Factors associated with performance in an internal medicine clerkship

The purpose of this retrospective study was to examine the relationship between demographic and educational variables and student performance on an internal medicine (IM) clerkship in order to target areas for educational intervention and potential early remediation. This study examined data associated with third-year medical student performance (N = 505) during the IM clerkship at Baylor Scott & White, Temple/Texas A&M Health Science Center College of Medicine from 2005 to 2011. Multiple regression analysis (N = 341) showed that a model containing the following variables was significantly associated with scores on the National Board of Medical Examiners (NBME) subject exam, accounting for 46.5% of the variance: Objective Structured Clinical Exam (OSCE), Medical College Admissions Test (MCAT), US Medical Licensing Exam (USMLE) Step 1, second-year grade point average (GPA), and clinical evaluation. A model containing USMLE Step 1, clinical evaluation, and NBME was significantly associated with OSCE score, accounting for 30% of the variance. Additionally, a model containing age, MCAT score, undergraduate GPA, NBME subject exam score, and OSCE was significantly associated with clinical evaluation score, accounting for 22% of score variance. Age accounted for the most unique variance in clinical evaluation score. Gender and IM interest group were not significantly associated with any outcome variable. In conclusion, in contrast to previous studies in the field, we did not find a significant association between undergraduate GPA and NBME score. Our findings supply further evidence that the OSCE, typically believed to be a clinical performance exam, actually assesses a broader set of domains. Interest group membership did not confer any academic benefit to medical students in IM clerkships in our study.