Increased health burden associated with comorbid depression in older diabetic Mexican Americans. Results from the Hispanic Established Population for the Epidemiologic Study of the Elderly survey.

OBJECTIVE

To examine the health burden associated with concomitant depressive symptoms and diabetes in older Mexican Americans.

RESEARCH DESIGN AND METHODS

Data from the Hispanic Established Population for the Epidemiologic Study of the Elderly were used to assess the association between high levels of depressive symptoms, measured with the Center for Epidemiologic Studies of Depression scale, and comorbid chronic health conditions, diabetic complications, functional disability, health service use, and medication use among 636 older diabetic Mexican Americans, in comparison with 2,196 older nondiabetic Mexican Americans.

RESULTS

Overall, 31.1% of the older diabetic individuals reported high levels of depressive symptoms. The risks of comorbid myocardial infarction, hypertension, arthritis, and angina were significantly higher in the presence of concomitant depressive symptoms, as were the risks of diabetic complications, functional disability, incontinence, vision impairment, poorer perceived health status, and health service use among both diabetic and nondiabetic individuals. Rates were substantially higher among depressed diabetic individuals, however, in comparison to depressed nondiabetic individuals. Importantly, this increased health burden was evident even when controlling for sociodemographic risk factors, including sex, age, level of education, marital status, immigrant status, and living arrangements.

CONCLUSIONS

The presence of concomitant depressive symptoms among older diabetic Mexican Americans is associated with a substantially greater health burden than is seen among diabetic individuals without depression or depressed individuals without diabetes. This association of depression with higher rates of chronic conditions, poorer functioning, and increased health service use is particularly significant in that this study was conducted among community-dwelling adults and was not confounded by the potential association of health care-seeking behavior that might occur in a medically ill sample.