Screening for depression in African American and Caucasian older women

Abstract

Objective: To assess the performance of a two-choice (yes/no), 10-item shortened form of the CES-D in both African American (AA) and Caucasian (CA) older women. The CES-D is a widely used screening instrument, but its use has been questioned for routine screening because of its length and the complexity of its four-choice format. There is also little data available about its suitability low-income AA respondents.

Method: Telephone screening for depression followed by in-home diagnostic interviews were conducted in a community sample of 256 CA and 186AA low-income older women who ranged in age from 64 to 94 years. Standard receiver operator curves were plotted to determine the sensitivities and specificities of the screening instrument at different cut-scores against a criterion of SCID-based diagnoses of current major depressive episode (CMDE).

Results: Sensitivity and specificity of the 10-item scale and an even shorter 5-item version was slightly higher for AA than for CA women. While both short forms produced significant numbers of false positives against a criterion of CMDE, many of the women identified by the screen did have significant depressive symptomatology. Significantly, fewer AA women received a diagnosis of CMDE primarily because they did not show diminution of functioning associated with their depressive symptoms.

Conclusion: Short, easy to administer forms of the CES-D can provide useful information in working with older patients. Clinicians should be aware of ethnic differences in symptom expression and levels of functional impairment that are likely to occur in follow-up medical and psychiatric exams.

Keywords:

• depression
• ethnic differences
• screening
• older adults
• short form CES-D

Acknowledgments

This research was funded by NIH RO1-MH49086-02 to Kenneth Heller and Ralph Swindle. Data collection for this study was initiated and conducted prior to Dr Swindle's employment at Eli Lilly and Company, and is not funded by Eli Lilly and company except for company permission for Dr Swindle to participate as a co-author. Analyses for this article and the initial draft were conducted at Indiana University by Dr Viken and Dr Heller.