Deep learning and novel predictive models outperformed traditional regression methods for predicting 30-day hospital readmissions across five CMS-penalized conditions.
Do deep learning and novel predictive models improve the prediction of 30-day hospital readmissions compared to traditional regression methods?
Deep learning and novel predictive models outperform traditional regression methods in predicting 30-day hospital readmissions for CMS-penalized conditions.
Risk sharing arrangements between hospitals and payers together with penalties imposed by the Centers for Medicare and Medicaid (CMS) are driving an interest in decreasing early readmissions. There are a number of published risk models predicting 30day readmissions for particular patient populations, however they often exhibit poor predictive performance and would be unsuitable for use in a clinical setting. In this work we describe and compare several predictive models, some of which have never been applied to this task and which outperform the regression methods that are typically applied in the healthcare literature. In addition, we apply methods from deep learning to the five conditions CMS is using to penalize hospitals, and offer a simple framework for determining which conditions are most cost effective to target.
Futoma et al. (Mon,) conducted a other in Early hospital readmissions. Deep learning and novel predictive models vs. Traditional regression methods was evaluated on Predictive performance for 30-day readmissions. Deep learning and novel predictive models outperformed traditional regression methods for predicting 30-day hospital readmissions across five CMS-penalized conditions.