Ilay Habaz1, Allan Okrainec1, Oscar M Crespin1, Andrea V Kwong1, Ethan Weiss1, Luis WC Liu2, Timothy D Jackson1, Silvana Perretta3, Lee L Swanstorm4, Eran Shlomovitz1. 1General Surgery, University Health Network, 2Division of Gastroenterology, University Health Network, Toronto, ON, Canada, 3Institute for Image Guided Surgery IHU-Strasbourg, Strasbourg, France, 4The Oregon Clinic, Division of GI/MIS, Portland, OR, USA
Introduction: The lack of a readily accessible and cost-effective model for the simulation, practice, and evaluation of endoscopic skills continues to present a major barrier for training residents. We have previously described a system for conversion of the Fundamentals of Laparoscopic Surgery box (FLS) for flexible endoscopic simulation. Six endoscopic tasks focusing on scope manipulation, and other clinically relevant endoscopic skills are performed within a 5-minute time limit per task. This study describes our experience and preliminary validation efforts with the first 100 participants.
Methods: A total of 100 participants were evaluated on the simulator. Thirty individuals were classified as experts (having done over 200 unsupervised endoscopic procedures), and 70 were classified as novices. Of the 100 participants, 55 individuals were retested on the simulator within a period of four months. These 55 individuals were also evaluated using the “Global Assessment of Gastrointestinal Endoscopic Skills” (GAGES). Pearson correlation was used to assess the association between task completion time and GAGES score.
Results: Experts completed all six tasks significantly faster than novices. Although all tasks had statistically significant test-retest reliability for experts and novices who did not practice in between tests, tasks 1-4 had the strongest retest correlation. Moderate correlations between lower completion times and higher GAGES scores was observed for all tasks except task 5, (see table).
Conclusions: The results of this 100-participant milestone provide initial evidence for the simulator’s validity under Messick’s framework. As expected, experts perform significantly better than novices. Additionally, preliminary supportive data is demonstrated for test-retest reliability, as well as GAGES score correlation. Additional studies to determine a scoring system for this simulator are ongoing.
Presented at the SAGES 2017 Annual Meeting in Houston, TX.
Abstract ID: 88285
Program Number: S084
Presentation Session: MIS Education Session
Presentation Type: Podium