Back to 2025 Karmody Posters
Blinded Video Analysis As A Tool To Reduce Gender Bias In Surgical Resident Evaluations
Jorge A. Sulca Flores, BS, Kenneth Lynch Lynch, PhD, APRN, Carla C. Moreira, MD, Adena J. Osband, MD, Pete Soden, MD, Katherine MacCallum, MD, Thomas Carruthers, MD.
The Warren Alpert Medical School, Providence, RI, USA.
Objective: Prior studies have shown an implicit gender bias when evaluating performance of general surgery residents, favoring male residents. Our study investigates this bias as it applies to the evaluation of general surgery residents during vascular anastomosis training.
Methods: 31 surgery residents participated in a 7-week annual vascular anastomosis training at a large Northeast academic center (2019-2021). Residents were video recorded performing end to side anastomosis for both a Pre and Post test. During recording, residents wore surgical gloves, the field of view remained on their hands and anastomosis only, and all audio was removed. Videos were sent to each resident and two independent attendings for evaluation and scoring. Scoring categories included: preparing graft (0-2), graftotomy (0-5), anastomosis (0-15), global rating scale (0-40), overall score (0-62), leak rate (mL), and completion time. Using resident self-evaluation and attending scoring, t-tests were calculated comparing resident pre vs post test scores to attending based on the resident’s gender.
Results: Residents’ technical skills significantly improved, with both resident and attending post-test scores significantly higher than pre-test across all categories (p < 0.0001). Resident self-scoring improved from a pre-test mean of 29.1 to 46.57 post-test, mean difference 17.47 (p < 0.0001). Attending scores improved from a pre-test mean of 29.6 to 49.92 post-test, mean difference 20.32 (p < 0.0001). Leak rate improved with mean difference of 15.32 (p < 0.0001). Completion time showed no significant difference. Based on 15 male (48.4%) vs 16 female residents (51.6%), attending pre-test score showed significant difference favoring females with respect to anastomosis and overall score, mean difference 1.53 (p = 0.0237, 95%) and 3.54 (p = 0.0402, 95%) respectively. Attending post-test scores showed no significant difference in any category or overall score with respect to gender, mean difference 0.93 (p = 0.541) (Table 1).
Conclusion: Vascular anastomosis training showed significant improvement in resident technical skills with no significant difference in scoring with respect to gender on the post test. While more research is needed in this area, our study shows blinded video analysis can be an effective way to eliminate bias from the resident evaluation process with respect to technical skill.
Back to 2025 Karmody Posters
