Validity of the CAGE questionnaire in an American Indian population.

Abstract

This study evaluated the performance of the CAGE questionnaire (a set of four questions about alcoholism) in an American Indian population. We analyzed data from a cross-sectional study of 275 individuals (179 women) aged 21 years or older. Alcohol dependence was diagnosed according to the Diagnostic and Statistical Manual of Mental Disorders, Third Edition-Revised (DSM-III-R), based on a detailed psychiatric interview using the Schedule for Affective Disorders and Schizophrenia-Lifetime Version. Accuracy of the CAGE questionnaire was quantified as sensitivity, specificity, likelihood ratios and the area under receiver operating characteristics (ROC) curves, using the DSM-III-R diagnosis as the reference. Of participants interviewed, 85% of men and 53% of women had a diagnosis of alcohol dependence by DSM-III-R. A CAGE score of > or = 2 had a sensitivity and specificity of 68% and 93%, respectively, in men and 62% and 79% in women, for the diagnosis of alcohol dependence. CAGE scores of 0, 1 and > or = 2 were associated with likelihood ratios of 0.3, 0.3 and 9.5, respectively, in men and 0.4, 0.7 and 1.5 in women. The area under the ROC curve was 81% for men and 75% for women. These findings suggest that the CAGE questionnaire is a valid screening method, in this population, for identifying people likely to have alcohol dependence.