PURPOSE: This study tested for an association between diabetes and depressive symptoms and assessed the effect of co-occurring diabetes and depressive symptoms on healthcare utilization outcomes among African-American patients. PROCEDURE: The sample consisted of 303 adult African-American patients age 40 and over from a primary care clinic serving the low-income population in Nashville, TN. Measures were based on self-reports during a structured interview. Multivariate analyses adjusted for age, gender, socioeconomic status, and comorbid chronic conditions. FINDINGS: African-American patients with and without diabetes did not differ on the presence or severity of depressive symptoms. However, the co-occurrence of major depressive symptoms with diabetes among African Americans was associated with nearly three times more reported emergency room visits and three times more inpatient days, but was only marginally associated with a lower number of physician visits. CONCLUSIONS: In contrast to previous studies with predominantly white samples that found a positive association between diabetes and depression, no association was found in this African-American sample. Nevertheless, the results did concur with research findings based on other samples, in that the co-occurrence of depression with diabetes was associated with more acute care utilization, such as emergency room visits and inpatient hospitalizations. This pattern of utilization may lead to higher healthcare costs among patients with diabetes who are depressed, regardless of race.