OBJECTIVES To estimate the modifying effect of age on the survival benefit associated with adjuvant chemotherapy receipt in elderly patients with a diagnosis of Stage III colon cancer. DESIGN Observational, retrospective cohort study using two samples: an overall sample of 7,182 patients to provide externally valid analyses and a propensity score-matched sample of 3,016 patients to provide more internally valid analyses by reducing the presence of treatment endogeneity. An interval-censored survival model with a complementary log-log link was used. Hazard ratios and 95% confidence intervals were obtained for all regressions. SETTINGS Data from the National Cancer Institute's Surveillance, Epidemiology and End Results database and the linked Medicare enrollment and claims database were used. PARTICIPANTS Selected patients were aged 66 and older and had a diagnosis of Stage III colon cancer. Patients were followed from surgery to time of death or censorship. MEASUREMENTS The outcome was colon cancer-specific death during the follow-up period. Receipt of adjuvant chemotherapy was measured according to the presence of a claim for 5-fluorouracil or leucovorin within 6 months after surgery. RESULTS All elderly patients had a significant survival benefit associated with adjuvant chemotherapy receipt, although the survival benefit of adjuvant chemotherapy was not uniform across all age groups. CONCLUSION These findings have important clinical and policy implications for the risk-benefit calculation induced by treatment in older patients with Stage III colon cancer. The results suggest that there is a benefit from chemotherapy, but the benefit is lower with older age.