PLOS Digital Health

Background: Clinical decision support systems (CDSS) are increasingly embedded in national telemedicine services, however how different stakeholders who use them experience these tools, particularly in low- and middle-income countries, is less explored. In the assisted model of India's eSanjeevani service, spoke health workers enter structured symptoms and hub clinicians review a symptom summary and an artificial intelligence (AI)-generated differential diagnosis. Objective: To describe role-specific experiences of eSanjeevani's CDSS functions and to estimate how acceptance coexists with specific reported difficulties within each role. Methods: We analyzed a program feedback web survey completed between 11 July and 29 July, 2025 by hub clinicians and spoke providers. Prespecified item profiles covered review of summaries and suggested diagnoses, the perceived relation of suggestions to the clinician's own diagnosis, ease of use and intention to use (hub), and symptom-capture utility, question quantity, question relevance, summary presentation and satisfaction (spoke). Within-role contrasts were reported as absolute differences in acceptance with Newcombe 95% CIs. Short open comments underwent content analysis. Results: The cohort comprised 190 hub and 662 spoke respondents from 13 and 15 states (21.4% and 19.3% of eligible providers). Among hub respondents, 76.8% regularly reviewed the symptom summary and 58.9% the suggested diagnoses; 72.6% judged suggestions matched (32.6%) or appropriate but different (40.0%), 7.9% mostly inappropriate, and 11.1% reported that suggestions were mostly not generated. Overall, 85.3% intended to keep using the tools. Affirmative intention was 98.4% when suggestions matched, 66.7% when they were mostly inappropriate (difference vs matched or appropriate -23.2 percentage points, 95% CI -48.5 to -4.0) and 61.9% when mostly absent (-25.7, 95% CI -47.1 to -7.3); technical issues, reported by 38.4%, made little difference (-2.8, 95% CI -14.1 to 7.3). Among spoke respondents, 63.4% rated question quantity appropriate, with miscalibration in both directions (15.6% too many; 16.6% too few); 89.9% found the questions relevant and 91.8% found the summary clear. Satisfaction was 75.3% and did not differ materially between those reporting too many questions and those reporting an appropriate number (-2.7, 95% CI -12.8 to 6.0). Comments located spoke difficulties in question burden and missing complaints, and hub difficulties in freezing and in the scope and availability of suggestions. Estimates were stable across cohort, ambiguity and state-exclusion sensitivity analyses. Conclusions: Acceptance of eSanjeevani's CDSS was high in both roles. For hub clinicians it tracked the perceived quality and availability of AI suggestions rather than technical friction; spoke workers accepted the question burden, but its two-directional miscalibration points toward complaint-specific, adaptive questioning. These provider reports do not measure diagnostic accuracy. They identify log-based and audit-based indicators that the next evaluation cycle should measure directly.

Background: Digital mental health applications offer scalable pathways to expand mental healthcare access in low and middle-income countries, yet limited evidence exists on factors predicting willingness to use such applications in South Asia. This study examined associations between digital mental health literacy (DMHL), Technology Acceptance Model (TAM) constructs, mental health symptoms, and willingness to use smartphone-based mental health applications among adults in Bangladesh. Methods: Cross-sectional study conducted from February to June 2026 across urban (Dhaka), semi-urban (Chattogram), and rural (Sylhet) divisions of Bangladesh. Adults aged 18-60 years attending primary care facilities with smartphone access (n = 400) completed a face-to-face interview in Bangla assessing sociodemographic, depression (PHQ-9), anxiety (GAD-7), DMHL (6-item scale, = .762), TAM perceived usefulness (5-item, = .937) and ease of use (5-item, = .924), and willingness to use mental health applications (5-item scale, = .946). Hierarchical binary logistic regression examined associations with willingness (dichotomized at mean [≥]3.5). Results: Greater DMHL was associated with higher willingness (adjusted odds ratio [aOR] = 1.24, 95% CI: 1.11-1.38). Both perceived usefulness (aOR = 1.29, 95% CI: 1.18-1.41) and perceived ease of use (aOR = 1.17, 95% CI: 1.07-1.28) independently predicted willingness. Anxiety symptoms were positively associated (aOR = 1.08 per 1-point increase, 95% CI: 1.01-1.15), whereas depression symptoms, age, and sex were not independently significant after adjustment for other variables. Conclusions: Digital literacy, perceived utility of technology, and symptom-specific phenomenology but not demographics alone are associated with willingness to use digital mental health applications. Building DMHL, optimizing the usefulness and usability of the app, and overcoming implementation barriers should all be given priority.

Background: Kano State has one of the highest concentrations of zero-dose (ZD) children in Nigeria, challenging equitable immunization coverage. Digital engagement tools meant to strengthen timely data to address such gaps often underperform due to limited adaptability to user needs. Learning-by-doing (LxD), which brings together human-centered design and adaptive management, offers more context-responsive solutions. Intervention and Implementation Approach: MCH-Track, a digital platform for tracking and referral for childhood vaccinations was used by 56 female community mobilizers in two local government areas (LGAs) out of the 44 LGAs in Kano State, Nigeria. We conducted a program case study from: 1) August 2024-April 2025 (initial implementation); and 2) adapted implementation (May 2025-November 2025), where LxD was incorporated to address emergent challenges. LxD included structured shadowing, monthly interviews with users, after-action reviews, and change log to track adaptations alongside routine services. Performance was monitored using four indicators: feedback actioned, response time-to-user feedback, ease-of-use score, and mobilizer performance in tracking ZD children. Findings: After incorporating LxD, 17 issues were identified and led to 12 prioritized adaptations. All logged issues received responses, with median response time declining to approximately two weeks, while community mobilizers reported higher usability of the digital platform (e.g., mean ease-of-use score of 4.1 out of 5). Enhanced root-cause analysis enabled trend tracking and clearer escalation pathways for systemic issues like micro-stockouts at immunization points. Empathy tool records revealed mobilizer struggles with platform. This was immediately logged, prompting user experience redesign into a seamless, single-step process from registration-to-referral. During the LxD implementation period, an average of 33 ZD children per month were identified via MCH-Track, a substantial increase from the pre-LxD period (an average of 11 ZD children identified per month). Average referral-to-access rates also improved during the LxD period, rising from 44% of identified and referred ZD children accessing immunization services pre-LxD to 70% after incorporating LxD practices. Conclusion: Embedding LxD within MCH-Track strengthened co-creation, streamlined workflows, and improved community mobilizer performance, demonstrating the value of adaptive management in reaching ZD children.

Background: Antimicrobial resistance (AMR) is a growing global health threat, and India's high antibiotic consumption makes public knowledge, attitudes, and practices (KAP) a critical determinant of stewardship outcomes. Existing Indian KAP evidence is drawn from small, localised cohorts, leaving a gap in nationwide, digitally-sourced data. This study aimed to describe antibiotic-related knowledge, attitudes and self-reported practices among users of a digital healthcare platform in India, examine demographic and regional variation, and identify potential areas for future stewardship interventions. Methods: A cross-sectional survey was conducted among adult users of the Tata 1mg platform (December 2025-February 2026), with participants invited via an in-app banner. Knowledge and attitude were each scored on a 0-4 scale (4 items); practice was scored as a percentage of applicable items answered favourably, with scoring adjusted for branching logic based on antibiotic-access route. Of 10,726 respondents who started the survey, 4,883 completed it (45.5%). City of residence was mapped to one of six regions (India's five statutory Zonal Council zones plus a North-Eastern region) for 4,187 respondents. Associations with demographic characteristics were assessed using multivariable linear regression, and regional differences were assessed using one-way ANOVA. A parallel survey of 256 physicians was conducted to enable patient-provider comparison. Results: Mean scores were: knowledge 2.28 (SD 1.19), attitude 0.91 (SD 1.14), practice 44.9% (SD 33.4); attitude, not knowledge, was the weakest domain. Older age was associated with higher scores across all domains (practice: B=12.52, p<0.001 for age [&ge;]65 vs. 18-34); city tier was independently associated with lower knowledge and attitude (Tier 3 vs. Tier 1: p=0.010 and p<0.001, respectively). Regional differences were non-significant for knowledge and practice but significant for attitude (F=2.56, p=0.026, ^2=0.003), driven by lower scores in the North-Eastern region. In the parallel physician survey, 55.5% of respondents estimated that more than half of their patients self-medicated before consultation. Conclusion: In this platform-based sample, attitudes were the weakest domain, not knowledge, and 42.0% said their last antibiotic did not come on a doctor's recommendation. The findings may inform digital stewardship touchpoints, such as point-of-purchase guidance and prescription verification, in line with India's National Action Plan on Antimicrobial Resistance 2.0. As a cross-sectional survey of platform users, it cannot establish the relative contributions of knowledge, beliefs and access pathways to antibiotic use.

Objective To evaluate the information quality, reliability, readability, and content deficits of ChatGPT, DeepSeek, and Doubao in answering frequently asked questions about bee stings. Methods Twenty-five high-frequency patient questions were selected based on Google Trends, domestic health platforms, and allergy guidelines, and submitted to the three LLMs, yielding 75 responses. Two emergency physicians independently assessed quality using GQS, mDISCERN, QAMAI, EQIP, and PEMAT-U/A. Readability was evaluated using six indices (ARI, FRES, GFI, FKGL, CLI, SMOG). Content deficits were classified qualitatively. Twenty emergency department patients evaluated the comprehensibility and credibility of AI responses. Results Significant differences were found among the three models on all quality metrics (all P < 0.01). DeepSeek achieved the highest mDISCERN score (median 21.00); ChatGPT led on QAMAI (25.00) and EQIP (75.00%); Doubao ranked lowest on all quality indicators but highest on PEMAT-U (73.00%). No model met the sixth-grade reading level standard. Eleven content deficits were identified: outdated medical advice (45.45%), omissions (36.36%), and hallucinations (18.18%). Doubao received the highest patient preference (45.0%). Conclusion The three LLMs exhibited distinct strengths: DeepSeek in source reliability, ChatGPT in overall quality, and Doubao in understandability. All models had excessively high readability levels, and outdated medical advice was the predominant content deficit. Objective To evaluate the information quality, reliability, readability, and content deficits of ChatGPT, DeepSeek, and Doubao in answering frequently asked questions about bee stings. Methods Twenty-five high-frequency patient questions were selected based on Google Trends, domestic health platforms, and allergy guidelines, and submitted to the three LLMs, yielding 75 responses. Two emergency physicians independently assessed quality using GQS, mDISCERN, QAMAI, EQIP, and PEMAT-U/A. Readability was evaluated using six indices (ARI, FRES, GFI, FKGL, CLI, SMOG). Content deficits were classified qualitatively. Twenty emergency department patients evaluated the comprehensibility and credibility of AI responses. Results Significant differences were found among the three models on all quality metrics (all P < 0.01). DeepSeek achieved the highest mDISCERN score (median 21.00); ChatGPT led on QAMAI (25.00) and EQIP (75.00%); Doubao ranked lowest on all quality indicators but highest on PEMAT-U (73.00%). No model met the sixth-grade reading level standard. Eleven content deficits were identified: outdated medical advice (45.45%), omissions (36.36%), and hallucinations (18.18%). Doubao received the highest patient preference (45.0%). Conclusion The three LLMs exhibited distinct strengths: DeepSeek in source reliability, ChatGPT in overall quality, and Doubao in understandability. All models had excessively high readability levels, and outdated medical advice was the predominant content deficit.

Background: Genetic conditions are common in congenital heart disease (CHD) and affect care, but are often missed or diagnosed late. We asked whether routine clinical data can shortlist the most likely genetic conditions. Methods: Retrospective single-center cohort of children with CHD and an echocardiogram who had either a confirmed genetic diagnosis or negative cytogenetic testing and CHD gene sequencing. Echocardiogram images and serial medical record data were added as model inputs in the order they become clinically available. Class balanced logistic regression (LR) and a multimodal deep learning fusion network were fitted to the same inputs at each step. The primary metric was balanced top-3 accuracy (correct diagnosis among the top three, averaged across conditions), with bootstrap intervals. Results: The cohort included 1,720 children (988 with common CHD-associated genetic conditions, 231 rare diagnoses, 501 negative testing). Using only inputs generated at the echocardiogram visit, LR reached a balanced top-3 accuracy 0.758 (chance 0.429). Serial records, radiology notes and ICD-10 codes, raised this to 0.809. Multimodal fusion performed similarly at each stage. With a genetic condition withheld from training, the fusion embedding placed affected children nearest to others sharing a gene group in 0.37 of cases versus 0.10 without the embedding (+0.268 [95% CI +0.242, +0.295]). Conclusions: Routine clinical data available at first echocardiogram can help identify likely genetic diagnoses, and later coded inputs ceiling. Multimodal fusion added no accuracy over LR, but places children sharing a genetic cause near one another, making nearest-neighbor retrieval informative for undiagnosed children.

Background: Passively collected smartphone and wearable data, combined with genomic information, may enable scalable early identification of youth who are at risk for substance use disorder. However, heterogeneous data collected in naturalistic settings limit researcher control over data quality, requiring substantial preprocessing. Data cleaning and feature engineering decisions can meaningfully influence analytic outcomes yet are often minimally described or omitted in published work. Objectives: We provide rationales and recommendations for common preprocessing challenges in passive sensing research, including establishing inclusion criteria when "outliers" may reflect true behavioral variability, analyzing differences across smartphone operating systems, handling observations collected outside protocol windows, and engineering features for machine learning applications. Methods: We illustrate principled approaches to these challenges using smartphone and Fitbit data from the Adolescent Brain Cognitive Development Study (ABCD). The sample was drawn from the available half of the year-4 follow-up (data release 5.1; N = 4,754; ages 13-14). Results: Extreme outlying observations were concentrated among low-engagement participants with few observed days. Smartphone usage among Apple users appeared under-recorded relative to Android users when compared with self-report measures. Fitbit-derived behavior differed systematically between observations collected within versus outside protocol windows. We also identified potential errors in the study's default preprocessing pipeline. Conclusion: We recommend prioritizing person-level inclusion criteria that preserve meaningful behavioral variability, examining temporal distributions rather than assuming observations are exchangeable, and thoroughly investigating and reporting data quality issues. Although illustrated using the ABCD Study, these principles broadly apply to research using smartphones, wearables, and other digital data sources.

This article critically considers how exhibition, as a modality for research evidence delivery, can be mobilised in the largely positivist and biomedical world of elite sport for the purpose of system change. Our interdisciplinary team composed of a health geographer and rehabilitation scientist together with non-academic specialists in athlete health and performance innovation, conceptualised an exhibition-as-intervention to reframe gender disparities in sports injury from being biologically induced to also socially influenced. Here, we reflect on our experience creating and implementing More Than Medals: Gendered Environments of Sports Injury Uncovered - a research-based, online, multimedia exhibition exploring the gendered environments of sports injury through the first-hand experiences of female athletes. Using evaluation interviews with both sports staff and athletes, we assess the impacts of our exhibition-as-intervention. We conclude that exhibition offers a novel way to bridge social and biomedical paradigms, innovate on evidence delivery in elite sport, and close the gap between social sciences research and practice.

Youth are increasingly engaging with generative AI beyond its use as a productivity tool, with a growing share turning to AI for social interaction and emotional support. This leads to unknown risks for their well-being and development. Online community discussions reveal that youth already perceive risks that caregivers have not recognized.

Introduction: Heart failure care requires coordination across hospital and community settings, yet information is often fragmented across electronic medical records and clinical systems. Clinical dashboards can bring together and display information to support care delivery and service management; however, existing dashboards have generally focused on specific measures, interventions and monitoring pathways. This study aimed to co-design, iteratively develop and user-test an integrated heart failure dashboard that links patient-level clinical decision-making with service-level management. Methods: A human-centred design approach comprising needs identification, collaborative ideation, iterative prototype development and end-user testing was applied across two complementary operational and clinical dashboard streams. Thirty-six clinicians, health service managers, data and implementation scientists, and consumers from metropolitan and regional services participated. Results: For operational decision-making, participants prioritised real-time visibility of patients across heart failure services, patient trajectories, service-performance information, and identification of variation in guideline-directed care and outcomes. Clinical priorities included rapid synthesis of longitudinal information, optimisation of guideline-directed medical therapy, continuity across care settings, and clinical workload prioritisation. These requirements informed the dashboard prototypes. Many prioritised information elements were incompletely represented in structured data and distributed across disconnected systems and structured and unstructured clinical data sources. Conclusion: Human-centred co-design identified complementary patient- and service-level information needs and translated them into linked dashboard prototypes. The proposed dashboard brings together current clinical status and longitudinal heart failure care history at the patient-level alongside service-level patterns. Implementation is required to evaluate whether these linked views improve care processes and patient outcomes.

Pain assessment remains challenging in clinical practice. Automated pain assessment using artificial intelligence (AI) and facial expression analysis has emerged as a potential approach to support more objective and continuous pain monitoring, particularly for patients who are unable to communicate effectively. However, successful implementation depends on clinician acceptance. We conducted a cross-sectional survey of physicians, advanced practice providers (APPs), and nurses at a tertiary academic medical center to evaluate current pain assessment practices and perceptions of a facial-analysis-based pain assessment system. Eighty-seven eligible clinicians participated. For current assessment practices, clinicians commonly relied on observational indicators of pain, including vital signs (88.5%), physical examination findings (85.1%), and facial expressions (79.3%). Attitudes toward automated pain assessment were generally neutral to positive. Half of respondents agreed that such a system would be useful for monitoring patients over time (50.6%), and many perceived value for nonverbal patients (86.5%), patients with dementia or cognitive impairment (78.4%), and patients with language barriers (62.2%). Most clinicians reported requiring high levels of accuracy before routine use, with nearly 60% requiring at least 90% accuracy. Key concerns included alert fatigue and unequal performance across patient populations. These findings suggest clinicians view automated facial-analysis systems as potentially valuable adjuncts for pain monitoring, particularly in communication-impaired populations, while emphasizing the importance of accuracy, fairness, explainability, and workflow integration during system development and implementation.

Hospital-based surveillance captures important components of antimicrobial resistance in India, but community-facing data linking prescribing and resistance remain limited. To address this, we analysed routinely generated prescription and diagnostic data from Tata 1mg from 2023 to 2025. Among 25.7 million medicines prescribed, 10.8% were antibiotics, with Watch antibiotics exceeding Access antibiotics in every state meeting the inclusion criteria. Among 156,000 diagnostic microbiology samples, urine cultures predominated, with Escherichia coli accounting for 60.6% of culture-positive isolates. In urinary E. coli, resistance was low to the Access antibiotics nitrofurantoin and amikacin (4.9% each), but high to the Watch antibiotics ceftriaxone (53.6%) and ciprofloxacin (57.0%). Exploratory cross-correlation analyses identified lead-lag relationships in which resistance to selected Watch antibiotics preceded changes in resistance to Access antibiotics by approximately 3-6 months. Routine digital healthcare data can therefore support community-facing AMR surveillance by jointly characterising platform-recorded prescribing and susceptibility within the observed care pathway.

The explosive growth of ambient AI scribes in healthcare has focused almost exclusively on relieving clinician documentation burdens and enhancing clinician efficiency and reducing burnout. While most research has focused on the clinician experience, these same types of tools, such as a patient held AI Personal Health Assistant, can be put in the hands of patients to facilitate communication with healthcare providers, to provide greater autonomy and activation for patients in managing their own healthcare, and to facilitate the task of receiving and retaining complex new information. However, to ensure these tools serve as more than just administrative aids, we partnered with a large technology company to conceptually codesign how such a tool might be studied in real world settings in a way that reflected outcomes of greatest importance to patients and families. We used the Nominal Group Technique and engaged thirty three patients and caregivers and ten multidisciplinary clinicians in structured virtual sessions to identify the outcomes on which subsequent studies should hinge. Our findings reveal a significant misalignment between current industry trends and patient priorities. While the market emphasizes the scribe function for billing, stakeholders overwhelmingly prioritized the AIPHAs role in longitudinal care management and post visit coordination. The most critical outcome identified was the impact on communication and respect, suggesting that patients view AI as a mechanism to improve the human to human dynamic and ensure a shared understanding of the care plan. Our findings further highlighted that the primary hurdle for the use of patient held AI is not technical utility but clinician trust and the need for formal consent processes. Importantly, as this technology is new and integration into the clinical experience is still being explored, patients cannot assume they will have access to the transcript of their own encounter when that record is created and held by the health system. This gap highlights the potential value of a patient held tool, preserving discussions from their visit and follow up steps in clearer detail than they may receive in their clinical note or traditional visit summary. From a system perspective, this transition requires a strategic shift from measuring workflow focused efficiency to measuring patient side advocacy and engagement. By centering the evaluation of AI on these patient defined metrics and fundamentally designing subsequent AI investigations using a partnership model, health systems can move beyond simple automation toward a model of care delivery that empowers patients to navigate complex clinical journeys with greater autonomy and clarity.

Cardiovascular diseases (CVD) are the leading cause of global mortality, responsible for 17.8 million deaths and 423 million prevalent cases annually. While the etiology of CVD is multifactorial, evidence confirms that multicomponent behavioral interventions regarding diet and lifestyle can substantially reduce risk. This study aimed to evaluate the effectiveness of a 12-month mHealth intervention, i.e., public health practice supported by mobile devices in reducing the 10-year CVD risk among overweight and obese adults aged 35 to 64 in an urban Sri Lankan population. A two-group parallel-arm randomized control trial (RCT) was conducted with 1,200 participants from the Colombo district, assigned either to an intervention arm (n=600) or a usual care control arm (n=600). The intervention group received automated text and voice messages in their native language at a frequency of two of each per week over 12 months. The primary outcome, 10-year CVD risk, was assessed using the WHO CVD risk prediction score. The intervention significantly reduced the proportion of participants with a 10-year CVD risk from 10.0% to 6.2% (p=0.003), whereas no significant change was observed in the control group (11.0% to 9.7%). Participants in the mHealth arm demonstrated substantial behavioral improvements, including increasing daily fruit and vegetable servings from 3.6 to 5.5 and reducing daily salt intake from 7.64g to 5.62g. Physical activity levels also improved, with 89.6% meeting minimum daily requirements compared to 77.4% at baseline. These findings suggest a potentially beneficial effect on predicted cardiovascular risk. The study concludes that an automated mHealth intervention is an effective, feasible, and acceptable approach to reducing cardiovascular risk in an urban South Asian setting. The findings provide a robust evidence base for integrating digital health strategies into national non-communicable disease (NCD) prevention policies. While the 12-month results are promising, further longitudinal research is required to evaluate the long-term sustainability of these behavioral changes and their ultimate impact on mortality rates.

Social media language and digital lifestyle behaviours are promoted as scalable signals of population mental health, but predictive models are often evaluated in ways that overstate accuracy. We analysed 20,000 tweets from 72 users, labelled by whether the account belonged to a user identified as depressed, and two synthetic datasets (100,000 and 30,000 records) linking screen time, sleep and social media use with self-reported stress. Under a conventional tweet-level split, a term-weighted logistic regression appeared accurate (area under the receiver operating characteristic curve 0.835), but when whole users were held out its performance fell to near chance (0.595; 95% confidence interval 0.505 to 0.686), because the model largely recognised users and their topics. Depression-associated tweets were more often negative (31.7% versus 24.2%), yet sentiment alone could not identify unseen users (0.539). The two behavioural datasets gave opposite answers: strong discrimination in one (0.926), driven by short sleep, and chance in the other (0.506), so synthetic results reflect how the data were generated. At a population prevalence of 10%, positive predictive values would be 0.10 to 0.30. Digital distress signals should be validated across individuals, on real linked data and at realistic prevalence before informing public health surveillance or prevention.

Abstract Background: Large language models (LLMs) are increasingly explored as adjunct tools in graduate medical education, but their reliability for answering discipline-specific certification examinations remains incompletely characterized. In dentistry in China, the national dental examination serves as a milestone for stomatology trainees. Prior studies predominantly rely on single metric, limiting reproducibility and granularity. Objective: To systematically evaluate the performance of five LLMs in answering the national standardized training examination for dental residents. Method: A total of 1,318 questions covering the six dentistry disciplines were selected. Five LLMs (DeepSeek-V3, DeepSeek-R1, Qwen3-32B, Claude 3.5, and GPT-4o) were evaluated via a single-response protocol on the Dify-based Chatflow framework. Performance combined human verification with automated scoring using the Retrieval-Augmented Generation Systems (RAGAS) framework across seven dimensions. Tasks involving question classification, key-point categorization, and instructional requirement summarization were designed to enable a horizontal comparison of models' cognitive capabilities, assessed using inter-model consistency scores. Results: DeepSeek-R1 achieved the highest correct choice rate (80.90%, p < 0.001) and answer correctness score (0.853, p < 0.001), significantly outperforming the others. All models exhibited limitations, including information insufficiency, redundancy, and hallucination-related issues (lowest 44.20% in DeepSeek-R1). In classification and summarization tasks, Claude 3.5 showed higher consistency scores (186/181) than Qwen3-32B (152/151). Conclusion: Evaluated LLMs demonstrated good performance on dental examination questions and in executing tasks. DeepSeek-R1 consistently outperformed the others. While LLMs showed potential as supportive tools in dental education, further refinement and validation are needed before formal adoption in graduate medical educational and clinical training settings.

Stroke is a leading cause of mortality and disability in Uganda, with a prevalence of approximately 2.5% and a 25-30% 30-day mortality rate, exacerbated by limited access to predictive tools and advanced diagnostics. Machine learning (ML) offers potential for early stroke risk stratification, but its application in resource-constrained settings like Uganda requires tailored models and rigorous performance evaluation, as well as assessment of misclassification patterns. This study aimed to develop and internally validate an ML-based predictive model for stroke using clinical and socio-demographic characteristics among patients at Mulago National Referral Hospital (MNRH) and residents of Wakiso District, Uganda, identify key predictors and assess misclassification patterns to enhance targeted prevention. A retrospective case-control study utilized secondary data from 1,016 participants (254 stroke cases, 762 controls) across three prior studies conducted between 2006 and 2015. Eight ML algorithms were evaluated using 5-fold cross-validation, with performance metrics including accuracy, sensitivity, positive predictive value, F1-score, and area under receiver operator characteristics curve (AU-ROC). McNemars test compared model classification performance of the top 2 models. Feature importance identified key predictors, and misclassification patterns were analyzed using classification matrix and line graphs. The Random Forest (RF) model achieved the highest performance (95.18% accuracy, 90.10% F1-score, 97% AU-ROC). McNemars test confirmed RFs superior classification (p=0.01). Age, systolic blood pressure, total cholesterol, and waist-hip ratio were the top drivers of RF stroke predictions. RF misclassified 8.3% of stroke cases and 3.3% of non-stroke cases, due to overlaps in risk profiles (e.g., lower cholesterol, systolic blood pressure and age in false negatives). The RF model demonstrates strong potential for stroke risk prediction in Uganda, leveraging routine data to prioritize high-risk individuals. Misclassification analysis highlights areas for model refinement, supporting its potential for integration into clinical workflows for early intervention and stroke prevention in resource-limited settings. Future external validation across diverse populations is recommended.

Background: Patient Decision aids (PtDAs) are important tools that support shared decision-making (SDM) between clinicians and patients, enabling patients to be more informed and engaged in decisions regarding their care. The use of PtDAs can increase patient knowledge, reduce decisional regret, and engage the clinician and patient in meaningful dialogue. Shared decision-making relies on decision aids, but how patients receive the PtDA can vary. This qualitative study elucidated patient familiarity with PtDAs and the barriers, needs, and requirements of patients to access and confidently use DAs, including when received via a clinician-facing EHR-integrated Decision Aid Navigator System (DEANS). Methods: We conducted six focus groups (FG) with participants who were recruited from two academic medical centers. Focus groups included an introduction to what a PtDA is and an exploration of how patients or family members, i.e., people who had not made a medical decision for themselves using a PtDA, but who have helped a family member, had previously used PtDAs in their healthcare. The FGs focused on PtDA-users, affordances and barriers to using PtDAs, and PtDA-user needs and requirements of a proposed DEAN patient portal. We applied Rapid Qualitative Analysis (RQA) to summarize the data, using a consensus making approach to identify emerging themes. Results: We identified three overarching themes: 1) Patients had varied familiarity with PtDAs and often confused them with general health tools, but found value in SDM, 2) Patients perceived technology accessibility, PtDA credibility, and time as barriers potentially impeding PtDA use, and 3) Patients valued flexible, user-friendly, and accessible PtDAs that support patient-led decisions, particularly when introduced by clinicians and delivered via a patient portal. Conclusion: This study provides patient-derived insights into factors that influence PtDA uptake, highlighting common misconceptions about PtDAs and identifying key patient priorities such as accessibility, flexibility, clear content, and clinician-guided usage. Patient also expected clinicians to be familiar with PtDAs, especially when the PtDA covers a complex decision. These findings can inform the development of more patient-centered SDM infrastructure and strategies to support PtDA use in clinical practice, including clinician-supported delivery, accessibility, and integration into routine care to strengthen shared decision-making.

Background and Objective: The volume of patient portal messages has grown substantially and artificial intelligence (AI) can support clinical staff and improve the timeliness of patient care. Our objective is to test if an AI tool for managing the electronic inbox In-Basket reduces response times to patient portal messages in neurology. Methods: We conducted an implementation study from January to May 2026 in the neurology department at two clinical sites within a regional medical system. A single AI tool was implemented to auto-categorize and auto-route all patient messages received in the electronic inbox. The tool also generates optional draft replies to messages. Primary process measures were the median weekly time (hours) from message creation to the first action and to closure. Impact was analyzed using a controlled interrupted time-series (ITS) design, comparing message-volume-matched control divisions. Results: Among 24,142 messages analyzed, pre-intervention median weekly time to first action was 11.3 hours [IQR: 5.6, 15.1] (Site A) and 16.1 hours [IQR: 13.8, 19.0] (Site B), falling post-intervention to 1.0 hour [IQR: 0, 3.1] (Site A) and 0.6 hours [IQR: 0.2, 2.4] (Site B). Controlled interrupted time-series analysis showed a significant immediate differential level change in median weekly hours to first action at Site A (-92.5%, 95% CI: -97.5% to -78.0%), and a significant downward trend at Site B (-43.5%, 95% CI: -65.2% to -8.2%). Changes in time to closure were directionally favorable but not statistically significant at either site. There was low use of the generated draft replies, only 0.5% at Site A and 3% at Site B. Discussion: Implementation of AI-assisted message categorization and routing was associated with a substantially reduced initial action time for patient In-Basket messages in Epilepsy. The utilization of AI-generated draft-replies to messages was low. These findings support the feasibility and effectiveness of AI-augmented workflows to improve inbox management.

BACKGROUND Timely cause-of-death (COD) information guides health policy, yet many deaths in low- and middle-income countries are not medically certified. In Africa, fewer than 10% of deaths are. Verbal autopsy (VA) yields a structured symptom questionnaire and a free-text narrative. Mortality surveillance increasingly records the narrative alongside the questionnaire, but the algorithms used to produce VA-based statistics were designed for structured responses only, so narratives inform physician COD assignment but do not contribute to algorithmic COD assignment. We evaluated whether supervised models could use both modalities to improve COD assignment. METHODS We analyzed 4,999 VA records for deaths at ages 12 years or older from a national sample of registered deaths in South Africa (RSA), using physician-assigned CODs as the reference standard. We trained unimodal and multimodal pipelines on the labelled records: pretrained language models (PLMs) fine-tuned on narratives alone, machine learning (ML) models on structured responses alone, and the two combined at the data, feature, and decision levels. We also prompted a locally deployed generative large language model (LLM) on the same unimodal and multimodal inputs, without parameter updates. All were compared with InSilicoVA, the operational comparator representing current automated practice. We evaluated individual-level assignment, population-level agreement, calibration, and generalization across provinces in RSA, reporting these metrics rather than ranking pipelines on any one of them. RESULTS Every supervised pipeline improved on InSilicoVA by 8.5-15.9 percentage points in top-1 accuracy and 3.5-7.1 points in cause-specific mortality fraction (CSMF) accuracy. The narrative and structured pipelines performed similarly in aggregate but differed by cause: 63.9% of deaths received identical predictions, 11.6% were correct only from the narrative pipeline, and 13.6% only from the questionnaire pipeline, with the narrative pipeline more accurate for noncommunicable causes and the questionnaire pipeline for communicable causes. Against the unimodal approaches, the decision-fusion pipelines added 4.2-5.4 percentage points in top-1 accuracy, concentrated in the most common causes, whereas feature fusion with cross-modal attention improved the less common causes, adding 6.6 points (95% CI, 1.7-11.5) in balanced accuracy and 5.5 points (95% CI, 0.6-10.3) in macro F1. Feature fusion reached the highest CSMF accuracy (95.7%). Paired differences from the better unimodal representative ranged from -0.2 to +3.4 points, with all intervals including zero. The LLM exceeded InSilicoVA on individual-level metrics and reached the highest balanced accuracy of any pipeline (60.6%), while the supervised pipelines retained higher CSMF accuracy. Supervised pipelines had lower expected calibration error than InSilicoVA. Supervised gains held across all nine held-out provinces, including every top-1 comparison and nearly every macro F1 and CSMF comparison. CONCLUSIONS PLMs using narratives alone substantially improved on current best practice and matched ML pipelines using structured responses, while multiple multimodal strategies produced further gains in individual COD assignment. The zero-shot LLM showed particular strength on rare causes, whereas supervised pipelines better recovered the population cause distribution. These findings support incorporating narratives as a core input to automated VA systems and evaluating representative fusion strategies for the intended output. External validation across settings and age groups will support broader implementation.