Predictors of Ophthalmology Resident Performance From Medical Student Application Materials

Tooley AA, Law J, Lelli GJ, Sun G, Godfrey KJ, Tran AQ, Kim E, Solomon JM, Chen JJ, Khan AR, Wayman L, Olson JH, Lee MS, Harrison AR, Espinoza GM, Davitt BV, Tao J, Hodge DO, Barkmeier AJ
J Surg Educ
Nov 2023
38036387

Objective: To determine whether elements in ophthalmology residency applications are predictors of future resident performance. Design: This multi-institutional, cross-sectional, observational study retrospectively reviewed the residency application materials of ophthalmology residents who graduated from residency from 2006 through 2018. Resident performance was scored by 2 faculty reviewers in 4 domains (clinical, surgical, academic, and global performance). Correlation between specific elements of the residency application and resident performance was assessed by Spearman correlation coefficients (univariate) and linear regression (multivariate) for continuous variables and logistic regression (multivariate) for categorical variables.

Setting: Seven ophthalmology residency programs in the US.

Participants: Ophthalmology residents who graduated from their residency program.

Results: High-performing residents were a diverse group, in terms of sex, ethnicity, visa status, and educational background. Residents with United States Medical Licensing Examination Step 1 scores higher than the national average for that year had significantly higher scores in all 4 performance domains than those who scored at or below the mean (all domains P < 0.05). Residents who had honors in at least 4 core clerkships and who were members of Alpha Omega Alpha Medical Honor Society also had higher scores in all 4 performance domains (all domains P ≤ 0.04). Step 1 score (ρ=0.26, P < 0.001) and the difference between Step 1 score and the national average for that year (ρ=0.19, P = 0.009) positively correlated with total resident performance scores. Residents who passed the American Board of Ophthalmology Written Qualifying Examination or Oral Examination on their first attempt had significantly higher Step 1/2 scores (P ≤ 0.005), Ophthalmology Knowledge Assessment Program scores (P = 0.001), and resident performance scores (P ≤ 0.004).

Conclusions: In this new landscape of increasing numbers of applicants to residency programs and changing of the Step 1 score to pass/fail, our findings may help guide selection committees as they holistically review applicants to select exceptional future residents in ophthalmology.