Concurrent prediction of in-hospital mortality and length of stay using single-task, multi-class, and multi-task machine learning

BackgroundAccurate predictions of discharge timing and in-hospital mortality could improve hospital efficiency, but clinician estimates are often inconsistent and imprecise. We evaluated if machine learning models could concurrently predict in-hospital mortality and length of stay (LoS) more reliably. MethodsWe used electronic healthcare data from 01-November-2021 to 31-October-2024 from Oxfordshire, UK, using two years of data for training and evaluating models using the final years data. The performance of task-specific extreme gradient boosting (XGB), logistic regression (LR), and multilayer-perceptron (MLP) models for the two tasks: (i) mortality prediction and (ii) LoS prediction, were compared with that of a single multiclass XGB model predicting combinations of LoS and mortality, and an MLP-based multi-task learning model predicting both outcomes simultaneously. Predictions from the best-performing models were compared to discharge predictions made by clinicians. FindingsClinicians provided relevant discharge predictions for only 3-5% of admissions, mostly close to discharge. Task-specific XGB models achieved an area under the receiver operating curve of 0.92 and 0.92 for predicting mortality, and 0.83 and 0.72 for predicting LoS quartiles, in elective and emergency admissions respectively, outperforming task- specific LR and MLP models. Neither the multiclass XGB nor the MLP-based multi-task models, predicting both outcomes simultaneously, consistently improved performance. The best-performing task-specific XGB models matched clinician LoS prediction accuracy in elective admissions, and significantly outperformed clinicians in emergency admissions (p<0.001). InterpretationMachine learning models can predict in-hospital mortality and LoS as well or better than clinicians and have potential to enhance discharge planning and hospital resource management. FundingNational Institute for Health Research (NIHR) Biomedical Research Centre, Oxford, and NIHR Health Protection Research Unit in Healthcare Associated Infections and Antimicrobial Resistance at Oxford University in partnership with the UK Health Security Agency (UKHSA).