Development of an electrosurgery-compatible simulation task for quantitatively assessing oral cancer resection skills: initial validity evidence

Abstract Background The quality of oral cancer resection is extremely important for patient outcomes, such as local control and survival. However, most current simulators either provide only rater-dependent feedback or are not compatible with electrosurgery. Therefore, we developed an electrosurgical simulation task for oral cancer resection that provides objective quantitative metrics and collected initial validity evidence. Methods We developed a soft tissue simulation task using a plant-derived model that supports electrosurgery. As quantitative measures demonstrating “ensuring appropriate margins” in oral cancer resection and “maintaining safety” during electrosurgery, we employed nine-directional margin error distance and tumor bed carbonization degree measured using a spectral colorimeter. As validity evidence of the task, 10 expert surgeons completed a questionnaire about the task. In addition, five experts and 12 novices performed the task, and quantitative data obtained from their performance was used for evaluation. Results The replication of oral cancer resection was highly evaluated (4.4 out of 5 points), and quantitative measures for evaluating the skills of surgeons (4.8 out of 5 points) were agreed upon by experts. The internal consistency of the measures was good (Cronbach's alpha: 0.803). Compared to novices, experts had smaller margin errors (0.79 mm vs 2.45 mm), lower carbonization (ΔE: 2.33 vs 8.70), faster resection times, and fewer grasping attempts (all P < 0.001). Conclusion This user-friendly plant-derived simulation task is compatible with electrosurgery and provides objective quantitative performance metrics. These findings support its use as a practical assessment tool for formative feedback in simulation-based training.