preprint · Zenodo (CERN European Organization for Nuclear Research)
An interpretable machine learning model has been developed to predict hospital admissions using only patient data gathered at emergency department triage. Seven classifiers were benchmarked using a rigorous validation and missing-data imputation pipeline, leading to the selection of an L2-regularised logistic regression model due to its clear interpretability. The model was trained and internally evaluated on one thousand encounters, then externally tested on an independent dataset of over one hundred and eighteen thousand encounters. It achieved strong and stable predictive accuracy across both cohorts, with triage acuity and patient age identified as the primary drivers of admission. Performance and calibration remained consistent across demographic subgroups, demonstrating that simple triage indicators can reliably forecast hospitalisation demands.
Emergency departments frequently face overcrowding and shortages of available beds. Predicting admissions immediately upon a patient's arrival allows healthcare facilities to anticipate bed demand, coordinate resources, and streamline patient flow earlier in the care pathway, helping to ease operational pressure without relying on complex, uninterpretable algorithms.
The model provides an applied and externally validated software tool intended for hospital operations teams, emergency staff, and bed planners. By accurately forecasting admissions at triage, it enables earlier resource allocation and bed-demand management. Having undergone successful external retrospective validation on a large dataset, the underlying methodology is well positioned for prospective clinical trials and integration into hospital operational workflows.
AI-generated from the published abstract. Always read the original work before citing.
Early prediction of hospital admission at emergency department triage may support clinical decision-making, patient flow coordination, bed planning and resource allocation. However, external validation of admission prediction models remains limited, while missing-data handling, calibration, clinical utility, and interpretability are inconsistently addressed. We developed an interpretable triage-only admission prediction model using 1,000 MIMIC-IV-ED encounters. Missing data were handled using 15 MICE-PMM imputations within a leakage-resistant, patient-grouped 5×5 nested cross-validation framework. Seven classifiers were benchmarked under the same analytical pipeline and showed closely comparable discrimination, and L2-regularized logistic regression was retained as the final model based on its interpretability and direct coefficient-based odds ratios. Performance was assessed using AUROC, calibration, Brier score, and decision curve analysis, while interpretability was examined using SHAP and odds ratios. The finalized model was externally validated in an independent cohort of 118,349 MC-MED encounters. Internal validation yielded an AUROC of 0.805 (95% CI 0.779-0.832) and a Brier score of 0.181(95% CI 0.169-0.193). In external validation, AUROC was 0.783 (95% CI 0.779-0.786) with a Brier score of 0.185(95% CI 0.184-0.187). Triage acuity and age were the most influential predictors. Discrimination was similar between females and males, while calibration remained close to 1 across gender and age subgroups. Overall, the findings support an interpretable triage model whose performance remained stable across validation settings and whose outputs could inform early bed-demand, patient-flow, and resource planning, providing a foundation for future hospital supply chain optimization.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.5281/zenodo.20720950
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
