Dupuytren's is a common problem, but little is known about its aetiology. We have undertaken a large case-control study to assess and quantify the relative contributions of diabetes and epilepsy as risk factors for Dupuytren's in the community. Cases were patients with a diagnosis of Dupuytren's disease and, for each, two controls were individually matched by age, sex, and general practice. Our dataset included 821 cases and 1,642 controls. Five hundred and eighty-eight (72%) of the cases were men. The mean age at diagnosis was 62 (range 24-97) years. Diabetes was a significant risk factor for Dupuytren's disease (OR=1.75) and there was an increased risk for medicinally treated diabetes (metformin--OR=3.56; sulphonylureas--OR=1.75) and particularly insulin controlled (OR=4.38) rather than diet-controlled diabetes. Epilepsy (OR=1.12) and anti-epileptic medications were not associated with Dupuytren's disease. Ascertainment bias in previous studies may explain the reported association with epilepsy.