Data source
The Yale institutional review board waived approval and the need for patient consent since the datasets are publicly available. The data was obtained from the publicly available 2014-2016 New York State Cardiac Data Reporting System and 2015-2016 California Report on CABG Surgery8,9. Both of these states have mandatory public reporting systems for surgeon-level cardiac surgery outcome data. We collected observed mortality rate (OMR) and expected mortality rate (EMR) for isolated CABG in both states. EMR is calculated from multivariable risk models developed by the New York State Department of Public Health and California CABG Outcomes Reporting Program, both of which account for various patient demographics and comorbidities8,9. Some of these demographics and comorbidities include ejection fraction, previous MI, cardiogenic shock, previous cardiac surgery or PCI, renal failure, liver disease, peripheral vascular disease, endocarditis, BMI, and others8,9. Based on these multivariable risk models, EMR captures the average risk profile of a surgeon’s cases. Operative mortality for both states is defined as death within 30-days from surgery or within the index hospitalization.
We determined surgeons’ number of years in practice by collecting each surgeon’s training history from The Cardiothoracic Surgery Network (CTSnet, ctsnet.org). Each surgeon’s final year of schooling was subtracted from 2016 (the latest year captured in the New York and California data) to determine number of years in practice. For surgeons whose training history was not listed on CTSnet, we searched other online resources including the website of the surgeon’s current hospital and healthgrades.com. International medical graduates (IMGs) were excluded from the study because IMGs may have practiced as surgeons overseas, which may have obscured the actual years in practice. We combined surgeon-level outcomes for individual surgeons practicing at multiple hospitals.