Does Removing the Photograph and Name Change the Reviewer's Perception of Orthopaedic Residency Applicants?

Abstract

Introduction: Orthopaedic surgery remains the least diverse field in medicine regarding female and minority representation. Scarce literature exists evaluating the role of implicit bias in the residency application process. We hypothesized that applicants perceived as underrepresented minorities in orthopaedic surgery (URMs) based on their photograph or name would have a decreased likelihood of being invited to interview.

Methods: Data from the 2018 to 2019 orthopaedic residency application cycle were collected from a single institution. Applications were classified URM or non-URM. After the application cycle was completed, the URM applications were propensity matched with non-URM applicants. Photographs and names were removed, and the applications were rereviewed by the Residency Applicant Review committee. Rank-in-group and the likelihood of being invited for an interview were compared.

Results: Four hundred eleven applications were included with 27.5% URM and 72.5% non-URM. During the regular application cycle, 34.7% of those invited to interview were URM and 50% of those who were ranked-to-match range were URM. After propensity matching, 90 matched pairs were rereviewed with their photograph and name removed. In the regular application cycle, the URM applicant was 3.8 times more likely to get an interview than the matched non-URM applicant (odds ratio, 3.8, 95% confidence interval, 1.7 to 8.8, P = 0.0014). In the "blinded" condition, the URM candidate was 2.5 times more likely to get an interview than the non-URM candidate (odds ratio, 2.5, 95% confidence interval, 1.1 to 6.2 P = 0.034). In the unblinded condition, the URM candidate had a higher ranking within their group than the corresponding non-URM applicant (P = 0.0005).

Discussion: Contrary to our initial hypothesis, URM applicants were invited to interview at a higher rate than non-URM applicants, both in the regular application cycle and in the propensity-matched "blinded" condition. This suggests that implicit bias based on the picture or name is not negatively affecting URM students during the application review process at our institution.

Level of evidence: 3.