• Jose F Figueroa, Yusuke Tsugawa, Jie Zheng, E John Orav, and Ashish K Jha.
• Department of Health Policy and Management, Harvard T.H. Chan School of Public Health, Boston, MA 02115, USA Department of Medicine, Brigham and Women's Hospital, Boston, MA, USA.
• BMJ. 2016 Jan 1; 353: i2214.

Objective To determine the impact of the Hospital Value-Based Purchasing (HVBP) program-the US pay for performance program introduced by Medicare to incentivize higher quality care-on 30 day mortality for three incentivized conditions: acute myocardial infarction, heart failure, and pneumonia.Design Observational study.Setting 4267 acute care hospitals in the United States: 2919 participated in the HVBP program and 1348 were ineligible and used as controls (44 in general hospitals in Maryland and 1304 critical access hospitals across the United States).Participants 2 430 618 patients admitted to US hospitals from 2008 through 2013.Main Outcome Measures 30 day risk adjusted mortality for acute myocardial infarction, heart failure, and pneumonia using a patient level linear spline analysis to examine the association between the introduction of the HVBP program and 30 day mortality. Non-incentivized, medical conditions were the comparators. A secondary outcome measure was to determine whether the introduction of the HVBP program was particularly beneficial for a subgroup of hospital-poor performers at baseline-that may benefit the most.Results Mortality rates of incentivized conditions in hospitals participating in the HVBP program declined at -0.13% for each quarter during the preintervention period and -0.03% point difference for each quarter during the post-intervention period. For non-HVBP hospitals, mortality rates declined at -0.14% point difference for each quarter during the preintervention period and -0.01% point difference for each quarter during the post-intervention period. The difference in the mortality trends between the two groups was small and non-significant (difference in difference in trends -0.03% point difference for each quarter, 95% confidence interval -0.08% to 0.13% point difference, P=0.35). In no subgroups of hospitals was HVBP associated with better outcomes, including poor performers at baseline.Conclusions Evidence that HVBP has led to lower mortality rates is lacking. Nations considering similar pay for performance programs may want to consider alternative models to achieve improved patient outcomes.Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://group.bmj.com/group/rights-licensing/permissions.

16 keywords...

hide…