Prospective and External Evaluation of a Machine Learning Model to Predict In-Hospital Mortality

AbstractThe ability to accurately predict in-hospital mortality for patients at the time of admission could improve clinical and operational decision-making and outcomes. Few machine learning models have been developed to predict in-hospital death that are both broadly applicable to all adult patients across a health system and readily implementable, and, to the best of our knowledge, none have been implemented, evaluated prospectively, or externally validated.The primary objective of this study was to prospectively and externally validate a machine learning model that predicts in-hospital mortality for all adult patients at the time of hospital admission. Model performance was quantified using the area under the receiver operating characteristic curve (AUROC) and area under the precision recall curve (AUPRC). Secondary objectives were to design the model using commonly available EHR data and accessible computational methods.A total of 75,247 hospital admissions (median [IQR] age 59.5 [29.0] years; male [45.9%]) were included in the study. The in-hospital mortality rates for the training validation, retrospective validations at Hospitals A, B, and C, and prospective validation cohorts, respectively, were 3.0%, 2.7%, 1.8%, 2.1%, and 1.6%. The area under the receiver operating characteristic curves (AUROCs), respectively, were 0.87 (0.83-0.89), 0.85 (0.83-0.87), 0.89 (0.86 – 0.92), 0.84 (0.80-0.89), and 0.86 (0.83-0.90). The area under the precision recall curves (AUPRCs), respectively, were 0.29 (0.25-0.37), 0.17 (0.13-0.22), 0.22 (0.14-0.31), 0.13 (0.08-0.21), and 0.14 (0.09-0.21).The results demonstrated accurate prediction of in-hospital mortality for adult patients at the time of admission. The data elements, methods, and patient selection make the model implementable at a system-level.