JMIRx Med

Background. Regulatory pressure to avoid confusingly similar drug names appears to have pushed manufacturers to coin more distinctive proprietary names, but no one has quantitatively analyzed this trend. Objective. To determine whether proprietary drug names became more orthographically distinctive (unusual in spelling) over time. Methods. We examined 1,321 US Food and Drug Administration (FDA) new molecular entity (NME) names (1985-2024), 2,769 FDA Orange Book trade names (1982-2026), 177,337 US pharmaceutical trademarks (1980-2026), and 1,273 generic names (1982-2025). Compared to prior names (including those already marketed) and ordinary English, we scored each name's character trigram (three-letter sequence) surprisal (statistical unexpectedness) and its edit-distance similarity to the closest existing names. Linear regression tested time trends, controlling for name length. Results. Mean NME trigram surprisal rose in every decade between the 1980s and 2020s, from 7.21 to 8.50 nats (natural-logarithm units, higher meaning less predictable), an 18% increase (b = 0.038, p < .001, R2 = 0.34). Half of 2020s names (7% in the 1980s) contained a letter sequence found in no common English word (e.g., Jesduvroq, Omvoh). Nearest-neighbor similarity fell after the early 1990s. Trademark data showed the same trend (b = 0.016, p < .001). Generic names showed a weaker trend (b = 0.011, R2 = 0.063). Conclusions. Over four decades, drug names became more orthographically distinctive, consistent with namespace crowding, deliberate efforts to avoid confusing names, and branding strategy. Because unfamiliar letter sequences can introduce other error pathways, increased distinctiveness does not necessarily imply improved safety.

The -Bonacci weighted sequence space {ell}(,) is introduced for [&ge;] 2, 1 [&le;] < {infty}, and > 0). It is proved to be a Banach space which is isometrically isomorphic to {ell}, with an unconditional Schauder basis. The space is also reflexive and uniformly convex. Its strict inclusion chain, continuous dual, and asymptotic truncation estimate of order - / are also established. Based on this framework, Scaled &#119898-BonacciNet is proposed as a lightweight CNN that rescales Fibonacci-Net channels using the dominant &#119898-Bonacci root and employs depth wise separable convolutions. Tribonacci-Net achieved 96.09% accuracy and a 0.9581 macro F1-score on the 15-class Brain MRI dataset using 0.85 million parameters and 0.366 GFLOPs. Pentanacci-Net attained 95.87% accuracy on the 44-class dataset while reducing parameters and FLOPs by 47.10% and 44.48%, respectively. On the CKD ultrasound dataset, Hexanacci-Net achieved 83.75% accuracy and a 0.7941 macro F1-score, with corresponding reductions of 50.37% and 45.10%. These findings support dominant-root scaling as a structured approach to lightweight CNN design.

Aims: To understand how users believe continuous glucose monitoring (CGM) changes their behaviour, which app features they credit, and whether endorsing more mechanisms of action (MoAs) is linked to reported diet and physical activity change. Methods: We surveyed 112 U.S. adults with type 1 or type 2 diabetes who had used a Dexcom or Abbott Libre CGM within 12 months. Participants rated 26 MoA statements mapped to the Capability, Opportunity, Motivation-Behaviour (COM-B) model, identified contributing app features, and reported CGM-attributed diet and physical activity change. Logistic regression examined MoA count and COM-B domains as predictors of behaviour change. Results: Participants (mean age 54.5 years; 60.7% female) reported dietary (82.1%) and physical activity (46.4%) change and endorsed a median of 18 MoAs, most often relating to knowledge, skills and behavioural consequences. Real-time glucose and graphs were attributed to many MoAs. Each additional MoA was associated with higher odds of dietary (OR 1.27, 95% CI 1.15-1.44) and physical activity change (OR 1.27, 95% CI 1.16-1.43). Motivation made the strongest distinct contribution among COM-B domains. Conclusions: Users perceive CGM as engaging many behavioural mechanisms; endorsing more was linked to behaviour change. Motivation may be a key target for CGM-based behavioural support.

Background: Lean Healthcare has been adopted as an approach to reducing waste, enhancing process efficiency, and optimizing clinical and administrative workflows. However, empirical evaluations aimed at establishing a causal relationship between Lean Healthcare interventions and outcomes remain limited. This evidence gap is particularly pronounced in the context of large-scale implementations in low- and middle-income countries, where more efficient allocation could yield substantial improvements in population health. Methods: We conducted an impact evaluation study employing a staggered difference-in-differences approach to estimate the effects of the Lean Healthcare in Emergency Departments (ED) initiative in the Brazilian context. The intervention was implemented between January 2017 and December 2025 across 309 hospitals. We analyzed hospital-level indicators of the number of hospitalizations and the average length of stay (in days) among patients admitted through urgent or emergency pathways. Results: Our estimates suggest that the intervention was associated with an average increase of 38.6 admissions per hospital per month (p = 0.033), corresponding to approximately a 5.9% increase relative to the monthly admission volume. We also observed a reduction of 0.184 days in the average length of stay (p = 0.058), equivalent to an approximately 3.0% decrease in hospitalization time. These effects exhibited distinct temporal dynamics: the effect on the number of hospitalizations emerged gradually and increased with the duration of the intervention, whereas the reduction in length of stay became evident earlier. Conclusion: This study provides evidence of a causal effect of Lean Healthcare implementation in emergency departments on operational performance. The simultaneous increase in admissions and reduction in length of stay suggests improved bed productivity and effective capacity without expanding physical infrastructure. These findings support the use of flow-management methodologies as an effective strategy for mitigating emergency department overcrowding while increasing service capacity.

Introduction: Fighting the ventilator (FV) is characterized by a patients active counteraction of respiratory gas flow across multiple breaths during mechanical ventilation. An effective ventilation is prevented during such episodes. The objective of this exploratory study was to evaluate clinicians knowledge, perception and agreement regarding this condition. Methods: In a cross-sectional online survey, medical professionals with experience in intensive care medicine (mostly neonatal and pediatric patients) responded to 11 questions covering various clinical aspects of FV. Results: All respondents were familiar with the phenomenon and assessed it to be clinically relevant. Neonates were perceived to be predominantly affected. Promoting factors of FV included inadequate sedation and ventilation-associated factors. In clinical practice, the participants identified FV by clinical evaluation, analysis of ventilator waveforms and capnography and used both pharmacological and non-pharmacological intervention strategies. There was no consensus concerning the English terminology of the condition. Only few respondents were familiar with local guidelines to manage FV. Conclusion: Although there was consensus on the relevance of FV, diagnostics and management strategies were highly heterogeneous. Many participants agreed that poor patient-ventilator synchrony was an important cause of FV. Future work should focus on characterizing and generating evidence on FV.

Objectives To evaluate discrepancies between radiologists and referring physicians in the interpretation of diagnostic probabilities conveyed by common Japanese expressions used in radiology reports. Materials and Methods Fifteen expressions commonly used to convey diagnostic probabilities in radiology reports were selected by five attending radiologists. An anonymous online survey was distributed to 26 radiologists and 33 referring physicians at a single institution. Participants assigned a diagnostic probability (from 0% to 100% in 10% increments) to each expression. Differences in median probabilities between radiologists and referring physicians were analyzed using the Mann-Whitney U test. Variability in interpretation was assessed using interquartile range (IQR), with IQR [&ge;] 30% defined as substantial variability. Results All radiologists (26/26) and 97% of referring physicians (32/33) completed the survey. Seven of the 15 expressions showed statistically significant differences in interpretation between radiologists and referring physicians: "c/w" (80% vs. 60%; p = 0.031), the Japanese equivalents of "present" (100% vs. 100%; p = 0.024), "suspect of" (80% vs. 70%; p = 0.007), "possibly" (50% vs. 70%; p < 0.001), "unlikely" (10% vs. 10%; p = 0.022), "cannot be ruled out" (25% vs. 40%; p < 0.001), and "not inconsistent with" (80% vs. 90%; p = 0.008). Substantial variability in interpretation (IQR [&ge;] 30%) was observed for four expressions: "cannot be ruled out" (IQR, 10%-40%), "s/o" (IQR, 60%-90%), "c/w" (IQR, 50%-90%), and "not inconsistent with" (IQR, 70%-100%). Conclusion This study revealed that nearly half of the Japanese expressions commonly used to convey diagnostic probabilities were interpreted differently by radiologists and referring physicians. These discrepancies highlight the need for a standardized lexicon to improve clarity and consistency in radiology reporting.

Background Chronic diseases are the leading causes of illness, disability, and death worldwide, with a disproportionate burden in low- and middle-income countries. Hospitalization is a critical outcome in chronic disease management, reflecting disease severity, treatment failure, and healthcare utilization. However, data on hospitalization burden and its associated factors among chronic disease patients in Ethiopia, are limited. This study aimed to determine the rate of hospitalization and its associated factors among patients with chronic diseases attending a tertiary hospital in Wolaita Zone, South Ethiopia. Methods A facility-based cross-sectional study was conducted among 321 patients with chronic diseases attending the chronic follow-up clinic at Wolaita Sodo University Comprehensive Specialized Hospital from November to December 2022. Consecutive recruitment was used, with unique medical record numbers tracked to prevent double recruitment. Data were collected through structured questionnaires and medical record review. Negative binomial regression was used to identify factors associated with hospitalization rate, with incidence rate ratios (IRR) and 95% confidence intervals (CI) reported. A p-value < 0.05 was considered statistically significant. Results Of the 321 participants, 66% reported at least one hospitalization while on treatment, with a mean of 1.70 {+/-} 1.81 hospitalizations per patient. Drug discontinuation was the most frequently documented reason for hospitalization (39.5%), followed by disease complications (22.3%) and infections (11.7%). Multimorbidity was prevalent, with the majority (59.8%) had 2-3 conditions, and a mean of 2.40 {+/-} 1.02 chronic diseases per patient. Male sex was associated with a higher hospitalization rate, while regular physical exercise was associated with a lower number of hospitalizations. Conclusion This study found a high burden of hospitalization among chronic disease patients, with drug discontinuation being the leading documented reason. Male sex and regular physical exercise were significantly associated highlighting the need for medication adherence support programs and structured exercise programs integrated into routine chronic disease care, as well as gender-sensitive interventions.

Background Influenza vaccines have seen many changes over recent years in terms of both policy recommendations and product types. Immune responses and vaccine effectiveness vary by vaccine strain composition and product type. This study aimed to explore this variation on a regional level through examining spatial and temporal autocorrelations of vaccine uptake of different product types within Michigan from 2017 to 2021. Methods Influenza vaccine administration data from the 2017/18 to 2021/22 seasons were obtained from the Michigan Care Improvement Registry. Population size estimates for 2017 and 2019 were obtained from the Michigan Department of Health and Human Services. Age-standardized and age-stratified vaccination rates were computed. Disparate county-level vaccine uptake clusters were constructed using spatial autocorrelation indicators and spatiotemporal clusters were identified using scan statistics. Linear mixed effect models were used to study the impact of social vulnerability and urbanicity level on seasonal influenza vaccination rate. Results Vaccination incidence rates for adjuvanted, recombinant, and nasal sprays vaccines and for any vaccines increased significantly through the study period. County-level spatiotemporal clusters of type-specific uptake rates were identified. Most notably, we identified an emerging cluster with high uptake of adjuvanted vaccine radiating from Ottawa County, located in suburban Grand Rapids, and the disappearance of high uptake cluster for high-dose vaccine uptake in the area. Vaccination rates by any product type were found to vary seasonally and showed weak negative association with more rural regions. The overall vaccination rates appear to be lower among counties with higher social vulnerability in certain urbanicity levels, but social vulnerability and urbanicity were not significantly associated with overall vaccination rate during the study period. Conclusions Vaccine type-specific uptake was spatially and temporally diverse in Michigan between 2017 and 2022. Mid and Northern Michigan had consistently relatively lower vaccination rate compared to other regions in Michigan.

Violence against women is a key contributor to mental health gender disparities, with greater gender gaps observed in low-and middle-income countries. This study examines state-level associations between gender-based violence against women and mental health burden in India over time. We linked 1) annual state-level age-standardised disability-adjusted life year rates (ASDR) from the Global Burden of Disease study for overall mental health disorders, depressive disorders, anxiety disorders, bipolar disorder, schizophrenia and self-harm, with 2) annual state-level gender-based violence rates from the Open Government Data Platform, India. We used descriptive ecological analyses and multilevel modelling to estimate the state-level association between gender-based violence and sex-stratified mental health outcomes over time (2001 - 2021). Results showed higher state-level gender-based violence (GBV) rates were associated with higher state-level ASDR of schizophrenia ({beta}women = 0.39 [95%CI 0.25 - 0.52]; {beta}men = 0.42 [0.29 - 0.55]) and bipolar disorder ({beta}women = 0.02 [0.01- 0.03]; {beta}men = 0.04 [0.03 - 0.05]), after adjustment for state-level social development, but not with other outcomes. Higher GBV rates were associated with a lower gender disparity in depressive disorders. Longitudinal findings revealed that the state-level association between GBV rates and ASDR of overall mental health disorders strengthened over time ({beta}women = 5.08 [3.96 - 6.20]; {beta}men = 3.38 [2.60 - 4.15]), whereas the association weakened for self-harm and schizophrenia. Lagged findings suggested that with an increase in GBV rate, ASDR of overall mental health disorders increased for men over two years and for both sexes over five years. Our findings highlight the importance of considering state-level trends in GBV and related policies, and considering impacts of GBV on both men and women's mental health.

INTRODUCTION: The digital clock drawing test (dCDT) is a cognitive screening tool capturing high-resolution pen-movement data. Existing approaches for classifying cognitive impairment from dCDTs rely on selected features or image-based methods. We propose representing dCDTs as structural networks to classify cognitive impairment and characterize spatial organization in clock drawings. METHODS: Weighted networks were constructed from 3,362 Framingham Heart Study participants, with and without cognitive impairment. Network representations were generated separately for command and copy conditions, with nodes representing pre-defined clock components (digits, hands, clock face) and edges capturing pairwise spatial relationships between components. Random forest models and graph convolutional networks were trained to predict cognitive impairment from the networks. Models were benchmarked against demographics-only and established summary feature baselines. RESULTS: Network-based models achieved AUC values comparable to the established summary feature baseline across command-only and copy-only conditions. Differences in spatial organization between the cognitively intact and impaired groups were greatest in the upper-right clock region, involving digits 1 and 2, the clock hands, and the surrounding clock face. Differences between the command and copy networks were larger among cognitively impaired participants. DISCUSSION: The proposed network representations provide an objective, scalable, and flexible framework for representing clock drawings for use in classification models. dCDT networks matched the predictive performance of established summary feature-based approaches and identified clock regions where spatial organization differed by cognitive status. After external validation, this approach has the potential to be incorporated within a clinical workflow to support automated and interpretable cognitive screening.

Financial well-being (FWB) encompasses objective and subjective financial status and may influence health. We developed and validated a parsimonious FWB index using data from 23,006 Health and Retirement Study (HRS) participants aged 50 years and older from 2006 to 2022. The FWB-5 Index was derived using correlated two-factor confirmatory factor analysis accounting for household clustering. Indicators included household wealth, household income-to-poverty ratio, monthly financial strain, satisfaction with one's financial situation, and perceived financial mastery. We assessed convergent and discriminant validity using the Consumer Financial Protection Bureau Financial Well-Being (CFPB-FWB) Scale and Satisfaction with Life Scale, respectively. Associations with pre-baseline socioeconomic factors and subsequent depressive symptoms and health conditions were examined to assess construct and predictive validity. Model fit indices supported the two-factor structure (robust comparative fit index = 0.99; robust root mean square error of approximation = 0.05). Correlations were stronger with the CFPB-FWB Scale (Spearman P = 0.80) than with life satisfaction (Spearman P = 0.43). Associations with socioeconomic factors and subsequent health outcomes were in the hypothesized directions. These findings support the FWB-5 Index as a parsimonious measure for examining financial well-being and its associations with health across HRS waves, including those predating the CFPB-FWB Scale.

Background: Conventional performance metrics do not indicate whether artificial intelligence (AI)-platforms and radiologists are right or wrong on the same examinations, nor whether their non-agreement is more often beneficial or harmful. Objective: To develop, illustrate, and apply directional metrics quantifying non-agreement between radiologists and AI-platforms against a common reference standard, and to adapt a fragility-based framework for quantifying the statistical robustness of these metrics. Methods: We defined Positive Non-Agreement (PNA, AI-platform correct, radiologist incorrect), Negative Non-Agreement (NNA, radiologist correct, AI-platform incorrect), and Net Non-Agreement (NENA, their difference). Metrics were illustrated in three hypothetical scenarios of 1,000 examinations each and applied to three previously published AI-radiologist studies on pulmonary embolism and mammography. Bootstrap confidence intervals were calculated, and a reclassification fragility index (FI) quantified the minimum case reclassification needed to remove statistical support for NENA's direction. Results: In illustration, NENA was positive in two scenarios and negative in one. In the three published studies, NENA ranged from -4.18% to +10.74% and decreased consistently with increasing radiologist experience (junior residents, +6.27%; consultants, +1.45%), a gradient driven primarily by declining PNA with modest increase in NNA. The reclassification FI was defined for all three illustrative scenarios (range, 0.10%-1.70%) and for six of the seven comparisons from these studies (range, 0.34%-4.31%), but not for the weakest comparison, whose confidence interval for NENA included zero. Conclusions: PNA, NNA, and NENA provide a directional, case-level complement to conventional performance metrics for AI-radiologist non-agreement. The accompanying reclassification FI offers a statistically grounded, generalizable method for quantifying their robustness.

Objectives: To determine how strongly clinical referral text, rather than image content, drives the decisions of multimodal large language models (LLMs) differentiating glioma from brain abscess on MRI. Materials and methods: This retrospective single-center study included 60 contrast-enhanced T1-weighted brain MRI examinations (30 gliomas, 30 abscesses; reference-standard diagnosis). One axial slice per case was paired with five referral texts (explicitly inflammatory to explicitly tumor-suggestive) under three prompt instructions and submitted 20 times each to GPT-4o, Claude Sonnet 5, Gemini 3.5 Flash, and Qwen 3.6 (72,000 answers). The primary outcome was the text-driven span (difference in tumor-call frequency between explicit inflammatory and explicit tumor text). Two blinded radiologists classified the images without text; sensitivity, specificity, and accuracy were compared descriptively. Results: Patients (29 women) had a mean age of 55.8 +/- 16.4 years. With neutral text, models answered tumor in 45-67% of queries (specificity, 0.38-0.78), versus 52% and 60% for radiologists (specificity, 0.77 and 0.63). The text-driven span was 0.79-1.00 (p < 0.001 for all models), whereas confirmed gliomas and abscesses differed by only 0.01-0.10 in three models and 0.25 in Gemini 3.5 Flash. Subtle wording produced at least 83% of the span in three models. An instruction to ignore the text reduced the span to 0.48-0.70 in three models and to 0.04 in Gemini 3.5 Flash, partly through reversion to a tumor default. Model accuracy was 0.52-0.73 versus 0.78 and 0.73 for radiologists. Conclusion: Multimodal LLMs deferred to referral text over imaging evidence in differentiating glioma from brain abscess, and prompt instructions did not reliably correct this.

The aim of this study was to develop a framework for assessing the incremental value of decisions on drug amount, timing, input structure, cessation and restart. The LOGOS (Longitudinal Optimization of Global Outcomes and Strategies) approach combines matched nested regimen classes, mechanism-specific benchmarks and evidence requirements in a single longitudinal comparison. Thus, the contribution of additional regimen freedom is assessed separately from improvements due to different resources, constraints or evaluation horizons. Nested feasible classes have nondecreasing objective suprema, whereas a strict gain requires a feasible witness above a valid bound for the smaller class. The approach is illustrated through analytical examples of periodic smoothing, matched desensitization, relaxed exposure allocation, finite-course loading and safety after cessation. In a normalized finite Emax model, integrated benefit increased from 0.629606 for three equal boluses to 0.672945 after optimization of amounts, 0.677043 for an amount-and-time candidate and 0.679553 for a globally certified loading-maintenance-coast solution. The adjacent observed gains were 6.88%, 0.61% and 0.37%, respectively, and all schedules satisfied the full-tail injury bound. In addition, a periodic factorial example separated response curvature from a joint recovery and repair process, including a null case and results dependent on feasibility constraints. The LOGOS framework allows assessment of whether additional regimen flexibility improves benefit, provides no added value or restores feasibility. These dimensionless examples support conditional mathematical and model-based conclusions; clinical superiority and safe restart in patients have not been established.

Health systems increasingly record ICD-10 Z59 social-needs codes for housing, food, transportation, and related hardship to trigger referrals, report quality, and adjust risk. However code documentation occurs at varying rates and no study has separated what the code says about a patient's hardship from what it says about being documented at all. Using a linked electronic health record and claims corpus covering 4.95 million patients in 21 identifiable health systems, we matched 26,153 adults with a first outpatient Z59 code to similar uncoded patients and, as a negative control, repeated the entire design around an administrative code. Coding ranged from 1 percent to 24 percent of patients across systems and remained highly variable after adjustment for patient mix. A hardship code was associated with 67% (95% CI=1.61, 1.74) more emergency department (ED) visits and 43% ( 95% CI=1.38, 1.49) more hospital admissions in the next year (62 versus 49 of every 100 patients with any acute event). The administrative-code reproduced 46 percent of the Z59 association for ED visits and 70 percent for admissions. In the five systems with the most complete stay dates, no hardship-specific admission excess was detectable. Homelessness codes forecast two and a half times the ED visit rate. At national planning prices, excess emergency and preventable hospital care totaled roughly $2,000 to $2,500 per coded patient-year ($2,516 base scenario; about $1,100 under an exploratory assumed-discount scenario (not an estimate of hardship-attributable spending)), with the largest emergency burden among Medicaid patients. The Z code added no increment to clinical and utilization-based risk prediction models in any system.

Background: Workplace discrimination and violence against health workers including sexual harassment is a societal problem worldwide. We aimed to assess the frequency, predictive factors, and effects of sexual harassment using a nationwide survey of physicians working in Switzerland. Materials and Methods: We conducted a self-administered questionnaire survey of physicians working in Swiss healthcare institutions. The questionnaire assessed experiences of sexual harassment at work, including frequency, types, and sources of harassment. It also asked about the impact of sexual harassment, whether incidents were reported, and what actions were taken following a report. Results: Among the 1837 physicians that participated in the survey, 575 (31.3%) had experienced sexual harassment at least once during their career. Among those reporting sexual harassment, 39.1% of women and 13.2% of men had experienced harassment at least once in their job. Women reported more verbal harassment, while men reported more physical harassment. Women were also more likely to face both forms combined. The main sources of harassment were colleagues of the same profession (46.7%), patients (42.3%), and superiors (41.0%). The most frequently reported impacts included stress (60.1%), emotional exhaustion (34.7%), reduced motivation at work (30%), and considering leaving the job (7.5%). Gender-men was associated with a lower likelihood of experiencing harassment 0.24 (0.18 to 0.31), whereas working in the German-speaking region was associated with a slightly higher likelihood of having experienced sexual harassment at least once (OR 1.35, 95% CI:1.02-1.81). Conclusions: Sexual harassment is common among physicians in Swiss healthcare institutions. It adversely affects the mental health and career trajectories of both men and women yet remains widely underreported due to fears of reprisals and inaction. Institutions should effectively intervene to ensure that reported incidents result in appropriate actions, and that those who report them receive adequate protection from reprisals.

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

Objective A diversity of approaches and cohorts have been published in research literature for suicide risk estimation, leading to some uncertainty around which factors and conditions are most strongly contributing to risk. Methods exist for weighting estimates by propensity score, removing simplifying the interpretation of model effects. Few studies have explored the use of these tools to assess when the effect of certain factors is uncertain or inconsistent. Method A mix of 1,001,737 suicide decedents and living patients were evaluated for suicide risk, relying on a subset of commonly identified factors: demographics, psychiatric conditions, substance use disorders, and social needs. Care utilization patterns were then assessed controlling for these factors in a hospital care setting. Four methods were considered for controlling multivariate confounding of odds predicting death by suicide: multivariate logistic regression, 1-to-1 sample matching, inverse propensity weighting (IPTW), and entropy balancing (EBAL). Results Test statistics showed suicide decedents were much more likely to have various psychiatric and substance use diagnoses, social needs and more likely to heightened care utilization across sampling intervals. Greater utilization was found close to date of death. Effect sizes after IPTW and EBAL weighting were also generally greater than would have been documented by regression alone. Conclusions Propensity weighting is a useful alternative to exploring nuanced effects, especially when the direction and strength of an association differ from traditional statistics. By weighting, the overall imbalance of cases remains consistent and effect terms might reflect more nuanced associations than regression or sample matching alone.

Type 1 diabetes mellitus (T1DM) remains a growing global health priority, with disproportionate underdiagnosis, delayed diagnosis, and mortality in sub-Saharan Africa. Despite the introduction of the Life for a Child (LFaC) programme in Kenya, which provides free insulin, glucose-monitoring supplies, and specialist education, no published studies have examined the determinants of glycaemic control among Kenyan children and adolescents with T1DM. This study aimed to determine the proportion of children and adolescents achieving good glycaemic control and to identify the socio-demographic and clinical determinants of glycaemic control at a tertiary hospital in Kenya. This cross-sectional study enrolled 160 children and adolescents with T1DM attending the paediatric diabetes clinic at Kenyatta National Hospital, Nairobi, between September 2025 and May 2026. Sociodemographic and clinical data were collected through structured interviews and physical examination. Poor glycaemic control was defined as HbA1c >7.5%. Associations between candidate determinants and glycaemic control were assessed using chi-square or Fisher's exact tests. Variables with p<0.2 on bivariate analysis were entered into a multivariable logistic regression model; Firth's penalized-likelihood method was used given the small number of participants achieving good control. Only 9 participants (5.6%) achieved the target HbA1c of [&le;]7.5% (mean HbA1c 11.02%, SD {+/-}2.29), and only 5 of 133 patients (3.8%) had [&ge;]70% of glucometer readings within the target blood glucose range (3.9-10.0 mmol/L).On bivariate analysis, household income (p=0.026) and self-monitored blood glucose (SMBG) frequency (p=0.018) were significantly associated with glycaemic control. In an exploratory multivariable regression analysis (Firth penalized logistic regression, given the small number of participants with good control), only SMBG frequency [&ge;]4 tests/day remained independently associated with lower odds of poor control (adjusted OR 0.19, 95% CI 0.04-0.84, p=0.028); the effect of household income was attenuated after adjustment. Glycaemic control remains markedly poor among children and adolescents with T1DM in this Kenyan tertiary hospital setting despite structured programmatic support. Frequency of self-monitoring, rather than household income alone, was the factor most strongly associated with glycaemic control, suggesting that strengthening access to glucose-testing supplies and supporting their consistent use should be prioritised in care, alongside attention to socioeconomic barriers. Longitudinal studies are needed to confirm these associations.

Computed tomography (CT) interpretation requires recognition of the anatomical level of each slice. Classification may facilitate preprocessing and AI-assisted fracture diagnosis. We developed deep learning models to classify craniofacial CT slices into six anatomical categories. A total of 11,769 images were collected from a multicenter database involving 10 Japanese university hospitals and the public Head-Neck-Radiomics-HN1 dataset. Patients were divided into training (70%) and evaluation (30%) sets. Five CNN architectures were trained from scratch, and hard- and soft-voting ensembles were evaluated. Individual-model accuracy ranged from 94.72% to 96.09%; the hard-voting ensemble achieved 96.46% accuracy (Cohen's k = 0.954). Although performance differed among individual models (Cochran's Q = 99.29, p < 0.001), the ensemble did not significantly outperform EfficientNetV2B1 (McNemar's p = 0.163). Errors occurred mainly between adjacent anatomical regions. Deep learning can classify craniofacial CT slices accurately and may support automated dataset organization and AI-assisted fracture diagnosis. External validation is needed.