Construct validity of the LapSim laparoscopic surgical simulator

Background Computer-aided simulators may increase the safety and efficiency of training in laparoscopic surgery. Before implementation of the Immersion LapSim (Gaithersburg, MD) simulator in our training curriculum, we wished to determine its construct validity (ie, whether the simulator could differentiate laparoscopic novices from trainees with greater experience). Methods Subjects were medical students (MS), residents (RES), and laparoscopic faculty (FAC). Subjects performed 10 repetitions of 6 LapSim tasks. The LapSim measured performance in 6 to 10 parameters per task, and performance was compared between groups. Post hoc t tests were used to make pair-wise comparisons among the 3 groups using the Bonferroni technique. Statistical significance was set at P < .05. Results The degree of prior laparoscopic experience was reflected in performance on at least 1 parameter for each task. Several patterns of performance between MS, RES, and FAC were observed. Conclusions The LapSim has performance parameters that reliably differentiate between subjects with varying laparoscopic experience. However, some performance parameters do not differentiate between groups. To accurately measure a trainee’s skill level, only parameters that sensitively measure the true level of performance should be used.