The proposed project is designed to provide information on the reliability and validity in clinical samples of an interview schedule designed for use in a huge-scale national epidemiologic study of alcohol consumption, alcohol use disorders and other potentially associated psychiatric conditions. Specifically, we propose to investigate the psychometric properties of the Alcohol Use Disorders and Associated Disabilities Interview Schedule (AUDADIS), a lay-administered interview schedule. The AUDADIS covers alcohol and drug dependence and abuse, affective and anxiety disorders, psychotic symptoms, and antisocial personality. Psychometric information from clinical samples is needed because it is hoped that the AUDADIS will be used to provide comparable information from clinical and general population samples.. An important component of validity research on an instrument such as the AUDADIS is the comparison of its results to reliable, systematic evaluations conducted by clinicians. In the proposed study, such an evaluation will be conducted. This evaluation will use our version of the Structured Clinical Interview for DSM-III-R (SCID), the SCID-A/D (Alcohol/Drug), developed by us for improved reliability, better coverage of alcohol and drug use disorders, and better coverage of the relationship of alcohol and drug use to other psychiatric disorders. To fully understand a clinical reappraisal, the reliability of each of the instruments must be established in the target populations. Thus, a test-retest reliability study will be conducted of both the AUDADIS and the SCID-A/D. Then, the AUDADIS and SCID-A/D will be compared in a series of patients. All sections of both instruments will be investigated. A total of 1200 patients from seven treatment facilities (alcohol, drug, and psychiatric) will be evaluated in the reliability and validity studies, including an oversampling of females and a substantial subsample of Spanish-speaking patients. Qualitative and quantitative analyses will be conducted to investigate concordance and reasons for discordance between interviews, and results will be compared to findings from parallel methodological research by NIAAA in the general population.
