AimsTo assess the diagnostic accuracy of four undiagnosed Type 2 diabetes mellitus risk scores accounting for erectile dysfunction status.MethodsThis was a population‐based cross‐sectional study. Type 2 diabetes was defined according to a oral glucose tolerance test and self‐reported physician diagnosis. Erectile dysfunction was defined according to the answer to the question, ‘Have you had difficulties obtaining an erection in the last 6 months?’ (yes/no). The risk scores used were the FINDRISC, LA‐FINDRISC, American Diabetes Association score and the Peruvian Risk Score. A Poisson regression model was fitted to assess the association between Type 2 diabetes and erectile dysfunction. The area under the receiver‐operating characteristic curve was estimated overall and by erectile dysfunction status.ResultsA total of 799 men with a mean (sd) age of 48.6 (10.7) years were included in the study. The overall prevalence of Type 2 diabetes was 9.3%. Compared with healthy men, men with Type 2 diabetes had 2.71 (95% CI 1.57–4.66) higher chances of having erectile dysfunction. Having excluded men aware of Type 2 diabetes status (N=38), the area under the receiver‐operating characteristic curve of three of the risk scores (not the American Diabetes Association score) improved among those who had erectile dysfunction in comparison with those who did not; for example, the area under the receiver‐operating characteristic curve of the LA‐FINDRISC score was 89.6 (95% CI 78.7–99.9) in men with erectile dysfunction and 76.5 (95% CI 68.5–84.4) overall.ConclusionsIn a population‐based study, erectile dysfunction was more common in men with Type 2 diabetes than in the otherwise healthy men. Screening for erectile dysfunction before screening for Type 2 diabetes seems to improve the accuracy of well‐known risk scores for undiagnosed Type 2 diabetes.