Abstract 17161: Type and Intensity of Strategies to Reduce Readmissions Under the Hospital Readmissions Reduction Were Not Associated With Improvements in Readmission Rates

Abstract

Background: Little is known about what strategies are associated with improvements in hospital readmissions under the Hospital Readmissions Reduction Program. Objective: To determine whether the type or intensity of readmission reduction strategies were associated with changes in readmission rates for heart failure, acute myocardial infarction, and pneumonia among acute care hospitals participating in the HRRP. Methods: We surveyed leaders of 1,600 U.S. acute care hospitals participating in the HRRP about their use of 13 specific strategies to reduce readmissions. Strategies were grouped into three domains: transitions of care (TOC, e.g. discharge checklist), quality improvement (QLT, e.g. medication reconciliation), and patient-centered (PC, e.g. patient engagement programs). Intensity of each domain was scored as high or low according to how many strategies were implemented. We calculated hospital-level readmission rates prior to (2011-12) and following (2014-15) HRRP implementation. We used linear regression to determine if there were associations between individual strategies, domains, or overall scope of strategies and a reduction in readmission rates. Results: Of the 1,600 hospitals surveyed, 926 participated (58% response rate). Hospitals reported using 6.1 (SD 2.5) strategies on average. TOC was the most commonly used domain: 69% of hospitals scored high intensity in TOC, compared to only 22% and 14% of hospitals scoring high intensity in QLT and PC domains, respectively. After adjusting for hospital size, type, teaching status, and location, there were no statistically significant associations between any individual strategy and changes in readmission rates, nor between domain intensity and changes in readmission rates. Nearly half of all hospitals, 49%, scored high in only one domain; only 22% scored high in two domains and 5% scored high in all three domains. In fully adjusted models, there was no association between scoring high in multiple domains and reducing readmission rates. Conclusions: Under the HRRP, hospitals focused most on transitions of care strategies. There was no evidence that any of the most commonly employed strategies for reducing readmissions were associated with differential changes in readmission rates.