In mid-1986 the District of Columbia Pretrial Services Agency decided to expand its adult pretrial drug testing program to include juveniles. A 24-month Federal operational grant was awarded to conduct a pilot juvenile urine-testing program. Program implementation began on October 21, 1986. The program had three components: initial (pre-hearing) testing, testing of released juveniles during the preadjudication period, and testing of juvenile probationers during the postadjudication period. The data used in the evaluation of the program came from the urine testing of juvenile arrestees conducted during a 15-month period. The major finding of this study is that urine testing of juveniles shortly after arrest in the District of Columbia did not improve risk classification; i.e, whether juveniles tested positive or negative for drugs did not help significantly in the identification of juveniles who would be rearrested. This conclusion was reached after performing multiple analyses that were conducted by two investigators working independently of each other. These findings contrast with the findings for the adult program, where urine test results did improve pretrial risk classification. The authors hypothesize that a positive test for adult arrestees is more likely to reveal heavy drug use that contributes to a criminal lifestyle; whereas, juveniles who test positive for drug use may not yet be heavily dependent on drugs. If this hypothesis is correct, the authors suggest that a second- stage screening may be needed for those juveniles who test positive for drugs, so that they receive detailed clinical assessments regarding the seriousness of their drug problems. 31 tables