Predicting Patient Discharge Disposition in Acute Neurological Care

Acute neurological complications are some of the leading causes of death and disability in the U.S. and the medical professionals that treat patients in this setting are tasked with deciding where (e.g., home or facility), how, and when to discharge these patients. It is important to be able to predict ahead of time these potential patient discharge outcomes and to know what factors influence the development of discharge planning for such adults receiving care for neurological conditions in an acute setting. The goal of this study is to develop predictive models exploring which patient characteristics and clinical variables significantly influence discharge planning with the hope that the models can be used in a suggestive context to help guide healthcare providers in efforts of planning effective, equitable discharge recommendations. Our methodology centers around building and training five different machine learning models followed by testing and tuning those models to find the best-suited predictor with a dataset of 5,245 adult patients with neurological conditions taken from the eICU-CRD database. The results of this study show XGBoost to be the most effective model for predicting between four common discharge outcomes of ‘home’, ‘nursing facility’, ‘rehab’, and ‘death’, with 71% average c-statistic. This research also explores the accuracy, reliability, and interpretability of the best performing model by identifying and analyzing the features that are most impactful to the predictions.