Impact of a hospital bounceback policy to reduce readmissions

Background Bounceback policies for readmitted patients are commonly used by hospitals in the U.S., but to date there have been no studies of their impact on patient care or outcomes. In July 2011, a large, tertiary-care, academic medical center introduced a stronger bounceback policy for readmitted patients cared for by the resident service. The primary objective of this study was to determine the impact of the policy on readmission rates, length of stay (LOS), and LOS index. The secondary objective was to assess the effect of time within clinical rotations on the same outcomes. Methods This retrospective cohort study included patients admitted to the resident and hospitalist medicine services from 6/15/09 to 6/16/13. Total number of hospitalizations analyzed was 66,118. Measures included 30-day readmission rates, LOS, and adjusted LOS index. Covariates included time within clinical rotation and patient-level characteristics. The hospitalist service was used as a control group. Results After the policy was implemented, the resident service had an 8% lower relative risk ratio of 30-day readmission compared to the hospitalist service (p =.05). There was no significant different in LOS or LOS index. LOS for patients on the resident service exhibited a weekly pattern, with LOS lowest on Mondays and increasing throughout the week. Conclusions A stronger bounceback policy resulted in a small decrease in 30-day readmission rates without increasing LOS. Implications Bounceback policies may be an effective and low-cost tool to reduce readmissions by leveraging non-financial incentives to change physicians’ behavior.