Many Asian-Americans are unfamiliar with depression and its treatment. When depressed, they generally seek treatment from their primary care physicians and complain about their physical symptoms, resulting in under-recognition and under-treatment of depression. This study evaluates the effectiveness of the Chinese version of the Beck Depression Inventory (CBDI) for screening depression among Chinese-Americans in primary care. A total of 503 Chinese-Americans in the primary care clinic of a community health center were administered the CBDI for depression screening. Patients who screened positive (CBDI > or = 16) were interviewed by a psychiatrist using the Structured Clinical Interview for DSM-III-R, patient version (SCID-I/P) for confirmation of the diagnosis. Patients who screened negative (CBDI < 16) were randomly selected to be interviewed using the depression module of the SCID-I/P. The results of the SCID-I/P interview were used as the standard for evaluating the sensitivity and specificity of the CBDI. A total of 815 Chinese-Americans in a primary care clinic were approached, and 503 completed the CBDI. Seventy-six (15%) screened positive (CBDI > or = 16), and the prevalence of major depression was 19.6% by using extrapolated results from SCID-I/P interviews. When administered by a native-speaking research assistant, the CBDI has good sensitivity (.79), specificity (.91), positive predictive value (.79), and negative predictive value (.91). Despite the commonly believed tendency to focus on physical symptoms rather than depressed mood, Chinese-Americans are able to report symptoms of depression in response to a questionnaire. The CBDI, when administered by research assistants, has good sensitivity and specificity in recognizing major depression in this population. Lack of interest among Chinese-American patients in using the CBDI as a self-rating instrument has limited its use for depression screening in primary care settings.