THE GENDER PAY GAP IN NON-PROCEDURAL CARDIOLOGY - EVIDENCE FROM A POPULATION-BASED STUDY

The gender pay gap in medicine is well documented. Frequently cited drivers of gender-based billings disparities are women entering lower-remunerated specialties and fewer hours worked. To date, there has not been a population-based analysis of the gender pay gap and gender-based practice differences in a highly remunerated specialty like cardiology in a fee-for-service system (FFS). Our objectives were to assess differences in billings, practice patterns, and patient outcomes between male and female non-procedural cardiologists in a FFS system.

We used linked population-based administrative health databases to conduct a retrospective cohort study of cardiologists practicing in Ontario, Canada between April 2011 and March 2016. All cardiologists submitting outpatient claims during the study period were included, with cardiologists practicing interventional cardiology or electrophysiology being excluded. Outpatients of study cardiologists were identified and followed as a patient cohort. Our primary outcome was overall annual physician billings. We assessed practice patterns by identifying days worked and volumes billed for outpatient visits, inpatient visits, and diagnostic test interpretation. Patient outcomes included healthcare utilization, all-cause mortality, all-cause hospital admission, and emergency department visit or hospitalization for cardiovascular cause. We developed multivariable regression models to estimate the effect of gender on billings and each patient outcome. We identified 394 cardiologists, of whom 20% were women. The patient cohort included 818,122 patients. Unadjusted annual billings were significantly higher among male cardiologists than female ($769,371 vs. $515,997 p < 0.0001). Men billed more annual outpatient visits (1,808 vs. 1,169, p < 0.0001) and outpatient visits per day (11 patients per day vs. 9, p = 0.0004) than women. Male cardiologists billed more days for diagnostic test interpretation than women (for echocardiograms, 172 vs. 146, p = 0.016) though the numbers of echocardiograms billed per day were similar. Billings were significantly associated with outpatient practice size and echocardiogram reading, but not cardiologist gender after adjustment for patient and provider factors. There were no differences by cardiologist gender for patient outcomes after adjustment for cardiologist and patient factors.

Our results demonstrate substantial gender-based billings disparity, which may be driven by practice styles in terms of days worked and patients seen per day in outpatient cardiology and diagnostic test interpretation. Our findings contribute to the growing body of evidence on the gender pay gap in medicine, as well as discussion of the equity of FFS remuneration. Further research is needed to understand what drives these differences in practice choices.