The number of individuals diagnosed with diabetes mellitus is increasing. The diabetic may present with complications involving all systems of the body. While onychomycosis is often observed in diabetics, there have been no large studies on the prevalence of the condition in this patient group. We examined the prevalence of onychomycosis in diabetics attending diabetes and dermatology clinics in London, Ontario, Canada and Boston, MA, U.S.A. Diabetic subjects seen in dermatology offices were for unrelated dermatoses; those referred specifically for the management of onychomycosis were excluded from the sample. A total of 550 diabetic subjects was evaluated (283 males and 267 females), age 56.1 ± 0.7 years (mean ± SEM). Patients with type I diabetes constituted 34% of the sample. The racial origin was: 531 Caucasians, 17 Asians, one African-American and one American-Indian. Abnormal-appearing nails and mycological evidence of onychomycosis (mostly due to dermatophytes) were present in 253 (46%) and 144 (26%), respectively, of 550 subjects. The development of onychomycosis was significantly correlated with age (P < 0.0001) and male gender (P < 0.0001). Males were 2.99 times more likely to have onychomycosis compared with females (95% confidence interval, CI 1.94–4 61). After controlling for age and sex, the risk odds ratio for diabetic subjects to have toenail onychomycosis was 2.77 times compared with normal individuals (95% CI 2.15–3.57). After controlling for age and sex, a stepwise logistic regression demonstrated that significant predictors for onychomycosis included a family history of onychomycosis (P = 0.0001), concurrent intake of immunosuppressive therapy (P = 0.035) and peripheral vascular disease (P = 0.023). Toenail onychomycosis was present in 26% of the sample and is projected to affect approximately one-third of subjects with diabetes. Predisposing factors include increasing age, male gender, family history of onychomycosis, concurrent intake of immunosuppressive agents and peripheral vascular disease.