Patients with extensive ulcerative colitis have a high risk of developing colon cancer. The etiology of mucosal dysplasia, a premalignant lesion that is used as a screening test in surveillance programs, is unknown. Previously, a case-control study [Lashner et al. (1989) Gastroenterology 97:255-259] suggested that folate supplementation was associated with a 62% reduction in the risk of developing dysplasia or cancer. The current case-control study was performed to obtain a better definition of this risk. All 67 patients with chronic ulcerative pancolitis having surveillance colonoscopy during a 1-year period were entered. There were 6 cases (4 with dysplasia and 2 with cancer) and 61 controls (no cancer or dysplasia). Red blood cell folate, reflecting intermediate-term stores, was a mean of 66.2 ng/ml lower in cases compared to controls. Serum folate, reflecting short-term stores, was not different between groups. Adjusting for confounding effects of age, sex, race, disease duration, and folate supplementation, the risk of dysplasia or cancer was significantly decreased by 18% for each 10 ng/ml increase in red blood cell folate (odds ratio 0.82, 95% confidence interval 0.68-0.99). Vitamins A, D, and E and carotene were lower in cases than in controls, but no water-soluble vitamin other than red blood cell folate was associated with an increased cancer risk. Depressed red blood cell folate is associated with an increased risk of dysplasia and cancer in patients with ulcerative colitis and may be a risk factor for neoplastic transformation.