Screening for depression in middle-aged and older puerto rican primary care patients.

BACKGROUND Brief depression screens have recently been developed, but their use in older or minority populations has not been studied. To date, optimal depression screens and optimal cutpoints have not been identified for middle-aged and older Hispanic primary care patients. METHODS This study compares multiple versions of four depression screening tools--Center for Epidemiologic Studies-Depression Scale (CES-D), Geriatric Depression Scale, Yale 1-question screen, and PRIME-MD 2-question screen--to the Composite International Diagnostic Interview (CIDI), the World Health Organization's diagnostic interview, which has been validated in adult Latino populations, to assess convergent validity. Three hundred and three Puerto Rican primary care patients age 50 and older completed all screens and the CIDI in a face-to-face interview. Sensitivity and specificity for each screen were calculated, and receiver operator characteristic curves were generated. RESULTS Between 34% and 61% of patients screened positive for depression, depending on the measure, with 12% meeting DSM-IV criteria for major depression (CIDI). The 10-item CES-D worked best to identify major depression in this population, with a sensitivity of 84% and specificity of 64% using a cutpoint of 3. CONCLUSIONS The 10-item CES-D, which takes about 2 minutes to administer, is a useful tool for identifying Puerto Rican patients in need of an in-depth mental health evaluation in a primary care setting. A lower cutpoint of 3 (instead of the conventional cutpoint of 4) is recommended for optimal sensitivity and specificity.