RESUMO
OBJECTIVE: To assess the association between the common quality metric of 30-day mortality and mortality at 60 days, 90 days, and one year after coronary artery bypass grafting. DESIGN: A retrospective cohort study, with multivariate logistic regression to assess association among mortality outcomes. SETTING: Hospitals participating in Medicare and reporting data within the Centers for Medicare and Medicaid Services Limited Data Set between April 1, 2016, and March 31, 2017. PARTICIPANTS: A total of 37,036 patients undergoing surgery at 394 hospitals. INTERVENTION: None. MEASUREMENTS AND MAIN RESULTS: Mortality rates were 1.0%-to-3.1% for the top and bottom quartile of hospitals at 30 days. At one year, the top 25th percentile of hospitals had mortality rates averaging 3.9%; while hospitals below the 75th percentile had mortality rates averaging 7.6%. Twenty-three percent of hospitals in the top quartile at 30 days were no longer in the top quartile at 60 days. At one year, only 48% of hospitals that were in the top quartile at 30 days remained in the top quartile. The correlation between mortality rates at 30 days and the reported points was assessed using Spearman's rho. The R value between mortality at 30 days and mortality at one year was 0.53, which improved to 0.7 and 0.76 at 60 and 90 days. CONCLUSIONS: Mortality at 30 days correlated poorly with mortality at one year. Hospitals that were high- or low-performing at 30 days frequently were no longer within the same performance group at one year.