Faculty Mentor(s)

Dr. Sarah Estelle, Economics & Business

Document Type


Event Date



According to the Healthcare Cost and Utilization Project, the rate of hospital readmissions for Medicare patients fell from 18.3 percent in 2010 to 17.1 percent in 2016. Still, since avoidable readmission within 30 days of discharge is a negative healthcare outcome and costly, whether the readmission is due to low-quality care or random complication, it captures the attention of healthcare cost-conscious policymakers. The Hospital Readmissions Reduction Program (HRRP), implemented in 2012, aims to decrease readmission following selected procedures. This research analyzes the effect of the HRRP on readmission rates that were not a part of the HRRP’s intended scope, referred to as non-targeted readmission rates. If hospital administrators are sensitive to payment changes, they have an incentive to reduce the probability of avoidable readmissions, potentially shifting resources, adopting new practices, and utilizing new technologies to improve health outcomes for targeted treatments. Moreover, if hospitals learn from their new practices or can make use of the improved technology in treating non-targeted conditions, the direct and intended effects of HRRP may spillover to readmission reductions for non-targeted procedures. Readmission spillover effects, if they exist, would provide suggestive evidence that HRRP can have an even broader benefit by shifting hospitals toward so-called “value-based care” thinking. With data from CMS Hospital Compare data archive and the American College of Surgeons National Surgical Quality Improvement Program surgical patient readmission data, this research models both targeted and non-targeted intervention readmission rates before and after the HRRP for hospitals bound by the HRRP and those that are not. Since hospitals that already exhibit lower than national average readmission rates in the targeted procedures face no incentives from HRRP, they provide a plausible control group for HRRP-treated hospitals and an opportunity for causal inference through a Difference in Differences approach.