SCORE-Based Simulated ABSITE Exam Performance as a Predictor of Performance on the ABSITE

Abstract

The American Board of Surgery In-Training Examination (ABSITE) is a crucial, objective assessment of surgical knowledge during training. In 2014, the American Board of Surgery (ABS) announced the alignment of the ABSITE to the SCORE® (Surgical Council on Resident Education) Curriculum Outline for General Surgery Residency. We hypothesized that implementing a pre-ABSITE SCORE-based exam would help identify underperforming residents and provide early guidance to improve performance on the ABSITE. In October 2014, our university-based surgical residency program began administering a yearly comprehensive pre-ABSITE SCORE-based exam consisting of 225 to 250 multiple-choice questions selected from the SCORE question bank to all our general surgery residents, preliminary and categorical. The 4-hour exam addresses both clinical management (80%) and applied sciences (20%). Residents receive reports with their scores (percentage correct). Residents performing at less than 60% meet with the Program Director for discussion and formulation of a study plan. Correlational analysis was performed between resident ABSITE scores, pre-ABSITE SCORE-based exam scores, gender, resident status (preliminary vs. categorical), postgraduate year (PGY), and the United States Medical Licensing Examination (USMLE) Step 1 and Step 2 scores. A total of 244 exam scores (122 pre-ABSITE SCORE-based exams and 122 matched ABSITE) were completed by 51 residents at different PGY levels (32 PGY1, 32 PGY2, 20 PGY3, 19 PGY4, and 18 PGY5). Fifty-seven percent were males, 62% were categorical residents, and 38% were preliminary residents. October pre-ABSITE SCORE-based exam scores were compared to the subsequent January ABSITE scores. Categorical residents completed 101 (83%) of the January exams, while preliminary residents completed 21 (17%) of these paired exams. We found strong correlations between the correct percentage on ABSITE and pre-ABSITE SCORE-based scores (r = 0.637, p < 0.001), between the correct percentage on ABSITE and PGY (r = 0.688, p < 0.001), and between ABSITE and resident status (r = 0.462, p < 0.001). Additionally, there was a weak to negligible correlation between the correct percentage on ABSITE and resident gender (r = 0.274, p = 0.001), USMLE-2 (r = 0.12, p = 0.16), and USMLE-1 (r = 0.04, p = 0.653). Multiple regression analysis, with all predictors, was performed to predict the percentage score on the ABSITE and produced R2 0.58, with an adjusted R2 of 0.57, with a large size effect, p < 0.001. After controlling for the other variables, three factors reached statistical significance (p < 0.05): pre-ABSITE SCORE-based exam scores, PGY, and resident gender. We found a strong correlation between performance on the pre-ABSITE SCORE-based exam and performance on the ABSITE exam. Surgery residents are encouraged to start studying earlier and to utilize SCORE contents as outlined by the ABS in their study plan.