Validation of an endoscopic part-task training box as a skill assessment tool

Abstract

There is no objective methodology to assess trainee progress in endoscopy. Our prior work hasdetailed the development of the endoscopic part-task training box. To assess validity evidence regarding relationship to other variables by evaluating a correlation between level of endoscopic experience and training box score. Prospective validation study. Three academic institutions. A total of 42 participants: 7 novices, 7 first-year GI fellows, 7 second-year GI fellows, 7 third-year GI fellows, 7 attending physicians, and 7 interventional attending physicians. The training box consists of 5 modules: retroflexion, knob control, torque, polypectomy, and navigation/loop reduction. Performance is scored for precision and speed. Each participant was required to complete the training box once. Additionally, 5 participants at different endoscopic levels completed the training box 3 times at 1-week intervals. A correlation between level of endoscopic experience and training box score. All 42 participants completed the 5 modules during a single session. Aggregate training box scores differed significantly between each training level (P values< .05). Individual modules significantly differentiated between experience-level groups (novices, fellows, and attending physicians; P values< .01). Participants who repeated the training box demonstrated score improvement over time, with persistence of separation between training levels. The training box focuses only on the technical aspects of endoscopy and does not address the cognitive elements of endoscopic training. The endoscopic part-task training box is able to objectively assess endoscopic ability by differentiating scores based on clinical experience. Further multicenter efforts are now needed to establish learning curves and to correlate use of the simulator with improved clinical aptitude.