JMIR Formative Research最新文献

{"title":"Exploring Benefits of and Barriers to Patient Involvement Through Digital Tools in Psycho-Oncology: Qualitative Study Within the Reduct Trial.","authors":"Caterina Schug, Alexander Bäuerle, Johanna Graf, Jana Heinen, Julia Barbara Krakowczyk, Andrea Borho, Katharina Koller, Marietta Lieb, Regina Herold, Martin Teufel, Yesim Erim","doi":"10.2196/73147","DOIUrl":"https://doi.org/10.2196/73147","url":null,"abstract":"<p><strong>Background: </strong>Patient and public involvement is essential for developing patient-centered and acceptable eHealth interventions, yet little is known about how digital collaboration with patient representatives can best be implemented in psycho-oncological research.</p><p><strong>Objective: </strong>This study aimed to identify the benefits and barriers of digital collaboration in the development of an e-mental health application and provide recommendations to optimize digital collaboration with patient representatives in psycho-oncology research.</p><p><strong>Methods: </strong>Conducted from July to September 2023, this study involved digital semistructured interviews with 5 patient representatives from the Reduct trial, a multicenter randomized controlled trial to evaluate the efficacy of the web-based psycho-oncological training Make It. The interviews were analyzed using qualitative content analysis.</p><p><strong>Results: </strong>The findings highlighted multiple advantages of digital collaboration. These included significant reductions in travel costs and effort, personal acceptance and preference for digital methods, enhanced flexibility and accessibility, a reduced health burden, increased efficiency, and scalability. Conversely, several challenges were identified: social impacts or impediments due to less face-to-face interaction, technical difficulties, compromised effectiveness and quality of communication, diverse personal preferences and acceptance levels, organizational issues, cognitive demands, socioeconomic barriers, and safety concerns. The following recommendations to optimize digital collaboration were identified: maintaining regular communication and information exchange, valuing and committing to the collaboration, using diverse communication channels, ensuring comprehensible communication, integrating feedback, fostering openness and understanding, diligent documentation and recordkeeping, and providing targeted training and support for patient representatives.</p><p><strong>Conclusions: </strong>These findings confirm and specify previously known opportunities and challenges of digital collaboration, adding crucial insights for its implementation in psycho-oncological research. This research contributes to enhancing patient-centered approaches in psycho-oncology.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e73147"},"PeriodicalIF":2.0,"publicationDate":"2026-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13132482/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147815396","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":"The Comprehensive Assessment of Social Media Use: Development and Validation Study.","authors":"Nathan Job Lowry, Rebecca R Gebert, Zachary Rogers, Christine B Cha, Colleen M Jacobson","doi":"10.2196/87599","DOIUrl":"10.2196/87599","url":null,"abstract":"<p><strong>Background: </strong>Nearly all youth use the internet daily, with many maintaining several social media accounts. As increasing numbers of young people engage online and the ways we communicate fundamentally change, it is increasingly important to ask: how do these shifts influence youth mental health? To better understand how social media may affect mental health, researchers require validated tools that capture young people's heterogeneous experiences with social media. However, few available measures evaluate the full range of positive and negative behaviors associated with its use, limiting our ability to meaningfully advance interventions promoting online hygiene.</p><p><strong>Objective: </strong>This study aims to develop and validate the Comprehensive Assessment of Social Media Use (CASM). The CASM is a self-report survey measure that moves beyond simple duration or frequency of use and captures how young people engage with social media. Importantly, the CASM assesses both the positive and negative dimensions of social media engagement.</p><p><strong>Methods: </strong>Two studies are outlined in this paper. Study 1 outlines the process of item generation and exploratory factor analysis. Study 2 outlines confirmatory factor analysis and validity testing. Both studies were conducted online and enrolled a convenience sample of college-aged young adults. Study 1 enrolled 260 participants (mean age 19.73, SD 2.91; n=172, 66.2% female; n=164, 63.1% White; n=38, 14.6% lesbian, gay, bisexual, transgender/transsexual, and queer [LGBTQ]). Study 2 enrolled 508 participants (mean age 18.99, SD 1.17; n=323, 63.6% female; n=272, 53.5% White; n=58, 11.4% LGBTQ).</p><p><strong>Results: </strong>Exploratory and confirmatory factor analysis resulted in a 29-item CASM scale that assesses 7 distinct aspects of young adult social media use: self-branding, compulsive use, disruptive use, impulsive sharing, social engagement, induce negative emotions, and induce positive emotions. This model accounted for 61% of the variance in responses. The chi-squared test of model fit was significant (χ²356=941, P<.001; root mean square error of approximation=0.064; comparative fit index=0.855; Tucker-Lewis index=0.848; standardized root mean squared residual=0.060). Factor internal consistency reliability ranged from 0.699 to 0.817. Validity testing suggested moderate discriminant, convergent, and criterion validity.</p><p><strong>Conclusions: </strong>The CASM measures a broad range of social media behaviors, enabling researchers to more effectively examine associations between online engagement and mental health outcomes. We hope the CASM will help researchers better understand how young people interact with social media, and that this knowledge will inform the development of more targeted interventions promoting healthy online habits.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e87599"},"PeriodicalIF":2.0,"publicationDate":"2026-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13128055/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147772085","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":"An Internet of Things-Based Audio and Radio Connected System Supporting Older Adults With Physical and Cognitive Health Challenges: Qualitative Stakeholder-Informed Design Study.","authors":"Jia Liu, Xiaomeng Wang, Bianca Shieu, Deanna Dolores Garcia, Amy Vondenberger, Jingye Xu, Yuntong Zhang, Mahathir Monjur, Wei Wang, Shahriar Nirjon, Robert Svatek, Mitzi Gonzales, Neela Patel, Lixin Song","doi":"10.2196/76341","DOIUrl":"https://doi.org/10.2196/76341","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Older adults managing chronic illnesses, such as cancer and Alzheimer disease and related dementias (ADRD), often experience significant physical or cognitive impairments that hinder daily activities and increase caregiver burden. Smart Internet of Things (IoT) technologies offer promising solutions by enabling passive monitoring, timely reminders, and personalized support at home. However, these technologies must be carefully tailored to accommodate users' individualized needs and preferences.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This formative qualitative study aimed to explore stakeholder perspectives, including patients, caregivers, health care providers, and technical experts, on the use of smart home-based IoT systems to support chronic illness management. The goal was to inform the early development of the audio and radio connected (AURA) system, an IoT prototype integrating Wi-Fi sensing, wearable trackers, and voice-assistive features.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Semistructured interviews were conducted with 6 patients who underwent postostomy creation for colorectal or bladder cancer treatment and 5 patients with ADRD and their caregivers. Input from additional stakeholders, including 2 health care providers, 2 community health workers, and 2 computer scientists, was also included in the report. Stakeholders reviewed a demonstration video depicting the conceptual features of the AURA system. Interviews explored stakeholders' needs and preferences for using such systems. Thematic analysis was guided by the extended Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) framework, with 5 adapted constructs: performance expectancy, effort expectancy, social influence, facilitating conditions, and hedonic motivation and habit.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Stakeholders identified distinct yet complementary needs across populations. Patients with cancer emphasized physical health monitoring, integration with health care systems, and customization; ADRD stakeholders prioritized routine support, emotional engagement, and simplicity; caregivers and clinicians emerged as key influencers of adoption. Barriers included privacy concerns, technology literacy, and fatigue, while facilitators included perceived caregiving support, streamlined interfaces, and electronic health record integration. Patients with cancer focused on motivational cues for physical activity, while emotional engagement and habit were more prominent for ADRD users.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;Stakeholder insights underscore the importance of designing adaptable, user-centered IoT systems that reflect the varied capabilities and care needs of older adults with chronic illnesses. These findings informed the design of the AURA prototype and highlighted theoretical considerations for technology acceptance in health care. Future work will test AURA in real-world settings to evaluate usability, acceptability, and clinical relevance.&lt;/","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e76341"},"PeriodicalIF":2.0,"publicationDate":"2026-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147771982","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sexual Assault Prevention via Bystander Intervention Using Instagram Reels as a Communication Channel: Experimental Design Study.","authors":"Leticia Couto","doi":"10.2196/86512","DOIUrl":"https://doi.org/10.2196/86512","url":null,"abstract":"<p><strong>Background: </strong>Bystander intervention is one of the most commonly used methods to curb the sexual violence crisis on college campuses. Most universities conduct training among their student bodies to ensure students are familiar with the procedure. However, it is necessary to remind and repeat messages to audiences to underscore their importance and solidify that knowledge among populations.</p><p><strong>Objective: </strong>In this study, the author developed Instagram-type reel messages that consider multiple frameworks used to develop bystander training and programs, such as the social norms approach and bystander barriers.</p><p><strong>Methods: </strong>These messages were tested via a 4×1 experimental design study with college students at a large public university (N=157). The conditions were messages that emphasize norms (norm reinforcement condition, based on the social norms approach; n=39), messages that emphasize behavior (norm readjustment condition, based on bystander barriers; n=39), messages that underscore the discrepancy between perceived peer norm and actual behavior (combination condition, based on both the social norms approach and bystander barriers; n=39), and a control condition (n=40).</p><p><strong>Results: </strong>ANOVA analysis revealed that the norms reinforcement condition and the combination condition seemed to have the greatest impact on participants' perceived norms and behavioral intentions. In particular, when assessing norms as the dependent variable, the reinforcement group was significantly different from the readjustment group (Δmean 0.682, SD 0.160; P<.001), the combination group (Δmean 0.675, SD 0.160; P<.001), and the control group (Δmean 0.432, SD 0.159; P=.04). For behavioral intentions to engage in bystander intervention, the reinforcement group was significantly different from the control group (Δmean 0.217, SD 0.073; P=.02), and the combination group was significantly different from the control group as well (Δmean 0.221, SD 0.073; P=.02).</p><p><strong>Conclusions: </strong>The findings indicate that these messages could work in conjunction with training as a way to underscore the importance of bystander intervention behavior. It also highlights the role that Instagram reels can play in the prevention of sexual violence on college campuses.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e86512"},"PeriodicalIF":2.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13124031/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147772044","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":"Restoring Engagement in Digital Self-Control Tools Using Nudge Reconfiguration Prompts: Quasi-Experimental Study.","authors":"Awen Kidel Peña-Albert, Sandy Ingram, Yasser Khazaal, Léo Litrico, Juan Carlos Farah, Denis Gillet","doi":"10.2196/85349","DOIUrl":"https://doi.org/10.2196/85349","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Digital self-control tools (DSCTs) have emerged as technological interventions to address excessive smartphone usage and promote digital well-being. However, these tools face persistent challenges with user attrition and sustained engagement, compromising their long-term effectiveness. Current literature lacks an understanding of how observable behavioral indicators, as opposed to self-reported measures, are associated with user engagement and readiness to change in DSCTs.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study addresses three research questions (RQs): (RQ1) whether prompting passive DSCT users to reconfigure nudges increases subsequent user-nudge interaction, (RQ2) how engagement evolves over time and what behavioral divergence emerges between accepting and rejecting users, and (RQ3) whether observable in-app behavioral indicators are more strongly associated with intervention acceptance than traditional self-reported measures.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;We conducted a quasi-experimental study (N=252) targeting users who had disabled nudges. Participants were randomly assigned to receive a prompt to reconfigure their nudge settings during daily check-ins (n=138, experimental group) or to a control condition (n=114, no intervention). The experimental group was further classified into acceptance and rejection subgroups based on their response to the intervention. Data collection included DSCT configuration logs, usage-triggered nudge logs, and self-reported questionnaire responses. We analyzed user-nudge interaction ratios using difference-in-differences with permutation tests (RQ1) and nudge configuration parameters and manual app blocking using independent-samples t tests with Cohen d (RQ2) and compared behavioral indicators against self-reported measures using t tests and chi-square tests (RQ3).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Of the experimental participants, 46% (63/138) accepted the nudge reconfiguration prompt. Post intervention, the acceptance subgroup's 7-day average user-nudge interaction ratio increased from 29.7% to 58.5% (peak of 65% on day 1), a significant increase even after controlling for the temporal decline observed in the control group (difference-in-differences=+36.3 percentage points, P&lt;.001). The rejection subgroup's decline was not significantly different from the control group's decline (P=.82). The acceptance subgroup showed preexisting behavioral indicators of higher readiness to change, including 21.53% shorter consecutive usage thresholds (P=.03) compared to the rejection subgroup, with a directionally consistent but nonsignificant difference in cooldown length (+20.56%). Behavioral divergence in consecutive usage thresholds widened post intervention, with Cohen d increasing from -0.47 to -0.67 (P=.002). Acceptance subgroup participants demonstrated a significantly lower tendency to select leisure-oriented daily goals (15.6% vs 26.2%; chi-square P=.001, Cramer V=0.13). Self-r","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e85349"},"PeriodicalIF":2.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13123756/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147771997","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 Formative Evaluation of Parental Perceptions Related to Acceptability, Appropriateness, Feasibility, and Reported Use of an e-Learning Resource Targeting Diet in the First 1000 Days: Survey Study.","authors":"Natalie Garzon Osorio, Frøydis Nordgård Vik, Elisabet Rudjord Hillesund, Sissel Heidi Helland, Penelope Love, Marianne Hope Abel, Harry Rutter, Andrew Keith Wills, Christine Helle, Tormod Bjørkkjær, Nina Cecilie Øverby, Anine Christine Medin","doi":"10.2196/84277","DOIUrl":"https://doi.org/10.2196/84277","url":null,"abstract":"<p><strong>Background: </strong>In October 2022, the Nutrition Now (NN) e-learning resource was implemented within Maternal and Child Healthcare centers and Early Childhood Education and Care centers of a southern Norwegian municipality. The e-learning resource targets expectant parents, parents of children aged 0-2 years, and Early Childhood Education and Care staff, aiming to promote healthy dietary behaviors during the first 1000 days of life.</p><p><strong>Objective: </strong>This study aimed to explore parental perceptions related to the acceptability, appropriateness, feasibility, and reported use of the NN e-learning resource among parents.</p><p><strong>Methods: </strong>From October 2022 to May 2023, expecting parents and parents of children aged 0-2 years were recruited from 2 Norwegian municipalities, one intervention group receiving access to the NN e-learning resource, and one control. Participants in the intervention group received a web-based follow-up questionnaire 7 months after gaining access to the NN e-learning resource. Data were analyzed using descriptive statistics.</p><p><strong>Results: </strong>Of the 179 participants in the NN study intervention group, 48 completed the web-based follow-up questionnaire administered 7 months after enrollment. Parents rated the e-learning resource positively on items assessing whether they liked and appreciated the resource, perceived it as an appropriate source of information, and found it doable and easy to use. Most respondents reported visiting the resource (38/48, 79%), although only 21% (10/48) reported frequent visits. Less than half of the participants answering the web-based follow-up questionnaire reported having watched the theme films (20/48, 42%), the recipe films (17/48, 35%), or making food using recipes provided in the e-learning resource (20/48, 42%).</p><p><strong>Conclusions: </strong>Parents rated the NN e-learning resource positively but reported limited use. These findings point to the need for strategies that enhance engagement with self-guided digital interventions among expectant parents and parents of young children. Future efforts should focus on identifying how to maximize potential adoption of the e-learning resource and evaluate its impact to promote healthy dietary behaviors during the first 1000 days of life.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e84277"},"PeriodicalIF":2.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13123635/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147772040","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":"Benefit of the N-of-1 Approach Versus Aggregate Analysis in Tracking Individual Trajectories During Pregnancy: Comparison of Longitudinal Wearable Observational Studies.","authors":"Tina Behrouzi, Jennifer Yu, Robin Yang, Adrien Boch, Anna Goldenberg, Sarah M Goodday, Stephen H Friend","doi":"10.2196/86203","DOIUrl":"https://doi.org/10.2196/86203","url":null,"abstract":"<p><strong>Background: </strong>Personal digital health technologies (DHTs) enable real-time monitoring of physiological metrics and behavioral data, including heart rate variability (HRV), supporting analysis of pregnancy-related conditions and personalized care throughout the perinatal period. While recent studies demonstrate the utility of personal DHTs in tracking pregnancy-related symptoms, they often rely on aggregate statistical methods that overlook individual variability.</p><p><strong>Objective: </strong>This study aims to compare aggregate and individual-level analyses of DHT data for pregnancy-related conditions, using the comprehensive BUMP (Better Understanding the Metamorphosis of Pregnancy) dataset to highlight the importance of individual variability and data heterogeneity.</p><p><strong>Methods: </strong>We analyzed wearable and self-reported data from 256 participants enrolled in the BUMP study (January 2021 to May 2022), including HRV, sleep, and fatigue measured via Oura Rings and smartphone surveys. Individual-level (N-of-1) trajectories were evaluated and compared with aggregate results to uncover personal and collective trends. A statistical method was developed to assess the influence of adverse events and severe symptoms, while case studies explored confounding and modifying factors underlying heterogeneity. Comprehensive statistical analysis included the coefficient of determination, Kolmogorov-Smirnov tests, likelihood ratio tests, and Welch t tests, with interindividual variability flagged based on high-variability thresholds.</p><p><strong>Results: </strong>Substantial interindividual variability was observed across all features. Only 4.76% (12/256) of participants exhibited an HRV inflection at the aggregate week-33 inflection point, with a coefficient of variation of 14.24%. The median value of the gestational week in individual fatigue troughs was 23 (IQR 8; range 8-38) weeks, differing from aggregate estimates. Distributional comparisons showed no statistically significant differences in individual-level model fit (R²) by pregnancy complications or age (P values ranging from .06 to .99 across all model fit comparisons). Case studies further highlighted both intraindividual and interindividual differences, emphasizing the importance of considering external factors, such as adverse events and severe symptoms.</p><p><strong>Conclusions: </strong>Our findings show that aggregate wearable data often fail to generalize across populations, oversimplifying pregnancy-related physiological and subjective changes. This simplification can obscure individual trajectories, leading to generalized insights that may not reflect many pregnant women's experiences. Our results highlight the impact of heterogeneity on pregnancy outcomes, emphasizing the need to move beyond one-size-fits-all models and leverage DHT for personalized care.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e86203"},"PeriodicalIF":2.0,"publicationDate":"2026-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13124032/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147771966","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":"RoBuster-Corpus Annotated With Risk of Bias Text Spans in Randomized Controlled Trials in Physiotherapy and Rehabilitation: Corpus Development and Annotation Study.","authors":"Anjani Dhrangadhariya, Roger Hilfiker, Karl Martin Sattelmayer, Nona Naderi, Katia Giacomino, Rahel Caliesch, Julian Higgins, Stéphane Marchand-Maillet, Henning Müller","doi":"10.2196/55127","DOIUrl":"https://doi.org/10.2196/55127","url":null,"abstract":"<p><strong>Background: </strong>Risk of bias (RoB) assessment of randomized clinical trials (RCTs) is vital to answering systematic review questions accurately. Manual RoB assessment for hundreds of RCTs is a cognitively demanding and lengthy process. Automation has the potential to assist reviewers in rapidly identifying text descriptions in RCTs that indicate potential risks of bias. However, no RoB text span annotated corpus could be used to fine-tune or evaluate large language models (LLMs), and there are no established guidelines for annotating the RoB spans in RCTs.</p><p><strong>Objective: </strong>The revised Cochrane RoB 2 test (RoB 2) tool provides comprehensive guidelines for RoB assessment; however, due to the inherent subjectivity of this tool, it cannot be directly used as RoB annotation guidelines. The study aimed to develop precise RoB text span annotation instructions that could address this subjectivity and thus aid the corpus annotation.</p><p><strong>Methods: </strong>We leveraged RoB 2 guidelines to develop visual instructional placards that serve as annotation guidelines for RoB spans and risk judgments. Expert annotators used these visual placards to annotate a dataset named RoBuster, consisting of 41 full-text RCTs from the domains of physiotherapy and rehabilitation. We report interannotator agreement (IAA) between 2 annotators for text span annotations before and after applying visual instructions on a subset (n=9) of RoBuster. We also provide IAA on bias risk judgments using Cohen κ. Moreover, we used a portion of RoBuster (n=10) to evaluate an LLM using a straightforward evaluation framework. This evaluation aimed to gauge the performance of an LLM (here GPT 3.5) in the challenging task of RoB span extraction and demonstrate the utility of this corpus using a straightforward framework.</p><p><strong>Results: </strong>We present a corpus of 41 RCTs with fine-grained text span annotations comprising more than 28,427 tokens belonging to 22 RoB classes. The IAA at the text span level calculated using the F1 measure varies from 0% to 90%, while Cohen κ for risk judgments ranges between -0.235 and 1.0. Using visual instructions for annotation increases the IAA by more than 17 percentage points. LLM (GPT-3.5) shows promising but varied observed agreements with the expert annotation across the different bias questions.</p><p><strong>Conclusions: </strong>Despite having comprehensive bias assessment guidelines and visual instructional placards, RoB annotation remains a complex task. Using visual placards for bias assessment and annotation enhances IAA compared to cases where visual placards are absent; however, text annotation remains challenging for the subjective questions and the questions for which annotation data are unavailable in RCTs. Similarly, while GPT-3.5 demonstrates effectiveness, its accuracy diminishes with more subjective RoB questions and low information availability.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e55127"},"PeriodicalIF":2.0,"publicationDate":"2026-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13120535/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147771957","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":"Multimodal Depression Detection Through Conversational Interactions with an Emotion-Aware Social Robot: Pilot Study.","authors":"Pu-Yu Liao, Yu-Quan Su, Xiaobei Qian, Yu-Ling Chang, Yun-Hsiang Lee, Li-Chen Fu","doi":"10.2196/84110","DOIUrl":"https://doi.org/10.2196/84110","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Depression affects more than 300 million people worldwide and is a leading contributor to the global disease burden. Traditional diagnostic methods, such as structured clinical interviews, are reliable but impractical for frequent or large-scale screening. Self-report tools like the Patient Health Questionnaire-8 (PHQ-8) require disclosure and clinician oversight, limiting accessibility. Recent artificial intelligence-based approaches leverage multimodal behavioral cues (linguistic, acoustic, and visual) for automated depression detection but remain constrained by limited adaptability, scarce annotated data, weak emotional expression in real-world settings, and the high computational cost of deployment of socially assistive robots (SARs).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Objective: &lt;/strong&gt;This study introduces Depression Social Assistant Robot (DEPRESAR)-Fusion, a lightweight multimodal depression detection framework designed for natural interactions with emotion-aware SARs. The objective of this study was to enhance detection accuracy in everyday conversations while addressing the challenges of data scarcity, weak emotional cues, and computational efficiency.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;DEPRESAR-Fusion integrates acoustic, linguistic, and visual features with an emotion-aware response module powered by large language models to adapt conversational strategies dynamically. To stimulate richer emotional expression, participants were exposed to emotionally evocative videos before SAR interactions. To overcome data scarcity, we augmented training with (1) public depression-related social media corpora and (2) synthetic samples generated via large language models. The proposed multimodal fusion architecture was evaluated on benchmark clinical datasets for both binary depression classification and PHQ-8 regression tasks. Performance was compared against prior multimodal baselines using root mean square error, mean absolute error, and standard classification metrics.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Participants who viewed emotional stimuli before interacting with SARs exhibited significantly higher emotional expressiveness, leading to improved model performance. Regression tasks showed lower root mean square error and mean absolute error, while classification tasks achieved significantly higher accuracy than the nonstimulus condition. DEPRESAR-Fusion outperformed prior multimodal baselines across multiple benchmark datasets, achieving state-of-the-art performance in both binary classification and PHQ-8 regression. The system maintained a lightweight architecture suitable for real-time deployment on SARs.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;DEPRESAR-Fusion demonstrates that integrating emotion induction, data augmentation, and lightweight multimodal fusion can enable accurate and scalable depression detection in naturalistic SAR interactions. By bridging the gap between structured clinical assessments and everyday conversations, this a","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e84110"},"PeriodicalIF":2.0,"publicationDate":"2026-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13120753/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147771989","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":"The Power of Multimodality in Multimodal Large Language Models, Unimodal ChatGPT 5.0, and Human Clinical Experts on a Wound Care Certification Examination: Cross-Sectional Comparative Study.","authors":"Mete Ucdal, Melike Elif Celik, Guliz Evik, Saniye Beyza Kuru, Saadet Ozer, Sultan Gungor","doi":"10.2196/88618","DOIUrl":"https://doi.org/10.2196/88618","url":null,"abstract":"<p><strong>Background: </strong>Multimodal large language models (MLLMs) capable of integrating visual and textual information represent a promising advancement for clinical applications requiring image interpretation. Wound care assessment, which demands simultaneous analysis of wound photographs and clinical data, provides an ideal domain to evaluate multimodal vs unimodal artificial intelligence capabilities against human expertise.</p><p><strong>Objective: </strong>This study aims to compare the performance of MLLMs, unimodal ChatGPT 5.0, and human clinical experts on a standardized wound care certification examination.</p><p><strong>Methods: </strong>This cross-sectional comparative study evaluated 3 participant groups on a 25-question wound care certification examination spanning 4 clinical domains (Diagnosis, Treatment, Complication Management, and Wound Subtype Knowledge). Participants included 3 MLLMs (Med-PaLM 2, LLaVA-Med, and BioGPT), 1 unimodal large language model (ChatGPT 5.0), and 4 human clinical experts (general surgeon, wound care nurse, and 2 internal medicine physicians). Statistical analyses included one-way ANOVA with Tukey post hoc tests and domain-specific Kruskal-Wallis comparisons.</p><p><strong>Results: </strong>Human experts achieved the highest accuracy (mean 86%, SD 9.1%), followed by MLLMs (mean 78.7%, SD 12.2%), while ChatGPT 5.0 achieved 64% accuracy, failing the 70% certification threshold. Significant overall group differences were observed (F2,5=8.42, P=.02, η²=0.74). MLLMs significantly outperformed ChatGPT 5.0 (difference=14.7 percentage points, P=.03, Cohen d=1.38), with the multimodal advantage most pronounced in visually dependent domains: Diagnosis (81% vs 43%, P=.008) and Complication Management (72% vs 50%, P=.03). No multimodal advantage was observed for text-based Wound Subtype Knowledge (both 67%). Med-PaLM 2 achieved 92% accuracy, matching that of the wound care nurse, while the general surgeon achieved the highest overall performance (96%).</p><p><strong>Conclusions: </strong>MLLMs demonstrate significant performance advantages over unimodal artificial intelligence in wound care assessment, particularly for visually dependent clinical tasks. While human experts with specialized wound care experience maintain overall superiority, the point estimate of the top-performing MLLM (Med-PaLM 2, 92%) fell within the observed range of human scores; however, the underpowered comparison (power=0.52) and wide CIs preclude definitive conclusions regarding noninferiority or equivalence to human experts. These findings support the potential role of MLLMs as clinical decision-support tools, warranting further adequately powered validation studies.</p>","PeriodicalId":14841,"journal":{"name":"JMIR Formative Research","volume":"10 ","pages":"e88618"},"PeriodicalIF":2.0,"publicationDate":"2026-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13120536/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147772067","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}