OBJECTIVES: This study focuses on the comorbidity between depression and diabetes, as well as depression and obesity, in a biracial community sample of older adults. METHODS: The data are drawn from a cross-sectional survey of five counties in North Carolina, USA, part of a longitudinal study of morbidity and mortality among elders in the urban and rural south. During the first wave of the survey, 4162 persons 65+ years of age participated in an interview at their homes. During this interview, data were collected to assess demographics, functional status, cognitive status, depression and self-report of diabetes. In addition, subjects were asked to estimate their height and weight for the interviewers, from which data body mass index (BMI) was estimated. RESULTS: In both uncontrolled and controlled analyses, female gender, lower education, functional impairment and cognitive impairment were associated with comorbid depression/diabetes and depression/high BMI. Age was not associated with comorbid depression/diabetes but younger age was associated with depression/cognitive impairment. African American race was strongly associated with depression/diabetes but not with depression/high BMI. CONCLUSIONS: More studies of comorbidity in the general population should be implemented to determine the relationship between these comorbid conditions and risk factors. Longitudinal studies are especially needed.