Objective: To determine the association between race, region and pre-diabetes. Method: The study used 2003-2007 United States baseline data from the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study for this cross-sectional analysis. Participants in this study were 45. years or older at recruitment. Logistic regression was used to assess whether race and region are associated with pre-diabetes independent of demographics, socioeconomic factors and risk factors. Results: Twenty-four percent of the study participants (n = 19,889) had pre-diabetes. The odds ratio (95% confidence interval) for having pre-diabetes was 1.28 (1.19-1.36) for blacks relative to whites and 1.18 (1.10-1.26) for people living in the Stroke Belt region relative to the other parts of the United States. The odds of having pre-diabetes for Stroke Belt participants changed minimally after additional adjustment for race (OR = 1.20; 1.13-1.28), age and sex (OR = 1.24; 1.16-1.32), socioeconomic status (OR = 1.22; 1.15-1.31) and risk factors (OR = 1.26; 1.17-1.35). In the adjusted model, being black was independently associated with pre-diabetes (OR = 1.19; 1.10-1.28). Conclusion: The prevalence of pre-diabetes was higher for both blacks and whites living in the Stroke Belt relative to living outside the Stroke Belt, and the prevalence of pre-diabetes was higher for blacks independent of region. © 2014 Elsevier Inc.