How Hospitals Responded to the Financial Incentives of the HRRP

The reliability of the assessment measure that determines performance in Pay for Performance (P4P) contracts can significantly impact the incentives to respond to such payment schemes. In this article, I Investigate the role of uncertainty in the performance measures that determine hospital reimbursements under the Hospital Readmission Reduction Program(HRRP) – a prominent P4P program that linked hospital reimbursements to hospital readmission rates. The HRRP penalizes Medicare serving hospitals with readmission rates above a given threshold and hospitals are continually assessed on readmissions. This circumstance, means that hospitals that were not initially penalized by the HRRP, might have an incentive to respond to the program incentives, if they face a future risk of a penalty. Similarly, hospitals penalized by the HRRP might have no incentive to respond if the assessment measure (readmissions) is arbitrary and driven by random chance. Empirically this uncertainty makes it difficult to conceptualize the treatment or exposure to the HRRP. To overcome this, I utilize the entire Medicare administrative inpatient claims and simulate hospital performance measures for the HRRP, in years that precede the HRRP. I then rely on the relationship between past and future performance in ex-ante periods to characterize the ex-ante risk of receiving a penalty under the HRRP. I then implement a difference in differences design and estimate the effect of the HRRP by comparing hospitals based on their ex-ante risk of receiving a penalty. The evidence from this research suggests that where hospitals did respond to the HRRP, they responded commensurately with expectations of future penalties. I find the largest reduction in readmissions due to the HRRP to be for AMI (heart attack) patients. I find no evidence that the reductions in readmissions for pneumonia patients post the HRRP can be causally attributed to the HRRP incentives. The results from this research highlight the need to further investigate why hospitals have selectively responded to the HRRP, and the differential cost associated with reducing readmissions across medical conditions.