"Factors Predicting Treatment Retention among High-Risk
Mexican American and African American Juvenile Offenders"
Investigator: David Springer, Ph.D.
Current Project Status and Findings
Many of the same factors that predict delinquent behavior also predict adolescent drug use, which has direct implications for intervention with substance-abusing juvenile offenders. Because delinquent and drug use behaviors overlap among juvenile offenders, interventions should focus on the predictors of the behaviors rather than on the behaviors themselves. There is insufficient information about factors that relate to treatment retention or the effectiveness of various treatment modalities with minority adolescent substance abusers. Furthermore, both African American and Mexican American youth are over-represented in the juvenile justice population, and their arrest rates for drug-related offenses have soared.
The foci of this study are:
· to explore the utility of the Transtheoretical model in understanding treatment retention,
· to examine additional factors that predict treatment retention, and
· to examine the interventions that maximize treatment retention among Anglo, African American, and Mexican American high-risk juvenile offending substance abusers.
Overall, the purpose of this study is to use data collected from multiple sources to examine whether Caucasian, African American, and Mexican American youth differ on their retention in substance abuse treatment programs using their first treatment admission. These three groups are compared for three different types of treatment: inpatient, day treatment, and outpatient. Understanding these differences can help in the design of intervention and treatment programs targeted toward minority youth.
Study subjects included 50 Caucasians (19.3%), 88 African Americans (28.3%), and 163 Mexican Americans (52.4%). The data set included 56 females (18%) and 255 males (82%). Treatment programs included: Residential (n = 102; 34%), Day Treatment (n = 90; 30%), and Outpatient (n = 105; 35%). A Stage of Change typology was constructed, classifying youth as being in one of three stages: ‘precontemplation’, ‘contemplation’, or ‘preparation’. The distribution of youth by Program and Stage of Change is shown below. There was no significant relationship between Program and Stage:
Generally, the majority of youth in each program were in the precontemplation stage.
Data analysis is in process. Preliminary survival analyses have been conducted to address the key research question of the study: Does the Stage of Change model predict treatment retention separately for high-risk adolescents in residential treatment?
Using a Cox regression model, we tested whether four sets of predictors were related to residential treatment retention: (1) gender and ethnic background; (2) Stage of Change (precontemplation, contemplation, or preparation), (3) CASI dimensions (number of negative education experiences, number of negative family experiences, number of positive extra-curricular activities, number of negative psychological experiences, number of stressful life experiences, number of substances ever abused, number of years engaged in abusing multiple substances), and (4) other intervention status (probation, case management, and mental health).
These preliminary analyses indicated the following:
There were no significant differences between programs in terms of treatment retention. Both the Log-Rank Chi-square(2)=2.24, p = .3265 and Wilcoxon Chi-square(2)=4.43, p = .1092 tests showed no statistically significant differences
· Mexican American youth were more likely to leave treatment relative to African Americans. More specifically, being Mexican American was related to a 23% increase in the hazard of leaving residential treatment [Odds ratio= .77, p = .0292]. This held true across all four models tested (i.e., inclusive of all four predictor variables defined above).
· For each additional family problem ever experienced, Mexican American adolescents were 15% more likely to leave treatment compared to African American families. Additionally, compared to Anglos, African American and Mexican American adolescents were 147% more likely to leave residential treatment when on probation. Finally, compared to African Americans, Mexican American adolescents were 15% more likely to leave residential treatment for each additional substance ever used.
We predicted that Mexican American and African American juveniles would be at greater risk of leaving treatment. However, based on both research questions presented above, it appears that Mexican American juveniles faired worse when compared to African American juveniles, especially when they experienced more family problems and used more substances. The treatment implication for juvenile court personnel and for treatment providers is that Mexican American juveniles appear to be at greater risk of leaving treatment prematurely. Thus, greater efforts must be made with regard to treatment retention early on. Additionally, given that experiencing family problems resulted in Mexican American juveniles being 15% more likely to leave treatment when compared to African American juveniles, family involvement in the treatment process for these juveniles appears critical.
Funded by: NIDA Social Work Research Development Program R24-DA13579.
Publications
and Presentations:
Springer, David. “Treatment with
Substance-abusing Juvenile Offenders: A Longitudinal Outcome Study”.
Presentation at the 6th Annual Conference of the Society for Social
Work and Research, January, 2002, San Diego.
Be sure and check the schedule of our First Friday Workshop Series for upcoming workshops. Session notes are posted on the schedule when available.
Center for Social Work Research
School of Social Work
University of Texas at Austin
1925 San Jacinto Blvd
Austin, TX 78712-1203
(512) 471-5457
Send comments and questions to: cmckinley@mail.utexas.edu CREDITS
