Validation of Two Novel and Complementary Training Platforms for Small Joint Arthroscopy.

Purpose: We developed 2 complementary low-fidelity models to be used to create the tool skills needed to perform small joint arthroscopy. The purpose of the study was to establish the face and construct validity of the 2 models.

Methods: The "foundation model" was constructed from lemon and radish sections, and the "advanced model" was constructed from a chicken knee. Using both models, novice, intermediate, and experienced participants were asked to perform specific tasks and were timed and scored on their performance. The experienced surgeons were given a 16-item survey to rate how closely each model emulated reality to determine face validity.

Results: For the foundation model, the mean total time for the completion of tasks was 1,138 seconds for novices, 1,059 seconds for intermediates, and 631 seconds for experienced, with significant differences between the groups for time to complete 2 of the tasks. With a maximum possible score of 50 points for the correct performance of all tasks, the mean total performance score was 23 for novices, 31.8 for intermediates, and 42.2 for experienced operators. For the advanced model, the mean total time for completion was 266 seconds for novices, 147 seconds for intermediates, and 72 seconds for experienced participants. With a maximum possible score of 31 points for the correct performance of all tasks, the mean total performance score was 1.9 for novices, 15.0 for intermediates, and 24.3 for experienced participants. The average scores for the face validity surveys using a 5-point Likert scale were 4.2 and 4.5 of 5 possible points for the foundation and advanced models, respectively.

Conclusions: Experienced operators completed the tasks more quickly and had higher performance scores than the operators in other groups. This correlation between experience and performance suggests that both models have construct validity. The face validity scores were on the upper end of the scale, suggesting that both models emulate reality for experienced operators.

Clinical Relevance: These novel models provide low-cost, available and valid simulations conducive to high-repetition training.