SIMPL Study of Surgical Feedback: Faculty Identity Outweighs Trainee Factors in Narrative Assessments

To identify factors associated with attending surgeons providing narrative feedback using the Society for Improving Medical Professional Learning assessment tool after operative procedures. We hypothesized that certain resident and attending characteristics, operative performance, and timing variables would be associated with the likelihood of narrative feedback.

This was a retrospective cross-sectional study of operative assessments completed through the Society for Improving Medical Professional Learning application between July 2020 and April 2024. Statistical analyses included univariate comparisons using chi-squared tests, followed by multivariate logistic regression with random effects to quantify variance due to clustering by individual attending physicians and residents.

A single midwestern general surgery residency program at a tertiary care academic hospital.

Attending surgeon evaluations of general surgery residents between July 2020 and April 2024 (n = 2869).

Overall, narrative feedback occurred in 66.5% of assessments. On univariate analysis, several resident and case characteristics (e.g., junior vs senior level, weekday vs weekend cases) appeared associated with receiving feedback. However, in the multivariable mixed-effects model, the variation due to the attending surgeon was the primary source of difference. After controlling for clustering, about 56% of the variance in narrative feedback was attributable to the attending surgeon's identity. In contrast, the variance attributed to individual residents was approximately 5%. In the adjusted model, only a few fixed effects remained significantly associated with narrative feedback. Significant predictors in the adjusted model included PGY2 status (odds ratio [OR] 0.59, 95% confidence interval [CI] 0.35-0.99), average case complexity (OR 1.56, 95% CI 1.19-2.05), assessments conducted in July (OR 3.41, 95% CI 1.67-6.97), and assessments from 2020 (OR 4.56, 95% CI 1.91-10.86).

Our findings highlight that whether a resident receives narrative feedback depends largely on which attending surgeon is completing the assessment, rather than on resident or case factors alone.