Proficiency-based progression training: quality assured preparation for the practice of surgery

Purpose: Investigate the effectiveness of proficiency-based progression (PBP) training coupled with simulation for the acquisition of surgical skills and to define the essentials for the development of a comprehensive, validated, PBP training curriculum. Methods: Task deconstruction of an arthroscopic Bankart repair (ABR) in which step, error, and sentinel (more serious) error metrics was conducted. Novice and experienced surgeons were evaluated using the metrics with a medium fidelity shoulder model (training tool) and separately with a cadaver shoulder (assessment tool). A randomized, controlled, blinded trial was conducted comparing operative performance for 4th and 5th year orthopedic residents for one of 3 training curriculums: 1) control: traditional Arthroscopy Association resident training (‘apprenticeship model’ N = 14); 2) a simulator enhanced curriculum (N = 14); and 3) a PBP curriculum coupled with a model simulator (N = 16). Results: Face and content validity were confirmed for the ABR metrics with a modified Delphi panel. Construct validity of the metrics coupled with the model simulator and with a cadaver shoulder was verified and proficiency benchmarks established. The PBP-trained group (Group C) made 56% fewer objectively assessed errors than the traditionally trained residents (Group A) and 41% fewer than Group B. Compared with Group A, Group B participants were 1.4 times and Group C, 5.5 times as likely to achieve the final proficiency benchmark for an ABR. Conclusions: The randomized trial comparing three different training curricula demonstrated unambiguous superiority of the PBP curriculum coupled with a medium fidelity simulator. Performance metrics must be unambiguously defined and able to be reliably scored. Error metrics are the most valuable in discriminating between levels of operative performance. Simulators are most useful when they serve as a vehicle to deliver a strong, metric based curriculum, which must be developed prior to the selection of specific simulations