This application proposes an 18 month renewal of NIDA support for analyses on the epidemiology of cocaine and other drug use by adults, including issues of psychiatric co-morbidity and risk of drug problems. The specific research question include: What is the current best estimate for the degree to which cocaine users are at increased risk of DSM-III mania and related serious psychiatric disturbances? What is the current best estimate for the degree to which cocaine users are at increased risk of DSM-III obsessive-compulsive disorder and related obsession-like or compulsion-like behaviors? What is the current best estimate for the relative risk of drug dependence and other specific types of drug problems in relation to the type of drug used illicitly, and in relation to daily versus non-daily use of these drugs? For example, which drug use pattern is associated with greater risk of drug dependence: daily marijuana use, or non-daily cocaine use? Are there differences in the patterns of drug problems reported by illicit drug users who have had drug abuse treatment encounters, as compared to all illicit drug users who have had drug abuse treatment encounters, as compared to all illicit drug users or to illicit drug users with no treatment history? If there are differences, what are their implications for new development of instruments to assess drug dependence and problems associated with psychoactive drug use? The applicants will use data from the 20,862 participants in the NIMH Epidemiologic Catchment Area Program in analyses to answer these questions, including data on 3925 illicit drug users who reported on their drug problem experiences. The data were gathered prospectively by administering the Diagnostic Interview Schedule at baseline and also at followup roughly one year later. The proposed conceptual models and related statistical models are based on work completed by the applicants during the initial award period and now being published, as well as new latent structure analyses appropriate to dichotomous and highly skewed diagnostic data.
Showing the most recent 10 out of 28 publications
