Healthcare

Background: System One Model is designed for structured decision-making and can provide probabilistic outputs, but their performance in domain-specific physical therapy tasks has not been established. Objective: To evaluate the performance and potential utility of Jev, a System One Model, for professional knowledge-based selection tasks in physical therapy by comparing it with GPT-4o. Methods: A total of 380 multiple-choice questions from the 2024 and 2025 Korean Physical Therapist Licensing Examinations were independently submitted to Jev and GPT-4o. The outcomes included answer accuracy, subgroup performance, output validity, API response latency, token usage, estimated API cost, and Jev's prediction uncertainty. A retrospective probability-based model-cascading simulation was also performed. Results: Jev correctly answered 282 questions (74.2%), whereas GPT-4o answered 326 (85.8%), a difference of -11.6 percentage points. Jev produced valid structured responses for all questions, whereas GPT-4o produced three invalid responses. Jev's selected-answer probabilities discriminated correct from incorrect answers (AUROC = 0.866), and accuracy reached 99.3% when probabilities were greater than or equal to 0.90. The median API response latency was 0.94 s for Jev and 1.06 s for GPT-4o, with estimated costs of US$0.008 and US$0.161, respectively. At a retrospective cascading threshold of 0.70, accuracy was 86.6% with 175 simulated GPT-4o calls. Conclusion: Jev showed lower overall accuracy than GPT-4o but provided reliable structured outputs, informative prediction probabilities, lower latency, and lower estimated cost. Its probabilistic outputs warrant further prospective evaluation for selective model escalation and decision-support workflows in physical therapy.

Background: The growing burden of lifestyle-related chronic diseases has increased the need for clinically interpretable decision-support tools capable of integrating artificial intelligence with evidence-based preventive nutrition. Although machine learning has shown considerable potential for health risk prediction, most existing approaches remain limited to isolated predictive models or conventional nutritional software, with little integration of multidimensional clinical assessment and personalized recommendations. Objective: To develop and internally validate NutrIA, a hybrid web-based Clinical Decision Support System (CDSS) that combines machine learning, validated clinical assessment, structured clinical reasoning and personalized nutritional recommendations for preventive medicine. Methods: NutrIA was developed using harmonized data from the National Health and Nutrition Examination Survey (NHANES, 1988 to 2018). A supervised machine learning model was trained to estimate 5-, 10- and 20-year all-cause mortality risk and subsequently integrated with an adaptive clinical questionnaire, validated screening instruments, nutritional indicators, dietary clustering, clinical phenotyping and a transparent rule-based recommendation engine within a unified web-based platform. Results: The predictive model achieved ROC-AUC values of 0.894, 0.914 and 0.923 for 5-, 10- and 20-year mortality prediction, respectively. The implemented CDSS incorporates an adaptive questionnaire (151 items), 39 validated clinical assessment instruments, 17 clinical phenotypes and 31 dietary clustering modules to generate individualized nutritional and lifestyle recommendations together with an automated clinical report. The integrated framework translates probabilistic risk estimates into clinically interpretable decision support for personalized preventive nutrition. Conclusions: NutrIA demonstrates the technical feasibility of integrating machine learning with knowledge-based clinical reasoning within a single web-based CDSS for preventive nutrition. Although external validation and prospective clinical evaluation are required before routine implementation, the proposed architecture represents a promising step toward clinically interpretable artificial intelligence for personalized nutritional care.

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.

Aim. To evaluate the accuracy, run-to-run consistency, and agreement of large language models on the Ukrainian Integrated Licensing Examination Krok 2 for the Specialty of Physical Therapy. Materials and Methods. Five cloud-based LLMs (ChatGPT-5.5, Claude 5 Sonnet, Gemini 3.5 Flash, Grok 4, DeepSeek-V3) were evaluated using 150 Ukrainian-language multiple-choice questions from the official Krok 2 database. Each model completed two independent runs with identical inputs. Accuracy (with 95% confidence intervals, Wilson method), run-to-run consistency, and agreement were assessed. Statistical analysis included Cohen's {kappa}, Cochran's Q and pairwise McNemar tests with Holm-Bonferroni correction. Results. All models achieved >80% accuracy in both runs, exceeding the passing threshold of [≥]64%. In the first run, accuracy ranged from 80.00% (DeepSeek-V3) to 93.33% (Gemini 3.5 Flash); in the second run, from 82.67% (Claude 5 Sonnet) to 88.67% (Gemini 3.5 Flash). In the first run significant differences were observed between Gemini 3.5 Flash and Grok 4 (93.33% vs 84.00%; p=0.0022) and DeepSeek-V3 (93.33% vs 80.00%; p=0.0001). There were no statistically significant differences between models in the second run. R2R consistency exceeded 90% for all models with the highest in ChatGPT-5.5 (98.67%) and lowest in DeepSeek-V3 (90.00%). Cohen's {kappa} ranged from 0.64 to 0.94, indicating substantial to almost perfect agreement. The stability of LLMs' response correctness across two runs ranged from 77.33% (DeepSeek-V3) to 88.00% (Gemini 3.5 Flash). Conclusions. All LLMs demonstrated high performance on Ukrainian-language Krok 2 examination tasks. Gemini 3.5 Flash showed the highest accuracy in both runs and response correctness stability while ChatGPT-5.5 exhibited the greatest response consistency and agreement. Although the accuracy and stability of LLM responses may be interrelated, they represent distinct aspects of model performance. This highlights the need for a comprehensive approach to evaluating LLMs. The findings indicate the potential of LLMs as an auxiliary tool in the professional practice of physical therapy specialists.

Background: Reducing maternal mortality and improving newborn and child outcomes are targets of the Sustainable Development Goals. Evidence on how these efforts are reaching women with disabilities is lacking. Objectives: to estimate global, relative inequalities in stillbirth, neonatal, infant, and maternal mortality for women with disabilities compared to women without disabilities. Methods: We searched MEDLINE, Global Health, PsycINFO, and Embase from 1 January 2015 to 28 January 2026, to identify articles on disability and stillbirth, neonatal, infant, and maternal mortality. We included studies that had a recognised measure of disability as an exposure, a control group of women without disabilities and at least one of the four outcomes. A pooled estimate for each outcome was done using a random-effects meta-analysis of the minimally adjusted results. Results: We identified over 4,300 titles, of which 17 papers were eligible for inclusion. Almost all data came from nationally-representative data sources in high-income countries. We found that women with disabilities were 4.66 times more likely (95% C.I. 1.70-12.75) to experience maternal mortality compared to women without disabilities. Women with disabilities were also 37% more likely to have a stillbirth and 27% and 48% more likely to have a neonatal or infant death compared to women without disabilities, respectively. Conclusions: Women with disabilities consistently have higher incidence of maternal, stillbirth, neonatal, and infant mortality, even in countries that have relatively low incidence of these outcomes. There is a lack of evidence globally and particularly from LMICs, and on effective interventions to improve maternal and infant outcomes for women with disabilities.

Against the backdrop of nationwide inclusive education promotion, children with autism spectrum disorder, intellectual disabilities and other special educational needs (SEN) in Jiangxi Province have raised growing demands for equitable schooling. Paraeducators serve as a critical on-site support mechanism enabling SEN children to access mainstream classrooms; the adequacy of shadow teacher service provision and the maturity of corresponding multi-stakeholder support systems jointly determine the overall quality of local inclusive education. This study adopted mixed quantitative-qualified methods, including questionnaire surveys and semi-structured interviews, to investigate SEN children, paraeducators, general and special education teachers, as well as SEN caregivers across multiple prefecture-level cities in Jiangxi. Grounded in provincial special education policies and local frontline inclusive education practices, we systematically unpacked multidimensional service demands from four core stakeholder groups, diagnosed prominent practical bottlenecks restricting the sustainable operation of shadow teacher services, and constructed regionally tailored multi-layered support strategies aligned with Jiangxi's educational realities. The findings of this research offer empirical evidence and actionable policy references to advance high-quality inclusive education for SEN children across central China's Jiangxi Province.

Aim: Little is known about patient safety in a less consolidated obstetric system where various facilities, such as perinatal medical centers (PMCs), general hospitals, and clinics, collaborate under risk-based role differentiation. We aimed to compare maternal complications after cesarean section by facility type across area types (rural, provincial, and metropolitan) in Hokkaido, Japan. Methods: This retrospective cohort study used insurance claims data from Hokkaido (2018-2025). Two comparisons were conducted for a composite outcome of postpartum hemorrhage, infection, and thrombosis: a two-category comparison (PMC vs. non-PMC, combining general hospitals and clinics) across all areas, with an interaction term between facility and area type; a three-category comparison (PMC vs. general hospital vs. clinic) restricted to metropolitan areas. Generalized estimating equations with a Poisson distribution, accounting for clustering within facilities, were applied to estimate risk ratios. Results: A total of 1,822 participants underwent cesarean section. PMCs were associated with lower maternal complication rates compared to non-PMC facilities (adjusted RR 0.37, 95% CI 0.16-0.88 in rural areas; adjusted RR 0.20, 95% CI 0.11-0.37 in provincial areas). In metropolitan areas, PMCs and general hospitals were associated with lower maternal complication rates compared to clinics (PMC vs clinic: adjusted RR 0.42, 95% CI 0.18-0.97; general hospital vs clinic: adjusted RR 0.25, 95% CI 0.09-0.70). Conclusions: Higher-level facilities were associated with lower maternal complication rates after cesarean section in Japan. These findings provide important evidence for regional consolidation of obstetric care.

Background: Whether corrective actions documented in medical safety incident reports rely on individual vigilance ("Safety-I") or on structural, system-level intervention ("Safety-II") has not been quantitatively evaluated on a national scale in Japan. We developed an automated classification pipeline to assign corrective-action free-text to a 7-level maturity scale (L0-L6) and computed two summary indices: the Safety Measure Quality Profile (SMQP), the full L0-L6 distribution, and the System-based Safety Measure Rate (SSMR), the proportion of non-L0 records classified L3-L6. Methods: We analyzed all 11,507 corrective-action free-text entries from the 2010 release of Japan's national medical accident and near-miss reporting database (Japan Council for Quality Health Care, JCQHC), comprising 8,804 near-miss (Hiyari-Hatto) and 2,703 accident (Jiko) reports. Records were classified using a five-stage hybrid pipeline: an expert-developed rule dictionary, TF-IDF + k-nearest-neighbor matching, cosine-similarity matching, a two-tier large-language-model (LLM) classifier, and a conservative priority-cascade fallback. SSMR was compared between near-miss and accident reports using a chi-square test, Wilson 95% confidence intervals, Cramer's V, and the risk difference (RD), against pre-specified minimal clinically important difference (MCID) criteria of RD >= 2 percentage points and Cramer's V >= 0.10. Results: Every record received a definitive L0-L6 label (0% unresolved). Overall, 16.6% of records were unclassifiable (L0); among the 9,599 classifiable (non-L0) records, individual-vigilance actions (L1) predominated (54.6% of all records), and only 11.82% (95% CI, 11.19-12.49%) met the SSMR criterion (L3-L6). SSMR was higher for accident reports than for near-miss reports (18.12% [95% CI, 16.69-19.64%] vs. 9.[95% CI,46% [95% CI, 8.80-10.17%]; RD = 8.66 percentage points; Cramer's V = 0.120; chi-square(1) = 136.97001), exceeding both pre-specified MCID thresholds. Conclusions: In this interim single-year analysis, the large majority of documented corrective actions in Japanese medical safety reports remained individual-vigilance-based rather than system-based, with accident reports showing a substantively, rather than merely statistically, higher proportion of system-based actions than near-miss reports. These findings support the feasibility of large-scale automated assessment of corrective-action quality and provide the rationale for the planned 16-year longitudinal analysis.

Abstract Background Community-based exercise may support physical and cognitive recovery after stroke, but the preferences of stakeholders involved in receiving, supporting and delivering these programmes are poorly understood. This study piloted a discrete choice experiment (DCE) that aimed to evaluate stroke survivors, carers, clinicians, and fitness instructors preferences for community-based exercise programmes for stroke survivors. The pilot aimed to assess the feasibility of conducting and analysing data for a larger-scale study and inform the sample size for a definitive DCE. Methods A DCE design was developed using literature reviews, stakeholder consultations, and evaluations of current services. Six attributes were included in the design: location, price, session frequency, individualisation, carer involvement, and recovery. Sixteen Bayesian D-optimal choice sets were developed, then blocked into two sets of eight to reduce cognitive burden. Main-effects were analysed using a mixed logit model, and stakeholder differences were explored using a conditional logit interaction model. Willingness to pay was also calculated. Results Overall, 161 respondents started the survey and 106 (66%) fully completed it. Stakeholders significantly preferred programmes that were delivered at a leisure centre, lower priced, individualised, had carer-specific sessions and offered greater chances of recovery. The largest willingness-to-pay estimates were {pound}49.41/ month for 80% chance of meaningful recovery vs 20%, and {pound}18.61/ month for leisure centre delivery vs online. Significant preference heterogeneity was identified in the main-effects model. Exploratory stakeholder interactions suggested clinicians were more price sensitive and placed greater importance on recovery and carer-specific sessions than stroke survivors, while fitness instructors placed greater importance on programme location and recovery. Carer interaction estimates were unstable due to the small subgroup sample. Conclusions The pilot supported the feasibility of the DCE design and analytical approach but identified limitations in the recruitment methods and survey completion. Addition of face-to-face recruitment methods are recommended for a larger-scale study. A larger sample size is required to investigate differences in preferences between stakeholder groups.

Background: Data from National Health and Nutrition Examination Survey (NHANES) cycles 1988-1994 and 1999-2004 show a declining trend in the adherence to Dietary Approaches to Stop Hypertension (DASH) among adults. Since then, DASH evidence, dietary guidelines, and public-facing dietary resources have advanced, but whether national DASH adherence has improved is uncertain. Methods: We extended the Mellen et al. NHANES analysis among 14,163 adults with self-reported hypertension (N=6,914, 2003-2010, N=7,249, 2011-2018). DASH score included 9 nutrient components: total fat, saturated fat, protein, cholesterol, fiber, calcium, magnesium, potassium, and sodium, with adherence defined as a score 4.5. Survey-weighted multivariable models compared DASH score, adherence, and nutrient-threshold attainment across periods. Results: In 2011-2018, U.S. adults with self-reported hypertension represented 76.2 million individuals (33.8%). Multivariable-adjusted DASH adherence declined from 22.1% in 2003-2010 to 18.6% in 2011-2018 (p=0.003), and mean DASH score declined from 2.98 to 2.85 (p=0.007). Adjusted odds of adherence were lower in 2011-2018 compared with 2003-2010 (OR, 0.78; 95% CI, 0.67-0.91; p=0.002). Adherence was lower among non-Hispanic Black adults (OR, 0.70; 95% CI, 0.59-0.84), adults with BMI 30 kg/m2 (OR, 0.61; 95% CI, 0.51-0.73), and adults with prior stroke (OR, 0.71; 95% CI, 0.56-0.89). Total fat and sodium threshold attainment declined from 20.9% to 18.7% and 13.8% to 11.8%, respectively. Conclusions: DASH adherence among U.S. adults with self-reported hypertension declined from 2003-2011 to 2011-2018. Implementation strategies should identify barriers to adoption and address the full DASH dietary pattern.

Importance: Dyslipidemia, hypertension, and diabetes are highly prevalent conditions for which medical nutrition therapy (MNT) is guideline-recommended, yet access remains limited, and real-world evidence on technology-enabled MNT models at scale is sparse. Objective: To characterize lipid, blood pressure, and glycemic outcomes among adults receiving virtual MNT (vMNT) across three cardiometabolic conditions. Design: Retrospective cohort study of adults with a first Nourish appointment before November 2025, followed for 6 months. Setting: National vMNT platform. Participants: Adults meeting enrollment criteria for dyslipidemia, hypertension, or diabetes and remaining engaged in care with [&ge;]1 follow-up measurement at 3 or 6 months. Final cohorts: dyslipidemia (n=6,501), hypertension (n=13,704), diabetes (n=2,444); 18,965 unique patients overall. Exposure: vMNT delivered by a registered dietitian via telehealth visits, with technology-enabled support between visits. Main Outcomes and Measures: Outcomes were LDL cholesterol, total cholesterol, HDL cholesterol, and triglycerides (dyslipidemia); systolic and diastolic blood pressure (SBP, DBP) (hypertension); and HbA1c (diabetes), assessed at enrollment, 3, and 6 months. Weight was a secondary outcome. The proportion achieving guideline-recommended targets (HbA1c <7.0%, SBP <130 mmHg) was also assessed. Results: Among 18,965 patients (mean [SD] age, 42.9 [11.6] years; 15,770 [83.2%] female), significant changes from baseline to 6 months included: LDL cholesterol (-14.0 mg/dL [SE, 0.56]); total cholesterol (-19.4 mg/dL [SE, 0.61]); HDL cholesterol (-0.5 mg/dL [SE, 0.15]); triglycerides (-28.6 mg/dL [SE, 1.06]); SBP (-5.5 mmHg [SE, 0.16]); DBP (-4.5 mmHg [SE, 0.11]); and HbA1c (-1.37 percentage points [SE, 0.04]) (all P<.001). Among those with elevated SBP, 57.3% achieved <130 mmHg by 6 months; among those with elevated HbA1c, 54.7% achieved <7.0%. Outcomes among patients with stable medication regimens were similar to the full cohort. Conclusions and Relevance: In this study of 18,965 patients, vMNT delivered through a technology-enabled care model was associated with clinically meaningful improvement across three prevalent cardiometabolic conditions. These findings support technology-enabled virtual care as a scalable approach to delivering effective MNT.

Background: Hypertension (HTN) and Type 2 diabetes mellitus (T2DM) are leading drivers of the global cardiometabolic disease burden and major contributors to preventable morbidity and mortality. In Hong Kong, where population ageing and lifestyle-related risk factors are increasingly prevalent, both conditions and their complications pose substantial challenges to the healthcare system. Although territory-wide prevalence figures are available, small area district-level evidence on the burden of these diseases and their associated complications remains limited. Disparities are likely due to heterogeneity in demographic composition, socioeconomic status, and primary care accessibility across Hong Kong's eighteen administrative districts. There is also a lack of accessible, age-adjusted, district-level disease visualization in Hong Kong. Objectives: This study estimates the age-standardized prevalence rates (ASPR) of HTN, T2DM, and their associated complications across the eighteen districts of Hong Kong, formally evaluate the statistical significance of inter-district disparities, and rank the districts according to disease burden to inform place-based health planning. Methods: This study analyzed the prevalence of chronic diseases from 2014 to 2023 in 18 districts among an adult cohort aged equal or above 35 years using data collected from the Hong Kong Hospital Authority (HA) electronic health records, using the International Classification of Diseases, Ninth Revision (ICD-9) diagnostic codes, the International Classification of Primary Care (ICPC) codes, and the disease flags to identify diseases. Sample weights are used to reflect the total population. HTN, T2DM, and their major complications were identified according to standardized diagnostic criteria. ASPR were calculated for each district with the WHO World Standard Population as the standard. Inter-district disparities in disease prevalence were assessed using Cochran's Q test, with statistical significance set at P <0.05. The slope index of inequality (SII) was also used to analyze the socioeconomic inequality between districts. Districts were subsequently ranked according to their ASPR for each condition to identify high-burden and low-burden areas. Results: 1,756,521 patients (52.71% females) contribute 9,906,729 patient-years from 2014 to 2023. Substantial inter-district variation and increasing trend were observed in the ASPR of HTN, T2DM, and their complications. The weighted ASPR of HTN varied by 27.3% from 46,701 (95% CI: 45,691-47,711) per 100,000 population in Tsuen Wan district to 59,445 (95% CI: 58,759-60,130) per 100,000 population in Yuen Long district, and the weighted ASPR of T2DM ranged from 20,464 (Tsuen Wan, 95% CI: 19,803-21,124) per 100,000 population to 28,275 (Yuen Long, 95% CI: 27,792-28,758) per 100,000 population across the districts in 2023. All diseases showed district disparities (all P <0.05). Cochran's Q test demonstrated statistically significant disparities across districts and showed no convergence through the period. The SII test also showed the most serious inequality exists in HTN, with the lower socioeconomic position districts have more burden. Conclusion: Significant geographic disparities exist in the prevalence of HTN, T2DM, and their complications across Hong Kong's eighteen districts. These findings highlight the need for district-specific public health strategies, equitable allocation of primary care resources, and targeted chronic disease prevention and management programs in high-burden areas.

Background. Refractive error is a leading cause of childhood visual impairment worldwide, yet longitudinal, multi-city data from Vietnam remain scarce, especially for children younger than school age. We aimed to describe the prevalence and secular trends of refractive error in a large urban Vietnamese school health program. Methods. We conducted a retrospective analysis of repeated cross-sectional school health screening data collected from 2021 to 2024 across three major Vietnamese cities (Hanoi, Ho Chi Minh City, and Hai Phong), among children aged 3 to 18 years. The dataset included 141,339 examinations, with a Hanoi sub-cohort of 19,788 examinations that had quantitative autorefractor measurements. We estimated crude prevalence and adjusted prevalence ratios using Poisson regression. Results. The prevalence of any refractive error stayed high throughout the study period, ranging from 53.9% to 57.1% (P for trend < 0.001), and was higher in Hai Phong (68.4%) and Hanoi (61.6%) than Ho Chi Minh City (35.9%), higher in girls than boys (58.5% vs 53.1%), and higher with older age (P < 0.001). In the Hanoi sub-cohort, confirmed myopia prevalence nearly tripled, from 3.7% in 2022 to 11.0% in 2024 (P < 0.001), while hyperopia declined. Conclusions. Refractive error is highly prevalent among urban Vietnamese schoolchildren, and confirmed myopia appears to be rising rapidly at younger ages. These findings support closer surveillance and earlier school-based intervention in Vietnam and similar settings.

Background and objective. The WHO growth reference judges a child's height against the population at one age, without using the child's own earlier growth. This study built and validated a quantile regression based conditional growth reference predicting a child's future height for age distribution from early growth status. Methods. Using deidentified longitudinal annual school children health check data from three major cities in Vietnam, this study linked a baseline height and BMI for age Z score near age 6 to outcome visits near age 12 or 15 in 4,857 children. A linear quantile regression and a pooled quantile generalized additive model, fit by sex, were compared against a linear benchmark and a naive persistence model on a held out test set, using pinball loss, interval coverage, and width against the WHO 90% band. This study also tested adding growth phenotype and checked generalization across the three cities. Results. The pooled model outperformed the discrete quantile regression in every group. Every individualized 90% interval was narrower than the WHO band, by 34.6 to 54.5%. Growth phenotype improved accuracy in most groups. The model generalized well to a held out Hanoi or Haiphong but poorly to a held out Ho Chi Minh City, where coverage fell to between 0.35 and 0.52, though the all-city model stayed well calibrated for Ho Chi Minh City itself. Conclusion. Conditional quantile modeling gives individualized growth predictions markedly narrower than a fixed population reference and gains further from growth phenotype, but needs local validation before use at an unrepresented site.

Cancer prevalence is distributed unevenly across regions and caused by the interaction of multiple risk factors. Previous studies focused on the use of global modeling techniques to predict cancer at the county level that overlooks important spatial differences. This study aims to develop geographically weighted machine learning models to predict cancer prevalence at the census tract level in the United States and identify local determinants of cancer burden. First, a scoping review was conducted to find a list of measurable drivers of cancer in the United States. Using this list, the data of these variables for 84415 census tracts were obtained from the Center for Disease Control and Prevention PLACES dataset and other publicly accessible resources. Then, several predictive models, including Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR), as well as Random Forest, XGBoost, and Deep Neural Network and their geographically weighted counterparts, were developed and compared using the Coefficient of Determination, Root Mean Square Error, and Absolute Error. Results presented that geographically weighted models outperformed other methods, and geographically weighted XGBoost achieved the strongest and most consistent overall performance with pseudo-R2 ranging between 0.89 and 0.98. Feature importance analysis of this model illustrated that most important cancer drivers changed location by location. Aged people, racial composition, preventative behaviors, and metabolic conditions such as diabetes, hypertension, and high cholesterol were determined as influential predictors, although their relative importance varied across regions. These findings revealed the value of localized models at a small geographic scale to identify regional cancer risk patterns and help the allocation of proper resources to hotspot areas. Keywords: Cancer prevalence, Census tracts, geographically weighted machine learning models, Deep neural network, XGBoost, Random Forest, Ordinary Least Squares, risk factor, determinant

Background: Artificial intelligence (AI) is increasingly used in healthcare, but data on its real-world use and perceptions among obstetricians and gynecologists remain limited. Objectives: To assess familiarity, use, perceptions, and expectations regarding AI among physicians practicing obstetrics and gynecology in France. Study Design: We conducted a cross-sectional survey between June 2 and July 30, 2025, mainly distributed through professional organizations or associations and social media. Participation was voluntary and self-selected. AI use was defined as any use of an AI tool in routine clinical practice, regardless of type or frequency. Comparisons were made between AI users and non-users and between junior and senior physicians. Results: Among 469 respondents, representing approximately 5% of the target population, 199 of 466 (42.7%) reported AI use in practice; three respondents were excluded from this analysis because of discordant questionnaire responses. AI use was associated with male gender and professional status, with residents more represented among users. Junior physicians reported greater AI use than senior physicians and expressed greater concern regarding medical confidentiality. Current applications primarily involved administrative tasks, medical writing, literature retrieval, and translation; anticipated future uses centered on clinical decision support and medical data interpretation. Perceptions of several AI applications differed between groups. Conclusions: AI was reported to be used by a substantial proportion of respondents, primarily for supportive and informational tasks. These findings support further evaluation of training, validation, governance, and safe implementation of AI in obstetrics and gynecology.

Background Inadequate spectacle correction limits the benefit of school vision screening, yet most evidence addresses only whether a child owns glasses rather than whether those glasses work. This study examined the prevalence, predictors, and persistence of inadequate spectacle correction among schoolchildren in three Vietnamese cities. Methods We analyzed annual vision screening records of students from a private school network in Ha Noi, Ho Chi Minh City, and Hai Phong, 2021 to 2024. Inadequate correction was defined as visual acuity below the age-appropriate threshold despite current glasses. A population-averaged modified-Poisson regression with robust standard errors, clustered by child, estimated adjusted prevalence ratios (aPR) for city, age, sex, and calendar year. Children linked across consecutive years allowed persistence analysis. Results Among 39,697 spectacle-wearing examinations, 30,936 (77.9%) were inadequately corrected, ranging from 64.8% (95%CI= 63.9, 65.8) in Ho Chi Minh City to 96.7% (96.0, 97.3) in Hai Phong. After adjustment, prevalence was lower in Ho Chi Minh City (aPR= 0.76; 95%CI= 0.75, 0.78; p<0.001) and higher in Hai Phong (aPR= 1.25; 95%CI= 1.23, 1.26; p<0.001) relative to Hanoi, and fell with age (aPR= 0.97 per year; 95%CI= 0.96, 0.97; p<0.001). Among 9,285 linked year pairs from 8,341 children, 85.3% (84.4, 86.1) of initially inadequately corrected children remained so the following year. Conclusions Most spectacle-wearing children in this Vietnamese cohort were inadequately corrected, with wide geographic variation and high persistence. School vision programs need mechanisms to verify correction adequacy after glasses are dispensed, not only to detect refractive error.

Background: Hypertension is the most important modifiable cardiovascular risk factor worldwide, and the prevalence is increasing in sub-Saharan Africa. Nigeria's newly released 2023-24 Demographic and Health Survey (NDHS) provides the first opportunity to explore national-level non-uptake of antihypertensive treatment using a machine-learning cascade framework, although survey fieldwork was conducted amid the unprecedented shock of fuel subsidy removal and currency devaluation in Nigeria. Objectives: To identify correlates of antihypertensive treatment non-uptake (Gap 2) among the diagnosed adults in Nigeria; test if the non-uptake varied based on when the survey was conducted during this macroeconomic shock; and compare four predictive algorithms. Methods: We used the 2023-24 Nigeria Demographic and Health Survey (NDHS) data (2,975 women diagnosed with hypertension and 529 men diagnosed with hypertension aged 15-49) to fit survey-weighted logistic regression models separately by sex and evaluated sex-by-predictor interactions in a single pooled model. We then compared the survey-weighted logistic regression model with the elastic net, random forest, and gradient boosting (XGBoost) models, all properly weighted, using held-out test-set AUC. Results: The timing of fieldwork interviews was a significant risk factor for women (OR=1.14 per month; 95% CI: 1.06-1.22; p<0.001) but not for men, and this sex difference was confirmed in the formal interaction test (p=0.016). This study tested the interaction of sex with each geopolitical zone, with the strongest result being a sex-reversed pattern in zones protective for women, showing these zones were significant risk factors for men (ORs 2.28-5.77); interaction testing (all p[&le;]0.007) confirmed this. No interaction was observed between diabetes and sex (p=0.170); instead, diabetes was protective for women (OR=0.34, p<0.001). For pooled samples, wealth was associated with non-uptake (OR=0.59; p=0.012), and this appeared significant only for women (OR=0.62; p=0.026), in keeping with the richest-vs-poorest estimate in Table 2. All three ML algorithms had similar AUC for women (0.61-0.63) and lower, near-chance performance for men (0.56-0.59), reflecting the much smaller number of diagnosed men. Conclusions: There is a large disparity in the rate of treatment and management of HTN across sexes, geopolitical zones, and macro-economic environments, making a single uniform national intervention model impractical. In contrast, an earlier study in Africa reported that similar algorithmic complexity was useful in predicting cardiovascular risk; this did not happen in the current study. The proposed programs are to build programs by sex and zone, integrate hypertension screenings into chronic disease and/or maternal health trigger points, and better understand the resilience of the pharmacy supply chain.

Background: Knowledge translation (KT) is essential for ensuring that research evidence informs policy and practice. However, little is known about researchers' intentions to engage in KT activities or their relationship with conventional bibliometric indicators. This study aimed to (1) assess researchers intentions to engage in KT activities using a questionnaire informed by the Theory of Planned Behavior (TPB), and (2) examine the association between these intentions and individual bibliometric indicators. Methods: A cross-sectional survey was conducted among faculty members at Tehran University of Medical Sciences. A questionnaire informed by selected constructs of the TPB was developed to assess researchers intentions to engage in activities across the seven components the action cycle of Knowledge-to-Action (KTA) framework. The instrument was validated through several rounds of expert review and then with a sample of potential participants. Bibliometric indicators (H-index, Number of Publications, Number of Citations, First-Author Publications, Corresponding Authorship, International Publications and Self-Citations) were extracted from the national academic registry. To assess representativeness, bibliometric characteristics of respondents were compared with those of a random sample of non-respondents drawn from the same registry. Confirmatory Factor Analysis (CFA) validated the intention construct, and Structural Equation Modeling examined associations with demographic and bibliometric variables. Results: A total of 143 researchers participated. Intention scores were highest for adapting knowledge to local contexts and identifying barriers and facilitators, and lowest for monitoring and sustaining knowledge use. CFA supported a unidimensional intention construct with favorable model-fit indicators. No bibliometric indicator was significantly associated with KT intention. Age was the only significant predictor. Respondents had stronger bibliometric profiles than non-respondents, indicating potential non-response bias. Conclusions: Researchers reported relatively strong intentions to engage in KT, but these intentions were not associated with conventional bibliometric indicators. Age was the only significant predictor. Overall, findings suggest that traditional academic metrics may not reflect KT intentions and highlight the need to strengthen institutional incentives for KT.

Despite the numerous benefits associated with higher dietary quality, many older Veterans do not meet national dietary recommendations. Existing programs are not typically tailored for older Veterans who may experience physical and environmental limitations that hinder their ability to access and prepare healthy foods. This project piloted the incorporation of age-friendly, tailored, virtual nutrition education into an existing health promotion program for older Veterans. Programming followed an iterative design process over two years. Year 1 included 14 weeks of virtual nutrition education on healthy aging paired with weekly produce bags and recipe suggestions. In Year 2, virtual nutrition education shifted to more hands-on learning emphasizing basic meal preparation and adaptive cooking strategies for mobility limitations. Produce distribution in Year 2 shifted to a farmer's market model, based on feedback from year 1, allowing participants to self-select items. Primary outcomes included program satisfaction, dietary intake assessed via 24-hour recall to calculate the Healthy Eating Index (HEI-2015), and health-related quality of life (PROMIS Global Health). Findings demonstrate the feasibility and acceptability of a virtual nutrition program incorporating age-friendly cooking instruction and adaptive strategies to improve dietary quality among older Veterans.