• Melina C Vassiliou, Pepa A Kaneva, Benjamin K Poulose, Brian J Dunkin, Jeffrey M Marks, Riadh Sadik, Gideon Sroka, Mehran Anvari, Klaus Thaler, Gina L Adrales, Jeffrey W Hazey, Jenifer R Lightdale, Vic Velanovich, Lee L Swanstrom, John D Mellinger, and Gerald M Fried.
• McGill University Health Centre, Montreal, QC, Canada. melinavassiliou@gmail.com
• Surg Endosc. 2010 Aug 1; 24 (8): 1834-41.

BackgroundSimulators may improve the efficiency, safety, and quality of endoscopic training. However, no objective, reliable, and valid tool exists to assess clinical endoscopic skills. Such a tool to measure the outcomes of educational strategies is a necessity. This multicenter, multidisciplinary trial aimed to develop instruments for evaluating basic flexible endoscopic skills and to demonstrate their reliability and validity.MethodsThe Global Assessment of Gastrointestinal Endoscopic Skills (GAGES) Upper Endoscopy (GAGES-UE) and Colonoscopy (GAGES-C) are rating scales developed by expert endoscopists. The GAGES scale was completed by the attending endoscopist (A) and an observer (O) in self-assessment (S) during procedures to establish interrater reliability (IRR, using the intraclass correlation coefficient [ICC]) and internal consistency (IC, using Cronbach's alpha). Instrumentation was evaluated when possible and correlated with total scores. Construct and external validity were examined by comparing novice (NOV) and experienced (EXP) endoscopists (Student's t-test). Correlations were calculated for GAGES-UE and GAGES-C with participants who had performed both.ResultsFor the 139 completed evaluations (60 NOV, 79 EXP), IRR (A vs. O) was 0.96 for GAGES-UE and 0.97 for GAGES-C. The IRR between S and A was 0.78 for GAGES-UE and 0.89 for GAGES-C. The IC was 0.89 for GAGES-UE, and 0.95 for GAGES-C. There were mean differences between the NOV and the EXP endoscopists for GAGE-UE (14.4 +/- 3.7 vs. 18.5 +/- 1.6; p < 0.001) and GAGE-C (11.8 +/- 3.8 vs. 18.8 +/- 1.3; p < 0.001). Good correlation was found between the scores for the GAGE-UE and the GAGE-C (r = 0.75; n = 37). Instrumentation, when performed, demonstrated correlations with total scores of 0.84 (GAGE-UE; n = 73) and 0.86 (GAGE-C; n = 45).ConclusionsThe GAGES-UE and GAGES-C are easy to administer and consistent and meet high standards of reliability and validity. They can be used to measure the effectiveness of simulator training and to provide specific feedback. The GAGES results can be generalized to North American and European endoscopists and may contribute to the definition of technical proficiency in endoscopy.

5 keywords...

hide…