A Predictive Model for Readmission of Patients with Congestive Heart Failure: A Multi-hospital Perspective

Mitigating preventable readmissions, where patients are readmitted for the same primary diagnosis within thirty days, poses a significant challenge in the delivery of high quality healthcare. Toward this end, we seek to understand whether health information technologies (IT) can help lower readmission risks. We develop a novel, predictive readmission model, termed as the beta geometric Erlang-2 (BG/EG) hurdle model, which predicts the propensity, frequency, and timing of readmissions of patients diagnosed with congestive heart failure (CHF). This unified model enables us to study the role of health IT applications, as well as patient demographics and clinical factors, in terms of their association with the risk of patient readmissions. The BG/EG Hurdle model provides superior prediction performance compared to extant models such as the logit, BG/NBD hurdle, and EG hurdle models. We test our model using a unique dataset that tracks patient demographic, clinical, and administrative data across 67 hospitals in North Texas over a four-year period. We find that health IT, patient demographics, visit characteristics, payer type, and hospital characteristics are significantly associated with readmission risk. We observe that implementation of cardiology information systems is associated with a reduction in the propensity and frequency of future readmissions, while administrative IT is correlated with a lower frequency of future readmissions. Our results indicate that patient profiles derived from our model can serve as building blocks for a clinical decision support system to identify CHF patients with high readmission risk.