Interview Invitations for Otolaryngology Residency Positions Across Demographic Groups Following Implementation of Preference Signaling

Abstract

Preference signaling is a new initiative in the residency application process that has been adopted by 17 specialties that include more than 80% of applicants in the 2023 National Resident Matching cycle. The association of signals with interview selection rate across applicant demographics has not been fully examined. To assess the validity of survey-based data on the association of preference signals with interview offers and describe the variation across demographic groups. This cross-sectional study examined the interview selection outcomes across demographic groups for applications with and without signals in the 2021 Otolaryngology National Resident Matching cycle. Data were obtained from a post-hoc collaboration between the Association of American Medical Colleges and the Otolaryngology Program Directors Organization evaluating the first preference signaling program used in residency application. Participants included otolaryngology residency applicants in the 2021 application cycle. Data were analyzed from June to July 2022. Applicants were provided the option of submitting 5 signals to otolaryngology residency programs to indicate specific interest. Signals were used by programs when selecting candidates to interview. The main outcome of interest was the association of signaling with interview selection. A series of logistic regression analyses were conducted at the individual program level. Each program within the 3 program cohorts (overall, gender, and URM status) was evaluated with 2 models. Of 636 otolaryngology applicants, 548 (86%) participated in preference signaling, including 337 men (61%) and 85 applicants (16%) who identified as underrepresented in medicine, including American Indian or Alaska Native; Black or African American; Hispanic, Latino, or of Spanish origin; or Native Hawaiian or other Pacific Islander. The median interview selection rate for applications with a signal (48% [95% CI, 27%-68%]) was significantly higher than for applications without a signal (10% [95% CI, 7%-13%]). No difference was observed in median interview selection rates with or without signals when comparing male (46% [95% CI, 24%-71%] vs 7% [95% CI, 5%-12%]) and female (50% [95% CI, 20%-80%] vs 12% [95% CI, 8%-18%]) applicants or when comparing applicants who identified as URM (53% [95% CI, 16%-88%] vs 15% [95% CI, 8%-26%]) with those who did not identify as URM (49% [95% CI, 32%-68%] vs 8% [95% CI, 5%-12%]). In this cross-sectional study of otolaryngology residency applicants, preference signaling was associated with an increased likelihood of applicants being selected for interview by signaled programs. This correlation was robust and present across the demographic categories of gender and self-identification as URM. Future research should explore the associations of signaling across a broad range of specialties and the associations of signals with inclusion and position on rank order lists and match outcomes.