Abstract
Commercial question banks (QBs) and post-baccalaureate (Post-Bac) pathways are widely used to prepare for COMLEX-USA (Comprehensive Osteopathic Medical Licensing Examination of the United States) Level 1, but their association with first-attempt pass rates in the pass/fail era is uncertain. We conducted a retrospective study at a single U.S. osteopathic medical school to examine whether (1) completion of UWorld and TrueLearn and (2) Post-Bac status are associated with first-attempt COMLEX Level 1 pass rates. Two analytic samples were used: a QB-engagement sample with usage data (n=541) and a Post-Bac comparison sample with status data (n=847). Exposures were the percentage of UWorld and TrueLearn completed before the exam and Post-Bac vs. non-Post-Bac educational background; the outcome was first-attempt pass (binary). Among passers vs. non-passers, the mean UWorld completion was 60.1% ±30.9 vs. 48.3% ±35.7 (Cohen's d=0.38; small-to-moderate), and TrueLearn completion was 42.9% ±21.0 vs. 38.5% ±20.3 (d=0.21; small). Pass rates were 97.5% (199/204) for Post-Bac vs. 96.0% (616/643) for non-Post-Bac (risk ratio (RR)=1.02, 95% CI: 0.99-1.05; risk difference (RD)=1.75%, 95% CI: -0.88% to 4.38%). Across analyses, p-values did not reach statistical significance. In this cohort, neither QB completion nor Post-Bac status showed a statistically significant association with first-attempt pass; precision was limited by a few failures. Based on the findings, we recommend collecting more granular engagement metrics (e.g., timed vs. tutor mode, practice-test trajectories) and adjusting for academic confounders in future multi-site studies.