Category: Nevin Manimala

JMIR Med Educ. 2026 Jun 16;12:e81250. doi: 10.2196/81250.

ABSTRACT

BACKGROUND: In 2021, the American Academy of Pediatrics released a policy statement spotlighting the health-promoting and stress-buffering effects of early relational health (ERH) and calling for universal ERH promotion in pediatric primary care. However, little educational content for the observation and promotion of ERH is available, highlighting the need for scalable ERH training modules.

OBJECTIVE: This study aims to investigate the acceptability, feasibility, and impact of the “Lens of Emotional Connection,” a self-paced, asynchronous, ~45-minute ERH training module codeveloped by Reach Out and Read and the Center for Early Relational Health at Columbia University. The module introduces practitioners to emotional connection, an observable component of ERH, through written and video didactic content and experiential rating of emotional connection in videos of parent-child dyads interacting face-to-face.

METHODS: The evaluation was conducted by the Carolinas Collaborative. Pediatric providers across 8 clinical sites were invited to participate and responded to embedded pre-post surveys. Focus groups conducted with participants further examined the educational experience.

RESULTS: Of 653 invited clinicians, 207 (31.7%) participated in the module. Individual survey responses were available for 44-75 participants, depending on the question. Of responders, 64 out of 69 (93%) reported the module was a good way to learn about emotional connection, and 63 out of 69 (91%) felt the module provided valuable knowledge. Overall, 60 out of 69 respondents (87%) reported satisfaction with the module length, and 36 out of 44 respondents (82%) reported they would recommend this training to other clinicians. Focus groups echoed these findings. Comparison of pre-post data showed the greatest changes were in familiarity with emotional connection (n=75, pre mean 54.20, SD 18.59; post mean 73.99, SD 14.73; Cohen d=1.14; P<.001) and confidence in observing the quality of the parent- or caregiver-infant relationship during well-child visits (n=75, pre mean 55.36, SD 18.49; post mean 74.20, SD 14.07; Cohen d=1.28; P<.001). Suggested areas for improvement included more thorough explanations of specific components of emotional connection identified in parent-child interaction videos, a desire for synchronous live training, and additional content addressing what to do if low emotional connection is identified.

CONCLUSIONS: In this evaluation of a training module designed to introduce pediatric practitioners to ERH and emotional connection, acceptability among participants was found to be high, with most responders reporting it as valuable and reporting they would recommend it. Statistically significant impact was noted in both perceptions of the importance of information about emotional connection and perceived knowledge acquisition. Feasibility of widespread implementation with voluntary participation, as here, was relatively low, with only a minority completing the module. Critically, Reach Out and Read’s commitment to iterative creation, validation, and eventual delivery of ERH training creates a scalable avenue for wide-scale implementation, given the organization’s presence in >6500 clinics across the 50 states.

PMID:42302308 | DOI:10.2196/81250

J Med Internet Res. 2026 Jun 16;28:e92852. doi: 10.2196/92852.

ABSTRACT

BACKGROUND: Public health medical education is increasingly important in the low-resource, high-altitude Xizang Autonomous Region (Tibet). Traditional authoritative textbooks do not meet modern needs for accessibility and interactivity, whereas general large language models (LLMs) may hallucinate in specialized medical domains. Developing specialized LLMs for low-resource regions is also expensive and difficult.

OBJECTIVE: This study aimed to explore a novel approach to high-altitude public health medical education in the low-resource Xizang Autonomous Region that integrates modern LLMs and authoritative textbooks, using a comprehensive benchmark evaluation across multiple dimensions and retrieval-augmented generation (RAG) technology.

METHODS: We conducted a 2-stage cross-sectional comparative evaluation study to benchmark publicly available LLMs and evaluate the added value of textbook-augmented retrieval under standardized generation settings and blinded expert assessment. First, 4 publicly available LLMs (GPT-5.2 [OpenAI], Gemini 3.0 Pro [Google], DeepSeek R1 [DeepSeek], and Tencent HY 2.0 [Tencent]) were benchmarked using an 80-question benchmark on high-altitude public health medicine developed by authoritative medical specialists. Each question was asked 3 times, yielding 960 outputs; first responses (n=320) were scored under blinded conditions by 2 independent 8-member physician panels. A clinically weighted evaluation of multidimensional first-response scores (including comprehensiveness, accuracy, clarity, and relevance) and a composite consistency metric (including semantic similarity and algorithmic similarity) was administered. Second, 4 specific and prevalent authoritative textbooks on high-altitude public health medicine-Ward, Milledge and West’s High Altitude Medicine and Physiology, High Altitude Medicine: A Case-Based Approach, High Altitude Medicine, and High Altitude Medical Protection-were deployed as the external knowledge base for the evaluation-optimized model. Statistical analyses included Spearman ρ, Cronbach α, intraclass correlation coefficients, Friedman tests with Dunn multiple comparisons, and paired Wilcoxon signed-rank tests. The significance threshold was set at α=.05.

RESULTS: DeepSeek R1 was selected as the optimal base model for achieving the highest weighted score (5.61/10.00), followed by GPT-5.2 (5.51/10.00), Gemini 3.0 Pro (5.39/10.00), and Tencent HY 2.0 (4.71/10.00). The deployed retrieval-augmented model integrating the authoritative textbooks and the optimal LLM DeepSeek R1, HPHME-Xplus-RAG, achieved remarkable improvement in multidimensional scores compared to baseline DeepSeek R1 (median 8.00 [IQR 7.88-8.00] vs median 7.63 [IQR 7.38-7.88]; P<.001, r_rb=0.68, indicating a large effect).

CONCLUSIONS: Integrating authoritative textbooks with an evaluation-optimized general LLM through an RAG framework showed strong performance for medical education in the low-resource Xizang Autonomous Region. Unlike prior studies that mainly evaluated general LLMs or used clinical guidelines to build RAG systems for diagnosis and treatment, this study used authoritative textbooks for the broader, guideline-scarce field of public health medical education. This work provides a replicable workflow-domain-authoritative knowledge+RAG+model optimization and evaluation-for low-resource settings, with practical implications for medical instructors and students, hospitals, and public health services seeking cost-effective, convenient, and trustworthy educational support.

PMID:42302307 | DOI:10.2196/92852

JMIR Aging. 2026 Jun 16;9:e90705. doi: 10.2196/90705.

ABSTRACT

BACKGROUND: Postoperative frailty is highly prevalent among older adults undergoing hip surgery and is closely linked to poor clinical outcomes. Despite growing interest in understanding its progression, the temporal patterns of frailty remain underexplored. Moreover, there is a lack of validated models that can predict frailty trajectories and stratify patients by risk in the early postoperative period.

OBJECTIVE: This study aimed to identify distinct frailty trajectories within 6 months following hip surgery in older adults and to explore their associated predictors. An interpretable machine-learning model was developed and internally validated for individualized risk prediction and was implemented as a clinically accessible web-based calculator.

METHODS: This prospective longitudinal observational study was conducted among older adults undergoing hip surgery at a tertiary hospital in China. Frailty assessments were performed preoperatively and at 1, 3, and 6 months postoperatively. A total of 209 participants who completed the 6-month follow-up were included in the analysis. Frailty was assessed using the Frailty Index, and group-based trajectory modeling was applied to identify distinct frailty progression patterns. Predictive variables were selected using the least absolute shrinkage and selection operator regression. An interpretable Extreme Gradient Boosting (XGBoost) model was developed using a 60:40 training-test data split. Model performance was evaluated in terms of discrimination, calibration, and clinical utility. Interpretability was assessed using SHAP (Shapley Additive Explanations) at both the global and individual levels.

RESULTS: Three distinct frailty trajectories were identified: low-fluctuation frailty (55/209, 26%), high-improvement frailty (81/209, 39%), and high-deterioration frailty (73/209, 35%). Twelve predictors grounded in the Health Ecology Model were selected, spanning individual characteristics, interpersonal networks, and the living environment. The XGBoost model demonstrated excellent discrimination, with a microaverage area under the receiver operating characteristic curve of 0.98 (95% CI 0.96-0.99) in the training set and 0.93 (95% CI 0.90-0.96) in the test set. Calibration was acceptable, with a weighted Brier score of 0.0852. Decision curve analysis showed favorable clinical utility across a range of threshold probabilities. A web-based risk calculator was developed to facilitate personalized frailty trajectory prediction.

CONCLUSIONS: The XGBoost model demonstrated strong predictive performance and interpretability, enabling the early identification of older patients at risk for adverse frailty trajectories following hip surgery. This tool may support targeted interventions and improve perioperative care in geriatric populations.

PMID:42302293 | DOI:10.2196/90705

JMIR Med Educ. 2026 Jun 16;12:e90064. doi: 10.2196/90064.

ABSTRACT

BACKGROUND: Feedback is essential for medical students’ learning during clinical clerkships; yet, supervising physicians often struggle to provide meaningful written feedback due to time constraints. Large language models offer a promising approach to supplement human feedback, but how artificial intelligence (AI)-generated and human feedback differ in authentic clinical settings remains unclear, as most comparisons have been conducted in classroom or simulation contexts.

OBJECTIVE: The aim of the study is to examine how AI-generated feedback and supervisor-provided feedback differ when applied to medical students’ clinical clerkship logs, by identifying the distinct characteristics and complementary strengths of each feedback type.

METHODS: This cross-sectional convergent mixed methods study included 161 weekly clinical clerkship logs from 47 fifth- and sixth-year medical students across 12 clinical departments at Nagoya University, Japan (January-May 2024). Of 164 eligible logs, 3 were excluded because supervisors entered contact messages rather than substantive feedback. AI feedback was generated using GPT-4o. In total, 10 faculty physicians and 10 medical students evaluated both feedback types in blinded, randomized order using a validated 5-category rubric (criteria-based, clear direction, accuracy, prioritization, and supportive tone), followed by open-ended comments and source identification. Quantitative analyses (paired 2-tailed t tests, cumulative link mixed-effects models; α=.05 with Bonferroni correction) were complemented by qualitative thematic analysis and integrated using joint display analysis.

RESULTS: AI feedback was significantly longer than supervisor feedback (mean 382.02, SD 81.82 vs mean 98.87, SD 73.66 characters; Cohen d=2.84, 95% CI 2.50-3.19; P<.001). Cumulative link mixed-effects models showed that AI scored higher on criteria-based (odds ratio [OR] 11.81, 95% CI 7.64-18.27; P<.001) and clear direction (OR 6.61, 95% CI 4.35-10.06; P<.001), with no significant differences on accuracy (OR 1.35, 95% CI 0.91-2.00; P>.99), prioritization (OR 1.70, 95% CI 1.16-2.50; P=.10), or supportive tone (OR 1.34, 95% CI 0.87-2.06; P>.99). AI feedback showed greater consistency (variance ratio 3.9:1; Levene F1,320=73.20; P<.001). All 20 evaluators correctly identified feedback sources. Qualitative analysis revealed that AI provided structured, text-anchored feedback addressing rubric criteria, while supervisors offered experience-based feedback grounded in clinical context and professional expertise.

CONCLUSIONS: This study extends the comparison of AI-generated and supervisor feedback to an authentic clinical clerkship environment, moving beyond classroom and simulation settings examined in prior work. Through integrated mixed methods analysis, a key distinction emerged between text-anchored AI feedback, which systematically addresses written log content in alignment with rubric criteria, and experience-based supervisor feedback, which draws on clinical observation and professional judgment. AI consistently delivered structured feedback addressing gaps that arise when time-pressured supervisors provide brief comments, while supervisors contributed clinically grounded insights that AI cannot replicate. These complementary strengths suggest that AI feedback should supplement rather than replace supervisor feedback, and that hybrid models leveraging each type’s advantages warrant investigation in clinical education.

PMID:42302283 | DOI:10.2196/90064

JMIR Cancer. 2026 Jun 16;12:e89823. doi: 10.2196/89823.

ABSTRACT

BACKGROUND: Lung cancer remains the leading cause of cancer deaths in the United States; however, uptake of lung cancer screening (LCS) with low-dose computed tomography (LDCT) among eligible individuals remains low. Evidence suggests that limited knowledge, stigma, and false health beliefs contribute to the underuse of LDCT screening.

OBJECTIVE: This pilot study aimed to examine an online educational intervention designed to improve knowledge, attitudes, health beliefs, behavioral intentions, perceived importance, and confidence related to LCS among high-risk individuals.

METHODS: A single-group preintervention and postintervention design was used. High-risk individuals who smoke, defined according to the US Preventive Services Task Force criteria, completed baseline questionnaires followed by 5 self-directed online educational modules delivered through Research Electronic Data Capture (REDCap). Postintervention questionnaires assessed changes in lung cancer and screening knowledge, lung cancer stigma, health beliefs based on the health belief model and precaution adoption process model, and intentions, perceived importance, and confidence regarding LDCT screening. LCS uptake was assessed via follow-up email 3 months after the intervention. Data were analyzed using descriptive statistics and paired-samples two-tailed t tests.

RESULTS: A total of 25 participants completed the intervention. Significant improvements were observed across all major study outcomes. Knowledge scores increased markedly (score=3.76-8.60; P<.001), while lung cancer stigma decreased (score=25.52-19.16; P<.001). Health belief model constructs showed significant improvements, including perceived susceptibility, perceived benefits, cues to action, and self-efficacy, alongside reductions in perceived barriers and perceived severity (all P<.001). Self-reported intentions, perceived importance, and confidence related to obtaining LDCT screening increased significantly. Of the 22 (88%) participants who completed the 3-month follow-up, 13 (59.1%) reported obtaining LDCT screening. Participant satisfaction with the intervention was high, with a mean score of 18.32 (SD 2.33) out of 20.

CONCLUSIONS: Findings from this pilot study support the feasibility, acceptability, and preliminary efficacy of an online educational intervention created to promote LCS among high-risk individuals. The intervention improved knowledge; reduced stigma; positively influenced health beliefs; and increased screening intentions, perceived importance, confidence, and uptake. Results provide a foundation for a larger-scale study and suggest that online educational platforms may be an effective strategy to reach geographically diverse high-risk populations and promote LDCT screening.

PMID:42302275 | DOI:10.2196/89823

Eur Stroke J. 2026 Jun 2;11(6):aakag047. doi: 10.1093/esj/aakag047.

ABSTRACT

INTRODUCTION: Guidelines recommend cognitive screening post-stroke, but there is no consensus on approach. Given the dynamic nature of cognition following stroke, acute screening should both detect prevalent issues (diagnosis) and predict persisting problems (prognosis). We describe the diagnostic and prognostic utility of brief cognitive screening tools.

PATIENTS AND METHODS: Patients were screened on admission with stroke using 12 modified screening tests: 10 and 4 question Abbreviated Mental Test, Cog-4, Clock Drawing test (CDT), Cognitive Impairment Test, informal bedside assessment, General Practitioner Assessment of Cognition, Minicog, Short Form Montreal Cognitive Assessment, Six-Item Screener (SIS), Harmonised Vascular Cognitive Impairment battery and 4-A’s Test. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value were calculated against a Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition adjudicated reference standard of neurocognitive disorders. Test accuracy was compared using area under the receiver operator characteristic curves.

RESULTS: Of 335 patients, 54 (16.1%) had pre-stroke neurocognitive disorder, 79 (23.6%) had 18-month neurocognitive disorder. Ten of 12 screening tests were more specific than sensitive. Informal bedside assessment had highest specificity (96%), but low sensitivity (9%); CDT had highest sensitivity (80%) but low specificity (33%). Negative predictive value ranged from 77% to 87%, PPV ranged from 27% to 54%. Area under the receiver operator characteristic curve ranged 0.53 (informal bedside assessment) to 0.69 (SIS).

DISCUSSION: In the acute setting, where the intention of screening is often to triage those who need further assessment, the pattern of high specificity at the expense of sensitivity is the opposite of what is desired.

CONCLUSION: Brief cognitive screening tools, used in isolation, may not be suitable for assessment in acute stroke settings.

PMID:42302274 | DOI:10.1093/esj/aakag047

Arch Clin Neuropsychol. 2026 May 29;41(5):acag046. doi: 10.1093/arclin/acag046.

ABSTRACT

OBJECTIVE: The study aims to examine the relationship between subjective cognitive complaints (SCC) and objective cognitive performance (OCP) in severe mental disorders, including schizophrenia (SCZ), bipolar disorder (BD), and major depressive disorder (MDD).

METHOD: Systematic literature searches were conducted using Web of Science and PubMed, including articles published until September 2025. Studies were included if they assessed both SCC and OCP using validated instruments in individuals diagnosed with SCZ, BD, or MDD and were peer-reviewed English research articles. A series of meta-analyses was conducted using random-effects models to examine the associations between SCC and composite OCP scores, as well as the subdomains of objective cognition. Analyses were repeated for each diagnosis.

RESULTS: The sample included 49 articles with 5,007 participants. Our analyses yielded a small but statistically significant correlation between SCC and global OCP (r = -0.145). Domain-wise associations indicated correlations between SCC and OCPs in processing speed, attention/vigilance, working memory, and verbal learning/memory (correlation coefficients ranged from -0.107 to -0.172). In diagnosis-specific analyses, individuals with SCZ showed significant associations in all domains except executive function. In contrast, the associations were restricted to only a few cognitive domains in other disorders, specifically processing speed and working memory in BD, and processing speed and attention/vigilance in MDD.

CONCLUSIONS: Although significant, the strength of these associations was small, suggesting that SCC explains only a limited proportion of the variance in OCP. This suggests that while SCC cannot substitute for objective testing, it provides complementary information that reflects patients’ experiences of cognitive dysfunction.

PMID:42302271 | DOI:10.1093/arclin/acag046

Arch Clin Neuropsychol. 2026 May 29;41(5):acag045. doi: 10.1093/arclin/acag045.

ABSTRACT

OBJECTIVE: Contemporary clinical neuropsychologists primarily provide assessment and consultative services rather than direct intervention. This study examined the extent to which cognitive rehabilitation remains part of contemporary neuropsychological practice and explored reimbursement pathways and perceived barriers to implementation.

METHOD: An anonymous national survey was distributed to members of the National Academy of Neuropsychology through listserv and newsletter announcements. The survey included structured multiple-choice and open-ended questions assessing whether respondents provide cognitive rehabilitation, years of practice, clinical setting, insurance reimbursement patterns, and experiences with insurance denials. Descriptive statistics were used to summarize responses.

RESULTS: Twenty-two respondents completed the survey. Eighteen (81.8%) reported currently providing cognitive rehabilitation services. Reimbursed Current Procedural Terminology (CPT) codes most commonly included 97129/97130 (54.5%) and 96116/96121 (45.5%), with additional endorsement of health and behavior intervention codes (96158/96159) and group therapy codes (97150). Insurance denial experiences were mixed, with most respondents reporting occasional or conditional denials rather than consistent rejection of claims. Respondents also described variability in session limits and billing approaches across practice settings and payer policies.

CONCLUSIONS: Cognitive rehabilitation was reported by a majority of respondents, though service models and reimbursement pathways varied considerably. Given the small, self-selected sample, findings should be interpreted as exploratory rather than representative of broader neuropsychological practice. Larger investigations are needed to clarify training expectations, billing practices, and interdisciplinary collaboration.

PMID:42302270 | DOI:10.1093/arclin/acag045

JMIR Aging. 2026 Jun 16;9:e79797. doi: 10.2196/79797.

ABSTRACT

BACKGROUND: Home telemonitoring programs are increasingly used to support older adults living with chronic conditions such as heart failure (HF). While these interventions show promise for improving health outcomes and reducing care burden, their effectiveness depends largely on how patients and caregivers integrate digital technologies into everyday life and care relationships. However, relatively few studies have examined these experiences using conceptual frameworks that capture both functional and relational dimensions of care.

OBJECTIVE: This study aimed to explore the experiences of older adults and their informal caregivers participating in a home telemonitoring program for HF. Drawing on the Person-Based Approach and the Person-Centered Practice frameworks, we examined how participants engaged with both the technofunctional and relational aspects of the intervention.

METHODS: We conducted a qualitative study involving 34 patients, 28 informal caregivers, and 20 nurses across 3 primary care organizations in Quebec, Canada. The 6-month intervention included 4 connected devices used by patients (smartwatch, Bluetooth-enabled scale, voice-activated tablet, and a smart pill dispenser [xPill; Domedic]) and a mobile app for caregivers, complemented by remote nursing follow-up. Nurses reviewed patient data through a clinical dashboard at least once daily during weekday daytime shifts. Data were collected through semistructured interviews and field notes and analyzed using directed content analysis.

RESULTS: Participants’ experiences revealed both enabling and constraining factors across 2 key dimensions. Technofunctional engagement was shaped by digital literacy, emotional responses to the technology, alignment with daily routines, and access to technical or caregiver support. Relational aspects of care were influenced by perceived professional presence, opportunities for communication and shared decision-making, and the degree of emotional reassurance provided by remote monitoring. While many participants reported increased confidence and a sense of being supported, others experienced frustration, fatigue, or disengagement when the system disrupted routines or when feedback from clinicians was perceived as limited.

CONCLUSIONS: Engagement with home telemonitoring technologies among older adults depends not only on usability but also on the relational context in which these technologies are embedded. Combining technofunctional and relational perspectives provides a more comprehensive understanding of how telemonitoring interventions are experienced and highlights the importance of personalized support, reliable technology, and sustained clinical engagement to promote meaningful adoption.

PMID:42302267 | DOI:10.2196/79797

JMIR Form Res. 2026 Jun 16;10:e89820. doi: 10.2196/89820.

ABSTRACT

BACKGROUND: An estimated 5 to 8 million US children live with a parent who uses cannabis, and most cannabis users report smoking cannabis inside their homes, placing children at risk for cannabis secondhand smoke (cSHS) exposure. Indoor air quality (IAQ) monitoring provides real-time feedback on airborne pollutants and has shown promise in reducing in-home tobacco secondhand smoke exposure, suggesting its potential as an effective harm reduction strategy for cSHS.

OBJECTIVE: This pilot study evaluated the feasibility, acceptability, and preliminary effectiveness of using low-cost, off-the-shelf IAQ monitors to increase caregivers’ awareness of children’s cSHS exposure risk and to change smoking behavior. Secondary aims were to assess participant engagement, perceived usefulness, and household communication regarding in-home cannabis smoking.

METHODS: Between February 2025 and April 2025, 14 adults who smoked cannabis indoors and lived with at least 1 child aged younger than 16 years were recruited primarily via targeted social media advertisements and completed a 3-week trial. Participants received an Awair Element IAQ monitor, printed health education materials, and text messaging prompts for brief surveys. The IAQ monitor continuously measured PM2.5, VOCs, CO₂, temperature, and humidity. Daily surveys captured self-reported PM2.5 readings and recent cannabis use, while baseline and end-of-study assessments evaluated IAQ perceptions, cSHS risk awareness, and in-home smoking behavior. Survey results were summarized via descriptive statistics, and linear mixed-effects models were used to characterize objective IAQ trends. Six additional adult household members provided parallel end-of-study data.

RESULTS: Reported engagement was high, with 85% (11/13) of participants indicating that they reviewed the monitor at least daily. The average number of days in the previous week that a caregiver reported a child being home while cannabis was smoked declined from 4.5 (SD 2.2) at the trial start to 2.8 (SD 2.9) at the end (6/13, 46% had a reduction; 1/13, 8% reported an increase). Furthermore, 62% (8/13) of participants reported that they reduced (4/13, 31%) or thought about changing (4/13, 31%) their smoking habits. Around 62% (8/13) of participants agreed or strongly agreed that IAQ monitoring helped drive conversations about changing indoor smoking rules, while 100% (13/13) reported no IAQ-driven disagreements among household residents regarding in-home smoking rules. A linear mixed-effects model did not indicate a consistent trend in PM2.5 levels across participants over time (β=-0.28; SE 1.13; P=.81), but there was heterogeneity in trends, and those with the largest reductions in PM2.5 over the trial had the largest reduction in reported children’s cSHS exposure.

CONCLUSIONS: In-home IAQ monitoring was feasible and perceived as useful among caregivers who smoked cannabis indoors. Real-time IAQ feedback supported risk awareness, promoted family dialogue, and coincided with reductions in in-home smoking around children. These findings suggest that IAQ feedback may represent a scalable tool for reducing children’s cSHS exposure and merits further testing in larger, controlled trials.

PMID:42302266 | DOI:10.2196/89820