be mitigated with the proposed time-series policy outcome analysis tool. To our knowledge, there is no such tool. There are two types of approaches against the COVID-19 pandemic such as pharmacological approach with vaccination and non-pharmacological approach such as test-isolation strategy. These two approaches must be integrated to control COVID-19 with the best possible strategy. The most important contribution in this paper lies in that the proposed single metric method can discover the best policy in prefectures in Japan regardless of the existing approaches. The outcomes of policies can be visualized by the proposed tools. The previously proposed scorecovid tool, a Python program can score individual policies by country and reveal the best country policy in the world (Takefuji 2021a). The score is actually calculated by dividing the number of COVID-19 deaths by the population in millions. The scorecovid tool generating sorted scores will reveal that which countries have been handling the pandemic well or not (Takefuji 2021b). In this paper, a newly developed jpscore is introduced for scoring COVID-19 prefecture policies in Japan. The jpscore is a Python package tool so that jpscore runs on Windows, MacOS, and Linux operating systems, respectively, as long as Python is installed on the system. The goal of the jpscore tool is for prefectures with poor scores to learn good strategies from prefectures with excellent scores. The jpscore tool will reveal which prefectures have been handling well against the pandemic. The result of sorted scores from scorecovid and jpscore will be discussed. Two web sites over the Internet show raw data and computed data using a variety of graphs. The significant difference between the proposed method with jpscore and scorecovid and two web sites lies in sorted outcomes of individual policies in countries and prefectures. Policymakers must learn the good strategies from excellent countries or prefectures. It is extremely hard for policymakers with two web sites to discover the best policy in the world and to learn it from countries or prefectures with excellent scores. In other words, providing a variety of data and calculated data on COVID-19 is not convenient for policymakers to make their right decisions. How can we discover which country or prefecture has the best score, the second-best score with using two sites? In other words, information on sorted scores plays a key role in revealing the best policy regardless of pharmacological and non-pharmacological approaches. Such a single metric sorted score method has never been proposed in previous studies. In the proposed scoring tools such as scorecovid and jpscore, it is intended to show sorted scores as policy outcomes. Experts must understand foundations of science. By comparing sorted scores, we will be able to discover new findings such as Niigata has the best score not only in Japan but also that in the world. The single metric for scoring policy outcomes (the number of deaths per million population) in prefectures or countries used in this study has been validated for COVID-19 policy evaluation (Takefuji 2021a, b, c; Alsolami et al. 2022). In the herd immunity debate in Sweden, the author proposed the single metric for investigating the policy effectiveness. The herd immunity failed in Sweden due to the large number of elderly COVID-19 deaths (Takefuji 2021c). In their response (Takefuji 2021c), researchers in Sweden agreed with the proposed issue by the author that infection testing plays a key role and that health policies need to be continuously updated (Takefuji 2021c). The less the number of COVID-19 deaths, the better the policy. The usscore tool was previously proposed for scoring individual state policy in the US (Alsolami et al. 2022). The usscore discovered that Vermont has the best score of 1099, while Arizona has the worst score of 4350 as of September 1, 2022. The hiscovid tool is newly proposed for evaluating time-transition scores to identify when policymakers made mistakes in their policies. This paper will discover when Japan made multiple mistakes, while Taiwan made only two mistakes and New Zealand made a single mistake from the beginning of the COVID-19 pandemic. Remember that mistakes in policies caused unnecessary COVID-19 deaths. The single metric score always monotonically increases in all tools such as scorecovid, usscore, jpscore, and hiscovid. In other words, policymakers must learn past lessons when they made mistakes not to repeat the same mistakes. The proposed scoring tools are extremely useful for policymakers to make right decisions against the COVID-19 pandemic. To enforce the foundations of science on the single metric for scoring policies, intensive literature survey was conducted. Gibney wrote an article on whose coronavirus strategy worked best (Gibney 2020 May). Gibney used the number of infections and that of deaths due to COVID-19. In other words, the number of deaths indicates the final outcome for evaluating coronavirus strategies in the world. Gibney supports the proposed single metric. The Lancet article conducted the intensive study on pandemic preparedness and COVID-19: an exploratory analysis of infection and fatality rates, and contextual factors associated with preparedness in 177 countries (COVID-19 National Preparedness Collaborators 2022). They analyzed infection and fatality rates. Their study also supports the proposed single metric as the final outcome evaluation. The UK government reported the article entitled “Coronavirus: lessons learned to date” (COVID-19 National Preparedness Collaborators 2022). In their report, the strong message was issued on the single metric: one of the key ways to measure a country’s success in fighting COVID-19 is to measure deaths from COVID-19. The UK official document supports the proposed single metric. Chang et al. studied on the determinants of COVID-19 morbidity and mortality across countries (Chang et al. 2022). Their study also supported the proposed single metric as the final outcome for evaluating policies. In existing studies, researchers did not attempt to compare countries' scores on a single indicator to derive the best policy. Fawaz Alsolami et al. investigated to determine which treatments for COVID-19 disease are the most effective and preferable (Alsolami et al. 2022). Fawaz Jaber Alsolami et al. analyzed two perspectives: the early approach and the late approach of COVID-19, and the consequent effects on different aspects of the society (Takefuji 2022). Although openness and continuous open data with daily updated datasets play a key role in analyzing the data and discovering new facts. Open means that the dataset is available to outside experts or scientists and the general public, not just to a limited set of experts and advisors chosen by government or local government. Being open is important so that outside experts can analyze from the dataset and complement the internal experts' inadequate analysis and mistakes. To solve intractable problems, openness can maximize the use of all available resources, including external and internal experts and the general public. Without data analysis, policymakers cannot decide whether the current policy should be updated, strengthened, or mitigated. Policymakers need to review all available decision-making data from internal and external experts to make the right decisions based on priorities. COVID-19 datasets in csv format of the world are available in public and can be downloadable over the Internet. In other words, daily updated data csv file needs to be scraped over the Internet and the data will be used for scoring individual policies. The US is one of the best countries on open datasets where datasets of individual states are updated daily and publicly available. In other words, the United States consists of 50 states and Washington, DC. Detailed daily datasets on COVID-19 for each state are available in public. In Japan, it is hard to find the latest dataset of prefectures on COVID-19. Although many countries have emphasized openness and open data, they did not disclose detailed daily datasets for each state or prefecture in public. Many datasets in the world are limited to a set of internal experts and scientists without being open. Unfortunately, as far as we know, the latest daily dataset on COVID-19 by prefecture is not available on the Japanese government website when the paper was submitted. However, NHK, stands for Nippon Hoso Kyokai (Japan Broadcasting Corporation) which has been providing the useful COVID-19 dataset in csv format in public on each prefecture. NHK is owned by government of Japan, statutory corporation chartered under the Broadcasting Act of 1950. NHK’s dataset was used in the proposed jpscore tool. However, NHK no longer updates that dataset, so it now uses the government's late-start dataset. In summary, the goal of the proposed scoring tools such as scorecovid, usscore, jpscore, and hiscovid with a single policy outcome indicator can be used for policymakers to learn good strategies against COVID-19 from countries with excellent scores for mitigating and ending the pandemic. Data science plays a key role in revealing the best COVID-19 policy with sorted scores. This paper will reveal that Niigata with the best score is the best state in the world while Japan is the best country in the world. However, there is still room for improvement in Japan with regard to COVID-19 policies. This paper will present policy analysis and data mining tools such as scorecovid, jpscore and hiscovid to mitigate the COVID-19 pandemic. The best and sustainable policy is based on the mandatory test-isolation strategy. The test-isolation strategy is to test and identify infected individuals at an early stage and to isolate them from uninfected people during the quarantine period. In other words, the more testing, the stronger the COVID-19 mitigation. The longer the quarantine period, the fewer the COVID-19 spreads. The shorter the quarantine period, the more COVID-19 spreads. This paper presents how to use the proposed visualization tools for controlling the mandatory test-isolation and the quarantine period for mitigating the pandemic. This paper is composed of “Methods” and “Results” Section on tools such as jpscore, scorecovid and hiscovid, “Discussion” Section, “Conclusion”, and Appendix on how to install and run the proposed tools. Methods PyPI packaging allows scorecovid for generating a list of sorted scores by country, usscore, jpscore for generating a list of sorted scores by prefecture in Japan and hiscovid for time-series policy outcome analysis by country to run on Windows, MacOS, and Linux operating systems, respectively, as long as Python is installed on the system. “Appendix-1” details how to install and run the proposed tools. There are two types of single metric policy scoring tools: a snapshot scoring tool and a time-series scoring tool. While snapshot analysis is a single event, time series analysis is a method of analyzing a sequence of snapshots over time. The time series analysis can look at the time transitions and changes while the snapshot analysis cannot. In the framework, scoring is based on the single metric or population mortality rate: the number of daily cumulative COVID-19 deaths divided by the population in millions. In all tools, dataset is automatically scraped over the Internet. The dataset on the number of daily cumulative COVID-19 deaths by prefecture in Japan is used at: https://covid19.mhlw.go.jp/public/opendata/deaths_cumulative_daily.csv. The dataset on that in countries on daily cumulative COVID-19 deaths is used at the following site: https://covid.ourworldindata.org/data/owid-covid-data.csv. To clarify the details of jpscore's scientific calculations, the open-source code of jpscore.py is attached to “Appendix-2”. This source code of jpscore-0.0.8.tar.gz file can be downloaded from the following site: https://files.pythonhosted.org/packages/59/d2/6f5c1f254151c4b1483a27b506d0d5b8078a5cb85f6b9fe496c1ee93c321/jpscore-0.0.8.tar.gz. Because jpscore is a snapshot policy outcome analysis tool, it uses the most recent data as a single event, while hiscovid, a time-series policy outcome analysis tool, uses the entire data to observe the passage of time and calculate time-series trends. In other words, jpscore is a subset of hiscovid. However, while hiscovid cannot calculate scores by prefecture in Japan, it can calculate time-series scores by country. Results Result of jpscore The result of four sorted scores is shown in Table 1. Niigata Prefecture has the best score of 56.7 while Osaka Prefecture has the worst score of 689.3. Niigata’s score is 12 times better than Osaka’s score. In other words, Osaka caused the unnecessary COVID-19 deaths due to the poor policy.Table 1 Sorted scores of four prefectures in Japan Prefecture Deaths Population Score Niigata 126 2.223 56.7 Tottori 59 0.556 106.1 Fukui 87 0.768 113.3 Hokkaido 2440 5.25 464.8 Hyogo 2661 5.466 486.8 Osaka 6072 8.809 689.3 Result of scorecovid Table 2 shows that Japan has the best score of 318 and Hungary has the worst score of 4895 in 15 countries.Table 2 Sorted scores of 15 countries in the world Country Deaths Population Score Japan 40,241 126.48 318.2 New Zealand 1910 4.82 396.3 Taiwan 9950 23.82 417.7 South Korea 26,940 51.27 525.5 Australia 13,999 25.5 549 Iceland 213 0.34 626.5 Canada 44,317 37.74 1174.3 Israel 11,620 8.66 1341.8 Germany 147,642 83.78 1762.3 Sweden 19,904 10.1 1970.7 France 154,203 65.27 2362.5 United Kingdom 205,288 67.89 3023.8 United States 1,047,006 331 3163.2 Brazil 683,965 212.56 3217.8 Hungary 47,291 9.66 4895.5 Result of hiscovid Figure 1 shows that New Zealand had the best score until February 2022. New Zealand made a single mistake and Taiwan made two mistakes. Japan made several mistakes during the pandemic.Fig. 1 Time-series scores of three countries such as Japan, Taiwan, and New Zealand as of September 1, 2022 Discussion Table 1 with jpscore shows that Niigata prefecture has the best score in the world, while Table 2 with scorecovid indicates that Japan is handling the pandemic well until July 2022. In Fig. 1, the vertical axis represents time series scores and the horizontal axis represents dates. Scores are monotonically increasing as they are scored by daily cumulative mortality. In general, the steeper the slope of the graph, the worse the policy outcome. The flat graph shows that the policy has successfully controlled and suppressed the COVID-19 pandemic. As shown in Fig. 1, the hiscovid tool discovered that the mandatory test-isolation policy by law is extremely effective against the COVID-19 pandemic. The test-isolation policy is to test and identify infected individuals at an early stage and to isolate them from uninfected people during the quarantine period. The mandatory test-isolation policy has been adopted only in Taiwan, New Zealand and other several countries. Since the score is calculated by dividing the number of deaths due to COVID-19 by the population in millions in time series, score is always monotonically increasing. In Fig. 1, the result of hiscovid (Japan, Taiwan, New Zealand) shows that until February 2022, New Zealand was the best country in the world handling the COVID-19 pandemic well. However, New Zealand made the single mistake of lifting boarder regulations (Mercer 2022), the score in New Zealand is rapidly increasing. Taiwan made two mistakes: May 2021 and May 2022. The first failure in Taiwan was attributed to the lack of testing of the aircraft crew and the crew's families in May 2021 (Davidson 2021). The second failure in Taiwan was caused by lifting border regulations since May 2022 (Reuters 2022). Most interestingly, the graph is linear and flat until Taiwan and New Zealand successfully implemented a mandatory test-isolation strategy, but two countries failed to mitigate the COVID-19 pandemic after the removal of border restrictions and shortening the quarantine period. Lifting border regulations can significantly affect the outcomes of the COVID-19 pandemic. The quarantine period plays a key role in mitigating the pandemic. The longer the quarantine period, the fewer the COVID-19 spreads. The shorter the quarantine period, the more the COVID-19 spreads. In Japan, the slope of the graph can always be seen and the scores are constantly increasing over time. In other words, a linear or flat graph as seen in Taiwan and New Zealand has never been observed for time series scores in Japan. Japan’s policy on the test-isolation strategy is voluntary and leaky. This is because Japan does not have the mandatory law on the test-isolation strategy, but the voluntary regulations. The mandatory regulations by law significantly play a key role in mitigating the COVID-19 pandemic. In other words, Japan, there is still room for improvement on the COVID-19 policy with the mandatory test-isolation strategy. The recent resurgence in Japan in Fig. 1 is due to relaxing the border regulations. The further investigation is needed since there is no official document on detailed policies in Niigata. Niigata’s policy against the COVID-19 is one of the best policies in the world. Japanese people wear face masks at all times, with or without warning from the government. In other words, the policy mentioned in this paper may be heavily affected by the general public’s behavior in Japan. Thus, policy effects on COVID-19 are a consequence of the policies of policymakers and the actions of the general public. However, we do not know the ratio of two characteristics on the consequence. The Japanese government's official residence provides daily data on the third vaccination rate by prefecture; the data as of October 21, 2022 shows that Niigata Prefecture's vaccination rate is 73.9%, which is in the top group, but not particularly high. However, in terms of immunization coverage by ordinance-designated cities, Niigata City ranks first in the nation at 70.7%. According to the population in the table, about one-third of the population of Niigata Prefecture is concentrated in Niigata City, suggesting that COVID-19 infection may have been suppressed by the was no unique policy in Niigata for non-pharmacological approaches such as test-isolation strategies, then the author may be right that the effect may have been due to herd instinct, such as the Japanese always wearing face masks. The tools can discover the best country in the world or prefecture in Japan. As a result, policymakers can understand that countries with poor scores can learn good strategies from countries and prefectures with good scores. The time-series policy outcome analysis tool allows policymakers to identify when they made mistakes for future policy updates. In the future, the focus should be on the worse off countries to mitigate the pandemic. Watson clearly mentioned that vaccination with boosting is not sustainable (Watson 2022). This is because we do not know how many boostings are needed. Conclusion This paper showed the results on data mining with datasets in public. The daily cumulative mortality scoring plays a key role in discovering five facts. The sorted scores and visualization of results are essential for policymakers to observe policy effectiveness, update policies, and make the right decisions to mitigate the pandemic. The jpscore tool is intended for policymakers to learn good strategies from prefectures with excellent scores. However, we need to investigate why Niigata has the best score in the world from a herd behavior perspective instead of individual policies. New findings are summarized as follows.Niigata has the best score in the world. New Zealand had the best score until February 2022. New Zealand made a single mistake against COVID-19 and Taiwan made two mistakes. hiscovid discovered that lifting border regulations and shortening the quarantine period significantly affect the outcomes of the COVID-19 pandemic. hiscovid also discovered that the mandatory test-isolation strategy by law is sustainable and extremely effective for mitigating the pandemic. Appendix-1: How to install and run jpscore, scorecovid, and hiscovid How to install and run jpscore For Windows, double-click the exe file: Miniconda3-py38_4.11.0-Windows-x86_64.exe. https://repo.anaconda.com/miniconda/Miniconda3-py38_4.11.0-Windows-x86_64.exe. For MacOS, run the following command: https://repo.anaconda.com/miniconda/Miniconda3-py38_4.11.0-MacOSX-x86_64.sh. ($) sign is a prompt given from system in the terminal. $ zsh Miniconda3-py38_4.11.0-MacOSX-x86_64.sh. or. $ bash Miniconda3-py38_4.11.0-MacOSX-x86_64.sh. For WSL on Windows or Linux operating systems, run the following command: https://repo.anaconda.com/miniconda/Miniconda3-py38_4.11.0-Linux-x86_64.sh. $ bash Miniconda3-py38_4.11.0-Linux-x86_64.sh. Before installing jpscore, the following libraries such as pandas and openpyxl are needed: $ pip install pandas. $ pip install openpyxl. Then, install jpscore by the following command: For WSL on Windows, MacOS, and Linux operating systems, $ pip install jpscore. Finally, run jpscore: $ jpscore. How to install and run scorecovid This paper will investigate scores of fifteen countries using scorecovid to compare with scores of prefectures in Japan. Scorecovid is a powerful tool to be able to reveal the effectiveness of policies. To run scorecovid, install scorecovid by the following command in the terminal on Windows, MacOS, or Linux operating systems. Scorecovid displays the sorted scores of countries. The file countries can be modified for more or less countries. For WSL on Windows, MacOS, and Linux operating systems: $ pip install scorecovid. Then, run the following command: $ scorecovid. How to install and run hiscovid The hiscovid tool is to identify when policymakers made mistakes against the COVID-19 pandemic. Scoring policies in hiscovid is based on time-series scores so that the outcomes can show clearly when they made mistakes. To run hiscovid, run the following command for installation. $ pip install hiscovid. And run the following command for scores of three countries such as Japan, Taiwan and New Zealand. $ hiscovid Japan Taiwan ‘New Zealand’. Appendix-2: jpscore.py source code Data availability Not applicable. Declarations Conflict of interest This research has no fund. The author has no conflict of interest. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Abbasi J Studies suggest COVID-19 vaccine boosters save lives JAMA 2022 327 2 115 10.1001/jama.2021.23455 Alsolami FJ Impact assessment of COVID-19 pandemic through machine learning models CMC-Comput Mater Contin 2021 68 3 2895 2912 10.32604/cmc.2021.017469 Alsolami F A unified decision-making technique for analysing treatments in pandemic context CMC-Comput Mater Contin 2022 73 2 2591 2618 10.32604/cmc.2022.025703 Chang D Chang X He Y The determinants of COVID-19 morbidity and mortality across countries Sci Rep 2022 12 5888 10.1038/s41598-022-09783-9 35393471 COVID-19 National Preparedness Collaborators Pandemic preparedness and COVID-19: an exploratory analysis of infection and fatality rates, and contextual factors associated with preparedness in 177 countries, from Jan 1, 2020, to Sept 30, 2021 Lancet (london, England) 2022 399 10334 1489 1512 10.1016/S0140-6736(22)00172-6 35120592 Davidson H (2021) How did Covid slip through Taiwan’s ‘gold standard’ defences? https://www.theguardian.com/world/2021/may/17/how-did-covid-slip-through-taiwans-gold-standard-defences Gibney E Whose coronavirus strategy worked best? Scientists hunt most effective policies Nature 2020 581 7806 15 16 10.1038/d41586-020-01248-1 32341558 https://pypi.org/project/jpscore/ https://ourworldindata.org/covid-deaths https://web.sapmed.ac.jp/canmol/coronavirus/japan_death.html Litvak E Keshavjee S Gewertz BL Fineberg HV How Hospitals Can Save Lives and Themselves Annal Surg 2021 274 1 37 39 10.1097/SLA.0000000000004871 Mercer P (2022) New Zealand Plans to Ease Its Tough COVID-19 Border Controls https://www.voanews.com/a/new-zealand-plans-to-ease-its-tough-covid-19-border-controls/6424766.html Miller MR Language choice about COVID-19 vaccines can save lives J Commun Healthc 2021 14 2 99 101 10.1080/17538068.2021.1892285 Ntoumi F Nachega JB Aklillu E Chakaya J Felker I Amanullah F Yeboah-Manu D Castro KG Zumla A World Tuberculosis Day 2022: aligning COVID-19 and tuberculosis innovations to save lives and to end tuberculosis Lancet Infect Dis 2022 10.1016/S1473-3099(22)00142-6 Pickersgill M Smith M Expertise from the humanities and social sciences is essential for governmental responses to COVID-19 J Glob Health 2021 11 03081 10.7189/jogh.11.03081 34221354 Reuters (2022) Taiwan cuts COVID quarantine for arrivals even as cases rise. https://www.reuters.com/world/asia-pacific/taiwan-cuts-covid-quarantine-arrivals-even-cases-rise-2022-05-03/ Ristea A Tucker R You S A multisource database tracking the impact of the COVID-19 pandemic on the communities of Boston, MA, USA Sci Data 2022 9 330 10.1038/s41597-022-01378-3 35725848 Organised by: DG Sante (2021) Chair persons: Massimo Fagnini (European Commission), 4.O. Workshop: Using data to save lives in times of COVID-19 and beyond. Euro J Public Health 31(Supplement_3), ckab164.306, doi:10.1093/eurpub/ckab164.306 Shah H COVID-19 recovery: science isn’t enough to save us Nature 2021 591 503 10.1038/d41586-021-00731-7 33758407 Spinney L Pandemics disable people — the history lesson that policymakers ignore Nature 2022 602 383 385 10.1038/d41586-022-00414-x 35173335 Takefuji Y SCORECOVID: A Python Package Index for scoring the individual policies against COVID-19 Healthcare Analytics 2021 1 100005 10.1016/j.health.2021.100005 Takefuji Y Analysis of digital fences against COVID-19 Health Technol 2021 11 1383 1386 10.1007/s12553-021-00597-9 Takefuji Y Correspondence: open schools, covid-19, and child and teacher morbidity in Sweden N Engl J Med 2021 384 e66 10.1056/NEJMc2101280NEJM(2021) Takefuji Y Discovering COVID-19 state sustainable policies for mitigating and ending the pandemic Cities (london, England) 2022 130 103865 10.1016/j.cities.2022.103865 35814189 Watson C Three, four or more: what's the magic number for booster shots? Nature 2022 602 7895 17 18 10.1038/d41586-022-00200-9 35091715 Yakusheva O van den Broek-Altenburg E Brekke G Atherly A Lives saved and lost in the first six month of the US COVID-19 pandemic: a retrospective cost-benefit analysis PLoS One 2022 17 1 e0261759 10.1371/journal.pone.0261759 35061722	0	PMC9734569	NO-CC CODE	2022-12-14 23:28:29	no	Netw Model Anal Health Inform Bioinform. 2023 Dec 5; 12(1):4	utf-8	Netw Model Anal Health Inform Bioinform	2,022	10.1007/s13721-022-00400-3	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 400 10.1007/s13721-022-00400-3 Original Article Policy analysis and data mining tools for controlling COVID-19 policies http://orcid.org/0000-0002-1826-742X Takefuji Yoshiyasu [email protected] grid.411867.d 0000 0001 0356 8417 Faculty of Data Science, Musashino University, 3-3-3 Ariake Koto-Ku, Tokyo, 135-8181 Japan 5 12 2022 2023 12 1 415 9 2022 20 11 2022 21 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. Much research has been done on the efficacy of vaccines against the COVID-19 pandemic, but the claims have not yet been realized in the real world. This paper proposes three COVID-19 policy outcome analysis tools such as jpscore for scoring and revealing the best prefecture policy in Japan, scorecovid for scoring and revealing the best country policy in the world, and finally hiscovid for visualizing and identifying when policymakers made mistakes in time-series scores. Poorly scored countries or prefectures can learn good strategies from the best country or prefecture with excellent scores. Three tools are based on a single metric dividing the number of COVID-19 deaths by the population in millions. Three tools suggest us that the sustainable mandatory test-isolation strategy should be adopted in the world for mitigating the pandemic. This paper also addresses what is lacking in Japan for scientific evidence-based research for mitigating the pandemic. Visualization tools and sorted and time-series scores of policy outcomes help policymakers make the right decisions. Keywords Scoring COVID-19 policies Evidence-based data COVID-19 policy PyPI tool issue-copyright-statement© Springer-Verlag GmbH Austria, part of Springer Nature 2023 ==== Body pmcIntroduction Data mining on continuous open data or time series datasets plays an important role in discovering scientific facts. Based on the scientific facts uncovered in this study, policymakers need to make the right decision on how to deal with the COVID-19 pandemic, whether to update, strengthen, or mitigate policies. Ristea et al. studied a database that tracked the impact of the COVID-19 pandemic on local communities (Ristea et al. 2022). Policymakers sometimes speak as if science has superhero powers; when it comes to COVID-19, they often speak as if they expect vaccines to return life to the way it used to be but it does not (Shah 2021). Pickersgill et al. addressed the similar research gap (Pickersgill and Smith 2021). Policymakers ignore the past lesson on polio and HIV (Spinney 2022). Long COVID is the latest reminder that epidemics have long tails—biologically, as well as psychologically, economically and socially. Since the persistent effects of COVID-19 were recognized 6 months into the pandemic, up to 200 symptoms have been reported in 10 organ systems, including the skin, brain, heart, and gut (Spinney 2022). Although many papers have emphasized the efficacy of the vaccine against the COVID-19 pandemic, their claim has not yet been realized in real society (Abbasi 2022; Yakusheva et al. 2022; Miller 2021; Sante 2021; Litvak et al. 2021; Ntoumi et al. 2022). This is because vaccine efficacy in the laboratory is different from vaccine effectiveness in the real world. The COVID-19 environment changes with human behavior and new COVID-19 variants. The purpose of this paper is to propose policy outcome analysis tools that score individual policies against COVID-19 and sort the list of scores to help policymakers navigate the pandemic problem. In other words, regardless of the vaccine effectiveness adopted by many countries, the proposed tools will be able to discover the best COVID-19 policy country in the world or prefecture in Japan based on the policy outcomes. Policy outcomes and results can be calculated by the number of COVID-19 deaths. This is because the number of cases is always proportional to the number of COVID-19 deaths. The more COVID-19 deaths the more cases. In other words, the better the policy, the fewer deaths there should be. The contribution of this paper is to proposed policy outcome analysis tools for policymakers to reveal the best policy among countries in the world or prefectures in Japan and they can learn good strategies from excellent scored countries or prefectures. A time-series policy outcome analysis tool allows policymakers to identify and quantify when they made mistakes. Past mistakes cannot be corrected, but mistakes in the future can be mitigated with the proposed time-series policy outcome analysis tool. To our knowledge, there is no such tool. There are two types of approaches against the COVID-19 pandemic such as pharmacological approach with vaccination and non-pharmacological approach such as test-isolation strategy. These two approaches must be integrated to control COVID-19 with the best possible strategy. The most important contribution in this paper lies in that the proposed single metric method can discover the best policy in prefectures in Japan regardless of the existing approaches. The outcomes of policies can be visualized by the proposed tools. The previously proposed scorecovid tool, a Python program can score individual policies by country and reveal the best country policy in the world (Takefuji 2021a). The score is actually calculated by dividing the number of COVID-19 deaths by the population in millions. The scorecovid tool generating sorted scores will reveal that which countries have been handling the pandemic well or not (Takefuji 2021b). In this paper, a newly developed jpscore is introduced for scoring COVID-19 prefecture policies in Japan. The jpscore is a Python package tool so that jpscore runs on Windows, MacOS, and Linux operating systems, respectively, as long as Python is installed on the system. The goal of the jpscore tool is for prefectures with poor scores to learn good strategies from prefectures with excellent scores. The jpscore tool will reveal which prefectures have been handling well against the pandemic. The result of sorted scores from scorecovid and jpscore will be discussed. Two web sites over the Internet show raw data and computed data using a variety of graphs. The significant difference between the proposed method with jpscore and scorecovid and two web sites lies in sorted outcomes of individual policies in countries and prefectures. Policymakers must learn the good strategies from excellent countries or prefectures. It is extremely hard for policymakers with two web sites to discover the best policy in the world and to learn it from countries or prefectures with excellent scores. In other words, providing a variety of data and calculated data on COVID-19 is not convenient for policymakers to make their right decisions. How can we discover which country or prefecture has the best score, the second-best score with using two sites? In other words, information on sorted scores plays a key role in revealing the best policy regardless of pharmacological and non-pharmacological approaches. Such a single metric sorted score method has never been proposed in previous studies. In the proposed scoring tools such as scorecovid and jpscore, it is intended to show sorted scores as policy outcomes. Experts must understand foundations of science. By comparing sorted scores, we will be able to discover new findings such as Niigata has the best score not only in Japan but also that in the world. The single metric for scoring policy outcomes (the number of deaths per million population) in prefectures or countries used in this study has been validated for COVID-19 policy evaluation (Takefuji 2021a, b, c; Alsolami et al. 2022). In the herd immunity debate in Sweden, the author proposed the single metric for investigating the policy effectiveness. The herd immunity failed in Sweden due to the large number of elderly COVID-19 deaths (Takefuji 2021c). In their response (Takefuji 2021c), researchers in Sweden agreed with the proposed issue by the author that infection testing plays a key role and that health policies need to be continuously updated (Takefuji 2021c). The less the number of COVID-19 deaths, the better the policy. The usscore tool was previously proposed for scoring individual state policy in the US (Alsolami et al. 2022). The usscore discovered that Vermont has the best score of 1099, while Arizona has the worst score of 4350 as of September 1, 2022. The hiscovid tool is newly proposed for evaluating time-transition scores to identify when policymakers made mistakes in their policies. This paper will discover when Japan made multiple mistakes, while Taiwan made only two mistakes and New Zealand made a single mistake from the beginning of the COVID-19 pandemic. Remember that mistakes in policies caused unnecessary COVID-19 deaths. The single metric score always monotonically increases in all tools such as scorecovid, usscore, jpscore, and hiscovid. In other words, policymakers must learn past lessons when they made mistakes not to repeat the same mistakes. The proposed scoring tools are extremely useful for policymakers to make right decisions against the COVID-19 pandemic. To enforce the foundations of science on the single metric for scoring policies, intensive literature survey was conducted. Gibney wrote an article on whose coronavirus strategy worked best (Gibney 2020 May). Gibney used the number of infections and that of deaths due to COVID-19. In other words, the number of deaths indicates the final outcome for evaluating coronavirus strategies in the world. Gibney supports the proposed single metric. The Lancet article conducted the intensive study on pandemic preparedness and COVID-19: an exploratory analysis of infection and fatality rates, and contextual factors associated with preparedness in 177 countries (COVID-19 National Preparedness Collaborators 2022). They analyzed infection and fatality rates. Their study also supports the proposed single metric as the final outcome evaluation. The UK government reported the article entitled “Coronavirus: lessons learned to date” (COVID-19 National Preparedness Collaborators 2022). In their report, the strong message was issued on the single metric: one of the key ways to measure a country’s success in fighting COVID-19 is to measure deaths from COVID-19. The UK official document supports the proposed single metric. Chang et al. studied on the determinants of COVID-19 morbidity and mortality across countries (Chang et al. 2022). Their study also supported the proposed single metric as the final outcome for evaluating policies. In existing studies, researchers did not attempt to compare countries' scores on a single indicator to derive the best policy. Fawaz Alsolami et al. investigated to determine which treatments for COVID-19 disease are the most effective and preferable (Alsolami et al. 2022). Fawaz Jaber Alsolami et al. analyzed two perspectives: the early approach and the late approach of COVID-19, and the consequent effects on different aspects of the society (Takefuji 2022). Although openness and continuous open data with daily updated datasets play a key role in analyzing the data and discovering new facts. Open means that the dataset is available to outside experts or scientists and the general public, not just to a limited set of experts and advisors chosen by government or local government. Being open is important so that outside experts can analyze from the dataset and complement the internal experts' inadequate analysis and mistakes. To solve intractable problems, openness can maximize the use of all available resources, including external and internal experts and the general public. Without data analysis, policymakers cannot decide whether the current policy should be updated, strengthened, or mitigated. Policymakers need to review all available decision-making data from internal and external experts to make the right decisions based on priorities. COVID-19 datasets in csv format of the world are available in public and can be downloadable over the Internet. In other words, daily updated data csv file needs to be scraped over the Internet and the data will be used for scoring individual policies. The US is one of the best countries on open datasets where datasets of individual states are updated daily and publicly available. In other words, the United States consists of 50 states and Washington, DC. Detailed daily datasets on COVID-19 for each state are available in public. In Japan, it is hard to find the latest dataset of prefectures on COVID-19. Although many countries have emphasized openness and open data, they did not disclose detailed daily datasets for each state or prefecture in public. Many datasets in the world are limited to a set of internal experts and scientists without being open. Unfortunately, as far as we know, the latest daily dataset on COVID-19 by prefecture is not available on the Japanese government website when the paper was submitted. However, NHK, stands for Nippon Hoso Kyokai (Japan Broadcasting Corporation) which has been providing the useful COVID-19 dataset in csv format in public on each prefecture. NHK is owned by government of Japan, statutory corporation chartered under the Broadcasting Act of 1950. NHK’s dataset was used in the proposed jpscore tool. However, NHK no longer updates that dataset, so it now uses the government's late-start dataset. In summary, the goal of the proposed scoring tools such as scorecovid, usscore, jpscore, and hiscovid with a single policy outcome indicator can be used for policymakers to learn good strategies against COVID-19 from countries with excellent scores for mitigating and ending the pandemic. Data science plays a key role in revealing the best COVID-19 policy with sorted scores. This paper will reveal that Niigata with the best score is the best state in the world while Japan is the best country in the world. However, there is still room for improvement in Japan with regard to COVID-19 policies. This paper will present policy analysis and data mining tools such as scorecovid, jpscore and hiscovid to mitigate the COVID-19 pandemic. The best and sustainable policy is based on the mandatory test-isolation strategy. The test-isolation strategy is to test and identify infected individuals at an early stage and to isolate them from uninfected people during the quarantine period. In other words, the more testing, the stronger the COVID-19 mitigation. The longer the quarantine period, the fewer the COVID-19 spreads. The shorter the quarantine period, the more COVID-19 spreads. This paper presents how to use the proposed visualization tools for controlling the mandatory test-isolation and the quarantine period for mitigating the pandemic. This paper is composed of “Methods” and “Results” Section on tools such as jpscore, scorecovid and hiscovid, “Discussion” Section, “Conclusion”, and Appendix on how to install and run the proposed tools. Methods PyPI packaging allows scorecovid for generating a list of sorted scores by country, usscore, jpscore for generating a list of sorted scores by prefecture in Japan and hiscovid for time-series policy outcome analysis by country to run on Windows, MacOS, and Linux operating systems, respectively, as long as Python is installed on the system. “Appendix-1” details how to install and run the proposed tools. There are two types of single metric policy scoring tools: a snapshot scoring tool and a time-series scoring tool. While snapshot analysis is a single event, time series analysis is a method of analyzing a sequence of snapshots over time. The time series analysis can look at the time transitions and changes while the snapshot analysis cannot. In the framework, scoring is based on the single metric or population mortality rate: the number of daily cumulative COVID-19 deaths divided by the population in millions. In all tools, dataset is automatically scraped over the Internet. The dataset on the number of daily cumulative COVID-19 deaths by prefecture in Japan is used at: https://covid19.mhlw.go.jp/public/opendata/deaths_cumulative_daily.csv. The dataset on that in countries on daily cumulative COVID-19 deaths is used at the following site: https://covid.ourworldindata.org/data/owid-covid-data.csv. To clarify the details of jpscore's scientific calculations, the open-source code of jpscore.py is attached to “Appendix-2”. This source code of jpscore-0.0.8.tar.gz file can be downloaded from the following site: https://files.pythonhosted.org/packages/59/d2/6f5c1f254151c4b1483a27b506d0d5b8078a5cb85f6b9fe496c1ee93c321/jpscore-0.0.8.tar.gz. Because jpscore is a snapshot policy outcome analysis tool, it uses the most recent data as a single event, while hiscovid, a time-series policy outcome analysis tool, uses the entire data to observe the passage of time and calculate time-series trends. In other words, jpscore is a subset of hiscovid. However, while hiscovid cannot calculate scores by prefecture in Japan, it can calculate time-series scores by country. Results Result of jpscore The result of four sorted scores is shown in Table 1. Niigata Prefecture has the best score of 56.7 while Osaka Prefecture has the worst score of 689.3. Niigata’s score is 12 times better than Osaka’s score. In other words, Osaka caused the unnecessary COVID-19 deaths due to the poor policy.Table 1 Sorted scores of four prefectures in Japan Prefecture Deaths Population Score Niigata 126 2.223 56.7 Tottori 59 0.556 106.1 Fukui 87 0.768 113.3 Hokkaido 2440 5.25 464.8 Hyogo 2661 5.466 486.8 Osaka 6072 8.809 689.3 Result of scorecovid Table 2 shows that Japan has the best score of 318 and Hungary has the worst score of 4895 in 15 countries.Table 2 Sorted scores of 15 countries in the world Country Deaths Population Score Japan 40,241 126.48 318.2 New Zealand 1910 4.82 396.3 Taiwan 9950 23.82 417.7 South Korea 26,940 51.27 525.5 Australia 13,999 25.5 549 Iceland 213 0.34 626.5 Canada 44,317 37.74 1174.3 Israel 11,620 8.66 1341.8 Germany 147,642 83.78 1762.3 Sweden 19,904 10.1 1970.7 France 154,203 65.27 2362.5 United Kingdom 205,288 67.89 3023.8 United States 1,047,006 331 3163.2 Brazil 683,965 212.56 3217.8 Hungary 47,291 9.66 4895.5 Result of hiscovid Figure 1 shows that New Zealand had the best score until February 2022. New Zealand made a single mistake and Taiwan made two mistakes. Japan made several mistakes during the pandemic.Fig. 1 Time-series scores of three countries such as Japan, Taiwan, and New Zealand as of September 1, 2022 Discussion Table 1 with jpscore shows that Niigata prefecture has the best score in the world, while Table 2 with scorecovid indicates that Japan is handling the pandemic well until July 2022. In Fig. 1, the vertical axis represents time series scores and the horizontal axis represents dates. Scores are monotonically increasing as they are scored by daily cumulative mortality. In general, the steeper the slope of the graph, the worse the policy outcome. The flat graph shows that the policy has successfully controlled and suppressed the COVID-19 pandemic. As shown in Fig. 1, the hiscovid tool discovered that the mandatory test-isolation policy by law is extremely effective against the COVID-19 pandemic. The test-isolation policy is to test and identify infected individuals at an early stage and to isolate them from uninfected people during the quarantine period. The mandatory test-isolation policy has been adopted only in Taiwan, New Zealand and other several countries. Since the score is calculated by dividing the number of deaths due to COVID-19 by the population in millions in time series, score is always monotonically increasing. In Fig. 1, the result of hiscovid (Japan, Taiwan, New Zealand) shows that until February 2022, New Zealand was the best country in the world handling the COVID-19 pandemic well. However, New Zealand made the single mistake of lifting boarder regulations (Mercer 2022), the score in New Zealand is rapidly increasing. Taiwan made two mistakes: May 2021 and May 2022. The first failure in Taiwan was attributed to the lack of testing of the aircraft crew and the crew's families in May 2021 (Davidson 2021). The second failure in Taiwan was caused by lifting border regulations since May 2022 (Reuters 2022). Most interestingly, the graph is linear and flat until Taiwan and New Zealand successfully implemented a mandatory test-isolation strategy, but two countries failed to mitigate the COVID-19 pandemic after the removal of border restrictions and shortening the quarantine period. Lifting border regulations can significantly affect the outcomes of the COVID-19 pandemic. The quarantine period plays a key role in mitigating the pandemic. The longer the quarantine period, the fewer the COVID-19 spreads. The shorter the quarantine period, the more the COVID-19 spreads. In Japan, the slope of the graph can always be seen and the scores are constantly increasing over time. In other words, a linear or flat graph as seen in Taiwan and New Zealand has never been observed for time series scores in Japan. Japan’s policy on the test-isolation strategy is voluntary and leaky. This is because Japan does not have the mandatory law on the test-isolation strategy, but the voluntary regulations. The mandatory regulations by law significantly play a key role in mitigating the COVID-19 pandemic. In other words, Japan, there is still room for improvement on the COVID-19 policy with the mandatory test-isolation strategy. The recent resurgence in Japan in Fig. 1 is due to relaxing the border regulations. The further investigation is needed since there is no official document on detailed policies in Niigata. Niigata’s policy against the COVID-19 is one of the best policies in the world. Japanese people wear face masks at all times, with or without warning from the government. In other words, the policy mentioned in this paper may be heavily affected by the general public’s behavior in Japan. Thus, policy effects on COVID-19 are a consequence of the policies of policymakers and the actions of the general public. However, we do not know the ratio of two characteristics on the consequence. The Japanese government's official residence provides daily data on the third vaccination rate by prefecture; the data as of October 21, 2022 shows that Niigata Prefecture's vaccination rate is 73.9%, which is in the top group, but not particularly high. However, in terms of immunization coverage by ordinance-designated cities, Niigata City ranks first in the nation at 70.7%. According to the population in the table, about one-third of the population of Niigata Prefecture is concentrated in Niigata City, suggesting that COVID-19 infection may have been suppressed by the was no unique policy in Niigata for non-pharmacological approaches such as test-isolation strategies, then the author may be right that the effect may have been due to herd instinct, such as the Japanese always wearing face masks. The tools can discover the best country in the world or prefecture in Japan. As a result, policymakers can understand that countries with poor scores can learn good strategies from countries and prefectures with good scores. The time-series policy outcome analysis tool allows policymakers to identify when they made mistakes for future policy updates. In the future, the focus should be on the worse off countries to mitigate the pandemic. Watson clearly mentioned that vaccination with boosting is not sustainable (Watson 2022). This is because we do not know how many boostings are needed. Conclusion This paper showed the results on data mining with datasets in public. The daily cumulative mortality scoring plays a key role in discovering five facts. The sorted scores and visualization of results are essential for policymakers to observe policy effectiveness, update policies, and make the right decisions to mitigate the pandemic. The jpscore tool is intended for policymakers to learn good strategies from prefectures with excellent scores. However, we need to investigate why Niigata has the best score in the world from a herd behavior perspective instead of individual policies. New findings are summarized as follows.Niigata has the best score in the world. New Zealand had the best score until February 2022. New Zealand made a single mistake against COVID-19 and Taiwan made two mistakes. hiscovid discovered that lifting border regulations and shortening the quarantine period significantly affect the outcomes of the COVID-19 pandemic. hiscovid also discovered that the mandatory test-isolation strategy by law is sustainable and extremely effective for mitigating the pandemic. Appendix-1: How to install and run jpscore, scorecovid, and hiscovid How to install and run jpscore For Windows, double-click the exe file: Miniconda3-py38_4.11.0-Windows-x86_64.exe. https://repo.anaconda.com/miniconda/Miniconda3-py38_4.11.0-Windows-x86_64.exe. For MacOS, run the following command: https://repo.anaconda.com/miniconda/Miniconda3-py38_4.11.0-MacOSX-x86_64.sh. ($) sign is a prompt given from system in the terminal. $ zsh Miniconda3-py38_4.11.0-MacOSX-x86_64.sh. or. $ bash Miniconda3-py38_4.11.0-MacOSX-x86_64.sh. For WSL on Windows or Linux operating systems, run the following command: https://repo.anaconda.com/miniconda/Miniconda3-py38_4.11.0-Linux-x86_64.sh. $ bash Miniconda3-py38_4.11.0-Linux-x86_64.sh. Before installing jpscore, the following libraries such as pandas and openpyxl are needed: $ pip install pandas. $ pip install openpyxl. Then, install jpscore by the following command: For WSL on Windows, MacOS, and Linux operating systems, $ pip install jpscore. Finally, run jpscore: $ jpscore. How to install and run scorecovid This paper will investigate scores of fifteen countries using scorecovid to compare with scores of prefectures in Japan. Scorecovid is a powerful tool to be able to reveal the effectiveness of policies. To run scorecovid, install scorecovid by the following command in the terminal on Windows, MacOS, or Linux operating systems. Scorecovid displays the sorted scores of countries. The file countries can be modified for more or less countries. For WSL on Windows, MacOS, and Linux operating systems: $ pip install scorecovid. Then, run the following command: $ scorecovid. How to install and run hiscovid The hiscovid tool is to identify when policymakers made mistakes against the COVID-19 pandemic. Scoring policies in hiscovid is based on time-series scores so that the outcomes can show clearly when they made mistakes. To run hiscovid, run the following command for installation. $ pip install hiscovid. And run the following command for scores of three countries such as Japan, Taiwan and New Zealand. $ hiscovid Japan Taiwan ‘New Zealand’. Appendix-2: jpscore.py source code Data availability Not applicable. Declarations Conflict of interest This research has no fund. The author has no conflict of interest. Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. ==== Refs References Abbasi J Studies suggest COVID-19 vaccine boosters save lives JAMA 2022 327 2 115 10.1001/jama.2021.23455 Alsolami FJ Impact assessment of COVID-19 pandemic through machine learning models CMC-Comput Mater Contin 2021 68 3 2895 2912 10.32604/cmc.2021.017469 Alsolami F A unified decision-making technique for analysing treatments in pandemic context CMC-Comput Mater Contin 2022 73 2 2591 2618 10.32604/cmc.2022.025703 Chang D Chang X He Y The determinants of COVID-19 morbidity and mortality across countries Sci Rep 2022 12 5888 10.1038/s41598-022-09783-9 35393471 COVID-19 National Preparedness Collaborators Pandemic preparedness and COVID-19: an exploratory analysis of infection and fatality rates, and contextual factors associated with preparedness in 177 countries, from Jan 1, 2020, to Sept 30, 2021 Lancet (london, England) 2022 399 10334 1489 1512 10.1016/S0140-6736(22)00172-6 35120592 Davidson H (2021) How did Covid slip through Taiwan’s ‘gold standard’ defences? https://www.theguardian.com/world/2021/may/17/how-did-covid-slip-through-taiwans-gold-standard-defences Gibney E Whose coronavirus strategy worked best? Scientists hunt most effective policies Nature 2020 581 7806 15 16 10.1038/d41586-020-01248-1 32341558 https://pypi.org/project/jpscore/ https://ourworldindata.org/covid-deaths https://web.sapmed.ac.jp/canmol/coronavirus/japan_death.html Litvak E Keshavjee S Gewertz BL Fineberg HV How Hospitals Can Save Lives and Themselves Annal Surg 2021 274 1 37 39 10.1097/SLA.0000000000004871 Mercer P (2022) New Zealand Plans to Ease Its Tough COVID-19 Border Controls https://www.voanews.com/a/new-zealand-plans-to-ease-its-tough-covid-19-border-controls/6424766.html Miller MR Language choice about COVID-19 vaccines can save lives J Commun Healthc 2021 14 2 99 101 10.1080/17538068.2021.1892285 Ntoumi F Nachega JB Aklillu E Chakaya J Felker I Amanullah F Yeboah-Manu D Castro KG Zumla A World Tuberculosis Day 2022: aligning COVID-19 and tuberculosis innovations to save lives and to end tuberculosis Lancet Infect Dis 2022 10.1016/S1473-3099(22)00142-6 Pickersgill M Smith M Expertise from the humanities and social sciences is essential for governmental responses to COVID-19 J Glob Health 2021 11 03081 10.7189/jogh.11.03081 34221354 Reuters (2022) Taiwan cuts COVID quarantine for arrivals even as cases rise. https://www.reuters.com/world/asia-pacific/taiwan-cuts-covid-quarantine-arrivals-even-cases-rise-2022-05-03/ Ristea A Tucker R You S A multisource database tracking the impact of the COVID-19 pandemic on the communities of Boston, MA, USA Sci Data 2022 9 330 10.1038/s41597-022-01378-3 35725848 Organised by: DG Sante (2021) Chair persons: Massimo Fagnini (European Commission), 4.O. Workshop: Using data to save lives in times of COVID-19 and beyond. Euro J Public Health 31(Supplement_3), ckab164.306, doi:10.1093/eurpub/ckab164.306 Shah H COVID-19 recovery: science isn’t enough to save us Nature 2021 591 503 10.1038/d41586-021-00731-7 33758407 Spinney L Pandemics disable people — the history lesson that policymakers ignore Nature 2022 602 383 385 10.1038/d41586-022-00414-x 35173335 Takefuji Y SCORECOVID: A Python Package Index for scoring the individual policies against COVID-19 Healthcare Analytics 2021 1 100005 10.1016/j.health.2021.100005 Takefuji Y Analysis of digital fences against COVID-19 Health Technol 2021 11 1383 1386 10.1007/s12553-021-00597-9 Takefuji Y Correspondence: open schools, covid-19, and child and teacher morbidity in Sweden N Engl J Med 2021 384 e66 10.1056/NEJMc2101280NEJM(2021) Takefuji Y Discovering COVID-19 state sustainable policies for mitigating and ending the pandemic Cities (london, England) 2022 130 103865 10.1016/j.cities.2022.103865 35814189 Watson C Three, four or more: what's the magic number for booster shots? Nature 2022 602 7895 17 18 10.1038/d41586-022-00200-9 35091715 Yakusheva O van den Broek-Altenburg E Brekke G Atherly A Lives saved and lost in the first six month of the US COVID-19 pandemic: a retrospective cost-benefit analysis PLoS One 2022 17 1 e0261759 10.1371/journal.pone.0261759 35061722	0	PMC9734571	NO-CC CODE	2022-12-14 23:28:29	no	psychopraxis. neuropraxis. 2022 Dec 8; 25(6):299-301	latin-1	null	null	null	oa_other

Subsets and Splits