Reliability of the Brazilian version of the CIDI in a case-control study of risk factors for drug abuse among adults in Rio de Janeiro

Abstract

Using a test-retest design, the reliability of DSM-III-R diagnoses generated by the Brazilian version of the Composite International Diagnostic Interview (CIDI) was evaluated in a sample of 30 subjects. This sample was drawn from first interviews conducted for a case-control study of risk factors for drug abuse among adults in Rio de Janeiro. The results indicated good to high levels of agreement between both sets of interviews with respect to all psychiatric diagnoses covered and all substance dependence/abuse diagnoses (kappa values of 0.61-1.00) except the diagnosis of alcohol abuse (kappa = 0.35). These findings show the portions of the new CIDI version tested to be sound and appropriate for use in future investigations of adult psychiatric and substance dependende/abuse diagnoses in Brazil

This article will also be published in Spanish in the BOSP. Vol. 116(5), May 1994