RESUMO
The study examined the consistency between retrospective self-reported drug use and urinalysis data among 281 male opioid dependent subjects attending out patient clinic of National Drug Dependence Treatment Centre from January 2001 to December 2001 at All India Institute of Medical Sciences, New Delhi. Preliminary analysis indicated that there was moderate to high concordance between the two measures among different drug types. On an average 85% of urine test results matched with self-report. Subject's over-reported drug use as indicated by the low positive predictive value. In contrast, subjects were more accurate when they were reporting no drug use as suggested by the high negative predictive value. The study suggests that urine analysis is a critical variable in substance abuse treatment programs. Clinicians should be cautious while prescribing agonist drug due to frequent over-reporting of drug use by patients in our setting. This will make the substance abuse program more meaningful.