Later Is Better: Projected USMLE Performance During Medical School

Journal Article

In this study, we investigated the performance of students from Years 1, 3, and 4 of medical school on sample United States Medical Licensing Examination (USMLE) questions to predict actual performance and advise about the optimal time to take these tests. A unique curriculum and changes in the USMLE suggested that (a) students would do better on Step 1 after clerkships and (b) fourth-year students would do best on Step 2. Approximately 50% of students from the first, third, and fourth years volunteered to take Step I and Step 2 sample questions without preparation. Passing on both tests was 58% correct. For Step 1, significantly more fourth- than first- and third-year students passed. Expectedly for Step 2, significantly higher proportions of third- and fourth-year students passed than first-year students; however, fourth-year students did not score higher than third-year students. Using percent correct, fourth-year students did best on both tests. These results suggest that clinical experience will likely improve performance on Step 1, and an additional clinical year may improve performance on Step 2. Medical schools should consider allowing the USMLE to be taken later, possibly in the fourth year. Curriculum differences and other factors may limit the generalizability of these results. © 1995, Taylor & Francis Group, LLC. All rights reserved.

Full Text

Duke Authors

Cited Authors

  • Petrusa, ER; Reilly, CG; Lee, LS

Published Date

  • January 1, 1995

Published In

Volume / Issue

  • 7 / 3

Start / End Page

  • 163 - 167

Electronic International Standard Serial Number (EISSN)

  • 1532-8015

International Standard Serial Number (ISSN)

  • 1040-1334

Digital Object Identifier (DOI)

  • 10.1080/10401339509539734

Citation Source

  • Scopus