Identifying predictors of drug court graduation: findings from an evaluation of the Milwaukee County Adult Drug Treatment Court

ABSTRACT

Adult drug treatment courts provide an alternative to incarceration that focuses on the treatment of substance abuse issues and other psychological dysfunctions. The literature thus far has generally indicated that drug courts are significantly more effective than other criminal justice interventions, yet rates of drug court effectiveness vary substantially across jurisdictions. The current study was able to peek inside the “black box” of drug treatment courts to understand which characteristics might be critical to the success of these courts. An evaluation of the Milwaukee County Adult Drug Treatment Court in Wisconsin revealed that age at intake, race and ethnicity, prior convictions, primary drug of choice, and custody sanctions were all significantly associated with an individual’s likelihood of graduating the drug court. The inclusion of these findings to the literature offers an additional examination of drug court effectiveness and further assists in providing a more comprehensive understanding of the factors that predict drug court graduation. As a result, the health of the individual, the community, and the justice system at large can be significantly improved.

KEYWORDS:

• drug court
• substance abuse treatment
• drug offenders
• rehabilitation

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. Various drug treatment courts measure participant outcomes in different ways. Some drug courts differentiate between participants who drop-out or do not complete the program from those who are revoked from the program, whereas other drug treatment courts combine these participants and label them as revoked-only. Furthermore, graduation and completion are occasionally used interchangeably for participants who complete the program. For consistency, this paper will utilize the terms ‘graduation’ to indicate individuals who completed the program and ‘revoke’ to indicate individuals who did not complete the program despite whether they dropped out or were revoked; the literature review will also reflect this despite terminology used in the original paper. For instances in which cited studies examined participants who dropped out as well as those who were revoked, only results pertaining to those categorized as revoked were reported.

2. During the court staffing meetings, the drug court team (i.e. presiding judge, court coordinator, attorneys, and case managers) discuss each participant’s progress in the program, including prosocial and noncompliant behaviors, since the previous court hearing. Participants are not present during these meetings, but instead meet with the presiding judge during the regular court appearances where they discuss the participant’s progress in the drug court.

3. A power analysis was conducted to determine the number of participants needed in this study to conduct multivariate analyses. Results of the power analysis (to achieve power of .80, with an alpha of .05) indicated a minimum sample size of 194 participants required to detect a significant model. There were no outliers or missing data; therefore, the sample size was determined to be adequate to identify medium and large statistical effects (see Cohen, 1988).

4. In Wisconsin, officials frequently employ the Level of Service Inventory – Revised (LSI-R), which is a quantitative assessment tool that incorporates various offender attributes on criminal history, education and employment, financial, family and marital status, accommodation, leisure and recreation, companions, alcohol and substance issues, emotional and personal health, and various attitudes (Andrews & Bonta, 1995). The LSI-R aids in predicting the risk of recidivism, as well as providing appropriate services and programming for individuals (Mellow et al., 2008).

5. The LSI-R suggests the following risk-levels based on the total score that an individual receives: low-risk (0–23), medium/moderate-risk (24–33), high-risk (34–40) (Andrews & Bonta, 1995).

6. Interaction effects were created for non-Hispanic White, non-Hispanic Black, Hispanic/Latino, heroin, cocaine, and alcohol.

7. Non-Hispanic White, heroin was used as the reference category. The models were also run using different reference categories and the results were the same.

Additional information

Notes on contributors

Alyssa Sheeran

Alyssa M. Sheeran is an Assistant Professor in the Criminal Justice and Criminology Department at the University of Wisconsin – Milwaukee.  Alyssa received her Ph.D. in Social Welfare from the University of Wisconsin – Milwaukee. Her research interests focus on adult drug treatment courts, veteran’s treatment courts, sentencing disparities in the criminal justice system, and correlates of jail recidivism.

Victoria A. Knoche

Victoria A. Knoche is a postdoctoral fellow in the Helen Bader School of Social Welfare criminal justice and criminology program at the University of Wisconsin - Milwaukee. She received her Ph.D. in interdisciplinary social psychology at the University of Nevada - Reno. Her research interests center around the intersection between criminal justice, psychology, sociology, and social welfare. She is particularly interested in the justice system as it relates to juveniles, including judicial impact on youth and families in the juvenile justice system.

Tina L. Freiburger

Tina L. Freiburger is a professor and Dean of the Helen Bader School of Social Welfare at the University of Wisconsin-Milwaukee. She has published several journal articles on racial and gender disparities in decision making and juvenile justice issues. Her recent publications have appeared in Feminist Criminology, Crime and Delinquency, and Youth Violence and Juvenile Justice.