Published in:
01-11-2012 | Review
Systematic review of validity testing in colonoscopy simulation
Authors:
James Ansell, John Mason, Neil Warren, Peter Donnelly, Neil Hawkes, Sunil Dolwani, Jared Torkington
Published in:
Surgical Endoscopy
|
Issue 11/2012
Login to get access
Abstract
Background
Simulation is a useful adjunct to skills-based training. It potentially avoids risk to patients during training and development of basic interventional techniques. This may be of particular relevance in colonoscopy where the learning curve can be long. Several endoscopic devices exist that simulate colonoscopy for training purposes. This study was designed to review the evidence for the validity of these simulators.
Methods
MEDLINE (1947 to present), PubMed, Embase classic + Embase, the metaRegister of Controlled Trials, and the Education Resources Information Center (ERIC) were searched for studies validating colonoscopy simulators. For each study, we recorded the type of simulator used, the tasks assessed, the endpoints reported, and the type of validity measured. Common endpoints between studies were compared, and the evidence was graded.
Results
Thirteen studies met the inclusion criteria. Construct validity was reported in five (41.7 %) studies for the Accutouch HT Immersion (cases 1, 3, and 4), four studies (33.3 %) for the GI mentor II (Simbionix) (Modules 1.1, 1.3, 1.7, 2.1, and 5), two studies (16.7 %) for the Olympus Endo Ts-1 2nd Generation, and one study for the Endo X bovine model. Face validity was reported for the Accutouch HT Immersion, the Olympus 2nd Generation, and the KAIST-Ewha. Content validity was reported for the all simulators, excluding the KAIST-Ewha. The only report of criterion validity was for the Endo X bovine model.
Conclusion
Evidence exists to support the face, content, and construct validity of several virtual reality colonoscopy simulators for specific diagnostic and therapeutic modules with selected endpoints. One study demonstrates content, construct, and criterion validity for an ex vivo animal platform. Further work is needed to demonstrate the criterion validity of all devices.