Abstract
Background Student scores on the Learning and Study Strategies Inventory (LASSI), an assessment of academic skills, have been shown in previous studies to be significantly different between U.S. medical students based on their scores in various examinations during the pre-clerkship curriculum. This study aimed to evaluate LASSI and other early academic performance markers for predicting the likelihood of shelf exam underperformance in third-year medical students. Methods A retrospective analysis of student-specific demographic information and medical school exam performance from 220 medical students from the University of Texas Southwestern was performed. Students were then categorized based on underperformance (score in <25th percentile) on each NBME shelf exam and statistical analysis was performed to identify predictors of shelf underperformance. Results For predicting Surgery shelf underperformance, pre-clerkship final exam average (PCA), STEP 1, and LASSI Time Management (TMT) were statistically significant in univariate analysis. Internal Medicine: PCA, STEP 1, LASSI Attitude (ATT), Test Strategies (TST), and TMT. Pediatrics: PCA and STEP 1 quartile. Obstetrics-Gynecology: PCA, STEP 1, and LASSI Anxiety (ANX), with ANX an independent predictor on multivariate analysis. Neurology: PCA, STEP 1, LASSI ANX, Information Processing (INP), TST, and average LASSI, with PCA, LASSI Concentration (CON), TMT, and ANX independent predictors on multivariate analysis. Family Medicine: PCA, STEP 1, LASSI ANX, TST, and Using Academic Resources (UAR), with PCA an independent predictor on multivariate analysis. Psychiatry: only STEP 1 was significant. Conclusion In contrast to previous studies, no single LASSI scale was significantly associated with underperformance on all 7 NBME shelf exams. Univariate analysis identified several LASSI scales that correlated with NBME underperformance, but the drastic inter-clerkship heterogeneity makes use of these scales in early academic intervention impractical. Conversely, PCA was found to be strongly associated with shelf exam underperformance.