The development of an objective model to assess arthroscopic performance.

BACKGROUND Many residency and fellowship programs have cadaver laboratories to teach and practice arthroscopic skills. However, there is currently no validated method of evaluating arthroscopic skill in this setting. The purpose of the present study was to develop and validate an objective model for evaluating basic arthroscopic proficiency on a cadaver knee in a bioskills laboratory. METHODS Two measures from the educational literature were adapted for use specifically for arthroscopy: a task-specific checklist and a global rating scale were combined to create the Basic Arthroscopic Knee Skill Scoring System. Fifty-nine residents, three sports medicine fellows, and six sports medicine fellowship-trained attending surgeons were recruited. After completing a demographic survey, including the postgraduate year and number of knee and shoulder arthroscopies performed, each subject performed a diagnostic knee arthroscopy and a partial meniscectomy on a cadaver knee while being assessed by a single evaluator using the Basic Arthroscopic Knee Skill Scoring System. RESULTS There was a strong positive correlation between global rating scale scores and both the postgraduate year (r = 0.93, p < 0.01) and the ranked number of knee arthroscopies performed (r = 0.88, p < 0.01). These scores detected significant differences between postgraduate years 1 and 2, and years 4 and 5 at the p <or= 0.01 level and between years 2 and 3 at the p <or= 0.05 level. Task-specific checklist scores were moderately correlated with both postgraduate year (r = 0.73, p < 0.01) and ranked number of knee arthroscopies performed (r = 0.64, p < 0.01). These scores detected significant differences only between postgraduate year-1 and year-2 residents at the p <or= 0.01 level, indicating that these skills are acquired early in training. CONCLUSION The Basic Arthroscopic Knee Skill Scoring System can capture and differentiate levels of arthroscopic skill and was validated to objectively evaluate basic arthroscopic proficiency in a bioskills laboratory. This model will allow benchmarks of surgical skill to be created for each level of residency training and individual progress to be monitored over time.