Letters to the Editor
To the Editor:
Dr. Jurich and colleagues1 concluded that taking the United States Medical Licensing Examination (USMLE) Step 1 after clerkships increases scores. While they adjusted for Medical College Admission Test scores and national Step 1 averages in their multiyear analysis, their findings may be confounded by changes to preclinical classes, addition of basic science to clerkships, and increased emphasis on Step 1 after the curricular change. We had hoped to address these limitations in a quasi-experimental study of Step 1 performance in a single class year.
In 2015, the University of Texas Southwestern Medical Center implemented a curriculum with an 18-month preclinical phase and requirement for students to take Step 1 within the first 6 months of the clerkship phase. This time frame for clerkships and Step 1 was primarily based on student-submitted preferences. Some were assigned a schedule by administrators because of the risk of Step 1 failure. We aimed to analyze the effect of clerkship exposure on Step 1 performance in a single cohort of 2017 test takers after controlling for preclinical grades, admissions qualifications, and demographics.
Each week of graded clerkships before Step 1 appeared to be associated with a 0.35-point increase in Step 1 scores. There was no significant difference in the average Step 1 score compared with prior cohorts. More students requested extensions for their exam date than in prior years.
The strongest association was between Step 1 scores and preclinical grades. Each point increase in the average of all preclinical grades was associated with a 2.66-point increase in Step 1 scores. Further analysis highlighted that clerkship phase sequence is predicted by preclerkship performance. While clerkship exposure appeared to be positively associated with Step 1 scores, it was not possible to fully control for the effect of preclerkship performance on scheduling. The effect of clerkship exposure may only be reflective of preclerkship performance.
We share our analysis of Step 1 performance as an important cautionary tale inherent in medical education research. Limitations such as ours are substantial and can lead to flawed conclusions about curricular redesign. In addition, current studies do not consider modest gains in Step 1 scores in the context of possible unintended consequences, such as increased stress related to a delay in taking Step 1 and the potentially negative impact on clerkship education and performance.
Acknowledgments: The authors wish to thank Shola Rogers, Holli Holbert, and Kelly Black for their assistance in gathering and processing the data.
Dustin Le, MD
First-year internal medicine resident, University of Michigan, Ann Arbor, Michigan; ORCID: http://orcid.org/0000-0003-3231-6758.
Donald Chan, EdM
Third-year medical student, University of Texas Southwestern Medical Center, Dallas, Texas; email@example.com; ORCID: http://orcid.org/0000-0003-0733-5009.
Blake R. Barker, MD
Associate dean of student affairs and associate professor of internal medicine, University of Texas Southwestern Medical Center, Dallas, Texas; ORCID: http://orcid.org/0000-0002-9939-5094.
1. Jurich D, Daniel M, Paniagua M, et al. Moving the United States Medical Licensing Examination Step 1 after core clerkships: An outcomes analysis. Acad Med. 2019;94:371–377.