Nevin Manimala

JMIR Form Res. 2026 Mar 23;10:e85452. doi: 10.2196/85452.

ABSTRACT

BACKGROUND: Clinical decision support systems (CDSSs) are widely used in various health care settings. In Japan, pressure ulcers are becoming a major concern in an aging society due to their increasing prevalence. However, management is often handled by nonspecialists in wound care due to regional disparities in specialist availability.

OBJECTIVE: To provide support for nonspecialists in wound care, we developed a prototype smartphone-based CDSS for pressure ulcer management. The system prompts users to answer questions about the wound’s condition and recommends appropriate ointments and wound dressings by using a safety-first approach. This study aims to evaluate the utility of this system.

METHODS: We conducted a randomized crossover pilot study involving 28 general internal medicine (GIM) physicians. Participants were randomly assigned to group A (intervention-control) or group B (control-intervention). Participants evaluated 10 standardized pressure ulcer photographs and selected the most appropriate ointment and wound dressing for each. The unit of analysis was the individual response to each question (N=280 total observations). We used generalized estimating equations with an exchangeable correlation structure to account for within-subject clustering and adjust for potential period and sequence effects.

RESULTS: The overall correct response rate during the intervention phase was significantly higher than that during the control phase (49.3% vs 4.3%, respectively). After adjusting for clustering and crossover biases, the use of CDSS was associated with a 29.1-fold increase in the odds of a correct response (95% CI 8.2-103; P<.001). Secondary analyses revealed significant improvements in ointment selection (adjusted odds ratio [aOR] 2.4, 95% CI 1.5-3.8; P<.001) and wound dressing selection (aOR 8.9, 95% CI 4.9-16.1; P<.001). However, no significant period (P=.11) or sequence (P=.25) effects were observed for the primary outcome.

CONCLUSIONS: The prototype CDSS improved the accuracy of treatment decisions made by GIM physicians in a pilot study that used photographs and fixed options. Within the parameters of this investigation, CDSS effectively guided participants toward standardized, safety-oriented choices as defined by our scoring criteria.

TRIAL REGISTRATION: UMIN Clinical Trials Registry UMIN000057294; https://tinyurl.com/36a6vvah.

PMID:41871340 | DOI:10.2196/85452

J Med Internet Res. 2026 Mar 23;28:e85583. doi: 10.2196/85583.

ABSTRACT

BACKGROUND: Mobile health (mHealth) apps can address health inequities and enhance access to care for individuals with immunocompromising conditions. Although hundreds of oncology apps exist, research on provider perspectives regarding their use in clinical care remains limited.

OBJECTIVE: This study aimed to describe oncology providers’ recommended apps, mHealth attitudes and beliefs, and perceived barriers to and facilitators of mHealth adoption. Exploratory aims examined differences based on provider type (medical vs psychosocial), provider age (<45 vs ≥45 years old), and patient population (pediatric vs adult).

METHODS: We conducted a cross-sectional survey administered via REDCap (Research Electronic Data Capture) to oncology providers across the United States between June and November 2024. Data were summarized using descriptive statistics. Pearson’s chi-square analyses examined exploratory group differences based on provider type, provider age, and patient age.

RESULTS: Of 188 respondents, the majority self-identified as female (150/188, 79.8%), White (161/188, 85.6%), and non-Hispanic/Latino (174/188, 92.6%). Nearly all providers (178/188, 94.7%) reported either recommending or using mHealth apps with their patients, with primary use for patient-provider communication (139/188, 73.9%). Providers perceived potential benefit across a broad spectrum of holistic care functions. Providers, on average, reported a growth mindset and confidence in their ability to learn mHealth tools and in its potential to improve care access. Key facilitators included alignment with patient needs, increased accessibility, and cost-effectiveness, while barriers included disparities in technology access, digital health literacy, and data security and privacy. Exploratory analyses showed some significant group differences by provider role, provider age, and patient age. Psychosocial providers were significantly more likely to recommend or use apps for pain management (χ²₁=14.34, P<.001, φ=0.28), mental health (χ²₁=50.54, P<.001, φ=0.53), and sleep health (χ²₁=25.47, P<.001, φ= 0.38). Psychosocial providers also perceived higher benefit for sleep health apps (χ²₁=6.40, P=.01, φ=0.19). Medical providers were significantly more likely to perceive medication management apps as potentially beneficial (χ²₁=10.93, P<.001, φ=0.25). Older providers (16/88, 18.2%) and adult care providers (8/32, 25%) were significantly more likely to recommend or use disease management apps compared to younger providers (5/100, 5%; χ²₁=8.20, P=.004, φ=0.21) and pediatric care providers (6/101, 5.9%; χ²₂=9.22, P=.01, Cramer V=0.22), respectively. Pediatric care providers (83/101, 82.2%) were more likely to recommend or use medical team communication apps compared to adult care providers (15/32, 46.9%; χ²₂=15.66, P<.001, Cramer V=0.29).

CONCLUSIONS: Our study underscores the opportunity to develop inclusive mHealth solutions tailored to the diverse needs of individuals across the cancer care continuum, including those in active treatment and survivorship care. Engaging diverse medical and psychosocial providers is essential to inform clinical integration of mHealth technologies in oncology care.

PMID:41871339 | DOI:10.2196/85583

JMIR Public Health Surveill. 2026 Mar 23;12:e76860. doi: 10.2196/76860.

ABSTRACT

BACKGROUND: Public health restrictive measures adopted during the COVID-19 pandemic led to significant changes in lifestyles. Global declines in physical activity (PA) and increases in sedentary behavior were noted. These trends were observed within different regions of the world, pointing toward potential long-term implications for PA behaviors.

OBJECTIVE: This mixed methods study aims to assess variations in daily step counts in Qatar using device-driven data throughout all 3 COVID-19 waves (February 2020 to February 2023) compared with a full pre-COVID-19 year. In-depth interviews were further conducted with randomly selected participants to gain insights into determinants, perceptions, and barriers of PA during the pandemic.

METHODS: A total of 362 participants (60/362, 16.6% female) from the Step Into Health community-based program reported daily step counts using pedometers (170/362, 47%) or a mobile phone app (192/362, 53%). Linear mixed models examined changes in daily step counts across 19 phases of implementation and lifting of restrictions. Overall, 9 participants also completed semistructured interviews that were analyzed thematically and phenomenologically. Triangulation of quantitative and qualitative data was applied to interpret convergences and divergences between device-measured activity patterns and lived experiences.

RESULTS: Significant declines in daily step counts (ie, from 689 to 1013 steps) were observed at the onset of each wave (P<.001), were especially marked at wave 2, and were followed by a recovery of step count following the lifting of restrictions at each wave (ie, increase of 609 to 1147 steps). Different patterns of change in step count emerged within sex (P=.03), age (P=.03), and BMI (P=.01) groups, where larger variations were seen among male individuals, pedometer users, and normal-weight participants. Qualitative themes (ie, disrupted routines, reliance on home-based exercise, and media influence) contextualized these patterns and explained subgroup differences.

CONCLUSIONS: The largest drops in daily step count coincided with increased case severity and Ramadan. Integration of quantitative and qualitative findings showed that declines in activity were shaped not only by restrictions but also by fear, motivation, and contextual factors. These results underscore the importance of designing interventions that encourage outdoor activity and provide reliable social media-based guidance during public health crises.

PMID:41871338 | DOI:10.2196/76860

JMIR Res Protoc. 2026 Mar 23;15:e82523. doi: 10.2196/82523.

ABSTRACT

BACKGROUND: The status of the axilla remains a significant prognostic factor and influences adjuvant systemic and locoregional treatment choices in early-stage breast cancer (EBC). Sentinel node (SN) biopsy continues to be the preferred technique for establishing axillary nodal status in clinically node-negative EBC. A multivariable prediction model with adequate accuracy and generalizability has been explored as a potential alternative to SN.

OBJECTIVE: This systematic review aims to evaluate the predictive performance, methodological quality, and risk of bias associated with the available mathematical models (MMs), excluding artificial intelligence (AI)-based models, for predicting SN status in patients with EBC.

METHODS: A systematic search will be conducted across PubMed, Cochrane CENTRAL, and Embase to identify studies reporting the development of an SN status prediction model. Only studies that report SN status using mathematical modeling techniques will be included. Two independent reviewers will screen the search results and extract data from the included articles. The primary outcome of this systematic review is to evaluate the methodological adequacy and generalizability of individual MMs and compare the reported predictive performances of methodologically robust MMs. The secondary objective is to identify key predictive factors contributing to SN status prediction in MMs. A narrative synthesis of all the included studies will be undertaken. The details of this protocol are accessible on PROSPERO, where it was registered on January 23, 2025. Ethics approval is not required for this study because only published data will be analyzed.

RESULTS: Funding for this review was obtained in 2022. The literature search was completed on December 15, 2023, and screening began in December 2023. Data extraction and assessment using the Prediction Model Risk of Bias Assessment Tool was completed by December 2025, with synthesis planned for March 2026. Of the 3458 screened records, 122 (3.5%) were selected for data extraction. Results will be prepared for submission for a peer review and publication in mid-2026.

CONCLUSIONS: This review will provide a consolidated evaluation of non-machine learning MMs for predicting SN status in EBC. By clarifying the predictive performance and methodological quality of traditional statistical approaches, the findings will serve as a benchmark against which emerging AI-based tools can be compared. This review is also expected to identify predictors that consistently contribute to accurate modeling, informing the development of future statistical and AI-enhanced prediction tools.

TRIAL REGISTRATION: PROSPERO CRD42025637632; https://www.crd.york.ac.uk/PROSPERO/view/CRD42025637632.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/82523.

PMID:41871336 | DOI:10.2196/82523

JMIR Form Res. 2026 Mar 23;10:e88905. doi: 10.2196/88905.

ABSTRACT

BACKGROUND: Conversational artificial intelligence (AI) systems offer potential solutions to traditional constraints in medical consultation skills training, including high costs, scheduling difficulties, and varied standardization. There is limited evidence evaluating medical professionals’ perceptions of AI-generated patient interactions across multiple fidelity dimensions and assessing the educational value of conversational AI for consultation skills training.

OBJECTIVE: This study aimed to evaluate perceptions of conversational AI patient simulations in primary care consultation training, examining functional fidelity, conversational realism, educational value, and implementation readiness.

METHODS: A cross-sectional evaluation study at a UK medical school (medical students and general practitioners) yielded 47 grouped and individual responses. Participants completed standardized clinical scenarios using the SimFlow conversational AI system, a conversational AI system, followed by a multidomain questionnaire evaluating AI realism, medical content, educational value, feedback, and usability. Data were analyzed using the Wilcoxon signed rank test, Spearman correlation, and Firth logistic regression to assess domain performance and participant characteristics.

RESULTS: Medical content received the highest ratings (median 4.5, IQR 4.0-5.0), with 97.8% (45/46) rating clinical plausibility highly. Educational value was rated positively (median 4.0, IQR 3.0-4.0), although AI realism received moderate scores (median 3.0, IQR 2.0-4.0). Participants with prior AI experience gave significantly higher ratings for AI realism than those without prior experience (mean 3.81, SD 0.63 vs 3.07, SD 0.72; P=.03). Concordance analysis demonstrated moderate-to-strong agreement between individual- and group-level domain rankings (mean Spearman ρ=0.685), supporting consistency between collaborative and individual survey evaluations. Qualitative analysis revealed 4 themes: clinical authenticity, interactional limitations, educational potential, and implementation considerations.

CONCLUSIONS: Conversational AI demonstrates strong capabilities in functional fidelity (clinical accuracy) despite limitations in conversational fidelity (realism). The technology shows promise as a supplementary tool for clinical skills training rather than higher-stakes assessment, with future development needed in dialogue naturalness and feedback capabilities.

PMID:41871335 | DOI:10.2196/88905

JMIR Form Res. 2026 Mar 23;10:e86714. doi: 10.2196/86714.

ABSTRACT

BACKGROUND: Despite the rapid expansion of internet infrastructure and digital health initiatives in China, there remains a lack of longitudinal, nationally representative analyses that track the concurrent development of general internet access and the specific adoption of online health services over the past decade. Understanding these parallel trends is crucial for evaluating the reach and equity of the ongoing digital health transformation.

OBJECTIVE: The aim of this study was to describe decade-long trends in internet infrastructure development and online health service adoption in China through a comprehensive secondary analysis of nationally representative, publicly available survey data from 2014 to 2025.

METHODS: Data were retrieved from the official website of the China Internet Network Information Center for the period from June 2014 to June 2025. The study was conducted in 31 provinces, autonomous regions, and municipalities, excluding Hong Kong, Macau, and Taiwan. The participants were citizens aged 6 years and older who had a telephone or mobile phone. The China Internet Network Information Center conducted a stratified 2-stage sampling survey using a computer-aided telephone access system.

RESULTS: From 2014 to 2025, an increasing trend was observed in the number of internet users and the internet penetration rate in China. It also showed an upward trend in the number of internet users, both in urban and rural areas. A consistent increasing trend was detected in the number of mobile internet users. In contrast, desktops and laptops showed a declining trend. The number of online health users in China showed a V-shaped change from 2020 to 2025. In June 2025, the total number of online medical users reached approximately 393 million, representing 35% of all internet users.

CONCLUSIONS: This decade-long observational study demonstrates sustained and significant growth in internet access across China, accompanied by a substantial rise in online health service adoption. A notable V-shaped trajectory in online health use emerged after 2020, indicative of a rapid COVID-19 pandemic-driven acceleration followed by market consolidation. The converging trends of near-universal smartphone-based access and the massive popularity of mobile-centric services, such as short videos, have fundamentally reshaped the digital landscape. Consequently, the findings suggest that for digital health strategies to achieve broad impact, policymakers and health care providers could consider prioritizing the integration of health promotion and services into existing high-penetration mobile platforms and communication formats that the population already uses daily.

PMID:41871334 | DOI:10.2196/86714

JMIR Med Inform. 2026 Mar 23;14:e86171. doi: 10.2196/86171.

ABSTRACT

BACKGROUND: Digital health literacy (DHL) is the ability to locate, understand, evaluate, and apply health information in digital environments. It is essential for older adults to effectively engage with contemporary health care. However, existing DHL assessments primarily rely on self-reported measures, which are susceptible to subjective bias and often fail to capture actual performance. There is a need for a comprehensive, data-driven approach that integrates objective performance indicators with self-assessments to accurately predict and explain DHL levels in older adults.

OBJECTIVE: This study develops and validates a machine learning approach to predict DHL levels in older adults by integrating performance-based and self-assessed evaluations.

METHODS: We applied a 2-stage methodological framework using 2 independent datasets. In the first stage, to identify performance-based determinants, we assessed actual digital and information comprehension in a separate pilot cohort of 30 older adults (aged 60-74 years). In parallel, to measure self-reported DHL, we conducted an online survey with a distinct group of 1000 older adults (aged 55-74 years) using the Digital Health Literacy Scale and the Korean version of the eHealth Literacy Scale (KeHEALS). Bayesian linear regression was applied to both datasets to identify significant explanatory variables. In the second phase, we trained and validated a binary classification model to predict KeHEALS levels using the survey dataset (n=1000), leveraging the features identified in the first stage. Five machine learning algorithms were evaluated, and the best-performing model was interpreted using Shapley Additive Explanations (SHAP) analysis.

RESULTS: In the pilot performance-based assessment, using a greater number of electronic devices and having higher educational attainment were positively associated with comprehension, whereas alcohol intake showed a negative association. In the self-assessed survey data, key correlates included interest in health-related apps, self-care confidence, age, smoking, alcohol intake, number of devices used, and exercise frequency. Among the machine learning models, categorical boosting demonstrated the most balanced performance (accuracy 0.785, precision 0.769, F1-score 0.765, area under the receiver operating characteristic curve 0.835), outperforming the dummy classifier (accuracy 0.540). SHAP analysis indicated that self-care confidence, health information search, interest in health-related apps, number of electronic devices used, and exercise frequency were the strongest positive contributors to high-DHL predictions, whereas older age and lifestyle factors (alcohol intake, smoking) contributed negatively.

CONCLUSIONS: By explicitly integrating performance-based and self-assessed indicators within an explainable machine learning framework, this study demonstrates that DHL in older adults is influenced by both digital engagement and health management factors. These findings suggest that the proposed framework can serve as a structured approach for evaluating DHL in older adults and inform the design of personalized digital health interventions in clinical and community settings.

PMID:41871331 | DOI:10.2196/86171

Commun Dis Intell (2018). 2026 Mar 24;50. doi: 10.33321/cdi.2026.50.015.

ABSTRACT

During winter months of temperate regions, concurrent epidemics of multiple respiratory pathogens can occur, causing periods of increased clinical burden. Case time series, which are predominantly used to monitor infection levels, can exhibit substantial noise and day-of-the-week effects, limiting the visual interpretation of trends in raw data. However, statistical methods can infer smoothed trends within case time series by quantifying and accounting for different sources of noise. Here we apply statistical models to estimate the epidemic dynamics of SARS-CoV-2, respiratory syncytial virus (RSV), and influenza subtypes (influenza A H3N2, influenza A H1N1, and influenza B) in Victoria, Australia, over the 2024 winter season. We model trends in daily reported cases and the daily growth rate over time for all pathogens/subtypes. We present: (1) retrospective analyses using the final dataset up to 10 September 2024 and (2) weekly real-time analyses from 19 March 2024 to 10 September 2024 using data up to each timepoint, including a retrospective performance evaluation. We estimated similar peak timing of SARS-CoV-2 and RSV epidemics in late May, followed by a H3N2-dominant influenza epidemic, which peaked in early July. Transient increases in SARS-CoV-2 activity coincided with the emergence of new variants and transient decreases in influenza activity corresponded to the timing of school holidays. Real-time estimates demonstrated good agreement with those produced at the end of the season, with significant overlap of the 95% credible intervals. Our findings demonstrate how statistical methods can be implemented in real time to synthesise noisy case time-series data into interpretable trends (including uncertainty), enabling quantification of the strength of evidence for whether epidemic activity is increasing, stable or declining. Our real-time outputs were reported weekly to the Department of Health, Victoria during June-September 2024, complementing other routine surveillance indicators.

PMID:41871329 | DOI:10.33321/cdi.2026.50.015

Commun Dis Intell (2018). 2026 Mar 24;50. doi: 10.33321/cdi.2026.50.016.

ABSTRACT

BACKGROUND: Endemic measles was verified as eliminated in Australia in 2014. We describe Australian measles epidemiology, 2014-2024.

METHODS: National measles notification data were analysed by age; sex; state/territory of residence; genotype; place of acquisition (overseas/Australia); vaccination status (number of doses); and outbreak reference ID (for clusters).

RESULTS: Between 2014 and 2024, there were 1,095 measles notifications (average annual notification rate 0.4 per 100,000 population per year). The highest annual notification rates were recorded in 2014 (1.4 per 100,000 per year) and 2019 (1.1 per 100,000 per year), when rates in the Northern Territory were 21.4 and 12.6 per 100,000 population per year, respectively. Although notification rates were highest among infants < 1 year of age (average 3.8 notifications per 100,000 population per year), people aged 20-49 accounted for 57.2% of total notifications (n = 626). Of cases with a known immunisation status (n = 766), there were 513 cases (66.9%) who reported being unvaccinated; 20.1% (n = 154) reported having received one dose of MMR vaccine prior to infection; and 12.1% (n = 93) reported two or more doses. For notifications where country of acquisition was available (n =1,077), just over half of cases (55.1%) were acquired in Australia. Where measles was acquired overseas (n = 493), the most common countries of acquisition were Indonesia (n = 99; 20.1%), the Philippines (n = 82; 16.6%) and India (n = 52; 10.5%). There were 47 clusters during the reporting period, of which the largest involved 74 linked cases in 2019. Of recorded clusters, 44 (93.6%) had a source country outside of Australia. Notifications tended to peak each year in the months coinciding with the end of Australian school holiday periods.

CONCLUSION: Australia has sustained measles elimination since 2014; this review of measles epidemiology, demonstrating a predominance of unvaccinated returning international travellers 20-49 years of age, provides strong motivation for maintaining high routine two-dose coverage and promoting measles vaccination to adults travelling internationally.

PMID:41871328 | DOI:10.33321/cdi.2026.50.016

J Geriatr Psychiatry Neurol. 2026 Mar 23:8919887261436696. doi: 10.1177/08919887261436696. Online ahead of print.

ABSTRACT

ObjectiveThis systematic review aims to evaluate recent evidence on interventions for sleep disturbances in dementia, a major clinical concern, with limited pharmacological and non-pharmacological strategies examined in controlled trials.MethodsWe included eight RCTs (N = 666 participants; 2020 – January 2025) in people with dementia and sleep disturbances. Records were identified via MEDLINE, screened using PRISMA-guidelines, and non-randomized or non-intervention studies were excluded. Risk-of-bias was assessed with Cochrane RoB2, and effect sizes (Cohen’s d) were calculated in R.ResultsZ-drugs indicated small-large benefits, with eszopiclone significantly improving sleep efficiency and latency. Orexin receptor antagonists indicated mixed evidence, with low-dose lemborexant (2.5 mg) most favorable, meeting benchmarks for nocturnal awakenings. Circadian modulation light therapy was associated with reduced nocturnal awakenings in small trials, while relaxation-based music therapy indicated little to no effect. Social stimulation with PARO indicated increased sleep time and large positive effects on sleep efficiency, despite missing clinical benchmarks.ConclusionsEvidence is limited by sparse reporting and few trials per intervention. Z-drug eszopiclone, low-dose ORA lemborexant, and social stimulation PARO show the most favorable profile across outcomes and benchmarks, but pharmacological options require careful risk-benefit consideration. Non-pharmacological approaches appear safer but under-researched. Future trials should standardize outcomes and tailor strategies to patient needs. No external funding. The review was registered in INPLASY: https://doi.org/10.37766/inplasy2025.1.0097.

PMID:41871321 | DOI:10.1177/08919887261436696