During 1984-88 a population-based case-control study was carried out in The Netherlands, in collaboration with the International Agency for Research on Cancer, to examine the possible relationship between aspects of medical history and exocrine pancreatic carcinoma in 176 cases and 487 controls. About 58% of patients were interviewed directly. We observed an inverse relationship between medical treatment for allergy-related conditions and the development of pancreatic cancer (30 cases vs. 130 controls, OR 0.57, 95% CI 0.36 to 0.90). A history of gallbladder problems, gallstones, cholecystectomy, stomach or duodenal ulcer, pancreatitis, appendicitis, diabetes or tonsillectomy was not related to risk. In direct responses, compared with once daily, a positive relationship was seen for stool frequency, 10 years ago, of less than once daily (18 cases vs. 40 controls, OR 2.10, 95% CI 1.09 to 4.04). In men, diabetes treated with insulin and diagnosed more than 1 year previously was significantly and positively related to risk (5 cases vs. 1 control, OR 11.66, 95% 1.28 to 105.95). In brief, the results of the present study suggest that a history of allergy-related conditions may protect, whereas a past stool frequency of less than once daily may enhance the risk of cancer of the pancreas. Other elements of the medical history were not consistently related to risk.