reports

A predictive model to identify patients at risk of unplanned 30-day acute care hospital readmission

In: 2013 IEEE International Conference on Healthcare Informatics, September 9-11, 2013, Philadelphia, PA, USA. Los Alamitos, CA, USA: IEEE:551-556.

Department of Health Policy and Management, Johns Hopkins University, Baltimore, MD, USA

Potentially preventable readmissions are burdensome on patients and those who pay for health care. Predictive models help hospitals and their partner organizations with a timely assessment of readmission risk to initiate interventions that are aimed at reducing avoidable readmissions. The purpose of this paper is to describe the development of a model for predicting unplanned 30-day readmissions. Our research objective is to develop an all-age, all-cause 30-day readmission risk model for unplanned acute care hospitalization with logistic regression on health plan claims data. We classify diagnoses and procedures to measure health status and health care utilization. Our individual readmission risk scores could be available at the time of admission and thus may have implications for individualized treatment plans and managing the discharge process at an early stage. Another application of our risk scores is to identify patients at high risk of readmission for outpatient care transition management.