Background: It has been observed that survival after colorectal cancer resection is longer in women than men. The majority of these studies are in non-U.S. populations and few use appropriate multivariate adjustment. We used the Surveillance, Epidemiology and End Results- Medicare database to examine disease-specific survival in women and men undergoing colorectal cancer resection in the United States, adjusting for patient, cancer, and hospital characteristics in an effort to identify disparities, not only in survival, but also in patterns of presentation, surgical resection, and treatment.

Methods: With use of the Surveillance, Epidemiology and End Results-Medicare-linked database, we performed a retrospective cohort study of 30,975 patients with colon cancer and 8,350 patients with rectal cancer who underwent surgical resection from 1996 to 2003. Kaplan-Meier curves, the log-rank test, and Cox regression compared survival between genders. Multivariate adjustment was performed by use of patient demographics; cancer variables including stage, medical treatment, and adequacy of nodal harvest; and hospital characteristics.

Results: In both cancers, women presented at an older age and more emergently than men. They also underwent less aggressive medical therapy for advanced disease; in particular, in the octogenarian population. In unadjusted analysis, there was no gender difference in survival (colon hazard ratio, 0.98; P = 0.74; rectal hazard ratio, 0.95; P =0.10). After full adjustment, however, women had significantly longer survival, in particular, after rectal resection (colon hazard ratio, 0.91; P< 0.001; rectal hazard ratio, 0.82; P< 0.001).

Conclusions: Women in this cohort have longer adjusted survival compared with men; however, they present more emergently and at an older age, and they receive less aggressive medical treatment. These are noticeable disparities that could serve as targets for continued improvement.