title
stringlengths 1
1.19k
| keywords
stringlengths 0
668
| concept
stringlengths 0
909
| paragraph
stringlengths 0
61.8k
| PMID
stringlengths 10
11
|
|---|---|---|---|---|
Population Stratification Into Demographic Environments | California’s population was stratified into demographic environments based on socioeconomic information. A total of 24 features, including age, gender, spoken language, nationality, access to the health system, income, education, as well as race and ethnicity, were gathered for the state of California from the American Community Survey [ | PMC10267788 |
||
Curating Digital Content About Organ Donation | The digital contents were short videos curated from YouTube by a clinical psychologist with experience in health education. The curation consisted of 4 steps: search, selection based on inclusion and exclusion criteria, classification, and tailoring. The main search strategy used the keywords organ donation, organ donors, One Legacy, and Donate Life in combination with the words United States, California, Asian, Latin American, and African American. The inclusion criteria for content to be included were: (1) English as the primary language or in the subtitles; (2) at least 1 out of the 3 perspectives (ie, story life, commercial, and educational); and (3) focus on living or deceased recipients or their families. The exclusion criteria were: languages other than English; and those that promote the idea of organ donation in a monetary exchange. A final set of 50 videos was used for content analysis.The analysis of organ donation internet videos and their contents was based on parameters of interest. The content was analyzed and classified into 6 domains: focus (on the donor, the recipient, or the family); the type of donor (living or deceased), recipient (relative, stranger, or exchange), and family (of the donor, of the recipient, or both); age and ethnicity of each subject; and type of content (story life, commercial, educational, or a mix of them). For each video included in the systematic review, we also extracted the year, source, search link, length, and number of views. A single digital content was selected for each demographic environment according to the cluster’s characteristics, with the addition of a reference content selected for having a high number of views on YouTube. | PMC10267788 |
||
Digital Content Delivery Using Facebook | In a CRT research design, ZCTAs from each uncovered demographic environment were randomly assigned to 2 intervention arms. The first arm, nonoptimal SNI, received the SNI with equal exposure to digital contents, and the second arm, optimal SNI, received the SNI with tailored exposure to digital contents (The framework targeted the optimal and nonoptimal intervention arms with digital content through Facebook. The SNI had a duration of 89 days, divided into 2 phases: preoptimization and postoptimization. In the preoptimization phase, 45 days, all clusters from the 2 arms, optimal and nonoptimal, were targeted with all educational contents with the same exposure (ie, providing the same budget for each one) and the engagement levels of each educational content at each cluster were gathered ( | PMC10267788 |
||
ACT Using Proportional-Integral-Derivative | PID | The ACT mechanism was used to enable optimal SNI (The ACT mechanism was modeled as a PID controller, which is widely applied in control systems engineering. This mechanism was implemented in the SNI software to enable the daily budget adjustments of each content in the Meta Business Ads Manager platform. On a daily basis, during the postoptimization phase, the software gathers the level of exposure and respective engagement for each content at each cluster and uses them as input to the PID controller, which evaluates the efficiency error and updates the budget of each content for each cluster accordingly (As shown above, the PID controller is a simple and effective control mechanism. Although strongly related to the industrial scenario, several apps use the PID on control tasks, for example, from arterial blood pressure regulation to electrical power generation [The PID has 3 parameters, which are related to the 3 control settings discussed in the last paragraph. Those parameters, also known as gains, are: Kp (proportional gain), Ki (integral gain), and Kd (derivative gain). They represent the weight of each control setting on the active control. In this work, based on preliminary analysis, we used the parametric configuration of: Kp=1.0, Ki=0.5, and Kd=0.1. Given the novel approach developed, we are considering a simple PID tuning parametrization [The adaptive content tuning (ACT) mechanism defined as a Proportional-Integral-Derivative (PID) controller. The component daily adjusts the contents budget based on the difference between the proportion of engagement and exposure from the contents for each cluster (eg, cluster 4). The PID controller error function e(t) was designed to ensure that the contents with more engagement would receive proportionally more budget. PID: proportional-integral-derivative. | PMC10267788 |
|
Statistical Analysis | REGRESSION | The average daily difference between the number of clicks between phases (C) was estimated using 1000 bootstrap samples with a replacement for each group. Ordinary least squares (OLS) regression was used to model the C/I as a function of both the intervention arm (β | PMC10267788 |
|
Ethics Approval | This study was approved by the institutional review board of the University of California, Davis, US (1596733-2). The study was registered on ClinicalTrials.gov (NTC04850287). | PMC10267788 |
||
Results | PMC10267788 |
|||
PID Controller Optimizes Social Network Interventions | PID | In the postoptimization phase, the PID control component daily adjusted the budget for the intervention units in the optimal arm (Taking cluster 4 as an example, it is possible to understand the behavior of the PID component given the levels of exposure (impressions) and engagement (link clicks). At the end of the preoptimization phase, in cluster 4, each digital content had a proportion of the total number of impressions (Comparing the average daily difference between the number of clicks between phases (C) for each intervention arm, it is possible to visualize how the optimization worked in each cluster. While C remains around 0 for each content of each cluster in the nonoptimal arm, a different pattern is observed in the optimal arm (Educational contents of daily budget updates, based on the PID controller, for each demographic environment in the optimal arm during the postoptimization phase. Clusters 1 and 2 had the majority of the budget redirected to deliver the educational content, EdC-Baseline. Cluster 3 had the budget adjusted to deliver the contents of EdC-1 and EdC-Baseline. Cluster 4 had the majority of the budget adjusted to deliver the contents of EdC-1 and EdC-3. EdC: educational content.PID component analysis for cluster 4 in the optimal arm over the postoptimization phase. Engagement proportion levels for each educational content (Top, Left). Exposure proportion levels for each educational content (Top, Right). The PID adjustments for each content based on the differences between exposure and engagement levels (Bottom, Left). The contents of the daily budget after the PID proportional adjustments (Bottom, Right). PID: proportional-integral-derivative.Daily content engagement’s mean difference between the pre- and postoptimization phases, for each cluster, for the nonoptimal (left) and optimal (right) arms. The engagement differences in the optimal arm (right) present which contents were delivered for each cluster in the postoptimization phase given the PID controller efficiency optimization in contrast with the nonoptimal arm (left), where the budget remained the same for each content. | PMC10267788 |
|
Discussion | PMC10267788 |
|||
Principal Results | PID | PID | In this work, we proposed an SNI mechanism that uses high-resolution assessments and controllability in adaptive interventions to increase the engagement in organ donation campaigns by tailoring educational content to different population groups. The Meta Business Ads Manager platform was used to deliver the SNI with 3 parameters: the ZCTA randomized for each controlled trial (optimal and nonoptimal), the daily budget for each intervention unit, and the platform optimization goal to increase impressions. The SNI targeted demographic environments in 2 phases: pre- and postoptimization. In the preoptimization phase, the SNI delivered all intervention units with an equal proportion of the daily budget. In the postoptimization phase, the efficiency control mechanism (PID controller) was applied to continually tune the optimization arm of the intervention.The optimization mechanism developed in this study enabled efficient budget allocation in the optimal intervention arm, resulting in a greater level of engagement per exposure. This has important public health implications as it allows highly specific targeting of educational health content to diverse populations to allow an equitable spread of information to populations in need. | PMC10267788 |
PID Controller Enables Tailored Educational Content per Demographic Environment and Increases Content Efficiency | diabetes | PID, DIABETES | PID controllers are loop-based control mechanisms used to maintain process variables close to desired set points. The PID controller acts as a physician analyzing the glucose level of a patient with diabetes; if the glucose level goes up, the insulin dosage should increase proportionally, and the opposite would happen if the glucose level decreases. Those mechanisms continuously calculate the differences between the current values of process variables and their respective set points to correct the system parameters concerning proportional, integral, and derivative terms. This methodology allows for the first time to plan, quantify, and optimize in real time public health education campaigns, making them more equitable, efficient, and cost-effective.Concerning social network interventions, the PID controller can increase people’s engagement in public health campaigns by tailoring educational content to different population groups. Thus, we need to define an appropriate performance measure to achieve the intervention goals, for example, the number of clicks per impression. In our study, we used organ donation registrations as the educational campaign, but the same can be applied to any other health awareness and education needs. | PMC10267788 |
Cluster Characterization and Content Optimization per Demographic Environment | EdC-3 | REGRESSION, PID | This study aimed to use systems control theory for ACT in a SNI to promote awareness regarding organ donation. In particular, an ACT mechanism enabled the efficient automation of budget adjustments given each content’s level of engagement.The population stratification prior to the ACT-enabled content tailoring for each demographic environment separately. In clusters 1 and 2, both rural demographic environments with a predominantly white population, content highlighting a personal story (EdC-Baseline) achieved a greater engagement level. In cluster 3, a demographic environment with high socioeconomic status, both contents highlighting personal stories regarding the relationship between parents and children (EdC-1 and EdC-Baseline) accomplished more engagement than the other contents. In cluster 4, a group with a greater proportion of Hispanic and African American populations, content with dense concepts about organ donation (EdC-3) had higher engagement at the beginning of the postoptimization phase; however, at the end of the intervention, content displaying a personal story attained more engagement. Cluster 4 was the only one that did not present a significant engagement increase in the optimal arm compared with the nonoptimal arm (β=.0936; We observed that using the PID controller on SNI promoted people’s awareness regarding organ donation since the total number of clicks increased by 15.47% from the nonoptimal arm to the optimal one. According to the regression, an additional 0.2187 (95% CI 0.162–0.276) click rate (C/I) was obtained in the optimal arm during the optimization phase.The precision public medicine assumption is that the effect of intervention varies across distinct subpopulations. Therefore, we need to unveil such subpopulations and tailor the intervention accordingly. We have demonstrated ( | PMC10267788 |
Limitations | PID | Even though the study addressed the California state disparities by stratifying the zip codes into distinct and meaningful demographic environments, the digital divide among minorities may still be biasing the results. This study did not assess the intervention’s effect on the number of organ donor registrations in California, which still needs to be tested in an adequately powered study.Another limitation concerns the PID parametric configuration, that is, the values of proportional, derivative, and integral constants. Finding suitable values for those constants is necessary because they interfere with the convergence rate and fine-tuning of the SNI mechanism. We defined those parameters after a preliminary analysis, which provided an acceptable performance for the mechanism proposed in this manuscript. However, each problem has an optimal set of constants that maximize the PDI’s performance in the respective context. Thus, we could improve the achieved solution by defining those constants through an optimization process driven by the SNI problem. Additionally, the error measure that drives SNI regarded only the difference between engagement and exposure. We could improve the SNI performance by using other relevant variables, such as complete video views, comments, and shares. | PMC10267788 |
|
Conclusions | PID, EVENT | We proposed an SNI framework with an ACT mechanism that learns and delivers, in real-time, for distinct subpopulations, the most tailored educational content and establishes new avenues to improve the future design of precision public health interventions using digital social media that are equitable, efficient, and cost-effective. In particular, the controller enabled an efficient automation of budget adjustment given the contents’ engagement level, prioritizing the most successful contents in each cluster. For clusters 1 and 2, the EdC-Baseline had more budget allocated by the controller, while for clusters 3 and 4, the EdC-1 was the prioritized content. That behavior shows how population stratification into demographic environments is a key step in the development of SNIs.The use of social media as a tool to promote health educational interventions moves toward more organized quantitative and personalized care as a pathway to improving the health care system using novel digital tools. The available social media’s ad management tool enables a level of control that allows the implementation of continuous randomized control trials, given the possibility of targeting people living with educational content in a set of specific zip codes and not in others, and every intervention serving as the reference for the next. On top of that feature, social media enables the real-time evaluation of the intervention in process; the number of people that view the content or clicked on the content’s link is available as soon as the event occurs. That high-resolution assessment makes it possible to use the PID controller to optimize the ongoing intervention instead of waiting until the end of the intervention to evaluate the results. The proposed SNI framework showed how precision public health can design novel health interventions with the use of social media, automation, and machine learning in a form that is more efficient and equitable. | PMC10267788 |
|
Future Works | PID, HEART | In future works, we intend to investigate other strategies to perform content optimization per demographic environment, including the parameter tuning for the PID controller and the use of other optimization techniques such as Genetic and Swarm-based algorithms. Although the PID controller optimization based on the difference between engagement and exposure increased people’s awareness, that initial approach could be improved by adding other relevant engagement measures, such as video views, comments, and shares. Moreover, the optimizer could also take into consideration other socioeconomic indicators such as race or ethnicity, educational level, and health insurance to guarantee equitable exposure between the distinct demographic environments.This work was funded by the Rosenfeld Heart Foundation. The sponsor had no role in the study.Conflicts of Interest: None declared.CONSORT-eHEALTH checklist (V 1.6.1). | PMC10267788 |
|
Abbreviations | adaptive content tuningnumber of clicks per 1000 impressionscluster-randomized trialeducational contentordinary least squaresproportional-integral-derivativesocial network interventionzip code tabulation area | PMC10267788 |
||
Objective | Alzheimer's disease, AD | ALZHEIMER'S DISEASE | Edited by: Madhan Balasubramanian, Flinders University, AustraliaReviewed by: Kofi Boamah Mensah, Kwame Nkrumah University of Science and Technology, Ghana; Chen Li, Shanghai University of Engineering Sciences, ChinaThis article was submitted to Aging and Public Health, a section of the journal Frontiers in Public HealthThis study aimed to investigate the knowledge status of Alzheimer's disease (AD) among community health service center (CHSC) staff in Jiaxing, China, and to compare the effects of online with offline training. | PMC9911520 |
Methods | Alzheimer's Disease Knowledge Scale | A total of 763 people from 12 community health service centers were investigated using a self-created general situation questionnaire and the Alzheimer's Disease Knowledge Scale (ADKS). Among the participants, 261 people who were willing to receive training were randomly divided into two groups according to the institution in which they worked to receive online or offline training, respectively. | PMC9911520 |
|
Results | DISEASE COURSE | The average ADKS score was 19.77, and the awareness rate was 65.92%; the results for every field were as follows: treatment and management (81.32%); life impact (77.76%); disease course (75.23%); assessment and diagnosis (68.94%); risk factors (65.05%); symptoms (57.90%); caregiving (44.06%). Education and profession had impacts on the total ADKS scores ( | PMC9911520 |
|
Conclusion | Community health service center staff in Jiaxing had limited knowledge of AD, particularly in the “symptom” and “caregiving” dimensions. One instance of training on AD-related knowledge to some degree helped to improve this but still fell short of meeting the national requirements. No significant differences were found between offline and online training effects. | PMC9911520 |
||
Background | dementia, behavioral impairment, cognitive dysfunction, Alzheimer's disease, AD | ALZHEIMER'S DISEASE, DISEASE | Alzheimer's disease (AD) is a neurodegenerative condition characterized by progressive cognitive dysfunction and behavioral impairment that occurs in pre-old age and old age (The focus on dementia and Alzheimer's disease has continued to increase in recent years. In September 2020, the National Health Commission announced the management of screening for Alzheimer's disease within basic national public healthcare services (National Health Office Disease Control Letter, 2020, No. 726) and instituted the requirement that public awareness rates regarding the prevention and treatment of AD in pilot areas be increased to 80% by 2022. All relevant medical staff in grassroots healthcare services should thus have knowledge about the prevention and treatment of dementia.According to the Statistical Bulletin of the Elderly Population and Undertakings for the Aged in Zhejiang Province,The ADKS is applicable to AD patients and their caregivers, elderly people in the community, students, and medical staff and has high reliability and validity ( | PMC9911520 |
Method | PMC9911520 |
|||
Sample and settings | MAY, DISEASES | Jiaxing includes three districts, as well as two county and three county-level cities, each with 4–12 streets or towns. From May to November 2020, 12 representative streets/towns were randomly selected by stratification according to geographical location and size. Following on, all the staff of community health service centers in these streets/towns was selected. A total of 860 questionnaires were returned; 97 were excluded for being too casual in nature; e.g., their answers were all “correct” or all “wrong,” the total time taken to complete them was very short, and questionnaires were repeated or had illegible handwriting. Finally, 763 valid questionnaires were collected. Among these, 261 individuals volunteered to participate in AD-related knowledge training and were randomly divided into an offline (142) and an online (119) training group, based on the institution in which they worked.The study's inclusion criteria were as follows: (1) staff who had worked in community health service centers for more than 1 month as of March 1, 2020; staff who had good compliance, and were willing to cooperate with the investigation and training.The study's exclusion criteria were as follows: (1) staff from community health service centers who treated AD and other neuropsychiatric diseases; (2) staff who had poor compliance and refused to cooperate with the investigators.All of the participants were informed about the research purpose, agreed to participate in the study, and actively cooperated with the investigation and training. | PMC9911520 |
|
Survey instruments | Alzheimer's Disease Knowledge Scale, dementia | DISEASE COURSE | A self-designed questionnaire for collecting general information, including name, gender, age, institution of employment, educational background, professional title, and current professions years of employment, whether any previous training related to AD had been received in the past, and whether participants' relatives/friends suffered from dementia or AD.The Alzheimer's Disease Knowledge Scale (ADKS), which comprises 30 true/false items, includes the following seven dimensions: risk factors (six items), assessment and diagnosis (four items), symptoms (four items), disease course (four items), life impact (three items), caregiving (five items), and treatment and management (four items). The total score ranges from 0 (worst) to 30 (best). | PMC9911520 |
Data collection and quality control | PMC9911520 |
|||
Questionnaire survey | Before and immediately after completing the training, the self-designed questionnaire and ADKS were completed in two ways; the offline training group filled in paper, and the online training group completed the questionnaire digitally using a provided link. Additionally, the online training group was invited to evaluate and provide feedback about the training they received. None of the participants were able to provide the correct answers after completing the ADKS for the first time. | PMC9911520 |
||
Training methods | The offline training group engaged in traditional learning, i.e., by gathering in a specific area where a lecturer presented a slide presentation and content explanations. For the online training group, the lecturer pre-recorded the training information and uploaded it to either the DingTalk or WeChat platforms. The members of this group finished the training on their own within the specified time. The content of the course was the same for both groups and the training duration was 1 h.Two neurology physicians were uniformly trained and qualified prior to conducting the survey. Following on, they conducted the questionnaire survey, and the AD-related knowledge training, and completed the data collection. The quality was controlled by at least one chief neurology physician. | PMC9911520 |
||
Statistical analysis | REGRESSION | Data were input into Microsoft Excel 2019, and the SPSS Statistics 25.0 software program was used to conduct statistical analysis. Quantitative data were presented in the form of median and interquartile spacing, which did not conform to normal distribution; a rank-sum test was conducted for making comparisons between the groups. Qualitative data were expressed as the number of cases (percentage), and a rank-sum test was used for comparison between groups in this context. The influencing factors were analyzed by multi-factor linear regression analysis. Statistical significance was based on a | PMC9911520 |
|
Results | PMC9911520 |
|||
Participant characteristics | AD, dementia | ALZHEIMER'S DISEASE | A total of 763 individuals from 12 community health service centers participated in the study. Most of the participants were female (73.0%), with an average age of 35.09 years; most had a bachelor's degree (71.6%), as well as junior or intermediate titles (74.2%), and 49.0% had majored in internal medicine. Their average employed time was 12.87 years. The majority reported that they had never received any training on AD-related knowledge (71.2%), and they had no relatives or friends who suffered from dementia or AD (80.5%). The sociodemographic characteristics of the sample are shown in Sociodemographic characteristics of the sample.AD, Alzheimer's disease. | PMC9911520 |
Alzheimer's Disease Knowledge Scale scores | anxiety, Alzheimer's disease” (19.79%), and “Tremor, Alzheimer's disease, Alzheimer's disease” (26.87%)., depression | DISEASE, ALZHEIMER'S DISEASE, DISEASE COURSE | The average ADKS score of 763 staff members was 19.77, and the awareness rate was 65.92%. The awareness rate of each specific field (from high to low) was as follows: treatment and management (81.32%); life impact (77.76%); disease course (75.23%); assessment and diagnosis (68.94%); risk factors (65.05%); symptoms (57.90%); caregiving (44.06%). The items with the best correct rate were “People whose Alzheimer's disease is not yet severe can benefit from psychotherapy for depression and anxiety” (96.20%), “A person with Alzheimer's disease becomes increasingly likely to fall down as the disease gets worse” (95.54%), and “Genes can only partially account for the development of Alzheimer's disease” (93.97%), while the poorest responses were “It has been scientifically proven that mental exercise can prevent a person from getting Alzheimer's disease” (11.40%), “If trouble with memory and confused thinking appears suddenly, it is likely due to Alzheimer's disease” (19.79%), and “Tremor or shaking of the hands or arms is a common symptom in people with Alzheimer's disease” (26.87%). Additional details are shown in | PMC9911520 |
Feedback information | A total of 111 of 119 participants of the online training group provided feedback information, most of whom agreed or agreed to some extent that “online training was an acceptable form of training,” “the training generally felt good,” “the training would be of great help to their future work,” and “they would also be willing to take part in online training in the future”; the above statements accounted for 77.48, 75.68, 77.48, and 77.48% of the respondents, respectively. Additional details are shown in | PMC9911520 |
||
Discussion | Cognitive Disorders, AD, dementia | REGRESSION, CORONAVIRUS | This is the first study to use ADKS to investigate the knowledge of Alzheimer's disease among community health service center staff in Jiaxing, and to conduct training and evaluate the effectiveness of the training. At present, the consultation rate of patients with dementia in China is not high, and most patients live in the community. In order that more patients with dementia can be identified, diagnosed, intervened and treated as soon as possible, the role of CHSC staff is of great importance. Understanding the current level of AD-related knowledge of CHSC staff in this area and their response and effectiveness to training is a favorable reference for designing training and formulating relevant policies in the future.The current study survey showed that the average ADKS score of 763 community staff members was 19.77, and the awareness rate was only 65.92%. This was similar to the results of studies conducted by Lin et al. (The multivariate regression analysis conducted for this study showed that education had an impact on ADKS scores, “assessment and diagnosis,” and “caregiving” aspects; additionally, a higher educational level yielded higher scores in the above areas, which was consistent with the results of various studies at home and abroad (Relatives or friends who suffered from dementia/AD had an impact on “symptoms,” and participants whose relatives/friends suffered from dementia or AD had higher scores. This was consistent with a study conducted by Liu et al. (The number of years staff members had been employed had an impact on “risk factors,” “assessment and diagnosis,” “life impact,” and “caregiving.” A longer period of employment indicated higher scores in the “assessment and diagnosis” and “life impact” categories but lower scores for “risk factors” and “caregiving.” The period of employment was positively correlated with age. Several foreign studies (The effect of gender on ADKS scores and all dimensions included was not determined. This study suggests that gender had an impact on “risk factors” and “symptoms,” and that males had higher scores in this instance. In a study conducted by Alacreu et al. (According to the present study, the awareness rate of AD knowledge among community health service center staff in Jiaxing was much lower than the national requirements. Therefore, targeted training is urgently needed to address this. This study showed that the ADKS scores for “treatment and management,” “assessment and diagnosis,” “risk factors,” and “symptoms” among the community health service center staff who volunteered to participate in the training had all been improved after completing the training. The ADKS scores of the offline and online training groups were both improved after the training compared with those before the training, with statistical significance. Accordingly, training indicated obvious importance. After completing the training, however, the awareness rate still did not meet the general requirements. Enhancement in different dimensions was unbalanced, as such, the role of one-time training was relatively limited. Hu et al. (Traditional training occurs offline and in person, and the place and time for its delivery are relatively fixed, and interaction during its delivery is relatively strong. The development of the Internet has gradually changed people's work, study and life, with online training has gradually entered people's attention. Particularly considering the outbreak of the novel coronavirus 2019, online teaching is useful for delivering training in a non-crowded (less contagious) environment. While the time and place in which online learning is delivered are highly selectivity, the interaction aspect is relatively poor and lacks supervision and management. In this study, staff members who volunteered to take part in the training were randomly divided into two groups (offline and online training, respectively). Due to the unit random grouping method, there were statistically significant differences between the two groups in terms of gender and current majors before the training, which is a common drawback of the current situation survey. Univariate linear regression analysis showed that gender and profession had no causal relationship with ADKS scores before and after training and, as such, they were not considered confounding factors. There was no significant difference in ADKS scores between the two groups before training and, as a result, the baseline data were comparable. There was no significant difference in ADKS scores between the two groups after training, either, suggesting a lack of significant difference in the training effect between the two methods (online and offline).An interprofessional team from the Department of Veterans Affairs, South Central Mental Illness Research (Houston, Texas, USA) designed the Program for Advancing Cognitive Disorders Education for Rural Staff to improve clinician competency and comfort when caring for individuals with dementia. Based on an interprofessional needs assessment, the team created six 1-h training modules, all of which are available for free The internet has become an important tool for learning and teaching. Training can be carried out online in the current pandemic environment, thereby reduce personnel gathering, save time, and complete the training task. In the present study, the majority of participants agreed that online training was acceptable. Xu et al. (The study also has some limitations. Firstly, a cross-sectional design of the present study could not determine the causal relationships, only associations between knowledge and related influence factors. Secondly, because of the stratified cluster sampling in this study, some bias might be present in the sample selection, thus the selected staff population may not reflect community health services centers (CHSCs) in other parts of Zhejiang Province or other parts of China when generalizing the findings. Thirdly, in the questionnaire survey, the offline training group completed hard copies of the survey, while the online training group completed it digitally using a provided link, which may have had a degree of influence. Fourthly, CHSC staff were also not enthusiastic about participating in training on AD knowledge. Fifthly, female participants outnumbered their male counterparts in the study; in the grouping study, the majority of the participants were female, which imbued the research with some limitations. | PMC9911520 |
Conclusion | 1. The staff of community health service centers in Jiaxing had a low awareness rate of AD-related knowledge, particularly in the “symptoms” and “caregiving” dimensions. 2. Education background and occupation are the influencing factors of ADKS score, and those with higher education background and physicians score more, and the years of working is not the influencing factor of ADKS score. 3. One-time training on AD knowledge could improve this shortcoming among community health service center staff to some extent but still failed to meet the national requirements. There were no obvious differences between offline and online training.Based on a variety of factors, offline, online, or a combination of both can be used to carry out training in the future; in doing so, assurances must be made that the training is repeated and updated, which will help to strengthen the training. This will help to improve the AD-related knowledge and skills of grassroots healthcare staff and provide a solid foundation for the early discovery, early diagnosis, and early treatment of AD in the community. | PMC9911520 |
||
Data availability statement | The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. | PMC9911520 |
||
Ethics statement | This study was conducted with approval from the Ethics Committee of the Second Affiliated Hospital of Jiaxing University (No. JXEY-2020JX065). The patients/participants provided their written informed consent to participate in this study. | PMC9911520 |
||
Author contributions | WM and XZ: conception and design of the research and obtaining financing. WM, LZ, JT, WD, LQ, and XF: acquisition of data. WM, LZ, JT, and XZ: analysis and interpretation of the data. WM, WD, LQ, and XF: statistical analysis. WM: writing of the manuscript. XZ: critical revision of the manuscript for intellectual content. All authors have read and approved the final draft.
| PMC9911520 |
||
Conflict of interest | The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. | PMC9911520 |
||
Publisher's note | All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. | PMC9911520 |
||
Supplementary material | The Supplementary Material for this article can be found online at: Click here for additional data file.Click here for additional data file.Click here for additional data file.Click here for additional data file.Click here for additional data file. | PMC9911520 |
||
References | PMC9911520 |
|||
Key Points | PMC10559185 |
|||
Question | schizophrenia | What is the therapeutic efficacy of multigenic pharmacogenomics–guided treatment in patients with schizophrenia? | PMC10559185 |
|
Findings | SYNDROME, POSITIVE | In this randomized clinical trial that included 210 Chinese Han men, patients treated with multigenetic pharmacogenomics–guided treatment had a greater symptom improvement than those treated with treatment as usual after a 6-week treatment, measured as the mean difference in percentage change of Positive and Negative Syndrome Scale score. | PMC10559185 |
|
Meaning | schizophrenia | Multigenetic pharmacogenomic testing can be an effective tool to guide the treatment of schizophrenia.This randomized clinical trial evaluates the clinical effectiveness of multigenetic pharmacogenomics–guided treatment in schizophrenia among Han men in China. | PMC10559185 |
|
Importance | schizophrenia | Limited evidence supports multigenetic pharmacogenomics–guided treatment (MPGT) in schizophrenia. | PMC10559185 |
|
Objective | schizophrenia | To evaluate the clinical effectiveness of MPGT in schizophrenia in a randomized clinical trial (RCT). | PMC10559185 |
|
Design, Setting, and Participants | schizophrenia | POSITIVE | This RCT was conducted from March 2020 to March 2022. Male Chinese Han inpatients aged 18 to 60 years diagnosed with schizophrenia with a Positive and Negative Symptom Scale (PANSS) score of 60 or more from 2 selected study hospitals were included. Patients and raters were masked to MPGT or treatment as usual (TAU) randomization. | PMC10559185 |
Interventions | Participants were randomly assigned in a 1:1 ratio to receive either MPGT or TAU for 12 weeks. | PMC10559185 |
||
Main Outcomes and Measures | SECONDARY, REMISSION | The primary efficacy outcome was the percentage change in PANSS total scores (range, 30 to 210) from baseline to week 6 analyzed by a modified intention-to-treat mixed model for repeated measures. The secondary outcome included response and symptomatic remission rates. | PMC10559185 |
|
Results | A total of 210 participants (mean [SD] age, 29.2 [8.8] years) were enrolled and analyzed, with 113 assigned to MPGT and 97 to TAU. Compared with those randomized to TAU, participants randomized to MPGT demonstrated a significantly higher percentage change in PANSS score (74.2% vs 64.9%; adjusted mean difference, 9.2 percentage points; 95% CI, 4.4-14.1 percentage points; | PMC10559185 |
||
Conclusions and Relevance | schizophrenia | In this RCT of MPGT, MPGT was more effective than TAU in treating patients with schizophrenia. These findings suggest that multigenetic pharmacogenomic testing could serve as an effective tool to guide the treatment of schizophrenia. | PMC10559185 |
|
Trial Registration | Chinese Clinical Trial Registry Identifier: | PMC10559185 |
||
Introduction | Schizophrenia, schizophrenia | DISORDER | Schizophrenia is a complex polygenic disorder characterized by disturbances in multiple mental modalities.In recent years, pharmacogenomic testing has been widely used to guide drug therapy and has yielded exciting results.Based on these foundations, we conducted an RCT to evaluate the efficacy of MPGT in male Chinese Han patients diagnosed with schizophrenia compared with treatment as usual (TAU). We hypothesized that MPGT would lead to a better improvement of symptoms. | PMC10559185 |
Methods | PMC10559185 |
|||
Study Design and Participants | schizophrenia | This study is a 12-week, 2-center, parallel RCT evaluating the therapeutic effects of MPGT compared with TAU in patients with schizophrenia. Patients and raters were masked to MPGT or TAU randomization. The trial protocol can be found in Participants were enrolled at the Peking University Sixth Hospital, Beijing, China, and the Second People’s Hospital of Guizhou Province, Guiyang, China, between March 2020 and March 2022. All participants were male Chinese Han inpatients aged 18 to 60 years who met the The protocols of this study were approved by the Medical Ethical Committee of the Sixth Hospital of Peking University and the Medical Ethical Committee of the Second People’s Hospital of Guizhou Province. Written informed consent from all participants or their guardians was obtained. The RCT conformed to the Consolidated Standards of Reporting Trials ( | PMC10559185 |
|
Randomization and Masking | Participants enrolled were randomly assigned to the MPGT or the TAU group in a 1:1 ratio using a preplanned randomization list. Raters were masked to the study group. Participants were masked to the study group and their pharmacogenomics report until the end of the trial. Clinicians who took care of participants in the MPGT group had access to the pharmacogenomics report to guide medication selection. | PMC10559185 |
||
Multigenetic Pharmacogenomic Testing | ADVERSE EFFECTS | In this study, we detected single-nucleotide variant loci of 26 alleles or variants across 11 genes that are reported to be associated with antipsychotic medication metabolism, efficacy, or adverse effects (eTable 1 in | PMC10559185 |
|
Trial Procedure | Detailed procedures are described in the eMethods in | PMC10559185 |
||
Outcome Measures | schizophrenia, first-episode schizophrenia | SECONDARY, REMISSION | The primary efficacy outcome of this trial was the percentage PANSS score change from baseline to the end of week 6. Percentage PANSS score change was calculated using the difference between the baseline and follow-up scores divided by the baseline score minus 30.The secondary outcome was response or remission rates at each time point. Early response was defined as a percentage PANSS score change of 20% or more at the end of week 2,There were several exploratory outcome measurements. Patients in the MPGT group were divided into subgroups of patients with first-episode schizophrenia and relapsed schizophrenia to compare their therapeutic efficacy. We also evaluate the distribution of pharmacogenomic recommendation levels in the TAU group and compared the differences in treatment efficacy among the recommendation levels. The dose of medications and metabolic profile, including triglyceride, cholesterol, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, fasting plasma glucose (FPG), and prolactin, were evaluated. | PMC10559185 |
Power Calculations | depressive disorder | Based on a previous study of pharmacogenomics-guided treatment on major depressive disorder, | PMC10559185 |
|
Statistical Analysis | For baseline characteristics analysis, variables were properly described as means and SDs, medians and IQRs, or counts and frequencies. For normally distributed data, unpaired The primary outcome (percentage PANSS score change) was evaluated by a mixed model for repeated measures (MMRM), which is a better approach than both last observation carried forwardStatistical significance was defined at a 2-tailed α level of .05. No correction for multiple testing of exploratory outcomes was applied. SPSS Statistics version 26.0 (IBM) and R version 4.2.1 (The R Foundation) were used for statistical analyses. All available postbaseline outcomes measures were included in the analysis. | PMC10559185 |
||
Results | PMC10559185 |
|||
Demographic Characteristics | first-episode schizophrenia | A total of 360 patients were screened for eligibility, of whom 213 patients entered the trial and 210 were included in the modified intention-to-treat analyses, with 113 randomized to MPGT and 97 to TAU. The mean (SD) age of the 210 included participants was 29.2 (8.8) years, and 63 participants (30.0%) had first-episode schizophrenia. The mean (SD) PANSS score at baseline was 102.1 (14.0). A summary of baseline characteristics is shown in | PMC10559185 |
|
Baseline Characteristics of the Modified Intention-to-Treat Population | high-density lipoprotein | SYNDROME, POSITIVE | Abbreviations: HDL, high-density lipoprotein; LDL, low-density lipoprotein; MPGT, multigenetic pharmacogenomics–guided treatment; PANSS, Positive and Negative Syndrome Scale; TAU, treatment as usual.SI conversion factor: To convert triglycerides to mmol/L, multiply by 0.0113; cholesterol to mmol/L, multiply by 0.0259; and glucose to mmol/L, multiply by 0.0555. | PMC10559185 |
Study Flowchart | mITT indicates modified intention to treat; MPGT, multigenetic pharmacogenomics–guided treatment; TAU, treatment as usual. | PMC10559185 |
||
Primary Outcome | schizophrenia | The time course of schizophrenia severity represented by PANSS scores and the percentage PANSS score change throughout the whole study are shown in | PMC10559185 |
|
Secondary Outcomes | REGRESSION | The early response rate was not statistically significantly different between treatment groups, as assessed by the odds ratio estimated by the logistic regression model. The response rates at the end of week 6 were 82.3% (93 of 113) for the MPGT group and 64.9% (63 of 97) for the TAU group (adjusted odds ratio, 2.48; 95% CI, 1.28-4.80; | PMC10559185 |
|
Secondary Outcomes | posturing, mannerisms, hallucinatory behavior, delusions | REMISSION, SYNDROME, POSITIVE, POSTURING | Abbreviations: MPGT, multigenetic pharmacogenomics–guided treatment; OR, odds ratio; TAU, treatment-as-usual.Early response was defined as a percentage Positive and Negative Syndrome Scale score change of 20% or more at the end of postrandomized week 2.Response was defined as a percentage Positive and Negative Syndrome Scale score change of 50% or more at the end of postrandomized week 6.Symptomatic remission was defined as a Positive and Negative Syndrome Scale score of 3 or less on items P1 (delusions), P2 (conceptual disorganization), P3 (hallucinatory behavior), N1 (blunted affect), N4 (passive/apathetic social withdrawal), N6 (lack of spontaneity and flow of conversation), G5 (mannerisms and posturing), and G9 (unusual thought content) at the end of week 12. | PMC10559185 |
Exploratory Outcomes | first-episode, schizophrenia | The difference in percentage PANSS score change between those with first-episode and relapsed schizophrenia in the MPGT group was not statistically significant, indicating a similar therapeutic effect of MPGT on both first-episode and relapsed schizophrenia (We found lower chlorpromazine equivalents were assigned to patients in the MPGT group than the TAU group throughout the study, though not statistically significant at the end of week 6 and week 12. MMRM of metabolic profile analysis showed no significant difference in lipid and glucose metabolism between treatment groups, but plasma prolactin levels were lower in the MPGT group (eTable 10 in | PMC10559185 |
|
Discussion | schizophrenia, psychiatric, Psychotic symptoms | REMISSION | To our knowledge, this study is the first RCT to evaluate the therapeutic efficacy of MPGT in Chinese Han patients with schizophrenia. Consistent with our hypotheses, patients with schizophrenia randomized to MPGT achieved greater improvements in symptoms and reported lower plasma prolactin levels.Several studies have evaluated the effects of single test–guided or limited multigenetic-test–guided pharmacotherapy in schizophrenia.In our study, patients treated with MPGT responded better to antipsychotic medications but with relevantly lower doses, which can be seen in the greater reduction of PANSS score and higher response rate. We found that 93 of 113 pharmacogenomics-guided patients (82.3%) reached a 50% reduction of PANSS score at the end of week 6 and 71 of 113 (62.8%) achieved symptomatic remission at the end of week 12. Psychotic symptoms were reported to be significantly associated with patients’ psychosocial functioning, cognitive function, and quality of life,The genetic variants included in the multigenetic pharmacogenomic testing of our study involved drug metabolism, transporters, and receptors. A significant number of the loci detected were also related to CYPs. Most of the psychiatric drugs currently available are metabolized by CYPs, especially CYP1A2, CYP2C19, CYP2D6, and CYP3A4 isoforms. | PMC10559185 |
Limitations | treatment-resistant | REMISSION | There are several limitations of our study. We only included male Chinese Han patients, and the response and remission rate in our study is relatively higher than in previous studies. Caution is needed in generalizing the findings to other populations, especially the treatment-resistant population. Moreover, to choose medications based on the pharmacogenomics report, clinicians were not masked to the study groups, which might increase the clinicians’ attention toward the patient’s medical treatment. Our study was a short-term study, and we did not evaluate the cost-effectiveness of pharmacogenomic testing. Long-term benefits and cost-effectiveness should be investigated in future studies with larger sample sizes and completer assessments.Evidence regarding the impact of CYP1A2 on the metabolism of antipsychotics is limited and inconsistent. As a result, we only considered its potential influence on clozapine and reduced its weight value. Nevertheless, due to the small number of patients using clozapine in our study, further research is needed to elucidate this matter more comprehensively. Moreover, while we restricted the use of CYP2D6 inhibitors, we did not limit the use of medications that might affect CYP1A2 and CYP3A4 enzymes. Factors such as smoking and coffee consumption, which can affect the activity of the CYP1A2 enzyme, were not accounted for in our study. Additionally, we did not document each nonpsychiatric medication that patients were concurrently taking. These factors must be considered in clinical practice when considering medication selection.It is important to acknowledge that our pharmacogenomic testing panel was primarily based on East Asian populations and did not encompass all potential loci and variants. For instance, our detection of | PMC10559185 |
Conclusions | schizophrenia | ADVERSE EFFECTS | In summary, our study showed that pharmacogenomics-guided treatment of antipsychotics significantly improved the clinical outcomes in terms of drug efficacy and adverse effects of prolactin elevation in patients with schizophrenia. Multigenetic pharmacogenomics testing can be an effective tool to guide the treatment of schizophrenia. | PMC10559185 |
2. Materials and Methods | PMC9914784 |
|||
2.1. Setting and Eligibility Criteria | Ovarian Cancer | DER, OVARIAN CANCER, BREAST | The study was conducted between 2019 and 2021 at six study centers in Germany that belong to the Familial Breast and Ovarian Cancer consortium (University Hospital of Cologne, University Hospital of Würzburg, University Hospital Schleswig-Holstein Campus Kiel, University Hospital Heidelberg, University Hospital Carl Gus-tav Carus Dresden, and University Hospital rechts der Isar Munich). Data presented in this paper were collected as part of a larger randomized controlled trial evaluating a decision support program in healthy | PMC9914784 |
2.2. Data Collection and Measures | depression, Depression, cancer, Anxiety | CANCER, EVENT | The data were collected from a self-administered questionnaire that participants received after giving informed consent. Participants were able to take the questionnaire home and were asked to return the questionnaire within one week. Data were collected on age, marital status, highest level of education, employment status, and children. Additionally, several measures on self-efficacy, decision making, and psychological burden were included: Coping self-efficacy. Self-efficacy to cope with the genetic test result was measured using the Coping Self-Efficacy Scale (CSE) [Anxiety and depression. The German version of the Hospital Anxiety and Depression Scale (HADS) was used to assess the level of psychological distress following disclosure of the genetic test result [Impact of genetic test result. To measure the subjective distress of the genetic test result, the Impact of Event Scale (IES) was used [Status of decision. The Stage of Decision-Making Scale (SDMS) is a single-item scale that assessed how far along women were in their decision-making process on how to deal with their cancer risk [Decisional conflict. Decisional conflict captures the uncertainty in health- or treatment-related decisions. It was measured using the German translation of the Decisional Conflict Scale (DCS) [ | PMC9914784 |
2.3. Statistical Methods | All statistical analyses were performed using SPSS Version 27.0 ) (IBM, Armonk, New York, USA). All results were interpreted using a two-sided | PMC9914784 |
||
3. Results | Overall, 130 women with a newly found | PMC9914784 |
||
3.1. Demographic Data | To identify potential covariates in the relationship between coping self-efficacy and psychological morbidity, age, BRCA mutation, and parity were analyzed. Age was not significantly correlated to any variable of interest. When comparing | PMC9914784 |
||
3.2. Anxiety and Coping Self-Efficacy | anxiety, non-elevated anxiety | The sample’s overall anxiety mean measured by the HADS was just under the threshold of borderline clinical anxiety at When categorized into groups, almost half of the sample (45.4%, For problem-based coping (CSE-PF), post hoc tests showed that participants with non-elevated anxiety were significantly better at problem-focused coping when compared to participants with clinical anxiety, For support from family and friends (CSE-SFF), post hoc tests revealed that those with non-elevated anxiety showed higher confidence in receiving support from friends and family compared to those with clinical anxiety, For stopping unpleasant emotions (CSE-SUE), post hoc tests revealed that those with non-elevated anxiety felt more equipped in stopping unpleasant emotions compared to those with clinical anxiety, Overall coping self-efficacy was better for non-elevated vs. borderline clinical anxiety, | PMC9914784 |
|
3.3. Depression and Coping Self-Efficacy | depression | The sample showed a mean score of depression of | PMC9914784 |
|
3.4. Impact of Event Scale | EVENT | The mean score for the impact of event scale was | PMC9914784 |
|
3.5. Decision Making and Coping Self-Efficacy | Frequency analysis of the SDMS (missing n = 7) yielded that very few participants (3.8%) had not thought about the preventive options yet (stage 1). Most participants (42.3%) reported that they were currently thinking about the different preventive options (stage 2), while 16.9% reported being close to making a decision (stage 3). About a third of participants (31.9%) indicated that they had already made a choice about which preventive option they were going to opt for (stage 4). While we hypothesized that those with higher coping self-efficacy would be further along in the decision-making process, univariate ANOVA yielded no significant differences between the groups regarding any of the coping self-efficacy subscales or the total scale.In terms of decisional conflict, over half of the sample (51.5%) had a total score ≥37.5, indicating the presence of decisional conflict. When investigating the subscales, 36.9% felt uninformed, 49.2% felt unclear about personal values, 46.2% felt unsupported in the decision-making process, and 60% felt uncertain about the best preventive option. As many participants had not made a final choice about the best preventive option, the subscale effective decision was not analyzed. Total decisional conflict was unrelated to coping self-efficacy, | PMC9914784 |
||
4. Discussion | anxiety, breast cancer, psychological distress, depressed, cancer worry, depression | BREAST CANCER | In this study, we identified high rates of anxiety, with over half of newly found cancer-unaffected The present study addresses a gap in the literature by investigating coping self-efficacy and its relationship with psychological and decision-related outcomes after genetic test result disclosure in cancer-unaffected female For depression, a similar pattern of results emerged, with lower coping self-efficacy being associated with a higher likelihood of clinical or borderline clinical depression. Participants with high confidence to receive social support from family and friends post genetic test result felt significantly less depressed. This pattern has similarly been reported in the literature; breast cancer patients with higher coping self-efficacy generally experience less psychological distress, including depression, anxiety, distress, and cancer worry [While coping self-efficacy seemed to have a strong negative relationship to psychological burden after genetic test result disclosure, it did not appear to have any impact on the current certainty in the decision-making process. Decisional conflict was present for a majority of the sample, but contrary to the hypotheses, higher coping self-efficacy was not associated with less decisional conflict, nor was it associated with being further along the decision-making process. Interestingly, decisional conflict was by and large not related to any of the psychological parameters. Psychological distress has previously been found to be a predictor of decisional conflict in longitudinal studies [Further limitations of this study need to be addressed. Participants were recruited after receiving their genetic test result. This could have led to biased sampling because women with seemingly higher psychological distress might not have been asked to participate in the study, as well as self-selection because distraught women might not have been willing to participate. Additionally, it has previously been reported that those with higher self-efficacy are more likely to undergo genetic testing for Despite these limitations, several clinical implications may be deducted from our results. The high rates of clinical anxiety support the notion for low-threshold psycho-oncological support, which should routinely be offered to newly diagnosed | PMC9914784 |
5. Conclusions | The aim of this study was to obtain insights into the relationship between coping self-efficacy and psychological morbidity in newly found | PMC9914784 |
||
Author Contributions | Conceptualization, S.S., R.S. and K.R.; methodology, S.S., R.S., K.R., F.V., A.S. and J.K.-N.; software, A.I. and Z.L.; formal analysis, A.I. and Z.L.; investigation, S.S., K.R. and R.S.; resources, F.V., H.F., A.S., B.B.-H., J.K.-N. and K.B.; data curation, M.T. and Z.L.; writing—original draft preparation, A.I.; writing—review and editing, F.V., J.K.-N., A.S. and B.B.-H.; visualization, Z.L.; supervision, S.S.; project administration, A.I., S.S. and M.T.; funding acquisition, S.S. All authors have read and agreed to the published version of the manuscript. | PMC9914784 |
||
Institutional Review Board Statement | The study was conducted according to the guidelines of the Declaration of Helsinki. Ethical approval to conduct the trial was obtained from the University Hospital of Cologne Research Ethics committee (no. 19-1110_1) on 19 August 2019. The trial was registered in the German Clinical Trial Register (WHO Registry Network; ID: DRKS00015527). | PMC9914784 |
||
Data Availability Statement | All data are available from authors upon reasonable request. | PMC9914784 |
||
Conflicts of Interest | The authors declare no conflict of interest. | PMC9914784 |
||
References | depression, Depression, Anxiety, anxiety | Demographic frequencies in the sample.Note. No demographic variable showed significant differences within its subgroups in terms of coping self-efficacy.Means, standard deviation, and correlations among all continuous psychological and decision-making variables.HADS—DepressionHADS—AnxietyDCS—InformedDCS—Value ClarityDCS—SupportDCS—UncertaintyDCS—TotalCSE—PFCSE—SFFCSE—SUECSE—TotalIES—IntrusionIES—AvoidanceIES—HyperarousalIES—TotalNote. HADS = Hospital Anxiety and Depression Scale [Mean scores on the CSE according to anxiety level.Note. HADS-A = Anxiety subscale of the Hospital Anxiety and Depression Scale [Mean scores on the CSE according to depression level.Note. HADS-D = Depression subscale of the Hospital Anxiety and Depression Scale [ | PMC9914784 |
|
Subject terms | grey matter alterations | SECONDARY | The combination of repeated behavioral training with transcranial direct current stimulation (tDCS) holds promise to exert beneficial effects on brain function beyond the trained task. However, little is known about the underlying mechanisms. We performed a monocenter, single-blind randomized, placebo-controlled trial comparing cognitive training to concurrent anodal tDCS (target intervention) with cognitive training to concurrent sham tDCS (control intervention), registered at ClinicalTrial.gov (Identifier NCT03838211). The primary outcome (performance in trained task) and secondary behavioral outcomes (performance on transfer tasks) were reported elsewhere. Here, underlying mechanisms were addressed by pre-specified analyses of multimodal magnetic resonance imaging before and after a three-week executive function training with prefrontal anodal tDCS in 48 older adults. Results demonstrate that training combined with active tDCS modulated prefrontal white matter microstructure which predicted individual transfer task performance gain. Training-plus-tDCS also resulted in microstructural grey matter alterations at the stimulation site, and increased prefrontal functional connectivity. We provide insight into the mechanisms underlying neuromodulatory interventions, suggesting tDCS-induced changes in fiber organization and myelin formation, glia-related and synaptic processes in the target region, and synchronization within targeted functional networks. These findings advance the mechanistic understanding of neural tDCS effects, thereby contributing to more targeted neural network modulation in future experimental and translation tDCS applications.The neural mechanisms underlying the beneficial effects of behavioural training in combination with transcranial direct current stimulation (tDCS) are not well understood. Here, the authors combine cognitive training with tDCS, showing a modulation of prefrontal white and grey matter microstructure, and increased prefrontal functional connectivity. | PMC10238397 |
Introduction | brain impairment | Developing effective cognitive interventions to reduce or even prevent age-associated brain impairment has received substantial scientific attention in aging societies worldwide. Preliminary evidence suggests that the combination of behavioral training and concurrent transcranial electrical stimulation (tES), one of the most widely used non-invasive brain stimulation (NIBS) techniques, may induce cross-task cognitive benefits, in young adults and advanced ageAs for learning-related brain plasticity, previous work has shown that the brain’s microstructure can be modified by learning. Seminal work in post-mortem monkey brains showed that learning of a new skill indeed induces generation of denser and more extensive white matter projectionsTES non-invasively modulates excitability and synaptic plasticity in neuronsSimultaneous tDCS-fMRI application in proof-of-principle studies revealed changes in local activity and functional connectivity (temporally coherent activity between brain regions) that predicted behavioral performance gainsEstablishing the underlying cellular (and molecular) mechanisms in the human brain can advance understanding of neuromodulatory plasticity. Animal models suggest modified tissue density due to altered neuronal morphology (e.g., size/shape of axons, dendritic spines and cell bodies), altered glial cells activity or reorganization/reshaping of synaptic connections as neuroplastic phenomena induced by tDCSIn this work, we tested the hypotheses that concurrent anodal prefrontal tDCS administered across repeated cognitive training sessions would modulate white matter microstructure in cortical target areas and associated neural networks compared to training with placebo (sham) stimulation. tDCS (1 mA) was administered for 20 min concurrently with two executive function training tasks (letter updating training, decision-making). While there were no between-group differences in the primary outcome (performance on letter-updating), we observed superior near-transfer effects (performance on N-back) in the tDCS group at post-intervention and follow-up, but in no other transfer tasks (please see ref. | PMC10238397 |
|
Study flow chart. | Following a pre-assessment of performance on the cognitive tasks, a pre-intervention MRI was conducted; the intervention commenced two days later and lasted for three weeks (with active (anodal) or sham tDCS + training administered three times per week). A post-intervention MRI session was conducted two days after the end of the intervention period. MRI magnetic resonance imaging, tDCS transcranial direct current stimulation. ‘Exclusion of | PMC10238397 |
||
White matter microstructure is modulated after brain stimulation | We performed individual probabilistic tractography seeding from the stimulation target (left middle frontal gyrus, defined to represent the gyrus below the anodal electrode, picked from the Harvard-Oxford atlas | PMC10238397 |
||
Gray matter microstructure is altered after brain stimulation | CORTEX | Gray matter regions in the cortex underneath the anode (left middle frontal cortex) were segmented using Freesurfer | PMC10238397 |
|
Functional connectivity is increased after brain stimulation | To investigate whether functional connectivity was modulated by anodal tDCS, we performed seed-to-voxel correlational analyses on resting-state fMRI data using CONN | PMC10238397 |
||
Pathways‘ microstructure change is associated with performance gain | performance gain | In order to explore linear relationships between the effects on different MR markers as well as with performance gain (LU and N-back change), correlation matrices were generated, illustrating scatterplots and Spearman correlation coefficients for all bivariate associations (Fig. | PMC10238397 |
|
Scatterplots for correlations between Post-Pre differences in FA, MD, and FC with individual performance gain (LU and N-back change). | BRAIN | Brain images were created with MRIcroGL (Bivariate scatterplots also revealed that microstructural plasticity in the stimulation target was associated with functional connectivity modulation: Higher decreases in gray matter MD were associated with increases in FC due to the intervention (As a control, we explored bivariate monotonic relationships between baseline integrity values (FA in white matter, MD in gray matter, and FC between the target and the resultant cluster) and behavioral performance gain. No substantial associations emerged (Supplementary Fig. | PMC10238397 |
|
Discussion | behavioral performance gain, stroke | STROKE, CORTEX | Three-week brain stimulation-assisted cognitive training in healthy older adults resulted in modifications of microstructure in white matter pathways and gray matter cortical target area as well as functional connectivity changes in a broader frontoparietal network. FA in prefrontal tracts originating from the stimulation target was increased in the group that had received anodal vs. sham tDCS, indicating higher integrity (i.e., directional preference of diffusion/directional coherence) of frontoparietal white matter tracts which was associated with higher (transfer task) performance gains following the intervention. Further, gray matter microstructure changes differed between the stimulation groups, mainly in individuals with higher microstructural integrity (i.e., molecular diffusion rate/magnitude of water molecule diffusion) at baseline that showed decreased MD values after anodal vs. sham tDCS. Increased resting-state FC between prefrontal areas indicated additional synchronization within frontoparietal networks induced by tDCS. Overall, we provide evidence for microstructural and network modifications through brain stimulation in the human brain, which may characterize the underlying mechanisms of functional benefits due to the intervention.We reconstructed individual tracts originating from the stimulation target in the left prefrontal cortex. Canonical images across our group of participants suggested that white matter fibers project from the stimulation target towards ipsilateral parietal and contralateral prefrontal areasThe DTI-derived index FA reflects the directional preference of diffusion and can be used to quantify the integrity of fiber organization in the human brain with higher values describing higher integrityIn a seminal study investigating structural changes induced by a combined tDCS-and-physical therapy intervention in stroke patients, Zheng and Schlaug observed increased FA in descending motor tracts in the treatment but not in the control groupCandidate cellular mechanisms reflected in FA variations include alterations in cell membrane and fiber density, fiber coherence, axon diameter, myelination, collateral sprouting. While intracellular directional coherence contributes to the FA metric, extracellular properties have been shown to affect the diffusion of water molecules as wellNo correlation was observed for microstructural alterations with behavioral performance gain in the training (LU) task. Differences in the task related to the procedure of administration (e.g., repeated vs. single sessions), content (e.g. letters vs. numbers) and involved executive processes (continuous updating in conjunction with memorizing temporal order vs. active comparison operations) may not only affect brain activation patterns and, thus, magnitudes of behavioral modulation, but also the relationships to neural plasticityIn order to examine microstructural changes within the gray matter of the stimulation target, MD values were extracted. A between-group comparison revealed an interaction between baseline MD values and the stimulation group effect, indicating a decrease after the intervention in the anodal compared to the sham group for individuals with initially lower values in the stimulation target.The DTI-derived index MD reflects the molecular diffusion rate and is used to quantify tissue microstructure. Higher MD values indicate reduced restriction of water molecule diffusion by cellular structuresOur finding of decreased MD in the anodal compared to sham group may indicate increases in tissue density (due to reshaping of neuronal or glial processes) or enhanced tissue organization (due to strengthened dendrites or axons) due to tDCSIt is important to note that DTI metrics are only indirect measures of microstructureIn our data, regional MD modulation was not related to performance gain, suggesting a more complex relationship with potentially other influencing factors, such as general training abilityIn order to examine potential functional connectivity modulations, we conducted seed-based FC analyses using resting-state fMRI. We found increases in FC in the prefrontal task-independent frontoparietal network in the anodal compared to the sham group. Similar FC modulations have been observed in task fMRI during single tDCS applications and after repeated tDCS sessions combined with working memory training in older adultsA limitation of our study is the relatively small sample size. In particular, in the context of brain-behavior associations, large sample sizes may be required for the observed relationships to be reliable/reproducibleIn sum, the present study advances the understanding of neurobiological after-effects of non-invasive brain stimulation combined with repeated training interventions and shows that tDCS exerts its effects on multiple levels, including microstructural properties of white matter tracts and gray matter regions and coordinated activitivity between distant brain regions. This rapid remodeling of neuro-glial networks and long-range signaling as the result of neuromodulation may underlie the functional effects, as indicated by their (partial) association with the observed performance gains | PMC10238397 |
Methods | PMC10238397 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.