Comparisons of self-report data and oral fluid testing in detecting drug use amongst new treatment clients

Drug testing is widely used and employed in diverse contexts, including drug treatment clinics. Building on previous research, this paper aims to (i) compare self-report data and oral fluid (OMT) testing in detecting drug use amongst individuals beginning a new episode of drug treatment and (ii) identify factors that may predict drug users who have discordant self-report and OMT test results. Two hundred and seventy-one new drug treatment clients completed a structured questionnaire that included questions relating to drug use during the preceding 3 days and provided an oral fluid sample that was independently tested for opiates, benzodiazepines, methadone and cannabis. Data were analysed using kappa statistics (Cohen, 1960) and univariate and multivariate logistic regression. Findings indicated a high level of consistency between self-reported drug use and OMT testing. However, agreement varied by drug type and respondents commonly reported consumption that screening failed to identify. Inconsistencies appeared to relate to a number of factors and were not necessarily a function of deliberate distortion by the drug user. Overall, it is concluded that OMT testing is a good indicator of the validity and reliability of drug users' self-report data. Nonetheless, its accuracy might be greater for some drug categories than for others. Equally, further study comparing test results and self-reported drug use amongst different populations and in different contexts is required. (C) 2003 Elsevier Science Ireland Ltd. All rights reserved.