The Society of Thoracic Surgeons Isolated Aortic Valve Replacement (AVR) Composite Score: A Report of the STS Quality Measurement Task Force
BACKGROUND:
Risk-standardized mortality rates provide a valuable but incomplete assessment of provider performance. Consequently, The Society of Thoracic Surgeons (STS) previously developed a multidimensional composite quality measure for coronary artery bypass grafting, the most frequently performed cardiac surgical procedure. The current study creates a similar composite measure for isolated aortic valve replacement (AVR).
METHODS:
Because there are few widely accepted process measures for AVR, the STS AVR composite score is based solely on outcomes, including risk-standardized mortality and any-or-none risk-standardized morbidity (occurrence of sternal infection, reoperation, stroke, renal failure, or prolonged ventilation). Isolated AVR is performed less frequently than coronary artery bypass grafting, and 1 year of data provided inadequate sample sizes for profiling. Therefore, we investigated observation periods of 3 years (July 1, 2007, to June 30, 2010: 67,138 records, 2,082 deaths, and 11,962 morbidity events) and 5 years (July 1, 2005, to June 30, 2010: 101,269 records, 3,123 deaths, and 17,514 morbidity events). We also compared results using 90%, 95%, and 98% credible intervals, corresponding to 95%, 97.5%, and 99% Bayesian probabilities, to determine “star ratings.”
RESULTS:
Differences between 3-year and 5-year results were small; the former was chosen because this time frame provides more current and relevant data. Using 3 years of data and 95% credible intervals, adjusted mortality and morbidity rates varied threefold from highest performing (3 stars) to lowest performing (1 star) programs. Approximately 3% of participants were 1-star, 6% were 3-star, and 91% were 2-star programs.
CONCLUSIONS:
STS has developed a composite mortality and morbidity outcomes measure for isolated AVR to be used in quality assessment, provider feedback, public reporting, and performance improvement.
