Construct validity: experts and novices performing on the Xitact LS500 laparoscopy simulator.

Abstract

This study was undertaken to establish construct validity of the Xitact LS500, a virtual reality laparoscopic cholecystectomy simulator. The primary research statement is: "The clip-and-cut task on the Xitact LS500 virtual reality laparoscopic cholecystectomy simulator mimics the surgical procedure of the clipping and cutting of the cystic duct and artery during the laparoscopic cholecystectomy adequately." According to the level of experience of the surgeon, an "expert group opinion" was formed resulting from 37 surgeons having performed over 100 laparoscopic cholecystectomies, and a "novice group opinion" was formed resulting of 37 surgeons having no experience at all with performing laparoscopic cholecystectomy. Participants received an instructed hands-on "tour" on the Xitact simulator and performed three formal simulation runs. The "novice group" is younger and more surgeons are female. Performance scores in the "expert" group are significantly higher on the second (p value 0.011) and third (p value 0.005) run, compared to the novices' scores. Experts are significantly faster on completion of all three runs. There is an increase in score over runs in both groups, which is statistically significant in the "expert group." Less than one-third of surgeons in either group are able to correctly predict their performance score as generated by Xitact. Both "experts" and "novices" feel it is useful to train with Xitact LS500 in the surgical curriculum. Three hypotheses, formulated to operationalize the primary research statement, could be answered affirmatively. Although further validation studies are needed, the Xitact LS500 simulator seems to be able to discriminate between expert and novice surgeons in this research setting, and thus the construct for this setting is considered to be valid.