This study seeks to fill a void in our knowledge about the validity of survey data on drug use among the general population. With current technology, it is possible to fairly accurately determine recent drug use by bioassaying body fluids or tissues. Several studies have examined the validity of self-reports of drug use in the criminal justice and/or treatment populations and have found significant under reporting of recent drug use as validated by bioassays of urine and/or hair specimens, but no study has examined the validity in a general population study using bioassays to validate self-report. The proposed study will replicate the methodology of the National Household Survey on Drug Abuse (NHSDA) in conducting a representative survey of drug use among the household population (n=5000)in one of the nine Census regions in the U.S. (the Pacific region). The NHSDA is the single most important source of information on the epidemiology of drug use in the U. S. Following completion of an abbreviated version of the NHSDA, respondents will complete a debriefing questionnaire asking about their candor on the survey, and again about their recent illicit drug use. Then they will be asked to provide urine and hair samples. We anticipate a minimum 80% response rate for the specimens based on offering a $50 incentive, emphasizing confidentiality and anonymity of results, and reintroducing informed consent procedures portraying the study as critical for further scientific understanding of drug use. Both the urine and hair samples will be analyzed for the presence of the following drugs or their metabolites: cannabis, cocaine, amphetamines, and opiates. Urine will also be analyzed for the presence of cotinine--a metabolite of tobacco. Data analyses will examine the consistency between self-report and biological specimen by drug type. Logistic regression, structural modeling, and event history analyses will evaluate potential variables that may impact on valid self-reporting of recent drug use. The efficacy of surveys in measuring recent drug use among the general population can be fairly accurately determined, and useful estimates of any needed corrections can be made. The proposed study will have particular relevance to the NHSDA, but the results will be generalizable to self-report surveys overall.
