Provider Profiling and Quality Improvement Efforts in Coronary Artery Bypass Graft Surgery

CONTEXT: In the last decade, a few states or regions in the United States have initiated efforts to publicly disseminate coronary artery bypass graft (CABG) surgery outcomes and/or formally initiate quality improvement programs for CABG surgery. OBJECTIVE: To compare CABG mortality rates and changes in CABG mortality rates in regions with quality improvement/public dissemination efforts with the remainder of the country. DESIGN, SETTING, AND PATIENTS: Medicare data from 1994 to 1999 were used to develop a logistic regression model that predicts patient mortality for CABG surgery on the basis of preoperative patient risk factors and region of the country. MAIN OUTCOME MEASURES: In-hospital, 30-day, and in-hospital/30-day mortality adjusted for preoperative patient risk factors. RESULTS: Four of the 5 regions with quality improvement/public dissemination programs had significantly lower unadjusted in-hospital/30-day, in-hospital, and 30-day mortality than the remainder of the country. The odds ratio for risk-adjusted mortality for the 6-year period in all study regions combined was significantly lower (odds ratio OR, 0.79; 95% confidence interval CI, 0.73-0.85) than in the remainder of the United States. The odds ratio was also significantly lower for each year and for the 6-year time period in New York (OR, 0.66; 95% CI, 0.57-0.77) and Pennsylvania (OR, 0.79; 95% CI, 0.73-0.86). The change in risk-adjusted mortality between 1994 and 1999 remained essentially constant for all regions except New Jersey, the only region to initiate their program during the study period, which exhibited a significant reduction in risk-adjusted mortality. CONCLUSIONS: Public dissemination of outcomes data/formal region-wide quality improvement initiatives appear to be associated with lower risk-adjusted mortality rates for CABG surgery.