American Surgical Association Home American Surgical Association Annual Meeting
Search Meeting Site Only



Current Meeting Home
Program
Past & Future Meetings

 

 





 

Back to Annual Meeting Program


Are Comprehensive Claims Data Accurate Enough for Pay-for-Performance and Public Reporting of Postoperative Complications?
Elise H Lawson*, David S Zingmond*, Rachel Louie*, Robert H Brook*, Clifford Y Ko
UCLA School of Medicine, Los Angeles, CA

OBJECTIVE(S):
Policymakers are increasingly focused on postoperative complication rates for pay-for-performance. There is debate regarding the optimal data source for such measures. We compare a national clinical registry and the Medicare claims datasets to evaluate the accuracy of 30-day postoperative complications for reporting of hospital-level surgical quality.
METHODS:
Patient records (2005-2008) from the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) were linked to Medicare inpatient and outpatient claims datasets. We assessed the ability of Medicare claims to detect the following postoperative complications as recorded in ACS-NSQIP: surgical site infection, urinary tract infection, pneumonia, deep venous thrombosis, and myocardial infarction. Agreement of patient-level complications by ACS-NSQIP vs. Medicare was assessed by kappa statistics. Hospitals were ranked by their complication rates and agreement of hospital rank as determined by ACS-NSQIP vs. Medicare was assessed using weighted kappa statistics.
RESULTS:
121,070 patients from 217 hospitals were studied. The sensitivity of the claims data for detecting ACS-NSQIP complications ranged from 0.45-0.86; the rate of false positives ranged from 58%- 88%. Agreement was routinely poor between clinical and claims data for patient-level complications and hospital rank.
CONCLUSIONS:This 217 hospital analysis demonstrates important differences between ACS-NSQIP and Medicare claims datasets for measuring surgical quality. Poor accuracy makes claims data suboptimal for evaluating surgical complications. These findings have meaningful implications for measures currently being considered for pay-for-performance.


Back to Annual Meeting Program

 



© 2021 American Surgical Association. All Rights Reserved. Privacy Policy.