JMIR Formative Research最新文献

{"title":"Determinant Factors of Stress in Caregivers of Patients With Schizophrenia: Cross-Sectional Study.","authors":"Isymiarni Syarif, Hasnawati Amqam, Saidah Syamsuddin, Veni Hadju, Syamsiar Russeng, Yusran Amir","doi":"10.2196/70648","DOIUrl":"10.2196/70648","url":null,"abstract":"<p><strong>Background: </strong>Caregivers of individuals with schizophrenia face ongoing psychological and emotional burdens due to the chronic and relapsing nature of the disorder and the complexity of caregiving. Prolonged exposure to caregiving stress characterized by emotional exhaustion, role overload, and lack of social support has been consistently associated with poor mental health outcomes among caregivers, including depression and anxiety.</p><p><strong>Objective: </strong>This study aimed to assess stress levels among caregivers of patients with schizophrenia and identify the key determinants of caregiver stress.</p><p><strong>Methods: </strong>This study used a cross-sectional survey that was conducted between June and August 2024 at the Labakkang District Health Center, South Sulawesi, Indonesia. A total of 110 female caregivers participated in the study. Data were collected using validated questionnaires to measure stress levels and related factors. Statistical analyses included chi-square tests to identify associations and partial least squares structural equation modeling to examine the strength and direction of relationships between variables.</p><p><strong>Results: </strong>This study included 110 female caregivers of individuals with schizophrenia. The majority were early older people (48/110, 44%), had a basic level of education (elementary to junior high school; 45/110, 46%), were unemployed (83/110, 75%), and had been providing care for more than 10 years (42/110, 38%). A total of 58 of 110 (53%) caregivers experienced mild levels of stress, while 63 of 110 (57%) caregivers reported a moderate caregiving burden. Additionally, 64 of 110 (58%) caregivers reported challenges related to patient treatment nonadherence, and 58 of 110 (53%) caregivers experienced low levels of social stigma. Most caregivers (69/110, 63%) adopted adaptive coping strategies; however, more than half reported low levels of knowledge (59/110, 54%) and limited access to health information (73/110, 66%). The chi-square analysis identified several statistically significant associations with stress: age (P=.03), education (P<.001), caregiving burden (P<.001), knowledge (P<.001), coping strategies (P<.001), treatment nonadherence (P=.004), and perceived stigma (P=.003). Further, partial least squares structural equation modeling analysis showed that caregiving burden (r=0.672), stigma (r=0.921), and limited knowledge (r=0.909) were positively correlated with stress. In contrast, social support was strongly negatively associated with stress (r=-0.872), indicating its protective role.</p><p><strong>Conclusions: </strong>These findings underscore the critical need for targeted interventions that enhance social support networks, reduce stigma, and strengthen caregivers' coping capacities. Strengthening these dimensions is essential to mitigating the psychological toll of caregiving and sustaining caregivers' functional well-being. Evidence increasingly supports","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e70648"},"PeriodicalIF":2.0,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12244270/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144560180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating a Large Language Model's Ability to Synthesize a Health Science Master's Thesis: Case Study.","authors":"Pål Joranger, Sara Rivenes Lafontan, Asgeir Brevik","doi":"10.2196/73248","DOIUrl":"10.2196/73248","url":null,"abstract":"<p><strong>Background: </strong>Large language models (LLMs) can aid students in mastering a new topic fast, but for the educational institutions responsible for assessing and grading the academic level of students, it can be difficult to discern whether a text has originated from a student's own cognition or has been synthesized by an LLM. Universities have traditionally relied on a submitted written thesis as proof of higher-level learning, on which to grant grades and diplomas. But what happens when LLMs are able to mimic the academic writing of subject matter experts? This is now a real dilemma. The ubiquitous availability of LLMs challenges trust in the master's thesis as evidence of subject matter comprehension and academic competencies.</p><p><strong>Objective: </strong>In this study, we aimed to assess the quality of rapid machine-generated papers against the standards of the health science master's program we are currently affiliated with.</p><p><strong>Methods: </strong>In an exploratory case study, we used ChatGPT (OpenAI) to generate 2 research papers as conceivable student submissions for master's thesis graduation from a health science master's program. One paper simulated a qualitative health science research project and another simulated a quantitative health science research project.</p><p><strong>Results: </strong>Using a stepwise approach, we prompted ChatGPT to (1) synthesize 2 credible datasets, and (2) generate 2 papers, that-in our judgment-would have been able to pass as credible medium-quality graduation research papers at the health science master's program the authors are currently affiliated with. It took 2.5 hours of iterative dialogue with ChatGPT to develop the qualitative paper and 3.5 hours to develop the quantitative paper. Making the synthetic datasets that served as a starting point for our ChatGPT-driven paper development took 1.5 and 16 hours for the qualitative and quantitative datasets, respectively. This included learning and prompt optimization, and for the quantitative dataset, it included the time it took to create tables, estimate relevant bivariate correlation coefficients, and prepare these coefficients to be read by ChatGPT.</p><p><strong>Conclusions: </strong>Our demonstration highlights the ease with which an LLM can synthesize research data, conduct scientific analyses, and produce credible research papers required for graduation from a master's program. A clear and well-written master's thesis, citing subject matter authorities and true to the expectations for academic writing, can no longer be regarded as solid proof of either extensive study or subject matter mastery. To uphold the integrity of academic standards and the value of university diplomas, we recommend that master's programs prioritize oral examinations and school exams. This shift is now crucial to ensure a fair and rigorous assessment of higher-order learning and abilities at the master's level.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e73248"},"PeriodicalIF":2.0,"publicationDate":"2025-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12244274/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144560181","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Implementing a Cross-Border Next-Generation Personal Health Record in the Philippines and Taiwan: An Implementation Case Report Using Health Level 7 International Fast Healthcare Interoperability Resources.","authors":"Hsiu-An Lee, Jui-Chun Huang, Shih-Wun Huang, Wei-Han Chen, Alvin B Marcelo, Miguel Sandino O Aljibe, Chien-Yeh Hsu","doi":"10.2196/56272","DOIUrl":"10.2196/56272","url":null,"abstract":"<p><strong>Background: </strong>Disparities in electronic health record systems hinder cross-border continuity of care, particularly where labor mobility and tourism intersect (eg, between the Philippines and Taiwan). Both nations collect claim data, yet neither fully aligns with international standards such as the Health Level 7 International, International Patient Summary (IPS).</p><p><strong>Objective: </strong>This implementation report aimed to convert health insurance data from Taiwan's My Health Bank (MHB) and the Philippine Health Insurance Corporation's Claim Form 4 (CF4) into a cross-border personal health record (PHR) aligned with the IPS using (Fast Healthcare Interoperability Resources) FHIR standards.</p><p><strong>Methods: </strong>We mapped each data element from CF4 (n=7 main sections) and MHB (n=12 major data items) to 17 sections of the IPS. We analyzed whether these elements matched IPS requirements (required or recommended or optional) and identified missing fields (eg, device use, social history, and advanced directives). We also designed a FHIR-based integration architecture, addressing system security with OAuth 2.0/SMART on FHIR and proposing a national uptake strategy for accelerating cross-border PHR implementation.</p><p><strong>Results: </strong>Of the 17 IPS sections, MHB covered 14 sections (82.4%), while CF4 covered 12 sections (70.6%). Both systems lacked sufficient data elements for medical devices, social history (eg, alcohol or tobacco), and advanced directives. We developed an implementation plan focusing on data interoperability, standardization, and privacy or security protocols. We propose a multiphase approach-beginning with the stakeholder engagement and pilot testing in both countries.</p><p><strong>Conclusions: </strong>Aligning CF4 and MHB data with IPS standards via FHIR can facilitate a robust cross-border next-generation PHR ecosystem. This approach may enhance patient safety, continuity of care, and policy development for the Philippines and Taiwan. Further collaboration, regulatory updates, and public awareness are vital to sustain integration and maximize patient benefits.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e56272"},"PeriodicalIF":2.0,"publicationDate":"2025-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12240210/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144553634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the Fit of a Digitally Delivered National Diabetes Prevention Program Among Rural Living Adults: Qualitative Study.","authors":"Gerit Wagner, Lyndsie M Koon, Patricia Smith, Kameron B Suire, Mary Hastert, Joseph E Donnelly, Melissa D Olfert, Paul Estabrooks, Anna M Gorczyca","doi":"10.2196/70406","DOIUrl":"10.2196/70406","url":null,"abstract":"<p><strong>Background: </strong>Rural living adults are disproportionately affected by type 2 diabetes compared to their urban counterparts. The Centers for Disease Control and Prevention's National Diabetes Prevention Program (National DPP) is an evidence-based intervention that reduces the risk of type 2 diabetes through increased physical activity and modest weight loss, but overall reach remains limited, specifically in rural communities.</p><p><strong>Objective: </strong>This qualitative study aimed to examine the fit of the National DPP delivered digitally using Zoom or Facebook to rural living adults at risk for type 2 diabetes.</p><p><strong>Methods: </strong>Focus group scripts assessed the characteristics and perceptions of rural adults at risk for type 2 diabetes, infrastructure supports for implementation and sustainability, and external factors that could influence program fit. A reflexive thematic analysis was conducted separately on coded transcripts for each focus group. Themes were then deductively linked to the Practical, Robust Implementation and Sustainability Model domains.</p><p><strong>Results: </strong>Two focus groups were conducted with 14 participants after participating in the National DPP for 6 months, delivered through Zoom (n=9) or Facebook (n=5). Participants highlighted positive relationships between Practical, Robust Implementation and Sustainability Model constructs related to participant characteristics (ie, value of health improvements, weight loss, and reduced medication dependence as primary motivators) and perceptions of compatibility (ie, content alignment with participant needs) as well as infrastructure (ie, digital platforms provided better access) with program success in reach and engagement. Conversely, both formats were negatively impacted by interruptions in internet connectivity. External factors, such as referral pathways from local health care providers, could improve program reach. When considering differences between implementation infrastructure, Zoom facilitated greater social engagement and accountability compared to Facebook.</p><p><strong>Conclusions: </strong>This study identified contextual factors influencing the fit of digitally delivering the National DPP to rural living adults, including opportunities for using existing connections and health motivations to help improve acceptability, while tailoring curriculum, modality, and technology may improve appropriateness for rural populations.</p><p><strong>Trial registration: </strong>ClinicalTrials.gov NCT05387434; https://clinicaltrials.gov/study/NCT05387434.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e70406"},"PeriodicalIF":2.0,"publicationDate":"2025-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12268215/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144540281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating the Quality of Psychotherapy Conversational Agents: Framework Development and Cross-Sectional Study.","authors":"Kunmi Sobowale, Daniel Kevin Humphrey","doi":"10.2196/65605","DOIUrl":"10.2196/65605","url":null,"abstract":"<p><strong>Background: </strong>Despite potential risks, artificial intelligence-based chatbots that simulate psychotherapy are becoming more widely available and frequently used by the general public. A comprehensive way of evaluating the quality of these chatbots is needed.</p><p><strong>Objective: </strong>To address this need, we developed the CAPE (Conversational Agent for Psychotherapy Evaluation) framework to aid clinicians, researchers, and lay users in assessing psychotherapy chatbot quality. We use the framework to evaluate and compare the quality of popular artificial intelligence psychotherapy chatbots on the OpenAI GPT store.</p><p><strong>Methods: </strong>We identified 4 popular chatbots on OpenAI's GPT store. Two reviewers independently applied the CAPE framework to these chatbots, using 2 fictional personas to simulate interactions. The modular framework has 8 sections, each yielding an independent quality subscore between 0 and 1. We used t tests and nonparametric Wilcoxon signed rank tests to examine pairwise differences in quality subscores between chatbots.</p><p><strong>Results: </strong>Chatbots consistently scored highly on the sections of background information (subscores=0.83-1), conversational capabilities (subscores=0.83-1), therapeutic alliance, and boundaries (subscores=0.75-1), and accessibility (subscores=0.8-0.95). Scores were low for the therapeutic orientation (subscores=0) and monitoring and risk evaluation sections (subscores=0.67-0.75). Information on training data and knowledge base sections was not transparent (subscores=0). Except for the privacy and harm section (mean 0.017, SD 0.00; t3=∞; P<.001), there were no differences in subscores between the chatbots.</p><p><strong>Conclusions: </strong>The CAPE framework offers a robust and reliable method for assessing the quality of psychotherapy chatbots, enabling users to make informed choices based on their specific needs and preferences. Our evaluation revealed that while the popular chatbots on OpenAI's GPT store were effective at developing rapport and were easily accessible, they failed to address essential safety and privacy functions adequately.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e65605"},"PeriodicalIF":2.0,"publicationDate":"2025-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12239686/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144540282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction: Identifying Themes for an Initial Beta Version of a Mobile Health App for Latino and Native Hawaiian/Pacific Islander Communities: Co-Design and Community-Based Participatory Research in a Code to Community Study.","authors":"Christina K Holub, Amy L Barrera, Rosalva Romero Gonzalez, Diane Hoang, Luna Prieto, Samuelu Fesili, Tiana Smith, Harleen Kaur, Cassandra Surban, Michael Markidis, Tana Lepule, Konane Martinez","doi":"10.2196/79194","DOIUrl":"10.2196/79194","url":null,"abstract":"<p><p>[This corrects the article DOI: 10.2196/76178.].</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e79194"},"PeriodicalIF":2.0,"publicationDate":"2025-07-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12268213/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144553633","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Remotely Delivered GLP-1RA-Supported Specialist Weight Management Program in Adults Living With Obesity: Retrospective Service Evaluation.","authors":"Rebecca Richards, Michael Whitman, Gina Wren, Peta Campion","doi":"10.2196/72577","DOIUrl":"10.2196/72577","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Limited access to specialist weight management services restricts the implementation of novel pharmacotherapies for obesity such as glucagon-like peptide-1 receptor agonists (GLP-1RAs) in the UK National Health Service (NHS). Second Nature, a commercial digital health company, offers a remotely delivered program combining a GLP-1RA medication (semaglutide) with digital behavioral support, potentially providing a scalable solution. However, evidence for long-term effectiveness in this real-world context is limited.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study aimed to evaluate the 12-month effectiveness, feasibility, acceptability, and potential cost-effectiveness of the remotely delivered, semaglutide-supported weight management program by Second Nature.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;This retrospective service evaluation analyzed data from participants who initiated the program between September and December 2023. The primary outcome was weight change at 12 months among participants with available data (completers). Secondary outcomes included retention, program engagement (measured by views of the Home screen in the app), behavioral changes, side effects, participant experience (qualitative analysis), and a comparative cost analysis against an NHS specialist weight management service. An \"active subscription\" was defined as maintaining a paid subscription for the full 12-month period. Descriptive statistics and paired 2-tailed t tests evaluated outcomes.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Data from 341 participants were included at baseline (282/341, 82.7% women; mean age 49, SD 11.1 years; mean baseline BMI 37.9, SD 6.9 kg/m&lt;sup&gt;2&lt;/sup&gt;). At 12 months, 39.6% (135/206) maintained an active subscription, while 60.4% (206/341) became inactive. Weight data at 12 months were available for 179 participants (52.5% of the baseline cohort; 100% of active and 19.4% of inactive participants). Among completers who maintained an active subscription, the mean weight loss was 20.0 kg (SD 8.7 kg; P&lt;.001), representing 19.1% of starting weight. Overall, 77.7% (139/179) of completers achieved ≥10% weight loss and 61.5% (110/179) achieved ≥15%. Program engagement declined over time. Side effects also decreased, with 69.6% (81/116) of respondents reporting none by month 12. Most participants completing the 12-month survey reported positive (41/120, 34.2%) or neutral (68/120, 56.7%) experiences.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;This evaluation suggests that remotely delivered GLP-1RA-supported weight management can achieve significant weight loss in participants remaining engaged for 12 months. However, the high rate of withdrawal limits generalizability. The program appears feasible, acceptable, and potentially cost-effective for completers. Further research, ideally in public health care settings using intent-to-treat analyses, is needed to confirm clinical outcomes, assess sustained results, and understand factors influencing re","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":" ","pages":"e72577"},"PeriodicalIF":2.0,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12260464/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144159198","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the Psychosocial Impact of Expressive Writing on Adults With Spinal Cord Injury: Qualitative Study.","authors":"Shelly M Xie, Molly McKenna, Kendall Veach, Sydney Williams, Mary Grace Jones, Elizabeth Vander Kamp, Salaam Green, Lauren Edwards, Kimberly Kirklin, Benjamin A Jones, Hon K Yuen","doi":"10.2196/71162","DOIUrl":"10.2196/71162","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Spinal cord injury (SCI) results in significant physical, emotional, and social consequences, often leading to profound grief and unresolved emotional burdens. While expressive writing has shown potential in facilitating emotional processing and in aiding coping after trauma, loss, and health-related adversity, its impact on individuals who have suddenly lost physical abilities due to SCI remains underexplored.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;The study aimed to examine the experiences of adults with SCI who participated in a 10-week web-based coach-guided expressive writing program to understand its impact using a qualitative research design with a phenomenological approach.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Participants were recruited through various outreach strategies based on a set of predefined criteria. A total of 50 individuals from 23 states expressed interest in the expressive writing program. Of those who met the eligibility criteria, 29 individuals completed the preprogram questionnaire. A total of 5 participants did not complete the postprogram assessments, including the exit interview. The program sessions were structured with reflective prompts designed to explore their emotions and life experiences related to their conditions. Qualitative data were collected through postprogram semistructured interviews and analyzed using thematic analysis to identify themes related to participants' experiences and program impact. The analysis was conducted without any preset theoretical framework of reference.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;A total of 24 adults with SCI, aged between 34 and 76 years (average age 51, SD 12 years), participated in the expressive writing program and the exit interview. Of these, 19 participants were White, and 17 were female. In total, 18 participants had sustained a traumatic SCI; of these, 10 had quadriplegia, 12 had paraplegia, and 2 had monoplegia. Qualitative analysis revealed three overarching themes: (1) supportive environment: the program provided participants a space that encouraged open reflection on past events and personal struggles, and the guidance of patient and empathetic coaches offered a sense of comfort, direction, and motivation. (2) Cathartic experience: the program helped them process complex emotions, reframe their perspectives, and cultivate a more positive outlook on life and their injury. Many participants, new to guided expressive writing, found the process therapeutic and transformative. (3) Acceptance of life: the cumulative impact of the sessions fostered self-compassion, forgiveness, empowerment, and self-advocacy. Participants reported reduced feelings of loneliness, a greater sense of community, and profound positive changes, expressing the desire to continue writing beyond the program.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;The sudden onset of lifelong disability due to SCI leads to profound physical, mental, and social challenges. The coach-guided expre","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e71162"},"PeriodicalIF":2.0,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12234395/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144528056","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing Diagnostic Accuracy of Ophthalmological Conditions With Complex Prompts in GPT-4: Comparative Analysis of Global and Low- and Middle-Income Country (LMIC)-Specific Pathologies.","authors":"Shona Alex Tapiwa M'gadzah, Andrew O'Malley","doi":"10.2196/64986","DOIUrl":"10.2196/64986","url":null,"abstract":"<p><strong>Background: </strong>The global incidence of blindness has continued to increase, despite the enactment of a Global Eye Health Action Plan by the World Health Assembly. This can be attributed, in part, to an aging population, but also to the limited diagnostic resources within low- and middle-income countries (LMICs). The advent of generative artificial intelligence (AI) within health care could pose a novel solution to combating the prevalence of blindness globally.</p><p><strong>Objective: </strong>The objectives of this study are to quantify the effect the addition of a complex prompt has on the diagnostic accuracy of a commercially available LLM, and to assess whether such LLMs are better or worse at diagnosing conditions that are more prevalent in LMICs.</p><p><strong>Methods: </strong>Ten clinical vignettes representing globally and LMIC-prevalent ophthalmological conditions were presented to GPT-4-0125-preview using simple and complex prompts. Diagnostic performance metrics, including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), were calculated. Statistical comparison between prompts was conducted using a chi-square test of independence.</p><p><strong>Results: </strong>The complex prompt achieved a higher diagnostic accuracy (90.1%) compared to the simple prompt (60.4%), with a statistically significant difference (χ2=428.86; P<.001). Sensitivity, specificity, PPV, and NPV were consistently improved for most conditions with the complex prompt. The simple prompt struggled with LMIC-prevalent conditions, diagnosing only 1 of 5 accurately, while the complex prompt successfully diagnosed 4 of 5.</p><p><strong>Conclusions: </strong>The study established that overall, the inclusion of a complex prompt positively affected the diagnostic accuracy of GPT-4-0125-preview, particularly for LMIC-prevalent conditions. This highlights the potential for LLMs, when appropriately tailored, to support clinicians in diverse health care settings. Future research should explore the generalizability of these findings across other models and specialties.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e64986"},"PeriodicalIF":2.0,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12261798/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144583914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Challenges and Mitigation Strategies in the Development and Feasibility Assessment of a Digital Mental Health Intervention for Depression (VMood): Mixed Methods Feasibility Study.","authors":"Leena W Chau, Jill K Murphy, Vu Cong Nguyen, Hai Nhu Tran, Harry Minas, Raymond W Lam, Kanna Hayashi, Thi Thanh Xuan Nguyen, Emanuel Krebs, John O'Neil","doi":"10.2196/68297","DOIUrl":"10.2196/68297","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Worldwide, the COVID-19 pandemic contributed to further gaps in mental health care, particularly in low- and middle-income countries such as Vietnam, where care is inaccessible for 90% of those who need it. There has subsequently been a considerable increase in the use of digital mental health interventions such as smartphone apps. Presently, the evidence for such interventions is limited, especially in cases in which the interventions have been adapted from evidence-based in-person formats. Implementation science aims to promote the incorporation of scientific findings into practice. A key determinant of implementation success is an intervention's usability. Hurdles to usability include an intervention being too confusing or time-intensive to use. Facilitators include incorporating a greater number of engagement features and integrating human support.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;The aim of this implementation science feasibility study was to describe the challenges and mitigation strategies used in the development, usability testing, and implementation of a digital depression intervention (VMood smartphone app) developed in Vietnam. VMood was adapted from an evidence-based in-person intervention originally developed in Canada that is grounded in principles of cognitive behavioral therapy with supportive coaching by a lay health or social services worker. The research team is currently testing the effectiveness and cost-effectiveness of VMood in a randomized controlled trial across 8 provinces in Vietnam informed by the results of this feasibility assessment.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;This mixed methods feasibility study was organized using an implementation outcome framework focused on acceptability, adoption, appropriateness, and feasibility. This study involved three data collection components: (1) usability testing (interviews and focus groups with app user and provider participants who tested VMood in 1 Vietnamese province), (2) app metrics (from the early phase of the randomized controlled trial in the same province but from different municipalities), and (3) discourse data (notes from various team meetings, communications, and reports on VMood's development and implementation). Qualitative data were analyzed using thematic content analysis. App use data were analyzed using basic descriptive statistics.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The findings of the 3 data components showed that there were seven main challenges: (1) challenges with recruitment and uptake of the app, (2) challenges with use and engagement, (3) screening challenges, (4) digital divide, (5) limitations to digital applications for mental health, (6) technological challenges, and (7) funding and policy constraints. Various solutions to help mitigate the challenges were used by the team.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;The findings contribute important evidence on the challenges to the development and feasibility assessment of ","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"9 ","pages":"e68297"},"PeriodicalIF":2.0,"publicationDate":"2025-06-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12260467/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144528057","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}