Abstract 15729: Performance on Hospitals on PCI Process Measures and Correlation With 30-Day Readmissions
Introduction: The National Quality Forum has endorsed the metric of 30-day risk-standardized readmissions rates (RSRR) following PCI as a measure of hospital quality. It is not known whether hospital performance on this publicly reported metric correlates with existing PCI process measures. This information could help hospitals decide on where to focus their quality improvement efforts.
Methods: Using data from the NCDR CathPCI Registry from 2010-2011 linked with Medicare administrative claims data, we calculated hospital performance on existing PCI process measures (aspirin, thienopyridines, and statins upon discharge; door-to-balloon time under ninety minutes; referral to cardiac rehabilitation; and an overall composite measure). We calculated hospitals’ 30-day RSRRs using hierarchical models as specified in the NQF-approved measure. We then analyzed the association between hospital performance on process measures and 30-day RSRRs. We subsequently calculated the explained variance of the process metrics on the hospital variation seen in RSRR.
Results: The median 30-day RSRR of all hospitals was 11.8 (interquartile range: 11.1-12.7); the median and interquartile range of hospital performance on the process metrics is shown below (Table). Hospital performance on both the individual and composite process was statistically significantly correlated with hospital performance on RSRR. However, differences in hospital performance on process metrics overall accounted for only 1.6% of observed variation in RSRR.
Conclusion: Hospital performance on PCI process measures is only modestly correlated with 30-day RSRR. Our findings suggest that a focus on process measures alone is unlikely to impact PCI readmissions.
Author Disclosures: P.W. Chui: None. C. Parzynski: None. F.A. Masoudi: None. J.C. Messenger: None. L.E. Slattery: None. J.P. Curtis: None.
- © 2014 by American Heart Association, Inc.