Background: The purpose of our study was to evaluate the construct validity of laparoscopic technical performance measures and the face validity of three laparoscopic simulations.
Materials and methods: Subjects (N = 27) of varying levels of surgical experience performed three laparoscopic simulations, representing appendectomy (LA), cholecystectomy (LC), and inguinal hemiorrhaphy (LH). Five laparoscopic surgeons, blinded to the identity of the subjects, rated the subjects on procedural competence on a binary scale and in four skills categories on a 5-point scale: clinical judgment, dexterity, serial/simultaneous complexity, and spatial orientation. Using a task-specific checklist, non-clinical staff assessed the technical errors. The level of surgical experience was correlated with the ratings, the technical errors, and the time for each procedure. Subject responses to a survey regarding the utility of the inanimate models were evaluated.
Results: Years of experience directly correlated with the skills ratings (all P < 0.001) and with the competence ratings across the three procedures (P < 0.01). Experience inversely correlated with the time for each procedure (P < 0.01) and the technical error total across the three models (P < 0.05). Nearly all subjects agreed that the corresponding procedures were well represented by the simulations (LA 96%, LC 96%, LH 100%).
Conclusion: The laparoscopic simulations demonstrated both face and construct validity. Regardless of the level of surgical experience, the subjects found the models to be suitable representations of actual laparoscopic procedures. Task speed improved with surgical experience. More importantly, the quality of performance increased with experience, as shown by the improvement in the skills assessments by expert laparoscopic surgeons.