Practice Profile: Adult Drug Courts

Aos and colleagues (2001) set out to evaluate the costs and benefits of certain juvenile and adult criminal justice policies, violence prevention programs, and other efforts to decrease certain at-risk behaviors of juveniles and adults. They used meta-analytic techniques to estimate the degree to which a program or policy can be expected to influence the outcome of interest (i.e., criminality).

The review focused on evaluations that measured a program’s effect on criminal behavior. Program evaluations were gathered from a wide variety of sources, including studies that were published in peer-reviewed journals, as well as other studies not published in journals (such as government or private agency sources). Once evaluations that met the criteria of inclusion were located, four types of information on the program’s effectiveness in reducing crime were recorded: (1) Did the program affect the percent of the population that offended or reoffend? (2) Of those that offended or reoffended, did the program change the average number of offenses? (3) Did the program affect the types of offenses of those that offended or reoffended? (4) Of those that offended or reoffended, did the program change the timing of offenses? Of the four effects, almost all of the evaluations provided information on the first effect. Far fewer evaluations reported information on the second effect, and even fewer reported information on the third effect. Almost none of the evaluations reported information on the fourth effect.
 

The search resulted in the inclusion of 26 studies of drug court programs. The 26 studies included approximately 14,691 adult participants. Three of the studies by Goldkamp (2000) were excluded due to repeated review of cohorts. The group of 26 studies included published and unpublished evaluations that spanned from 1993 to 2000. No information was provided on the age, gender, or racial/ethnic breakdown of the studies’ samples, or on the location of the programs.

The mean difference effect size was calculated for each program. Adjustments were made to the effect sizes for small sample sizes, evaluations of “non-real world” programs, and for the quality of the research design (the quality of each study was rated using the University of Maryland’s five-point scale; only studies that received a rating of ‘3’ or higher on the scale were included in the analysis). Once effect sizes were calculated for each program effect, the individual measures were added together to produce a weighted average effect size for a program or practice area. The inverse variance weight was calculated for each program effect and those weights were used to compute the average. The fixed effects model was used for the analysis.

Latimer, Morton-Bourgon, and Chrétein (2006) located 54 individual studies between 1993 and 2005 to determine if drug treatment courts for youth and adults reduce recidivism compared to traditional justice system responses. Similar to Aos and colleagues (2001), to be included in this meta-analysis, studies needed to have (1) examined the effectiveness of a drug treatment court, (2) used a control group that did not experience the drug treatment court, (3) searched for sufficient statistical information to establish an effect size, and (4) measured the impact of the drug treatment court on recidivism rates (Latimer et al., 2006). Age, length of the program, follow-up period, and methodological versus random assignment were identified as variables in the analysis.

Because studies often contained information on more than one program, data from 66 individual drug treatment court programs were aggregated and analyzed. The 66 studies included 17,214 individuals who had successfully completed a drug treatment court program. The comparison group comprised of 14,505 who had not received the intervention. The meta-analysis evaluated mean participant characteristics. Average program size consisted of 260 participants who were 28.4 years old. No information was provided on the gender or racial/ethnic breakdown of the studies’ samples, or on the location of the programs.

The phi coefficient (Pearson’s r product moment correlation applied to dichotomous data) was used as the effect size estimate. Once the effect size estimate from each study was calculated for recidivism, the overall mean effect size estimate, along with the corresponding confidence intervals, and a weight effect size estimate were calculated.

Mitchell and colleagues (2012) conducted an extensive search process for studies that (1) evaluated a drug court program (defined as specialized courts for handling drug-involved cases that are processed in a non-adversarial manner, refer them to appropriate treatment programs, regularly test for drug use, and have a judge who actively monitors progress and provides sanctions for misbehavior), (2) included a comparison group in the evaluation that was treated in a traditional fashion by the court system, (3) measured criminal behavior (such as arrest or conviction), and (4) provided enough information to compute an odds ratio.

The search process resulted in the inclusion of 154 independent evaluations of drug courts. Of those, 92 assessed adult drug courts, 34 examined juvenile drug courts, and 28 investigated driving while intoxicated (DWI) drug courts. The outcomes reported here are for the 92 evaluations that assessed adult drug courts. The first data set preference was given to effect sizes that were general, based on arrest, dichotomous, and followed sample members for 12 months. For each evaluation contrast, an odds ratio was calculated. Effect sizes were calculated for three outcomes: (1) general recidivism; (2) drug related recidivism); and (3) drug use. The study utilized the effect size. In addition, a standardized mean different effect size was used and transformed to odds ratios. A random-effects model was used.

To control for the rigor of study design, the 154 studies were placed into four possible categories: (1) weak quasi-experiments, (2) standard quasi-experiments, (3) rigorous quasi-experiments, and (4) randomized experiments. Measures were also coded into three dimensions pulled from a study by Longshore and colleagues (2001). They included leverage, population severity, and program intensity.

Shaffer (2010) sought to identify characteristics of effective drug courts. The author used meta-analytic techniques to investigate the relationship between six structural and five process dimensions of drug court effectiveness. The six structural dimensions included target population, leverage, service delivery, staff, funding, and quality assurance. The five process dimensions included assessment, philosophy, treatment, predictability, and intensity. The objective was to merge survey data with existing study data to determine the relative influence of each dimension listed above.

Studies of drug court programs using an experimental or quasi-experimental design were searched for in various locations, including databases and government reports (i.e., Academic Press, Criminal Justice Abstracts, and Government Publications Monthly). This search included a comparison group and at least one measure of criminal behavior with a six month follow-up period.

A total of 115 studies were found, but only 60 studies evaluated separate drug courts and were eligible for inclusion. From the 60 outcome evaluations, there were 76 individual samples of drug courts and 6 multi-site evaluations where the results could not be disaggregated. From the eligible studies, there were 61 samples of adult drug courts that contributed to the overall effect size. This included 20,830 individuals contributing to the overall effect size.

Drake (2012) conducted a meta-analysis to review the effectiveness of various types of chemical dependency treatment in the adult and juvenile justice systems to determine whether the programs reduce crime and substance abuse. For adults, the different types of treatments included drug treatment during incarceration, drug treatment delivered in the community, case management for individuals with substance use disorders, therapeutic communities for those with co-occurring disorders, and drug courts. There were four primary methods used to locate studies: (1) bibliographies of systematic and narrative reviews in various topic areas were consulted; (2) citations from the individual studies were examined; (3) independent searches of research databases were conducted; and (4) authors of primary research were contacted. Studies were included if they had a comparison group (random or non-random assignment), contained enough information to calculate an effect size, and had a standardized measure of the primary outcome of interest (crime). A total of 51 evaluation studies of adult drug courts were included in the analysis.

The search process resulted in the inclusion of 51 evaluation studies of adult drug courts. However, some studies represented multiple sites under evaluation. The 51 individual studies represented 67 effect sizes. The standardized mean difference effect size was calculated for each program effect. Once the effect sizes are calculated, the individual measures are summed to produce a weight average effect size for the program area (in this instance, adult drug courts). Adjustments were made to the effect sizes to account for small sample sizes, evaluations of “non-real world” programs, and research design quality (the quality of each study was rated using the University of Maryland’s five-point scale; only studies that received a rating of ‘1’ or higher on the scale were included in the analysis). A random effects model was used to calculate the weighted average effect size.

Sevigny, Fuleihan, and Ferdik (2013) conducted a meta-analysis to determine the impact of adult drug courts on overall incarceration. A comprehensive search of the literature was conducted to identify eligible studies. To be eligible for review, studies were required to 1) be experimental or quasi-experimental evaluations of U.S. adult drug court programs, 2) use a comparison group that is subject to routine criminal justice processing such as probation, 3) report incarceration outcomes, and 4) be published in English. All studies also had to be published after 1989, the date the first drug court was established.  

The search process resulted in 229 potentially eligible publications, of which 201 were available for further review.  However, a large majority of these studies did not report incarceration outcomes, leaving only 19 studies that contributed at least one independent effect size related to incarceration outcomes, and only 11 that focused on the overall use of incarceration. Approximately 11.1 percent of the included studies were categorized as randomized experimental, 50 percent as strong quasi-experimental, and 38.9 percent as weak quasi-experimental. The majority of studies were evaluated between 1995–1999 (72.2 percent), with 5.6 percent evaluated between 1990–1994. Approximately 22.2 percent were evaluated in 2000 or later. A logs odds ratio was calculated to examine the impact of drug courts on overall incarceration.