Stein and colleagues (2006a) examined participants from a northeastern State juvenile facility where adolescents were identified immediately following adjudication. To be included, participants had to be between the ages of 14 and 19, have been sentenced to between 4 and 12 months, have presented regular (or binge) marijuana or alcohol use (monthly or in the month before the offense or in the month before incarceration), and consent to participate (from parents, guardians, or adult adolescents). This study was part of a larger treatment outcome study. The sample (n= 130) was 34.6 percent African American, 28.5 percent Hispanic, and 30.8 percent white. Ninety percent were male. The average age was 17.16 years, and on average they had been incarcerated 2.92 times before the current imprisonment. Sixty-two percent of the sample qualified for alcohol use disorder, while 89 percent qualified for marijuana use disorder in the previous year. The sample was then randomized into a treatment group, who were administered Motivational Interviewing (MI), and a comparison group who received Relaxation Therapy (RT). The purpose of the study was to examine the effects of MI on youth engagement with substance use treatment.
The MI was administered by research counselors delivering the four components of MI adapted as appropriate to each adolescent. The RT was delivered by research counselors instructing participants on relaxation and meditation techniques and providing general advice and information on reducing risky behavior and alcohol and marijuana use. Both interventions were 90 minutes at baseline and 60 minutes at booster. Assessment was made at baseline and then at 2 months postadjudication. In addition to these interventions, adolescents received the standard services offered by the facility.
Adolescents receive group and individual treatment on a variety of behavioral problems (drug use, sex offending, violence prevention, anger management, etc.). Drug programs at the facility concentrate on increasing the adolescents’ knowledge of the negative effects of drug use, including HIV risks, mental, dental, and psychological services also are available.
The measures used at baseline by the study were a record review, a background questionnaire, and the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders (or DSM–IV). At the 2-month follow-up, the average daily behavioral rating was used (adolescents’ behavior is rated with a possible 100 points to accumulate per week) and a treatment participation questionnaire (evaluating their engagement at baseline and at 2-month follow-up). The authors performed Analysis of Covariance (ANCOVA) for each dependent variable (negative treatment engagement, positive treatment engagement, and behavioral ratings).
The second Stein and colleagues (2006b) study was also part of a larger treatment outcome study using the same inclusion criteria as the previous study. It examined the effects of MI on driving under the influence (DUI) offenses of adolescents after release. The final sample (n= 105) was 34.3 percent African American, 27.6 percent Hispanic, and 32.4 percent white. They were 89.5 percent male, and their average age was 17.06 years old. Sixty-one percent of the sample qualified for alcohol use disorder, while 89 percent qualified for marijuana use disorder in the previous year.
Participants were randomly assigned to either the MI group (n= 59) or to a RT comparison group (n= 46). Assessment was made at baseline and then at 3 months postrelease. The measures used at baseline by the study were a record review, a background questionnaire, and the Structured Clinical Interview for DSM–IV. At both points in time, the Center for Epidemiological Studies—Depression Scale and the Risky Behaviors Questionnaire were administered (concentrating on DUI and being a passenger with a driver under the influence—PUI).
The methods applied to the data were Analysis of Covariance (ANCOVA) for each dependent variable (number of DUI–marijuana, DUI–alcohol, PUI–marijuana and PUI–alcohol), using baseline measure as a covariate with the intervention condition (MI or RT) and depressive symptoms as independent variables.