Factors that correlate with the U.S. Medical Licensure Examination Step-2 scores in a diverse medical student population.

OBJECTIVE: To assess factors that correlate with performance on U.S. Medical Licensure Examination (USMLE) Step-2 examination. Our hypothesis was that demographic factors, faculty assessments and other standardized test scores will correlate with students' performance on USMLE Step 2. STUDY
DESIGN: A comparison of standardized examinations and demographic factors with USMLE Step-2 scores as the outcome variable was accomplished using the educational records of 171 medical students.
RESULTS: Mean USMLE Step 2, USMLE Step 1, NBME-OB/GYN and MCAT scores, respectively, were 190.63, 194.53, 67.47 and 24.03. Positive correlations of USMLE Step 2 were: USMLE Step-1 scores (r=0.681, p=0.000); MCAT scores (r=0.524, p=0.000); NBME-OB/GYN scores (r=0.614, p=0.000); year of OB/GYN rotation (r=0.432, p=0.000); faculty grades (r=0.400, p=0.000); undergraduate GPA (r=0.287, p=0.000); and science GPA (r=0.255, p=0.002). Negative correlations of USMLE Step 2 were students' increasing age (r=-0.405, p=0.000), increasing number of MCAT attempts (r=-0.182, p=0.000) and increasing number of NBME-OB/GYN attempts (r=-0.310, p=0.000). There was no correlation with gender or race. Logistic regression analysis showed that a failing NBME-OB/GYN score (p=0.008), failing USMLE Step-1 score (p=0.01), failing faculty grade (p=0.029) and multiple MCAT attempts (p=0.033) independently increased the risk of failing USMLE Step 2.
CONCLUSIONS: Premedical test-taking abilities (MCAT scores), results of preclinical standardized tests (USMLE Step 1) and performance on clinical rotations (NBME-OB/GYN score and faculty grade) correlate significantly with USMLE scores. These findings may assist educators in selecting medical students at risk of performing poorly on the USMLE Step-2 examinations.