“Juvenile Drug Courts and Recidivism: Results from a Multisite Outcome Study”

Abstract

This study reports findings from a study of nine juvenile drug courts (JDCs) from across the US. A quasi-experimental design, with one-to-one matching on possible confounders and sociodemographics, was used for the outcome assessment (n = 1372). Baseline and outcome data were drawn from justice system records. Although there is variation across sites and, to some extent, outcomes, these JDCs were generally ineffective in reducing recidivism. Similar findings have emerged in other recent studies of JDCs. Given the results of this study and others, it is essential that juvenile courts work to improve the effectiveness of JDCs by increasing adherence to known principles of effective intervention.

Keywords:

• juvenile justice
• drug courts
• evidence-based practice

Acknowledgments

This study was supported by Grant No. 2007-DC-BX-0016 awarded by the Office of Juvenile Justice and Delinquency Prevention, Office of Justice Programs, US Department of Justice. Points of view or opinions in this article are those of the authors and do not necessarily represent the official position or policies of the US Department of Justice. The authors wish to acknowledge and thank research staff in the University of Cincinnati Corrections Institute, as well as those at the various study sites for their help in coordinating and carrying out research activities. We also thank the Editor and anonymous reviewers for helpful comments on an earlier version of this article.

Notes

¹ For a more thorough review of cognitive-behavioral and family therapies for adolescents, see Milkman and Wanberg (2012).

² Mitchell et al. used the Maryland Scientific Methods Scale (see Farrington, Gottfredson, Sherman, & Welsh, 2006) to categorize studies based on rigor along a four-point ordinal scale, ranging from weak quasi-experiments to randomized experiments.

³ This was meant to overcome some of the comparison group issues highlighted by Belenko and Logan (2003) and Butts and Roman (2004); it also preserves the nature of the policy as implemented (Shadish, Cook, & Campbell, 2002).

⁴ Per the original Request for Proposals (RFP), sites were selected only from those receiving some funding from OJJDP.

⁵ Initially, 10 courts agreed to participate, but one was eventually dropped due to enrollment issues.

⁶ This information was taken from a larger process study conducted at the same time as the current research. This involved site visits to each of the JDCs to collect data on the drug court process and available treatment. Those details are beyond the scope of the current study; please see the full report for details (Latessa, Sullivan, Blair, Sullivan, & Smith, 2013).

⁷ 196 youth (279%) in the probation group were enrolled in the study through this process, which allowed for each drug court to identify a comparison case by accessing already collected court data to match cases and collect official record data without specific consent or assent. In sites where blanket consents were sought, the University IRB and the Court in question was required to approve the process. These cases are equivalent to administrative record controls.

⁸ In all sites except one, comparison groups comprised solely youth on probation. A portion of that site’s drug court is a diversionary program, so comparison cases for the diversionary portion of the drug court were obtained from a nondrug court diversion program (n = 26 in DC and C groups).

⁹ Further analysis of matching quality by site revealed no association between treatment effects and the degree to which youth were/were not perfectly matched. This issue is also investigated further in the Results and Discussion sections.

¹⁰ A self-report follow-up survey was also administered, but data are not analyzed here due to a low overall response rate.

¹¹ The main analyses presented below were repeated with full information maximum likelihood estimation (see Schafer & Graham, 2002) in order to better account for missing data. The results were similar to those reported here.

¹² The date of termination in the analyses focused only on outcomes following program involvement.

¹³ Multicollinearity diagnostics indicate tolerance values above .90 and variance inflation values ranging from 1.0 to 1.8 suggesting no problems with linear dependence among the covariates.

¹⁴ These analyses were also conducted using negative binomial regression models for the number of new referrals (Mean = 1.66, sd = 2.48) and number of new convictions (Mean = .95, sd = 1.82). The relative risk ratios were 1.31 in both scenarios, suggesting that the Drug Court group had roughly 30% higher expected counts on the outcome measures–controlling for risk level, time at risk of a new offense, age, sex, and race.

¹⁵ Although frequency of substance use (alcohol, other drugs) was collected for the majority of youth (n = 1060), it was not included in the main models due to data loss (particularly among comparison youth). Given that it is a relevant covariate and the baseline differences observed between groups, the main outcome analysis was repeated using that measure and the results were found to be similar to those reported here.

¹⁶ This analysis was based on 750 cases with available data on the official diagnosis measure.

¹⁷ Further analysis of those two sites in terms of the characteristics of the youth involved and the processes of the drug courts revealed relatively few differences from the other seven. Most of those variables were either (a) controlled for in the analyses (e.g. time at risk) or elements of the drug court process that tended to vary between treatment and comparison youth at all sites (e.g. number of case hearings, number of drug tests). More information on site-level differences is provided in Latessa et al. (2013).

¹⁸ There was, however, missing data in the drug screen variables that predominately affected the comparison group.

Additional information

Notes on contributors

Christopher J. Sullivan

Christopher J. Sullivan is an associate professor in the School of Criminal Justice at the University of Cincinnati. His research interests include developmental criminology; juvenile delinquency and prevention policy; and research methodology and analytic methods. His recent work has appeared in Justice Quarterly, Journal of Research in Crime and Delinquency, British Journal of Criminology, Prevention Science, and the Journal of Experimental Criminology.

Lesli Blair

Lesli Blair is a doctoral candidate at the University of Cincinnati (UC). Her dissertation is on community gardens and their potential impact on crime via either collective efficacy theory or routine activities theory. Her research interests include testing and integrating criminological theories; the effects of imprisonment; and the psychology of criminal behavior. She worked for three and a half years at the UC Corrections Institute where she co-managed a national multisite study on juvenile drug courts.

Edward Latessa

Edward J. Latessa received his PhD from the Ohio State University in 1979 and is a professor and director of the School of Criminal Justice at the University of Cincinnati. He has published over 140 works in the area of criminal justice, corrections, and juvenile justice. He has directed over 150 funded research projects including studies of day reporting centers, juvenile justice programs, drug courts, prison programs, intensive supervision programs, halfway houses, and drug programs. He and his staff have also assessed over 600 correctional programs throughout the United States.

Carrie Coen Sullivan

Carrie Sullivan, M.A is a Research Associate in the Center for Criminal Justice Research and serves as the Associate Director for the Corrections Institute at the University of Cincinnati (UC). She received her master’s degree in criminal justice from the University of Massachusetts Lowell in 2001. Ms. Sullivan has 14 years of correctional experience and has been a project manager for numerous national, state, and local correctional projects.  Ms. Sullivan conducts correctional program evaluations to assist agencies in providing effective correctional interventions and assists agencies in implementing evidence-based practices. Ms. Sullivan also trains correctional professionals on numerous evidence-based practices.