Resident evaluations: what are the predictors of future negative outcomes in surgical education?

Abstract

Background: Evaluation variables that correlate with future negative outcomes in surgery allow for actionable steps and early intervention. Diverse evaluation systems have been used to quantify progression of residents. This study assessed our internal evaluation system to evaluate for variables that could predict negative outcomes in surgical education and, secondarily, whether differences in resident rating existed based on faculty–resident gender dyad. Methods: This is a retrospective single-institution review of general surgery resident evaluations (2008–2020). Demographics, simulation lab, technical skill, and percentile for overall clinical performance are given by attendings. Logistic regression models were used to determine whether any factors were associated with negative resident outcomes and to determine if the faculty–resident gender dyad influenced evaluation ratings. Results: A total of 3117 evaluations of 116 residents were included (66% male). Twenty negative outcomes were identified. In univariate logistic regression model American Board of Surgery In-Service Training Exam (ABSITE) scores, simulation lab scores and technical evaluations ≤ 20th percentile were associated with negative outcomes (p = 0.04). In multivariate analysis, no variables were associated with negative outcomes. Male staff tended to grade residents lower than female staff, independent of resident gender. Conclusion: Although multiple subjective and objective variables, including national standardized tests, of surgical trainees are measured and reported, none of the variables collected over 10 years in 116 residents including over 3000 evaluations were independently associated with negative outcomes. This points to a need for development of new objective variables that can predict negative outcomes in residency. Preemptive identification of “at-risk” residents remains elusive.