AIMS: Anxiety disorders may cause substantial impairment in patient functioning and well-being. Little is known about the relationship between diabetes and anxiety. We estimated the prevalence of lifetime diagnosis of anxiety in adults aged > or = 18 years with and without diabetes in the USA. METHODS: We analysed data from the 2006 Behavioral Risk Factor Surveillance System (total, N = 201 575; 20 142 with diabetes; 39.4% men, 77.9% non-Hispanic Whites, 8.1% non-Hispanic Blacks and 7.7% Hispanics; mean age 52.4 years). Diabetes and lifetime diagnosis of anxiety were self-reported. A multivariable log-binomial model was used to estimate prevalence ratios (PR) and associated 95% confidence intervals (CI) of anxiety based on diabetes status. RESULTS: The overall age-adjusted prevalence of lifetime diagnosis of anxiety was 19.5 and 10.9% in people with and without diabetes, respectively. After adjustment for educational level, marital status, employment status, current smoking, leisure-time physical activity and body mass index, people with diabetes had a 20% higher prevalence of lifetime diagnosis of anxiety than those without (PR 1.20; 95% CI 1.12, 1.30). There were no significant differences in the PR by gender (P = 0.06). However, the ratios differed significantly by age (P = 0.04) and by race/ethnicity (P < 0.01), indicating that people aged 18-29 years (PR 1.70; 95% CI 1.19, 2.43) and Hispanics (PR 1.69; 95% CI 1.33, 2.15) had a higher ratio than their counterparts. CONCLUSION: Diabetes was significantly associated with anxiety in adults in this large population-based sample, particularly in Hispanics and young adults.