This R21 proposal seeks 3 years of funding to conduct a randomized trial with pre-adjudicated juvenile offenders. Substance use is a key factor in the high crime rate and high recidivism rate by juvenile offenders. Currently, first and second offenders are referred to a diversion program, while second or third offenders are sent to an eight-week substance education program. Further substance-use offenses may result in a referral to drug court, or, when there are no openings in drug court, probation, inpatient treatment, or commitment. The proposed project would offer an intermediate step before drug court referral and an opportunity for court-referred adolescents to change behaviors that could potentially result in incarceration. Adolescent offenders aged 14-17 referred for a third, fourth, or fifth time, with at least one substance-related offense, will be screened for drug or alcohol dependence, and those not referred to treatment will be eligible for the study. A total of 288 juveniles will be randomly assigned to either a cognitive behavioral skills training program intervention group (CBTP) or a drug education alternative intervention group (DE). Adolescents refusing participation in the study will comprise a third """"""""no-contact"""""""" group that will be used to model recidivism rates. We propose a two-factor (clinician by treatment) experimental design to evaluate the effectiveness of the CBTP on adolescent behavior. The two-factor design allows us to evaluate the effectiveness of the CBTP program over the DE program, as well as the portability of CBTP across clinicians. The CBTP groups, will meet over a 3-month period for 12, 90-minute weekly sessions and the DE groups will meet over a 3-month period for 12, 60-minute weekly sessions. Participants will complete a baseline assessment prior to randomization, a post-intervention assessment, and a six-month follow-up assessment. Court records will be reviewed at three and nine months following randomization to determine the occurrence and nature of re-arrests. The main intervention effect examined in this study is whether juveniles exposed to the CBTP intervention report significantly lower drug use than alternative DE intervention group. We hypothesize that the CBTP group will have lower rates of substance use and lower recidivism rates, compared with the DE group, after adjusting for important pre-intervention variables. The main method of analyzing intervention effects, the general linear model approach, permits adjustments for unbalanced non-orthogonal designs. The GLM analyses will control for pretest scores as well as additional factors that may contribute to a lack of pretest equivalence (e.g. differences in age). This study holds promise for the application of primary prevention strategies in secondary """"""""indicated"""""""" settings using evidence-based cognitive-behavioral strategies to reduce substance use and delinquency.
| McMillan, Garnett P; Bedrick, Edward; C'deBaca, Janet (2009) A Bayesian model for estimating the effects of drug use when drug use may be under-reported. Addiction 104:1820-6 |
