The Lancet Digital Health

Predicting disease risk from prior diagnoses is fundamental to clinical decision-making, particularly during health emergencies such as the COVID-19 pandemic, when individuals with long-term conditions may be disproportionately vulnerable to adverse outcomes. Despite intense interest in developing models to predict disease risk from prior diagnoses (1-3), most prediction models do not estimate effects of each prior diagnosis on disease risk conditional on other diagnoses, limiting interpretability and clinical utility. We developed the Comorbidity Risk Score (CRS), trained on 13 million individuals (age 40-69) from linked electronic health record (EHR) datasets of the entire population of England, to predict COVID-19 hospitalisation and 87 other disease outcomes. CRS was trained at close to saturated sample size and precisely estimated the effects of 212 prior diagnoses on the 88 disease outcomes, conditional on all other prior diagnoses. Correlations of CRS effect sizes across outcomes (e.g. 0.76 for myocardial infarction vs. hyperlipidaemia) matched the corresponding genetic correlations (e.g. 0.79 for myocardial infarction vs. hyperlipidaemia), confirming that comorbidity architectures capture disease aetiology. On average, CRS identified 5% of the population with 3.4-fold higher disease risk, including myocardial infarction (4.4-fold), lung cancer (6.5-fold), and COVID-19 hospitalisation (6.3-fold). Using prior diagnoses alone, CRS outperformed state-of-the-art clinical COVID-19 models (4). Furthermore, CRS (N=13 million) substantially outperformed state-of-the-art AI (1) (N=0.5 million) and linear (3) (N=0.5 million) models in predicting disease risk, suggesting that training sample size outweighs model complexity. CRS attained near-perfect transferability across self-reported ethnicities (e.g., Black vs. White: AUROC ratio = 97.3%). Finally, CRS distinguished independently predictive comorbidities from indirect associations, e.g., lipid metabolism disorder was a strong predictor of myocardial infarction risk but not ischaemic stroke, after conditioning on other prior diagnoses. In conclusion, CRS provides a comprehensive resource for understanding the impact of comorbidities on COVID-19 and other future diseases, revealing disease aetiology while enabling powerful prediction of disease risk.

Agentic workflows can coordinate modelling, but balancing predictive performance, measurement burden and reproducibility is unclear. We developed DXA Agent, an agentic workflow for dual-energy X-ray absorptiometry (DXA) outcomes integrating planning, feature-model refinement, tools, provenance and hypothesis-generating interpretation. Models were independently developed and tested in UK Biobank (5,318 participants) and the National Health and Nutrition Examination Survey (NHANES; 3,777 participants), using cost-efficient and no-limit strategies. Across 20 UK Biobank and three NHANES bone mineral density sites, cost-efficient models achieved lower RMSE and higher R2 than the best conventional comparator, with median relative RMSE reductions of 10.9% and 9.9%, respectively. Classification was task dependent: UK Biobank osteoporosis averaged AUROC 0.839 and PR-AUC 0.182, whereas NHANES performance was comparable with conventional models. Higher-burden features did not consistently improve prediction. These retrospective, cohort-internal findings position DXA Agent as an inspectable, measurement-burden-aware research workflow requiring independent prospective validation.

Background. Routine service databases are attractive sources of training labels for clinical prediction models, but the processes that write those labels are rarely audited before the labels are used. In a deployed community cognitive-screening programme, we audited the routine cognitive-status label, built a matrix of twenty-four model arms over the same patients under a specialist reference standard, and measured what each supervision choice bought or cost. Methods. The study cohort is the 672 individuals whose cognitive status was recorded by a titled (attending-or-above) physician, that record being the reference standard; after holding out one institution entirely, a development panel of 642 individuals at 38 institutions. The routine cognitive-status label these individuals also carry was first audited at the operator level: for each data-entry account we counted diagnoses entered and the proportion recording any impairment, and tested a competing bulk-timestamp explanation. Twenty-four arms span the supervision choices such a programme faces: an incumbent 21-variable logistic regression; local language models (Qwen2.5-1.5B/3B, Qwen3-4B/8B) zero-shot, with chain-of-thought, fine-tuned on physician labels, on routine labels with and without decontamination, or on a proxy scale-band task; preference-optimised (DPO) and reinforcement-trained (GRPO) variants; a proprietary frontier model queried zero-shot; and knowledge distillation of that frontier model into the regression and into the local 4B, using 943 teacher-labelled records from the programme's unlabelled pool. All arms are scored out-of-fold under one five-fold split grouped on registry-resolved institution clusters (no cluster spans a fold); paired contrasts use a 2,000-draw cluster bootstrap. Results. 181 operator accounts (each entering at least 100 diagnoses with zero recorded impairments) account for 45,315 rows - 40.5% of the outcome column; recorded impairment falls monotonically with account volume (15.7% for 1-9 rows to 0.7% for 500-999); a bulk-timestamp explanation was tested and refuted, identifying the write-time column as a migration artefact. Under the specialist standard, no locally fine-tuned arm beat the incumbent regression (AUROC 0.926): physician-label SFT reached 0.924 (4B), DPO 0.881, and GRPO 0.789; the pre-registered two-stage proxy-then-RL recipe was worse than its single-stage contaminated baseline (-0.030, 95% CI -0.077 to -0.004). Chain-of-thought reduced discrimination at every size (-0.072, -0.080, -0.041 at 1.5B/3B/4B; -0.012, n.s., at 8B). The frontier model scored 0.932 (vs. regression +0.007, n.s.). The distilled 4B reached 0.940 - above the incumbent (+0.014, 0.004 to 0.031) and above its own teacher (+0.008, 0.001 to 0.017) - with near-teacher calibration; it reached the teacher's level by 50 teacher labels and changed little beyond 200. Conclusions. The audit and the arm matrix support one deployment recipe: audit the routine label at the operator level before training on it; do not expect fine-tuning, preference optimisation, or reinforcement learning on a few hundred specialist cases to beat a well-calibrated regression; and if a frontier model is available but undeployable, spend a bounded number of queries on it as a labelling instrument and distil. A companion paper uses these frozen predictions to quantify how evaluation design choices compare with model choice.

Background: The condition of a patient with acute stroke often changes within hours of ICU admission. Prognostic work here targets fixed endpoints predicted from admission data, and trajectory phenotyping assigns one label per patient. We used longitudinal ICU data to identify interpretable dynamic clinical states, characterize transitions between them, and relate the current state to later events. Methods: Retrospective cohort study of 6368 adults with acute stroke in MIMIC IV v3.1. The first 72 h were divided into twelve 6-hour windows, and a hidden Markov model was fitted to 21 neurological, physiological and organ support variables. State number was chosen against criteria fixed before fitting: statistical fit, restart stability, state occupancy and clinical interpretability. Generalized estimating equations related the current state to new mechanical ventilation and vasopressor use within 12 h, and to ICU death within 72 h. Eleven sensitivity analyses assessed the robustness of the state solution. Results: Four states were selected: neurologically preserved-low support, neurological impairment low support, impairment renal dysfunction and impairment-respiratory support (63.3%, 7.8%, 11.8% and 17.1% of windows). Within 72 h, 40.3% of patients changed state at least once, and transitions ran in both directions rather than along a single severity gradient. States were identified without outcome data, yet ICU mortality by last state ranged from 2.9% to 43.9%. Adjusted for age, sex, subtype and Charlson index, the current state remained associated with organ-support escalation and death. State prevalence differed by at most 1.1 percentage points between training and test sets, and 10 of 11 sensitivity analyses gave a stable four-state solution (ARI 0.754 0.955). Conclusions: The early ICU course of acute stroke can be represented as movement among a small number of clinically interpretable states. The representation was reproducible in a held out set and across admission eras, but requires validation in an independent database before any clinical use.

Background. Videolaryngoscopy still requires adjuncts or hyperangulated rescue in a clinically important minority, and bedside screening discriminates modestly. Point-of-care ultrasound (POCUS) of the anterior airway is a promising alternative, but existing prediction models are opaque or assume a pre-specified functional form. We developed and internally validated a parsimonious, fully disclosed POCUS risk equation whose form is recovered from data and whose structural properties are machine-checked by formal proof - to our knowledge the first formally verified clinical risk predictor - following TRIPOD+AI 2024. Methods. In a prospective single-centre, single-operator cohort of 259 adults undergoing elective videolaryngoscopy (no-Easy airway 68/259, 26.3%), Sequentially Thresholded Least Squares with bootstrap stability selection (B=300) screened a 71-term library of nine POCUS features and retained a seven-term logistic equation; a two-term bootstrap-stable model was pre-specified as robustness analysis. Internal validation used 5x10 repeated cross-validation plus temporal and device hold-outs, with pre-specified overfitting and optimism assessments. Five behavioural properties of the deployed equation were machine-checked in Lean 4. Results. Two interactions met the |c|/sigma_c>2 stability criterion: skin-to-epiglottis x skin-to-hyoid-bone distance and tongue volume x sagittal tongue area. The seven-term equation reached a 5x10 cross-validated C-statistic of 0.966 (optimism-corrected 0.968) and held across temporal and device hold-outs (0.94-0.97). Calibration-in-the-large matched prevalence, with cross-validated slope 0.90 attenuating to 0.625 out-of-time; standard recalibration restored 0.92 without loss of discrimination. The pre-specified two-term robustness model reproduced this performance (C-statistic 0.964-0.968; events-per-parameter 34; shrinkage 0.99), confirming the result is not an artefact of the screening stage. Net benefit over a clinical baseline was positive across 10-50% thresholds. All five Lean 4 theorems compiled without sorry. Conclusions. A sparse, formally verified POCUS equation predicts difficult videolaryngoscopy with high internally validated discrimination and quantified, modest overfitting. Because the equation was developed in a single-operator cohort and its inputs are operator-dependent, external validation requires prior harmonisation of the measurement protocol and operator credentialing.

Study Objective Little is known about where in hospital care for multiple long-term conditions (MLTC) is delivered. We aimed to describe pathways of care (ward transfers) and outcomes for people admitted to hospital for unscheduled care by MLTC status and other key sociodemographic characteristics. Design and setting Analysis of routinely-collected electronic health records from a large acute UK hospital. Participants Adult unscheduled care admissions from 1st July 2018 to 30th June 2019. The presence of two or more of 59 long-term conditions was ascertained using ICD-10 codes from previous hospital discharges. Main outcome measures Markov state transition probabilities were derived for ward moves and compared for MLTC vs no MLTC, age, sex, ethnicity and neighbourhood deprivation. Outcomes (length of stay, death, readmission, move from definitive ward) and time spent in emergency and assessment departments were compared between subgroups. Results A total of 33,252 adults, mean age 56.0 (SD 21.9) years were analysed; 14,834 (42.4%) had MLTC. People with MLTC were more likely to die in hospital (4.2 vs 1.9%, p<0.001), transfer to internal medicine wards or older peoples medicine wards, were less likely to transfer to surgical wards, had longer median length of stay (1.83 vs 0.69 days, p<0.001), stayed longer in acute medical units (15.5 vs 9.6 hours, p<0.001), and were more likely to move from their definitive ward (18.2 vs 16.4%, p=0.002). Conclusion Unscheduled hospital care pathways are complex and differ for people with MLTC, who have worse outcomes and may be less likely to receive optimal care.

Background: High-sensitivity cardiac troponin (hs-cTn) assays enable safe early discharge of patients at very low risk for myocardial infarction. We previously developed a single-sample rule-out pathway using the ARCHITECT hs-cTnI assay to risk stratify patients with suspected acute coronary syndrome. In a secondary analysis of the POC-ET (Point of Care Evaluation of High-sensitivity Cardiac Troponin) study, we evaluated performance of risk stratification with the Alinity hs-cTnI assay. Methods: Patients presenting with possible myocardial infarction in the POC-ET (NCT05665127) study were included. The primary outcome was type 1, 4b or 4c myocardial infarction or cardiac death at 30 days. Cardiac troponin I (cTnI) was measured in stored materials using the ARCHITECT and Alinity hs-cTnI assays. The sex-specific 99th percentile upper reference limit (URL) are 34 ng/L in men and 16 ng/L in women for both assays. Agreement was assessed with Bland-and-Altman limit of agreement method, Passing Bablok regression, and Pearson's correlation coefficient. Distributions of presentation measurements were compared with Kolmogorov-Smirnov test. Performance was evaluated in the overall population and prespecified subgroups. The negative predictive value (NPV) and sensitivity were determined and proportion of patients identified as low, intermediate, and high risk were calculated and modelled using ordinal logistic regression. Results: In 986 patients (60 [51-70] years, 38% female), 78 (7.9%) had a primary outcome. Strong agreement was found in the raw cTnI measurements (99% samples within the Bland-Altman limit of agreement; correlation coefficient r: 0.967 (95% CI 0.964-0.969, P<0.001); Passing Bablok regression: slope 1.12 [1.11-1.13], intercept -0.16 [-0.18 to -0.13]). At presentation, distributions of cTnI measurements by the two assays were similar (P=0.810). Both assays showed comparable diagnostic performance using a risk stratification threshold of <5 ng/L and the sex-specific diagnostic threshold, with the same NPV (Alinity 100 [99.7-100]% versus ARCHITECT 100 [99.7-100]%) and sensitivity (Alinity 100 [97.3-100]% versus ARCHITECT 100 [97.3-100]%). Similar proportions of patients stratified as low- (Alinity 67% versus ARCHITECT 67%), intermediate-risk (23% versus 24%) and high-risk (10% versus 9%) at presentation with minor reclassification. Similar efficacy was observed across subgroups stratified by sex, age, history of myocardial infarction, renal function, and symptom duration. Conclusions: The Alinity hs-cTnI and the ARCHITECT hs-cTnI assays can be used interchangeably in the assessment of suspected myocardial infarction with comparable safety and efficacy.

Background: Consumer use of AI chatbots for health advice is rising, yet triage safety relative to established services remains unclear. Australia's Healthdirect, a government-backed symptom checker with 2.4 million uses in FY2024-25, remains unevaluated against frontier large language models (LLMs), and whether premium subscriptions improve triage safety remains unexplored. This study compared the triage accuracy and safety of Healthdirect against six LLM configurations across ChatGPT, Claude, and Gemini, assessed whether paid subscriptions improve triage safety, and characterised each system's error patterns. Methods: Forty-five clinical vignettes from the Semigran et al. benchmark spanning emergency, non-emergent, and self-care categories (15 each) were evaluated across seven systems. Healthdirect was tested following a seven-rule interaction protocol. LLMs were evaluated using first-person patient-language prompts under free-tier and paid-tier conditions. Outcomes were triage accuracy, emergency sensitivity, under-triage, and critical misses, analysed using Cochran's Q, Bonferroni-corrected McNemar tests, Cohen's kappa, and Wilson intervals. Findings: Triage accuracy differed significantly (Cochran's Q = 36.79, p < 0.001). Healthdirect achieved 48.9% accuracy (95% CI 35.0% to 63.0%; kappa = 0.233) versus 73.3% to 86.7% for LLMs (kappa = 0.600 to 0.800). Healthdirect operated under conservative interactive defaults while LLMs received complete information in a single prompt, which may have disadvantaged Healthdirect. Emergency sensitivity was 46.7% versus 80.0% to 86.7% for LLMs. Healthdirect produced two critical misses; no LLM produced any across 270 evaluations (95% CI 0% to 1.4%). When LLMs undertriaged, they recommended GP care rather than self-care. No tier differences were significant (all p > 0.05), and most systems over-triaged self-care cases. Interpretation: Frontier LLMs demonstrated higher triage accuracy and safer error profiles than Healthdirect. All LLMs avoided critical misses; Healthdirect did not. Premium subscriptions did not significantly improve triage safety. These findings support clinical governance decisions about whether LLMs warrant formal evaluation alongside government-backed symptom checkers.

Active Feature Acquisition (AFA) adaptively selects which diagnostic test to order next and offers a route to reduce unnecessary laboratory testing in acute care. Existing clinical AFA evaluations, however, assume every feature can be retrieved on demand and split data at the visit level, both of which inflate apparent performance. We re-evaluate cost-aware AFA under constraints designed to reflect deployment. From MIMIC-IV we constructed a cohort of 64,766 acute admissions (39,884 patients; 21 conditions; 55 features in 30 test panels) with a patient-level split, a 12-hour decision cutoff, and a per-patient availability mask from what was actually measured, and priced panels using the 2026 Medicare fee schedule under panel-level billing. We evaluated EIG-Cost, which scores each panel by Monte-Carlo Expected Information Gain penalised by its dollar cost, against eight published methods across budgets \30--$60 over five patient-level resamples. At a $30 budget, EIG-Cost achieved the highest macro-F1 (0.188, 95% CI [0.185, 0.191]) at the lowest cost ($17.28), exceeding the strongest baseline in all five resamples (p<0.001; Cohen's d=4.0), and led at every budget. Three of the eight methods collapsed to a vitals-only baseline (macro-F1 approx 0.040), acquiring nothing even at higher budgets, a genuine failure to adapt to availability rather than a budget limitation. Despite modest absolute accuracy, EIG-Cost's probabilities were well-calibrated (expected calibration error $0.048$). Under realistic availability constraints, clinical AFA is substantially harder than full-availability benchmarks imply, several published methods fail outright, and cost-aware information-gain scoring is a robust choice in this harder setting.

Objective To measure the accuracy of clinical coding, clinician review, and a locally deployed large language model (LLM) in identifying alcohol, drug, and self-harm involvement in emergency department (ED) attendances, and quantify prevalence. Design Two-phase diagnostic accuracy study. In a validation week, the identification strategies were assessed against a conflict-adjudicated reference standard (n=2,256); the LLM was then applied to n=105,096 annual attendances at the same site. Setting UK Type 1 Emergency Department treating patients [&ge;]16yrs. Main outcome measures Prevalence quantification compared with the reference standard; sensitivity, specificity, and balanced accuracy of each strategy; monthly identification rates and adjusted annual prevalence. Results The reference standard identified 12.1% of attendances as involving alcohol, drugs, or self-harm (coding 6.0%; clinician 10.0%, LLM 15.6%). LLM balanced accuracy matched or outperformed clinician review in all three domains (alcohol 0.942 v 0.930, p=0.635; drug 0.959 v 0.791, p<0.001; self-harm 0.982 v 0.908, p=0.004). Coding recorded 1.07 domains per identified patient against 1.32 in the reference standard. Adjusted annual prevalence corresponded to 12,890 domain involvements per year not identifiable in coded data. Subdomain classification found at least 81.6% of self-harm attendances required medical assessment for injury or overdose before psychiatric review. Conclusions Clinical coding identified fewer than half of presentations involving alcohol, drugs, and self-harm and rarely captured co-occurring domains; under-recording was present across a full year. A locally deployed LLM generated more complete structured data from existing clinical text within NHS infrastructure, at a scale which is not feasible for manual review.

Background: Systematic reviews of clinical prediction models increasingly include studies using artificial intelligence (AI) and machine learning (ML) methods alongside traditional multivariable regression approaches. A previously published Excel tool enabled standardised data extraction using the CHARMS checklist and risk of bias assessment using PROBAST. The recent publication of the PROBAST+AI framework, which distinguishes the assessment of model development quality from the assessment of model evaluation risk of bias and assesses applicability in both parts, necessitates an updated digital instrument applicable across prediction modelling methods. Methods: We updated an open-access Excel tool to incorporate the full PROBAST+AI framework. The updated template incorporates structural separation between assessment of model development quality and model evaluation risk of bias, with applicability assessed in both parts. It also incorporates updated signalling questions, including those addressing methodological issues particularly relevant to AI/ML, and automates the generation of summary tables and graphical displays. Results: The updated tool (CHARMS & PROBAST+AI Template) contains 11 worksheets and supports data extraction and appraisal for up to 30 prediction models. Dedicated, linked worksheets enable separate assessment of model development and model evaluation, with Domain 4 distinguishing among Apparent, Internal, and External evaluation settings. Key updates include dedicated assessments for predictor pre-processing, class imbalance handling and recalibration, data leakage prevention, and replication of the full model development pipeline within resampling procedures. Automated sheets dynamically format tables and summary charts covering PROBAST+AI parts. Conclusions: The CHARMS & PROBAST+AI Excel template provides a standardised, user-friendly, and rigorous digital framework for systematic reviewers appraising traditional statistical and AI-driven clinical prediction models.

Artificial intelligence applied to routine electrocardiograms (ECGs) has largely focused on detecting existing disease or predicting individual cardiovascular outcomes. Whether ECGs can support prediction of multiple future diseases across organ systems remains unclear. We developed ECG-RISK, a multitask survival model for 67 incident three-character ICD-10 endpoints using ECG waveforms, demographic characteristics and routinely collected laboratory data from 86,673 MIMIC-IV patients. Discrimination was highest for heart, brain, kidney and lung endpoints, with organ-level C-indices ranging from 0.796 to 0.825, whereas liver and pancreatic endpoints showed lower discrimination. The ECG-only model achieved strong discrimination across most endpoints, whereas the incremental improvement gained by incorporating ECG and laboratory inputs beyond demographic information varied substantially across endpoints. Across the nine exploratory aggregated outcomes, Kaplan Meier curves showed clear separation among model-score tertiles. Discrimination was highest for dementia (C-index, 0.891) and heart failure (C-index, 0.857). These findings support the feasibility of ECG-based longitudinal risk prediction across multiple diseases. External validation and competing-risk analyses are required to assess generalisability and clinical utility.

Large language models perform well on medical examinations, but users routinely challenge their answers and invoke professional roles, and it is unclear what a system does when a medical credential and a stated task-specific accuracy point in opposite directions. In a factorial experiment on 480 items from four Polish specialty examination sets and three consumer large language model systems (ChatGPT, Claude, Gemini), each item and system received eleven independent conversations. Conditions crossed attributed source role (medical student, experienced specialist), stated prior accuracy on similar questions (2/10, 8/10) and suggestion correctness. The primary outcome was adoption of a prespecified incorrect option when the baseline answer matched the official key, comparing a specialist described as 2/10 with a student described as 8/10. Baseline agreement with the key was 87.2% across 15,683 analyzable conversations. The incorrect option was adopted more often from the specialist described as 2/10 than from the student described as 8/10 (10.2% vs. 7.6%; adjusted risk difference +2.82 percentage points, 95% CI +0.65 to +4.99). Estimates varied across the three systems and only one system-specific interval excluded zero. In a prespecified exploratory analysis with a shared eligibility rule, correct suggestions were adopted far more often than incorrect ones (risk difference +35.7 percentage points, 95% CI +30.8 to +40.7), indicating selective rather than indiscriminate compliance. An incorrect suggestion from a specialist with low stated accuracy was therefore slightly more influential than the same suggestion from a student with high stated accuracy, although the difference was modest and varied across systems. Agreement reached only after a user has disclosed a preferred answer should not automatically be treated as an independent second opinion, and medical large language model systems should be evaluated on how they revise answers after such disclosure, not solely on initial accuracy.

The analysis uses a global 2010-2023 panel comprising 3,038 economy-years across 217 economies. It explicitly separates between-economy and within-economy estimands and tests the longitudinal interpretation using an identical-sample temporal analysis with cluster-aware coefficient contrasts, a formal isometric log-ratio sensitivity analysis, independent fixed-effects replication, and wild-cluster-bootstrap inference. The central finding is deliberately calibrated: cross-economy agreement cannot validate national sugar availability for longitudinal obesity surveillance. The study identifies temporal and construct instability without claiming that sugar is protective or that the mechanisms producing the instability have been identified. The manuscript aligns well with PLOS ONEs emphasis on technically sound, transparent and reproducible research of broad relevance. All data required to reproduce the findings, complete metadata, executable code, full-precision results, diagnostic outputs and a completed STROBE checklist are provided as S1-S5. Figures are provided separately as compliant 350-dpi TIFF files. The study used only publicly available, aggregated economy-year statistics and involved no individual participants, identifiable information or biological specimens; institutional ethics review and consent were therefore not required. This is original work; it is not under consideration elsewhere, and the sole author has approved the submission and accepts responsibility for its content. Funding and competing-interest declarations will be entered accurately in the submission portal. An Academic Editor with expertise in nutritional epidemiology, global health metrics, longitudinal panel methods, or food-system surveillance would be well placed to assess the work.

Routine childhood growth surveillance is a cornerstone of pediatric care. Growth pattern abnormalities are often early manifestations of chronic disease. Yet subtle abnormalities are frequently underrecognized, leading to diagnostic delays and avoidable morbidity. We introduce SPROUT (System for Pediatric Recognition Of Undiagnosed Trajectories), a generalized, multi-agent large language model (LLM) reasoning system designed to identify a broad spectrum of pediatric growth-related conditions from longitudinal electronic health records (EHRs) earlier than standard clinical practice. Using a large pediatric primary care EHR dataset, we developed and validated SPROUT as a two-stage system. First, a highly specific LLM screener flags concerning longitudinal growth patterns. Second, an Orchestrator module coordinates a multidisciplinary panel of LLM agents to generate a ranked differential diagnosis. To correct systemic reasoning errors, a Trainer module injects meta-knowledge into the panel via a dedicated "Learner" agent. Diagnostic capability was evaluated using a walk-forward, visit-by-visit simulation leading up to the diagnosis date. The SPROUT screener model achieved 98% (83/85) specificity and 28% (9/32) sensitivity on a gold-standard dataset of pediatric primary care patients when evaluated one year before the index date, and 100% specificity and 47% sensitivity when evaluated using longitudinal data up to the day of diagnosis. When applied to 300 control patients (i.e., healthy or undiagnosed), the screener flagged 15. Subsequent expert panel review confirmed high suspicion for undiagnosed pathology in 33% (5/15) of these cases. In chronological walk-forward validation on disease cases, the diagnostic engine identified conditions well before standard-of-care documentation. One year prior to clinical diagnosis, the system achieved sensitivities of 81% for type 1 diabetes mellitus, 56% for pituitary disorders, and 44% for celiac disease. The SPROUT multi-agent system demonstrates the ability to detect a significant portion of latent growth-related pediatric conditions months to years before current clinical standards while minimizing false positives. These results support its potential as a decision support tool for reducing diagnostic delays in pediatric care.

Large Language Models (LLMs) are increasingly used by clinicians and patients for medical queries, yet their accuracy and safety at the specialist level in hematology remain insufficiently characterised. We benchmarked ten frontier proprietary and open-weight LLMs across two generations on 1,477 board-style hematology multiple-choice questions (MCQs) derived from five educational datasets spanning nine disease areas and six clinical skill domains, including text-only and multimodal case vignettes. Claude Opus 5 had the highest mean accuracy (92.7% text, 76.9% multimodal), followed closely by Gemini-3.1 Pro (91.4% and 78.7%), Gemini-3.6 Flash (91.0% and 74.8%) and GPT-5.6 Sol (89.9% and 76.7%). Accuracy significantly correlated with model size both for text-only and multimodal MCQs. Between model generations, the largest improvements in accuracy were seen for open-weight models whereas proprietary models showed only marginal gains. In error analysis, top-performing models exhibited highly concordant failure patterns, suggesting shared limitations on challenging cases. Frontier LLMs exhibit substantial specialist hematology knowledge across diverse subspecialist domains and clinical skill sets. Yet, despite high accuracy on board-style questions in hematology, continuous expert-on-the-loop output monitoring is paramount.

Purpose: This study assesses the association between colorectal cancer (CRC) screening and a validated, housing-based measure of individual-level socioeconomic status (SES, called HOUSES hereafter) within rural communities and determines whether HOUSES-integrated geospatial analysis can be used to tailor interventions. Methods: We used CRC screening data from a subset of Mayo Clinic Midwest patients living in cities without ready access to routine care in the Mayo Clinic Health System in 2019 to represent rural communities. At the individual level, we assessed the association between CRC screening rates and the HOUSES index, adjusting for age, sex, race/ethnicity, comorbidity, distance from home address to clinic, and area deprivation index, using a multilevel mixed-effects logistic regression model. Additionally, we conducted geospatial analysis to examine the correlation between hotspots of 1) lower CRC screening rates and 2) lower SES of the subject population (HOUSES quartile 1). Findings: Among 34,489 individuals (median age 64.0 years, 52.4% female), those with the lowest SES (HOUSES Q1) had 37% lower odds of being CRC screening adherent than those with the highest SES (HOUSES Q4) (adj. OR [95% CI]: 0.63 [0.58-0.69]). In the 14 identified HOUSES Q1 hotspots, there was a significant correlation in counts of HOUSES Q1 and low CRC screening (correlation coefficient=0.81). Conclusion: Lower SES was significantly associated with lower CRC screening among rural populations. HOUSES-enabled geospatial analysis identified geographic hotspots with lower CRC screening rates for targeted interventions to address disparities in CRC screening in rural communities. HOUSES may be a useful digital tool for cancer preventive care and research.

Contemporary national-scale electronic health record (EHR) trends in documented acute myocardial infarction (AMI) rates during the high-sensitivity cardiac troponin (hs-cTn) and Type 2 myocardial infarction (T2MI) era are not well characterized. We conducted a serial cross-sectional analysis of U.S. adults aged 18 years in Epic Cosmos from 2016-2024, encompassing 821,859,867 patient-years. Age- and sex-standardized AMI diagnosis rates increased 75.7%, from 343.1 to 602.7 per 100,000 patients. This increase was predominantly driven by T2MI, which increased 133.8% from 99.9 per 100,000 in 2018 to 233.4 per 100,000 in 2024; NSTEMI increased 13.8% while STEMI decreased 4.1%. Annual hs-cTn-tested encounters increased 34.5-fold from 2017 through 2024. The proportion of tested encounters associated with any AMI remained relatively stable after 2021, whereas T2MI continued to increase and surpassed NSTEMI in 2024 as the most frequently diagnosed AMI subtype per hs-cTn-tested encounters. Males had higher absolute AMI rates across all age groups, although relative increases were greater among females. Documented AMI epidemiology shifted substantially toward T2MI during expanding hs-cTn utilization, underscoring the need for evidence-based approaches to the evaluation and management of T2MI.

Acute kidney injury (AKI) is a frequent complication within intensive care units, with its sudden onset often missed. This is especially important because a timely window for intervention is required as delayed detection leads to progressively worse outcomes. Existing machine learning and deep learning models have contributed to closing this gap, but their complexity, requiring hundreds to thousands of features, and lack of generalisation pose a limitation that prevents them from being integrated into clinical workflows across different electronic health-record ecosystems. This study presents a 37-feature XGBoost model trained on the MIMIC-IV dataset with 5.4% positive cases, with hyperparameters optimised via Optuna and probabilities calibrated using isotonic regression, designed for transportability across clinical settings. Validation was conducted internally using a temporal patient-level split simulating prospective deployment, training on 2008-2016 data and testing on 2017-2022 data"External validation was performed on the eICU Collaborative Research Database, a multi-centre dataset spanning 208 US hospitals, using the trained model without retraining. SHAP TreeExplainer was used to provide feature-level explainability for individual predictions. Internal testing yielded an AUROC score of 0.794 for predicting AKI onset within a 12-24 hour window. External validation produced a 0.750 AUROC without retraining. Equitable discrimination was observed across gender, age, chronic kidney disease presence, race, and AKI stages on both datasets, with a 95% internal CI of 0.789-0.799 confirming the model's estimate stability. These results suggest that clinically useful prediction systems are achievable with substantially fewer features than current models require.

The openEHR standard provides an open, vendor-neutral architecture for clinical data repositories (CDRs), yet its real-world deployment has not been systematically documented. We conducted a dual-perspective survey combining a vendor survey of openEHR CDR providers with a community survey of openEHR practitioners. Eleven vendor organisations reported deployments across 22 countries and over 100 institutions and health regions. A complementary community survey (n=29, 17 countries) provided context on regulatory environments, adoption drivers, and barriers. Combined, the surveys cover 28 countries, 26 of them with a reported openEHR CDR deployment. Three findings emerge: openEHR has achieved national-scale presence through two distinct channels. Through vendor-market convergence, openEHR-based systems cover the majority of regional health authorities without a national mandate, including 19 of 21 Swedish regions, 3 of 4 Norwegian health regions, and 16 of 21 Finnish wellbeing services counties. Through national health record adoption, governments have built or procured national systems on openEHR as their technical foundation, including Ireland, Malta, Greece, Jamaica and Slovenia. Across Europe, this constitutes an openEHR-based interoperability infrastructure already in place across multiple EU member states. We identified no country in which openEHR is named in binding national regulation, creating structural fragility and an unrealised opportunity for alignment with the European Health Data Space (EHDS). Second, 61% of deployments serve primary use only, and 12% support both primary and secondary use. Third, lack of openEHR-specific knowledge is the most consistent adoption barrier across all geographies and deployment tiers. Adoption is driven by practitioner need and innovation, not by regulatory mandate.