Racial/ethnic disparities in prevalence of screened depression among community-dwelling older women

Background: Depression is a significant public health problem among older adults, especially among women. People with screened depression are at higher risk of developing major depression. Objective: To estimate racial/ethnic disparities in the prevalence of screened depression and explore associated correlates among community-dwelling older women. Design: A multiple-year cross-sectional study using Medicare Current Beneficiaries Survey for 2001-03 and the Area Resource File for 2002. Participants: 11,756 community-dwelling person-years for women age 65+ in traditional Medicare for a full year. Methods: A two-item screen (anhedonia and/or sadness) was used to identify subjects who screened positive for depression (“screened depression”) in the past year. Racial/ethnic groups included non-Hispanic whites, non-Hispanic blacks, Hispanics, and others. Bivariate analysis and a multivariate logistic regression were utilized. All estimates were adjusted by sampling weights and study design effects. Results: Bivariate analyses indicated that a significantly higher proportion of Hispanics reported screened depression than did non-Hispanic whites (21.1% vs. 13.7%, p=0.001). No significant difference was found between non-Hispanic blacks and whites (15.8% vs. 13.7%, p=0.24). Similar differences were found after controlling for demographic characteristics and health status. After additionally controlling for income, education and health insurance, non-Hispanic blacks were significantly less likely to report screened depression than non-Hispanic whites (OR=0.77, P=0.03). No significant difference was found between Hispanics and non-Hispanic whites (OR=1.24, p=0.18). Conclusion: Among community-dwelling older women, minorities reported more screened depression than non-Hispanic whites. Improving women's socioeconomic conditions may decrease screened depression and help reduce racial/ethnic disparities in developing major depression.