BACKGROUND: Previous studies have shown an association between diabetes mellitus (DM) and urinary incontinence (UI) in women, especially severe UI. The purpose of this study was to investigate whether diabetes related variables could explain this association. METHODS: The study is part of the EPINCONT study, which is based on the large Nord-Trøndelag Health Study 2 (HUNT 2), performed in the county of Nord-Trøndelag, Norway, during the years 1995 - 1997. Questions on diabetes and UI were answered by a total of 21 057 women aged 20 years and older. Of these 685 were identified as having diabetes, and thus comprise the population of our study. A variety of clinical and biochemical variables were recorded from the participants. RESULTS: Blood-glucose, HbA1c, albumine:creatinine ratio (ACR), duration of diabetes, diabetes treatment, type of diabetes, cholesterol and triglycerides did not significantly differ in women with and without UI in crude analyses. However, the diabetic women with UI had more hospitalizations during the last 12 months, more homecare, and a higher prevalence of angina and use of oestrogene treatment (both local and oral/patch). After adjusting for age, BMI, parity and smoking, there were statistically significant associations between any UI and angina (OR 1.89; 95% CI: 1.22 - 2.93), homecare (OR 1.72; 95% CI: 1.02 - 2.89), and hospitalization during the last 12 months (OR 1.67; 95% CI: 1.18 - 2.38). In adjusted analyses severe UI was also significantly associated with the same variables, and also with diabetes drug treatment (OR 2.10; 95% CI: 1.07 - 4.10) and stroke (OR 2.47; 95% CI: 1.09 - 5.59). CONCLUSION: No single diabetes related risk factor seems to explain the increased risk for UI among women with diabetes. However, we found associations between UI and some clinical correlates of diabetes.