Clinically relevant levels of depressive symptoms in community-dwelling middle-aged African Americans

OBJECTIVES: To identify the prevalence of and potentially modifiable risk factors for clinically relevant levels of depressive symptoms in a population-based sample of community-dwelling African Americans and the prevalence of treatment by prescription and alternative medications. DESIGN: Cross-sectional survey, 2000-01. SETTING: Community-based. PARTICIPANTS: Nine hundred ninety-eight noninstitutionalized African Americans in St. Louis, Missouri, born between 1936 and 1950. MEASUREMENTS: Depressive symptoms were measured using the 11-item Center for Epidemiologic Studies Depression scale (CES-D). Clinically relevant levels of depressive symptoms were defined as nine or more (equivalent to greater than or equal to16 on the 20-item CES-D). A comprehensive set of risk factors was considered that included three demographic variables, eight socioeconomic-access measures, four environmental factors, seven measures of functional status, 15 biomedical markers, one service utilization indicator, and three psychosocial measures. All analyses were weighted to the represented population. Treatment with an antidepressant was determined by examining subjects' medications compiled in their homes. RESULTS: Two hundred ten subjects (21.1%) had clinically relevant levels of depressive symptoms. Several multivariate logistic regression approaches were used for model building, which identified a consistent set of nine predictive factors: female sex (odds ratio adjusted (AOR) for all factors in the final model=1.52; 95% confidence interval (CI)=1.01-2.27), lower objective income (AOR=1.62, 95% CI=1.08-2.43), perceived income inadequacy (AOR=2.33, 95% CI=1.49-3.65), lower assessment of home environment (AOR=1.07 per scale point, 95% CI=1.01-1.12), limitations in visual acuity (AOR=1.12 per scale point, 95% CI=1.04-1.21), being severely underweight (AOR=2.52, 95% CI=1.02-6.20), being obese (AOR=1.72, 95% CI=1.16-2.54), being hospitalized in the previous year (AOR=2.25, 95% CI=1.45-3.49), and lower social support (AOR=1.20 per scale point, 95% CI=1.16-1.26). Of these, social support was the most important (adjusted standardized odds ratio =2.41). Forty-one (19.5%) of the subjects with clinically relevant levels of depressive symptoms were taking prescription antidepressants. CONCLUSION: The prevalence of clinically relevant levels of depressive symptoms in middle-aged African Americans was greater than that for the general U.S. population. Community-based health programs that screen for depression and refer individuals to clinical care sites with appropriately designed systems of care for depression management should be developed. For optimal effect, these programs should concentrate their efforts in socioeconomically disadvantaged areas and address socioeconomic factors such as income inadequacy and social support in addition to the biomedical risk factors. Given the pervasive adverse effects of depression, such interventions have the potential for significantly enhancing the health of African Americans in their later years and reducing current health disparities.