The purpose of this study was to determine the degree to which colon cancer treatment in rural North and South Carolina in 1991 and 1996 conformed to national treatment recommendations. Data came from medical records of colon cancer patients residing in rural North and South Carolina. The National Cancer Institute's Physician Data Query (PDQ) database was used to define state-of-the-art care and to categorize receipt of primary and/or adjuvant treatment. Changes in treatment over time, location, and stage and bivariate relationships between treatment and selected covariates were assessed with chi-square and Fisher's exact tests. Regression was used to control for possible interactions between patient and/or disease characteristics and treatment. The majority of colon cancer cases received primary therapy as suggested by the PDQ which was not significantly related to other factors examined. There was variation in provision of adjuvant therapy. Stage III patients received adjuvant therapy significantly more often than did stage II patients (p </= 0.01). Receipt of appropriate adjuvant therapy among stage III patients was significantly associated with younger patient age and white race (p </= 0.05). Rural colon cancer patients are likely to receive primary therapy as recommended by the PDQ, but may be less likely to receive suggested adjuvant therapy. Further understanding of variations in the rate of adjuvant therapy for colon cancer is needed to ensure appropriate treatment regimens are obtained for rural colon cancer patients.