ABSTRACT
Objectives: A prospective observational trial to develop and gather validity evidence using Messick's framework for a simulator-based test in TURB. Methods: Forty-nine doctors were recruited from urology departments (Herlev/Gentofte University Hospital, Rigshospitalet Copenhagen University Hospital and Zealand University Hospital Roskilde) and enrolled from April to September 2018. The TURB Mentor™ virtual reality (VR) simulator was assessed at an expert meeting selecting clinically relevant cases and metrics. Test sessions were done on identical simulators at two university hospitals in Denmark. All participants performed three TURB procedures on the VR simulator. Simulator metrics were analysed with analysis of variance (ANOVA) and metrics with the ability to discriminate between groups were combined in a total simulator score. Finally, a pass/fail score was identified using the contrasting groups' method.Results: Eleven simulator metrics were found eligible and four had significant discrimination ability between competency levels: resected pathology (%) (p = 0.008); cutting in bladder wall (n) (p = 0.004); time (s) (p = 0.034); and inspection of the bladder wall (%) (p = 0.002). The internal structure of the total simulator score [(resected pathology*inspection of the bladder wall)/time] was high with the intraclass correlation coefficient, Cronbach's alpha = 0.85. The mean total simulator score was significantly lower in the novice group than in the intermediate, 15.9 and 25.6, respectively (mean difference = 9.7, p = 0.011) and experienced group, 30.6 (mean difference = 14.7, p < 0.001). A pass/fail score of 22 was identified.Conclusion: We found validity evidence for a newly developed VR simulator-based test and establised a pass/fail score identifying surgical skills in TURB. The TURBEST test can be used in a proficiency-based TURB simulator training programme for accreditation prior to supervised procedures on patients.
