The CES-D in Chinese American women: construct validity, diagnostic validity for major depression, and cultural response bias.

Previous studies of the Center for Epidemiologic Studies Depression Scale (CES-D) in Chinese Americans describe internal reliability and factor structure. We report CES-D construct validity and diagnostic validity for major depression in a probability sample of 168 community-dwelling Chinese American women. Internal consistency was satisfactory (Cronbach's alpha=0.86). Good construct validity was indicated by significantly higher mean CES-D scores for respondents who reported lower social support, worse self-perceived general health, or stressful life events, including intimate partner violence. Cultural response bias was found, with positively-stated CES-D items (e.g. "I was happy") producing higher depression scores in immigrants and subjects who preferred to speak Chinese. Diagnostic validity for major depression was assessed using the Composite International Diagnostic Interview. A CES-D cut-off score of 16 had sensitivity of 100% (95% CI: 44% to 100%), specificity of 76% (95% CI: 69% to 82%), PPV of 7% (95% CI: 3% to 19%) and NPV of 100% (95% CI: 97% to 100%). Our findings suggest that the CES-D is useful for screening out non-depressed subjects in a first-stage assessment. However, it should be followed by a diagnostic tool in Chinese American women with scores above the cut-off in order to identify those with clinical depression.