Association of hospital coronary artery bypass volume with processes of care, mortality, morbidity, and the Society of Thoracic Surgeons composite quality score.

OBJECTIVE: This study examines the association of hospital coronary artery bypass procedural volume with mortality, morbidity, evidence-based care processes, and Society of Thoracic Surgeons composite score. METHODS: The study population consisted of 144,526 patients from 733 hospitals that submitted data to the Society of Thoracic Surgeons Adult Cardiac Database in 2007. End points included use of National Quality Forum-endorsed process measures (internal thoracic artery graft; preoperative beta-blockade; and discharge beta-blockade, antiplatelet agents, and lipid drugs), operative mortality (in-hospital or 30-day), major morbidity (stroke, renal failure, reoperation, sternal infection, and prolonged ventilation), and Society of Thoracic Surgeons composite score. Procedural volume was analyzed as a continuous variable and by volume strata (< 100, 100-149, 150-199, 200-299, 300-449, and > or = 450). Analyses were performed with logistic and multivariate hierarchical regression modeling. RESULTS: Unadjusted mortality decreased across volume categories from 2.6% (< 100 cases) to 1.7% (> 450 cases, P < .0001), and these differences persisted after risk factor adjustment (odds ratio for lowest- vs highest-volume group, 1.49). Care processes and morbidity end points were not associated with hospital procedural volume except for a trend (P = .0237) toward greater internal thoracic artery use in high-volume hospitals. The average composite score for the lowest volume (< 100 cases) group was significantly lower than that of the 2 highest-volume groups, but only 1% of composite score variation was explained by volume. CONCLUSION: A volume-performance association exists for coronary artery bypass grafting but is weaker than that of other major complex procedures. There is considerable outcomes variability not explained by hospital volume, and low volume does not preclude excellent performance. Except for internal thoracic artery use, care processes and morbidity rates were not associated with volume.

Duke Scholars

Author Sean Michael O'Brien Biostatistics & Bioinformatics, Division of Biostatistics