The validity of self-reported cocaine use in a criminal justice treatment sample.

Recent studies comparing self-admitted cocaine use with hair and urine test results have raised concerns about underreporting due to variations across situations and settings. Because of the frequent need for self-report data in conducting treatment evaluations, more information is needed on factors that affect the credibility of this information. The present study examines records of cocaine use collected as part of an evaluation of prison-based treatment (N=396). Specifically, self-reported cocaine use from 6-month postrelease follow-up interviews, completed with treatment graduates and a comparison sample of parolees who were eligible but not sent to treatment, was examined in relation to urine and hair test results. Overall, cocaine use was underreported when compared to hair test results, but underreporting was lower for program graduates than for the untreated comparison group. Furthermore, program graduates originally admitted to prison on a drug-related arrest were the most likely to underreport cocaine use, seemingly due to a heightened concern about potential consequences, such as returning to prison on another drug-related charge. Given that the validity of self-report for high-risk individuals varies greatly and is difficult to predict across studies, future research with criminal justice samples should continue to assess the validity of these measures under specific research conditions and for diverse types of individuals.