A case-control study was conducted in the Minneapolis-St Paul (Minnesota, United States) area to assess the role of dietary factors in the etiology of pancreatic cancer. Cases were White males aged 40 to 84 whose death certificate listed pancreatic cancer (exocrine only). White male controls were ascertained through random-digit dialing. Family members were interviewed about the subject's dietary usage in the two years prior to death (cases, n = 212) or prior to interview (controls, n = 220). Energy-adjusted, nutrient-intake, risk estimates were calculated. Among all respondents, negative trends were observed for polyunsaturated fat, linoleic acid, vitamin C, and beta-carotene. Positive trends were observed for riboflavin and retinol. Point estimates were, in general, comparable between the analyses of all respondents and spouse-only respondents. The nutrients associated with a decreased risk for pancreatic cancer occur primarily in vegetables and fruits, of which the consumption of cruciferous and beta-carotene-rich vegetables and citrus fruits provided the greatest reduction in risk.