Association of Resident Independence With Short-term Clinical Outcome in Core General Surgery Procedures

Abstract

Importance: Prior studies evaluating the effect of resident independence on operative outcome draw from case mixes that cross disciplines and overrepresent cases with low complexity. The association between resident independence and clinical outcome in core general surgical procedures is not well defined.

Objective: To evaluate the level of autonomy provided to residents during their training, trends in resident independence over time, and the association between resident independence in the operating room and clinical outcome.

Design, setting, and participants: Using the Veterans Affairs Surgical Quality Improvement Program database from 2005 to 2021, outcomes in resident autonomy were compared using multivariable logistic regression and propensity score matching. Data on patients undergoing appendectomy, cholecystectomy, partial colectomy, inguinal hernia, and small-bowel resection in a procedure with a resident physician involved were included.

Exposures: Resident independence was graded as the attending surgeon scrubbed into the operation (AS) or the attending surgeon did not scrub (ANS).

Main outcomes and measures: Outcomes of interest included rates of postoperative complication, severity of complications, and death.

Results: Of 109 707 patients who met inclusion criteria, 11 181 (10%) underwent operations completed with ANS (mean [SD] age of patients, 61 [14] years; 10 527 [94%] male) and 98 526 (90%) operations completed with AS (mean [SD] age of patients, 63 [13] years; 93 081 [94%] male). Appendectomy (1112 [17%]), cholecystectomy (3185 [11%]), and inguinal hernia (5412 [13%]) were more often performed with ANS than small-bowel resection (527 [6%]) and colectomy (945 [4%]). On multivariable logistic regression adjusting for procedure type, age, body mass index, functional status, comorbidities, American Society of Anesthesiologists class, wound class, case priority, admission status, facility type, and year, factors associated with a complication included increasing age (adjusted odds ratio [aOR], 1.19 [95% CI, 1.16-1.22]), emergent case priority (aOR, 1.41 [95% CI, 1.33-1.50]), and resident independence (aOR, 1.12 [95% CI, 1.03-1.22]). On propensity score matching, AS cases were score matched 1:1 to ANS cases based on the variables listed above. Comparing matched cohorts, there was no difference in complication rates (817 [7%] vs 784 [7%]) or death (91 [1%] vs 102 [1%]) based on attending physician involvement.

Conclusions and relevance: Core general surgery cases performed by senior-level trainees in such a way that the attending physician is not scrubbed into the case are being done safely with no significant difference in rates of postoperative complication.

Figure 1.. Proportion of Cases Completed Without the Attending Surgeon Scrubbed by Procedure Over Time

Figure 2.. Results of Multivariable Logistic Regression for Attending Physician Involvement and Occurrence of a Postoperative Complication by Procedure

Adjusted for age, body mass index class, functional status, chronic obstructive pulmonary disease, tobacco use, alcohol use, American Society of Anesthesiologists class, admission status, emergent nature, procedure type (entire cohort only), wound class, facility level, surgical approach, and year of procedure. aOR indicates adjusted odds ratio.