A case-control study comprising 216 cases of pancreatic cancer and 279 controls was conducted to investigate the relationship of pancreatic cancer with certain chronic medical conditions and with the consumption of tea, coffee and alcoholic beverages. Significant positive associations with pre-existing diabetes mellitus and gall-bladder disease were observed and there was weak evidence of association with liver disease. The relative risks for diabetes mellitus and gallstones diagnosed at least one year previously were 4.1 (p = 0.005) and 2.8 (p = 0.01) respectively. Cases drank significantly more beer than controls (p = 0.005) and there was evidence of a positive trend in risk with total alcohol consumption. Smoking was a clear risk factor, but cases and controls were very similar with respect to tea and coffee drinking habits.