Preclinical Assessment Performance as a Predictor of USMLE Step 1 Scores or Passing Status.

PURPOSE

To determine the association between student performance on preclinical pass/fail assessments in an allopathic medical school curriculum and Step 1 scores or passing status.

MATERIALS AND METHODS

This observational retrospective study involved preclinical assessments, including National Board of Medical Examiners Customized Assessment Services (NBME CAS) exams, faculty developed exams, and the United States Medical Licensing Examination (USMLE) Step 1 from 582 medical students in four cohorts (2018-2021). Analyses included descriptive statistics, Pearson's correlation coefficient (ρ) and logistic regression, presented as odds ratios (ORs) and associated values.

RESULTS

Mean scores on Component 4 end-of-block NBME CAS examinations positively correlated with Step 1 scores (ρ = 0.83,  < .001), as did mean scores on both Component 1 weekly faculty-created assessments and Component 3 end-of-block faculty-created assessments (ρ = 0.70,  < .001; ρ = 0.73,  < .001). Passing all Component 3 end-of-block faculty-created assessments in all blocks was associated with passing Step 1 (OR = 8.66,  < .001). Independently, passing all Component 4 NBME CAS exams or passing all Component 1 weekly faculty-derived assessments in all blocks did not correlate with passing Step 1 (OR = 2.40,  = .12.; OR = 0.29,  = .30). Passing all assessment types in all blocks was among the strongest correlators with passing Step 1 (OR = 9.026,  < .001).

CONCLUSIONS

Scores on faculty-derived and NBME CAS end-of-block assessments were positively correlated with Step 1 scores. Passing status on institution-derived end-of-block assessments was associated with passing Step 1, whereas passing status on weekly institution-derived assessments or end-of-block NBME CAS assessments was not associated with passing Step 1. End-of-block pass/fail NBME CAS and faculty-derived preclinical examinations may help prepare students for Step 1 and predict their outcomes. Weekly faculty-created assessments should primarily be used to continuously reinforce educational material.