Screening for Depression Among American Indians

Somervell, Philip

Abstract

This Small Grant application requests funds for secondary data analyses to assess the performance of methods of screening for depressive disorders, as well as alcohol dependence, in an American Indian community sample. The epidemiologic and programmatic value of this work stems from indications that depressive disorders are relatively common among American Indians, and the very high prevalence of alcohol use disorders among them. Very little is known about the performance of screening instruments for depression and alcoholism in these populations. The data to be analyzed were collected in 1987-88 on a sample of 122 Makah Indian adults, more than 25% of the adult Makah population of the village of Neah Bay, WA. Screening data available include the Center for Epidemiologic Studies Depression Scale (CES-D) and the Michigan Alcoholism Screening Test (MAST). Psychiatric diagnostic data are from the Schedule for Affective Disorders and Schizophrenia, Lifetime Version (SADS-L). Cronbach's alpha (a measure of internal consistency reliability) will be calculated for the CES-D and MAST. Their criterion validity will be analyzed by calculation of sensitivity and specificity as well as by receiver operating characteristics (ROC) analysis. The criterion for these analyses of validity will be diagnosis according to the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (DSMIII-R), based on SADS-L data and confirmed by consensus of three psychiatrists (one of whom administered the SADS-L interview). The dataset represents a unique opportunity to analyze the performance of two screening measures in relation to a state-of-the-art diagnostic interview, in a sample from a particular tribal group in one American Indian community.