Correlates of underreporting of post-discharge cocaine use among therapeutic community clients

This study had two goals: first, to assess the validity of self-reported re cent cocaine use at follow-up among therapeutic community (TC) clients and second, to identify pre- and post-admission factors that are related to und erreporting of cocaine use. Research interviews and urinalyses were used fr om the follow-up sample of 307 people participating in the District of Colu mbia Treatment Initiative (DCI) study 19 months after treatment discharge. Results show that the type of measure used to assess validity is very impor tant to determining the results. Overall percentage agreement between self- reports and urinalysis was greater than 70 percent; however, after restrict ing the sample to include only those clients who were cocaine positive, the agreement decreased to 35 percent. Bivariate analyses revealed no pre-admi ssion correlates of underreporting cocaine use at follow-up and a few post- admission correlates. We conclude that if is not possible to identify in ad vance clients who are likely to underreport cocaine use. To avoid underesti mating cocaine use, treatment outcome studies should use objective measures of drug use along with self-reports.