Surgeon knowledge contributes to the relationship between surgeon volume and patient outcomes in rectal cancer

OBJECTIVE: To determine whether surgeon knowledge contributes to the relationship between surgeon procedure volume and patient outcomes in rectal cancer. BACKGROUND: Although previous research has shown that treatment by high-volume surgeons is associated with improved outcomes among patients with rectal cancer, the mechanisms for such an association are not well understood. METHODS: In 2009, a mail survey with 8 questions pertaining to rectal cancer care was created, modified for content validity, and sent to all general surgeons in Nova Scotia, Canada. Patients with rectal cancer, who were treated by the survey respondents between July 1, 2002, and June 30, 2006, were identified retrospectively, and a comprehensive standardized review of medical records was used to collect outcome data for this population-based cohort. The association between surgeon survey score (dichotomized into high- and low-score groups on the basis of the median score), surgeon procedure volume, and patient outcomes was examined. RESULTS: Of 521 patients who underwent treatment with curative intent from July 1, 2002, to June 30, 2006, 377 patients (72%) were treated by 25 surgeons who responded to the survey. After controlling for patient and tumor factors, patients treated by high-volume surgeons were more likely to receive a total mesorectal excision (TME) [odds ratio (OR) = 3.89; 95% confidence interval (CI), 2.20-5.83], more likely to undergo an adequate lymph node harvest (OR = 3.67; 95%CI, 2.36-5.70), less likely to have a permanent colostomy (OR = 0.53; 95%CI, 0.30-0.93), and less likely to develop local recurrence (HR = 0.54; 95%CI, 0.29-0.99). When surgeon survey score was included in the multivariate regression models, the relationship between surgeon procedure volume and permanent colostomy was diminished. There was a significant interaction between surgeon survey score and surgeon volume for the outcomes of use of TME (P < 0.01) and local recurrence (P = 0.01). CONCLUSIONS: These data suggest that surgeon knowledge may, at least in part, explain surgeon volume-associated differences in rectal cancer outcomes.