Assessing differences in physician associate/assistant program admissions: observing the influence of demographic characteristics on likelihood of acceptance

Patient and clinician racial concordance has been shown to improve therapeutic trust and health outcomes. Like other health professions, the Physician Associate/Assistant (PA) profession has struggled to improve the diversity of the profession compared to the United States population. By analyzing the applicant pool from application to acceptance, this study aimed to apply a novel assessment methodology to evaluate the association between admissions policies and procedures and the diversity of the PA entering class. De-identified demographic and academic performance data were culled from 2022 to 2023 admissions applications to the George Washington University PA program from the Centralized Application Service for Physician Assistants (CASPA). Descriptive statistics were used to characterize applicants. Logistic regression models were used to identify applicant characteristics associated with the odds of being interviewed and/or accepted. A total of 2518 applications were analyzed, describing 8% of the CASPA applicant pool from 2022 to 2023. Bivariate analysis showed that race/ethnicity and overall GPA were significant predictors of being interviewed (G2(4) = 18.92, p < 0.001; OR = 1.66, 95% CI [1.54, 1.79]) and being accepted (G2(4) = 11.29, p = 0.02; OR = 1.41, 95% CI [1.26, 1.59]). Prior to controlling for covariates, Black applicants showed significantly lower odds of being accepted into the PA program compared to their White counterparts. After adjusting for overall GPA, the association between race/ethnicity and the odds of interview and acceptance became non-significant. Overall GPA remained strongly associated with odds of interview and acceptance after controlling for other covariates. Strong associations between Asian, Black and Hispanic applicants and the following backgrounds: (1) first-generation college student, (2) living in a health professions shortage area, (3) school district where 50% or less attend college and (4) economically disadvantaged family income were identified when compared to White students. This study proposes a unique method for evaluating the effects of admissions policies on accepted students from diverse backgrounds. Given some student background factors are associated with race/ethnicity, they may be useful to healthcare educational programs during the admissions cycle to facilitate increasing the diversity of the PA student body.