Self-directed learning by video as a means to improve technical skills in surgery residents: a randomized controlled trial.

BACKGROUND: With their demanding schedules, surgical residents have limited time to practice techniques. The aim is to evaluate the pedagogic model of self-directed learning using video in surgery residents. METHODS: Informed consent was obtained from all the participants. A randomized controlled trial was conducted in 2018 at Hôpital Maisonneuve-Rosemont (University of Montreal). Participants were general surgery residents. There were 27 eligible residents; 22 completed the study. They were filmed performing an intestinal anastomosis on cadaveric pig bowel. The self-directed learning by video (SDL-V) group was given an expert video, which demonstrated the technique performed by an experienced surgeon. The control group continued with their regular duties. Three weeks later, participants performed a second filmed anastomosis. Two attending surgeons evaluated the residents' filmed anastomosis using the Objective Structured Assessment of Technical Skills scale. After their second anastomosis, all participants had access to the expert video and completed a survey. RESULTS: Score did not differ significantly between groups during the first (control: 23.6 (4.5) vs. SDL-V: 23.9 (4.5), p = 0.99, presented as mean (SD)) or second filmed anastomosis procedure (control: 27.1 (3.9) vs. SDL-V: 29.6 (3.4) p = 0.28). Both groups improved significantly from pre- to post-intervention (mean difference between the two anastomosis procedure with 95% CI for control: 3.5, [1.1; 5.9] and for SDL-V: 5.8, [3.4: 8.2]). Correlation between the evaluators for score was moderate (r = 0.6, 95% CI: [0.3: 0.8]). The pass/fail global evaluation exhibited poor inter-rater reliability (Kappa: 0.105, 95% CI: [- 0.2:0.4]). On the survey, all participants wanted more expert-made videos of specific surgical techniques. CONCLUSIONS: Despite a higher final OSATS score for the intervention group, self-directed learning by video failed to produce a statistically significant difference on the overall OSATS scores between the two groups in this small cohort.