Correlation of Supervised Independence and Performance with Procedure Difficulty amongst Surgical Residents Stratified by Post Graduate Year

This study investigates the role of procedure difficulty on attending ratings of supervised levels of independence and procedural performance amongst general surgery residents, while accounting for case complexity. Attending ratings for residents were obtained from System for Improving and Measuring Procedural Learning (SIMPL) database. Current procedural terminology (CPT) codes were used to match procedures to a corresponding work relative value unit (wRVU) as a surrogate for procedure difficulty. Three categories of wRVU (22) were identified using recursive partitioning. Procedures were also divided into ‘Core’ or ‘Advanced’ as defined by the American Board of Surgery Surgical Council on Resident Education (SCORE). Temporal advancement in resident skill was accounted for through academic quarterly analysis. A generalized estimating equations (GEE) approach was used to form separate multivariable logistic regression models for meaningful autonomy (MA) and satisfactory performance (SP) adjusted for potential clustering by program, subject, and rater. Models were further adjusted for core/advanced procedures, attending rated complexity, and academic quarter. A total of 33,281 ratings were analyzed. Overall, 51.6% were rated as MA and 44.4% as SP. For core procedures, surgical residents rated as MA (53.5%) and SP (45.7%), which was twice as high as those for advance procedures (MA-29.2%, SP-29.0%). MA and SP both decreased with increasing wRVU (Figure 2 &3). Using a wRVU22.00 (Table 3). Post graduate year (PGY) 5 residents in the final quarter of training obtain MA in 95.5% and SP 92.9% for core procedures with wRVU