What Program Characteristics Are Associated with Resident Racial Diversity in Orthopaedic Surgery? An Analysis of Association of American Medical Colleges Data

In orthopaedic surgery, there are fewer Black or African American (4%) and Hispanic or Latino (4%) residents compared with general surgery, internal medicine, family medicine, and pediatrics (5%-7% Black residents and 7%-9% Hispanic/Latino residents, respectively). There are also fewer underrepresented in medicine minority (URiM) faculty in orthopaedic surgery (6.1%) compared with general surgery (8.9%), otolaryngology (7.8%), internal medicine (9.7%), and obstetrics and gynecology (15.6%). Identifying program characteristics that are associated with the percentage of URiM residents could reveal strategies for improving diversity. METHODS: Using Association of American Medical Colleges orthopaedic resident and faculty race/ethnicity data from 2007 to 2016, we analyzed the racial diversity of 166 of 207 residency programs. The primary outcome was program racial diversity, measured as the percentage of URiM residents per program. The top quartile of programs was compared with the other quartiles. Characteristics analyzed included percentage of URiM faculty, affiliation with a university/top 40 medical school/top 40 orthopaedic hospital, geographic region, city type, and city size. We used a multivariable linear regression model to evaluate program characteristics associated with diversity and a linear mixed-effects model with program-specific random effects to evaluate time trends. RESULTS: The mean percentage of URiM residents per program was 9.3% (SD = 10.5%). In the top quartile of programs, URiM residents composed 20.7% ± 2.5% of the program compared with 5.8% ± 0.3% in other quartiles (p < 0.001). After adjusting for program and faculty size, the only factor associated with the number of URiM residents per program was the number of URiM faculty. For every 5 additional URiM faculty members, there was an associated increase in the number of URiM residents per program by 3.6 (95% confidence interval [CI]: 2.3-5.0). There was a small but statistically significant annual increase in the percentage of URiM residents per program of 0.207 (95% CI: 0.112-0.302) percentage points during the study period. CONCLUSION: URiM representation remains low among orthopaedic residents. Efforts to increase the URiM faculty base represent a potential strategy for programs to increase URiM representation among residents by attracting more diverse applicants.