Reducing Diabetes Hospital Readmissions With Interactive Predictive Modeling
Sep 7, 2017 1:15 PM
Mary Loveless, JMP Senior Systems Engineer, SAS
Ruth Hummel, PhD, JMP Academic Ambassador, SAS
Diabetes is a chronic disease, affecting nearly one in 12 United States residents and costing approximately $250 billion annually in the US alone. In this talk we use data on diabetes patient and hospital outcomes to identify key factors related to diabetic readmissions and to predict the probability of readmission using modern modeling techniques. The data set, which is publicly available from the UC Irvine Machine Learning Repository, includes 10 years of data (1999-2008) on clinical care and demographic information for approximately 70,000 patients at 130 hospitals and integrated delivery networks. This data originated from the Health Facts database (Cerner Corporation) and was de-identified and sampled by the Center for Clinical and Translational Research at Virginia Commonwealth University before being made publically available. Using this data, we illustrate tools for data preparation, demonstrate interactive visualization techniques, and discuss the importance of validation to gauge model accuracy. We then build several exploratory and predictive models, including logistic regression, bootstrap forests and boosted trees, neural networks, and penalized regression (LASSO, elastic net, and ridge). We interactively compare competing models, evaluate the risks of misclassification and select the best model(s). Finally, we generate scoring code to deploy the selected model as an interactive web-based application.