Poster5-03: WHAT, WHERE, AND WHEN: AN ASSESSMENT OF STUDY HABITS AND NBME® SURGERY SUBJECT EXAM PERFORMANCE
Rachel L Slotcavage, MD, James P Selig, PhD, Jason S Mizell, MD, Carol R Thrush, EdD; University of Arkansas for Medical Sciences

Background: Material tested on the NBME® Surgery Subject Exam (SSE) can be disparate from clinical clerkship education, causing difficulties in advising students regarding successful study strategies.  The purpose of this study was to identify study habits, study resources, and personal characteristics of third year medical students that may be associated with higher SSE scores.

Methods: This study was IRB approved and exempt.  A 25-item survey, adapted from prior published research and evaluating study habits, resources, and personal characteristics, was administered to third-year medical students rotating on the Surgery Clerkship during the 2017-2018 academic year.  The response rate was 93% (n=139/150).  Univariate and multivariate analyses were performed to evaluate associations between study habits, resources, and characteristics, and students’ SSE scores.

Results: Frequency of reading (number of days/week) was associated with significantly higher SSE scores (p=0.013), but total number of hours reading was not (p=0.163).  However, hours spent reading in the two weeks preceding testing showed significance (p=0.029).  Where students studied and group vs. independent studying were not associated with SSE scores. 

No single study resource or combination showed association with SSE scores.  The use of a newly recommended textbook approached significance (p=0.064); this was used more frequently by high-performing students.  Use of the assigned clerkship textbook, online medical references/videos, prep books, and flashcards were not significant.

Regarding student characteristics, performance on USMLE® Step 1, attitude about SSE importance, reported self-discipline, and prior course remediation were each associated with SSE score (p<0.0001).  Interest in surgical vs. non-surgical fields (p=0.084) and focus on SSE preparation vs. patient care (p=0.064) approached significance.  Gender and prior advanced degrees were not significant.

When all variables showing evidence of association (p=≤0.08) in univariate analyses were placed in a multivariate linear regression model, variables remaining significant were performance on Step 1 (p<0.0001), frequency of reading (p=0.047), and attitude regarding SSE importance (p=0.047).

Conclusion: Students’ past performance on standardized medical testing is associated with future performance, but is a fixed factor.  Modifiable factors associated with higher SSE performance include frequent exam preparation rather than widely-spaced study blocks, regardless of duration, and improving attitudes on the importance of the exam.