Context: Hispanics are at increased risk for diabetes, while rural residents have historically had decreased access to care. Purpose: To determine whether living in a rural area and being Hispanic confers special risks for diagnosis and control of diabetes. Methods: We analyzed the Third National Health and Nutrition Examination Survey (1988-1994). Hispanics and non-Hispanic white adults were classified according to rural/urban residence to create 4 ethnicity-residence groups. Investigated outcomes were previously diagnosed and undiagnosed diabetes. Among those with diagnosed diabetes, we investigated control of glucose, hypertension, and lipids. Findings: The prevalence of diagnosed diabetes was greatest for rural residents, especially for rural Hispanics (8.2%) versus that for urban whites (4.6%), rural whites (6.5%), or urban Hispanics (4.5%), (P less than 0.01). However, urban Hispanics were most likely to have undiagnosed diabetes at 3.7%, versus 2.3% of rural whites, 2.8% of urban whites, and 2.7% of rural Hispanics (P = 0.04). Among people with diagnosed diabetes, there was no difference in glycemic control between the 4 groups. Rural Hispanics with diagnosed diabetes had the greatest prevalence of elevated systolic blood pressure at 45%, compared to 37% of urban whites, 29% of rural whites, 28% of urban Hispanics (P = 0.01). In regression models controlling for potential confounders, there were no differences among urban and rural whites and Hispanics in the likelihood of undiagnosed diabetes or in glycemic control for those with diagnosed diabetes. Conclusions: Initiatives that target Hispanic health, and especially diabetes, should acknowledge rural/urban Hispanic health differences.