Abstract
Background: The NSQIP database enables measurement of postoperative outcomes across a spectrum of practice settings. This allows for observations about potential effects of resident participation in surgical care during training. Study Design: We queried the NSQIP database for 6 index laparoscopic surgical procedures performed during 2005-2008. Selected procedures require varying skill level (eg, appendectomy, cholecystectomy, gastric bypass, fundoplication, colectomy, and inguinal hernia), and 79,720 cases were identified. Preoperative, operative, and postoperative outcomes for each procedure were tabulated. Operative and postoperative outcomes assessed included operative time, hospital length of stay, mortality, morbidity, and return to the operating room. Initial analysis compared cases done with a resident present with cases done without residents. Subset analysis was done to determine possible differences in outcomes based on the level of resident participating, divided into Junior (PGY1-2), Senior (PGY3-5), or Fellow (PGY>5). Groups were scrutinized for both clinical and statistical differences. Results: Preoperative characteristics were similar between groups. Operative times were 20% to 47% longer with resident participation, with bigger differences seen in more basic procedures. Mortality and return to the operating room were not clinically different between the groups. Morbidity rates were higher in all procedures with resident participation. More senior residents were associated with longer operative times, without adverse impact on outcomes. Conclusions: Resident participation increases operative times for laparoscopic surgery considerably. Morbidity is statistically higher with resident participation but differences are unlikely to be clinically significant. Resident participation is a surrogate for the learning environment. These findings provide impetus for additional development of training techniques that occur outside the operating room.