Table of Contents
Impact of Juvenile Drug Courts on Drug Use and Criminal Behavior
Audrey Hickert, Erin Becker, Moisés Próspero, and Kristina Moleni
Utah Criminal Justice Center
University of Utah, Salt Lake City, Utah
Audrey Hickert, Utah Criminal Justice Center, University of Utah; Erin Becker, Utah Criminal Justice Center, University of Utah; Moisés Próspero, Utah Criminal Justice Center, University of Utah; and Kristina Moleni, University of Utah.
Keywords: drug courts, recidivism, delinquency, drug use, probation
Acknowledgements: The authors wish to acknowledge the Utah Commission on Criminal and Juvenile Justice (CCJJ) and the Utah Administrative Office of the Courts (AOC) for their funding and support of this research.
Correspondence concerning this article should be addressed to Audrey Hickert, Utah Criminal Justice Center, University of Utah, 395 South 1500 East, Rm. 230, Salt Lake City, UT 84112-0260; E-mail: Audrey.firstname.lastname@example.org
Juvenile drug courts (JDC) have borrowed the philosophy and models of adult drug courts but the success of JDCs in reducing drug use and criminal behavior has been mixed. This study compared JDC youth with youth receiving standard probation on alcohol and other drug (AOD) and delinquency/criminal re-offending at three through 30 months post-exit from the JDC program or probation. This quasi-experimental study tested JDC effectiveness by examining re-arrests for AOD and criminal offenses 30 months post-intervention and into adulthood. Participants included youth who participated in either JDC (n = 622) or probation only (n = 596) between the years 2003 and 2007. JDC and probation youth did not significantly differ at any of the follow-up time intervals on AOD offending. On the other hand, JDC youth had statistically significantly fewer delinquency/criminal offenses than probationers at all follow-up points, with the difference between the groups getting larger with longer follow-up periods. Implications for practice, policy, and future research with JDC are discussed.
Juvenile drug courts (JDCs) were created in the mid-1990s following the initial success of problem-solving courts in the adult system and an overall shift toward therapeutic justice (American University, 1999; Applegate & Santana, 2000). As of December 2007, there were 455 juvenile drug courts in operation throughout the country (Huddleston, Marlow, & Casebolt, 2008). Evidence supporting the effectiveness of adult drug courts in reducing recidivism has been consistent (Belenko, 2001; Gottfredson, Najaka, & Kearley, 2003; U. S. Government Accountability Office [GAO], 2005). However, recent studies note that the support for the effectiveness of JDCs is limited (Fradella, Fischer, Hagan Kleinpeter, & Koob, 2009; Ruiz, Stevens, Fuhriman, Bogart, & Korchmaros, 2009). Several evaluations report that JDC youth fare better than non-JDC youth on measures of during and post-program recidivism (Anspach & Ferguson, 2005; Dickie, 2000; Latessa, Shaffer, & Lowenkamp, 2002; Thompson, 2001; Willard & Wright, 2005). On the other hand, some studies show no better outcomes for JDC youth than those receiving more traditional interventions, particularly when increased supervision among JDC youth leads to increased detection of substance use and delinquent behaviors (Rodriguez & Webb, 2004). Furthermore, because JDCs vary widely in their program structure and target populations, their relative success compared with youth processed through traditional court procedures also varies (Thompson, 2004).
The present quasi-experimental study seeks to contribute to the growing body of literature on JDC effectiveness by examining post-program alcohol/other drug (AOD) offenses and delinquency/criminal offenses. Participants from four JDCs were compared with juvenile AOD offenders who received probation. We hypothesized that JDC youth will fare better than similar probationers on post-program AOD offenses and delinquency/criminal offenses when tracked into adulthood.
Juvenile Drug Courts
Although the specific treatment and content of the programming is different for each JDC, the primary drug court philosophy and components are consistent. There are a number of elements to the drug court model. These include: 1) screening and assessment, 2) an individualized treatment plan, 3) judicial supervision, 4) community-based treatment, 5) a designated courtroom, 6) regular status hearings, 7) accountability and compliance monitoring, 8) sanctions and incentives, 9) comprehensive services, 10) a non-adversarial team approach, and 11) case dismissal or reduction for successful completers (National Association of Drug Court Professionals (NADCP), 1997; Office of Justice Programs (OJP), 1997).
In the case of JDCs, changes were made to the model to address the differences between juveniles and adults. While the model still emphasizes addressing the underlying issues, there are some specific considerations unique to adolescence substance abuse treatment, such as: the conflict between a desire for independence and the juvenile’s dependence on their family, limit testing, physical and emotional maturation and development, and peer pressure (Belenko & Dembo, 2003; Kaminer, 2001). Family involvement, especially, has been noted as a key aspect of the JDC in relation to predicting both graduation (Fradella et al., 2009) and long-term success (Hills, Shufelt, & Cocozza, 2009).
JDC treatment is not standardized. Rossman, Butts, Roman, DeStefano, and White (2004) note that JDC “treatment approaches (e.g., therapeutic models, individual versus group settings, frequency and duration of treatment) vary from one program to another” (p. 57), in large part due to the availability of local treatment providers and resources. Additionally, Hills et al. (2009) found that most of the JDCs they surveyed were not following evidence-based treatment practices.
Because of adolescent developmental issues and lack of standardized evidence-based treatment, JDCs may not be as effective as adult drug courts. Even so, there is a growing body of literature on JDC outcomes that has demonstrated some qualified success for the therapeutic jurisprudence model when used with juveniles (Anspach & Ferguson, 2005; Dickie, 2000; Latessa et al., 2002; Thompson, 2001; Willard & Wright, 2005; see also Marlowe (2010) for a research update on JDCs).
Juvenile Drug Court Impact on Juvenile Offending
Published articles that support the effectiveness of JDCs by comparing them to similar youth who receive typical juvenile court processing are limited; however, several evaluations of JDCs have shown at least short-term or qualified successes. Studies have shown reduced re-arrests, number of charges or offenses, or court referrals for JDC youth compared with similar youth who did not receive this intervention.
Henggeler et al. (2006) tested the JDC model in four different types of interventions: family court, drug court, drug court with multisystemic therapy, and drug court with multisystemic therapy and contingency management. JDC participants in multisystemic therapy and contingency management reported significantly less alcohol use than juveniles receiving the other three interventions. For marijuana and polydrug use, JDC participants in multisystemic therapy and contingency management and those in multisystemic therapy alone displayed significant extended treatment effects at 12 months (lower drug use), compared with juveniles receiving the other treatments. These findings suggest that the use of evidence-based practices in JDC is more likely to have sustainable positive treatment effects than interventions without such practices. This same pattern of results was found in self-reported delinquency and crime; however, there were no significant differences between treatment conditions in re-arrest outcomes (Henggeler et al. 2006).
Dickie (2000) found that six months after program completion, JDC participants averaged one arrest compared with an average of 2.3 arrests for similar offenders who were randomly assigned to usual court processes. In addition, only 11% of JDC participants had three or more new charges during this time, compared with 46% of those who did not participate in the JDC. In another study, participants in three Ohio JDCs (in Belmont, Summit, and Montgomery counties) were compared with similar offenders who were not treated in a JDC. The study revealed statistically significant differences between the groups in re-arrest rates, with 75% of the comparison group re-arrested compared with 56% of the JDC group (Latessa et al., 2002).
A study of the Delaware Juvenile Drug Court Diversion Program comparing JDC youth with untreated juvenile offenders with substance abuse issues found that JDC graduates experienced better outcomes related to post-program recidivism than comparison youth 12 months after program completion (Miller, Scocas, & O’Connell, 1998). In a follow-up study 18 months after the end of the treatment period, these positive results for JDC participants were retained, with 67% of comparison youth recidivating compared with 48 % of successful JDC participants (O’Connell, Nestlerode, & Miller, 1999).
In an evaluation of Maine’s Statewide Juvenile Drug Treatment Court, Anspach and Ferguson (2005) found that fewer JDC participants (44%) than a matched probationer comparison (52%) were re-arrested in the year after program completion. Furthermore, JDC participation was significantly associated with a decreased risk of re-offending. Thompson (2001) found that, after controlling for demographic characteristics and court history, JDC youth had a 69% lower risk of recidivating than a group of substance abusing juveniles not participating in drug court in the year following JDC start or court referral (for comparison youth). Finally, Willard and Wright (2005) revealed a lower re-arrest rate for JDC participants (43%) than juvenile drug offenders in jurisdictions without JDCs (60%).
In contrast, other studies have shown mixed findings for JDC effectiveness when compared with other interventions for youth in the juvenile court system. Rodriguez and Webb (2004) found that JDC participants were less likely to commit a subsequent criminal act than similar juvenile offenders assigned to standard probation in the same county. However, drug screenings indicated that JDC participants were using marijuana as much as juveniles assigned to standard probation and using cocaine 2.7 times more than probation participants (Rodriguez & Webb 2004). The authors note that this finding could indicate an increase in supervision and, therefore, detection, rather than an actual higher rate of drug use among the JDC participants. However, in a later study of the same JDC examining the role of social bonds in JDCs, the positive finding of reduced delinquency for JDC youth compared with juvenile probationers no longer remained (Gilmore, Rodriguez, & Webb, 2005). The findings that JDC youth were more likely to test positive for drugs and fail to successfully complete program requirements remained.
Finally, some studies have shown worse outcomes for JDC youth compared with those processed in traditional juvenile court. Sloan, Smykla, and Rush’s (2004) study found that, after controlling for the significant effects of age, criminal history, ethnicity, gender, and termination status, participants in the JDC and the comparison substance abuse treatment program did not significantly differ in re-arrest during the 24 months following program exit. Furthermore, average time to re-arrest for the JDC group (M = 8 months) was significantly shorter than for the comparison group (M = 15 months). Hartmann and Rhineberger (2003) found that although JDCs reduced criminal offending among participants both during and after exiting the program (compared with pre-JDC crime rates), these reductions were not as great as those recorded for youth who opted out of the program. JDC participants had a pre-program crime rate of 1.64 and a post-program crime rate of 0.62, compared with a pre-program crime rate of 1.67 and a post-program crime rate of 0.49 for those who opted-out of the JDC.
Juvenile Drug Court Impact on Adult Offending
Two studies have followed JDC participants into adulthood and compared their outcomes with a comparison group of similar youth (Thompson, 2004; Pitts 2006). An evaluation of two JDCs in North Dakota compared JDC participants with a comparison group of youth who did not participate in JDCs (Thompson, 2004). The JDC and comparison groups were similar on the majority of sample characteristics (e.g., demographics, court history); recidivism was measured as an adult arrest or conviction for a Class A misdemeanor or more severe offense. Terminated JDC youth from both JDCs had a higher adult re-arrest rate (52%) than the comparison group (44%). Graduates from the more rigorous (and longer-lasting) JDC had the lowest adult arrest rate (21%), while graduates of the less rigorous JDC had the highest re-arrest rate (60%) of any group (Thompson, 2004). This study demonstrates that the positive impacts of JDCs may last into adulthood. In addition, the results demonstrate that the success rates of JDCs can vary widely based on the structure and characteristics of the JDC program itself; specifically, the largest effects may be found among graduates of programs that faithfully follow evidence-based drug court principles.
Pitts (2006) examined both juvenile and adult recidivism outcomes for JDC youth compared with a similar group of juvenile probationers. In this retrospective study of a New Mexico JDC, comparison youth who received juvenile probation (and were never screened or referred to drug court) were matched with JDC participants on demographic, substance abuse, and juvenile court history factors. Pitts (2006) noted that all JDC youth were also on probation. Recidivism measures included any new referral to juvenile court and any new adult arrest. All study participants had at least 16 months follow-up post-exit from JDC or probation. When examining only juvenile recidivism, Pitts found no significant differences between the two groups (JDC = 23%, Comparison = 30%). When the study examined adult recidivism alone, it also found no statistically significant group differences (JDC = 18%, Comparison = 30%). However, when Pitts examined juvenile and adult recidivism together, JDC participants recidivated at a significantly lower rate (37%) than participants in the comparison group (56%). Further detailed analyses found that JDC graduates had a combined (juvenile and adult) recidivism rate of 28%, compared with 43% for terminated JDC participants (Pitts, 2006). By combining juvenile and adult recidivism measures and allowing for a sufficient follow-up period, this study demonstrates the potential long-term impacts of JDCs. However, this study relied on a small sample size (JDC n = 62, probation n = 61) and did not examine differences in delinquency/criminal versus alcohol/drug recidivism.
The present study will build upon the nascent JDC literature by examining both juvenile and adult recidivism for JDC participants compared with youth with a similar AOD offense who received probation. This study will test the following hypotheses:
- Juvenile drug court participants will have significantly fewer post-program alcohol and other drug (AOD) offenses than similar youth who receive probation.
- Juvenile drug court participants will have significantly fewer post-program delinquency and criminal offenses than similar youth who receive probation.
The JDC group consisted of participants in Utah’s four largest JDCs from January 2003 to May 2007 (n = 622). These JDCs were located in primarily urban and suburban juvenile court districts. We obtained participation lists from each JDC; we included all participants during this time period, regardless of exit status, in the JDC study group. Two of the JDCs primarily served youth who were also on probation (see Table 1), while two were considered primarily an alternative to probation.
We selected the AOD probationer comparison group from the state juvenile court database. We identified youth who had an AOD offense (e.g., minor in possession, driving under the influence [DUI], controlled substance possession) that resulted in a probation placement between 2003 and 2007. For those youth who had more than one offense, we randomly selected one offense as the primary event. We removed from the comparison group youth who had ever been in a JDC, resulting in 596 comparison youth. A major limitation of this comparison group was that they had significantly more severe juvenile court histories than the JDC youth (see Table 1). Although we made various other attempts to identify a more appropriate comparison group from juvenile court data, including propensity score matching, we could not find a more similar group with sufficient sample size. Probationer youth may also receive substance abuse treatment (sometimes even at the same providers as JDC youth), but a reliable record of this was not available for comparison.
Independent variable. The independent variable was program participation, coded as 0 = Probation and 1 = JDC. Probation participation, as well as start and end dates, came from the state juvenile court database. JDC start dates, end dates, and exit statuses came from individual JDC records. Each JDC also provided brief qualitative information about their program structure, such as number of phases, drug testing and judicial hearing frequency, and available treatment options, such as modality and and intensity (see Table 2).
Control variables. We collected youth demographic information and court involvement measures from the state juvenile court database. Demographic variables were date of birth, gender, and race/ethnicity. Court involvement measures included offense type, offense date, and referral date, as well as the dates of contempt and probation violations. Due to the small number of participants from individual racial and ethnic groups, we combined race/ethnicity information into a minority flag with 0 = White, Non-Hispanic and 1 = Hispanic and other race categories (e.g., African American, Native American, Pacific Islander, and Asian). We used date of birth in conjunction with the offense date of each youth’s first offense to calculate age at first offense. We subtracted date of birth from JDC and probation start dates to calculate age at start for each program.
We identifited prior offenses (priors) as any offense occurring prior to JDC or probation placement. The count of offenses included each unique offense type on any offense date; therefore, if three offense types (e.g., possession of controlled substance, possession of drug paraphernalia, and shoplifting) were referred to the juvenile court, whether they occurred on the same or different offense dates, the count of prior offenses would be three.
We split juvenile offenses into two types: 1) AOD, which included DUI offenses, and 2) delinquency, which included person, property, and public order offenses. We excluded status, infraction, and traffic (except DUI) offenses, as well as non-compliance with court orders (e.g., contempt and probation violations), from the count of juvenile offenses. We computed a separate count of contempt and probation violations occurring during JDC or probation placement as an additional control variable measuring program non-compliance.
Dependent variables. We defined recidivism dichotomously as a new juvenile or adult AOD or delinquency/criminal offense in the 30 months post-exit from JDC or probation. We defined juvenile recidivism as a new court referral for an AOD or delinquency offense following JDC or probation exit. Information on adult recidivism came from the state department of public safety’s (DPS) criminal history record; we defined this as the arrest of any adult for a criminal or AOD offense, including DUI. We did not include arrests for non-DUI traffic offenses. We sent the two study groups (JDC and AOD probationer) that we identified using the juvenile court data to the DPS for a match of their records. We searched multiple name and date of birth combinations (aliases) for participant matches from juvenile and adult records. Just over one-half of the JDC (51%) and AOD probationer (52%) samples matched with an adult criminal record. We identified those who did not match across systems as non-recidivists in the adult system.
We used bivariate analyses, including chi-square and independent sample t-tests, to compare the JDC and AOD probationer groups on demographics and juvenile court histories. We used chi-square and one-way analysis of variance (ANOVA) for comparisons across the four JDCs. We also used chi-square tests to test statistical significance of the differences in the percentage of youth recidivating from each group at various follow-up points and we used bivariate tests to examine the relationship between youth characteristics and recidivism. Finally, we conducted two logistic regressions to examine the influence of group membership (JDC versus probation) on AOD recidivism and delinquency/criminal recidivism after controlling for other significant individual factors (e.g., demographics, youth court history) that were initially identified in the bivariate tests and the literature.
The overall sample was primarily male, White/non-minority, and on average age 14 at the time of their first juvenile offense. The AOD probationer group had a juvenile court history that was statistically significantly more severe than the JDC group (see Table 1), including younger age at first offense (AOD probationers M = 14.3, JDC M = 14.6), more prior AOD offenses (AOD probationers M = 2.9, JDC M = 2.2), and more prior delinquency offenses (AOD probationers M = 4.5, JDC M = 2.4).
Group characteristics also varied widely among the four JDCs. Although all four JDCs screened and accepted youth who had a recent AOD offense, JDC “A” accepted youth who had more severe court histories (M = 6.6 prior offenses) and who were primarily also on probation (89%); JDC “B” was considered an earlier intervention and targeted youth with minor court histories (M = 2.5 prior offenses; M = 15.3 years at first offense). As shown in Table 1, JDCs “A” and “D” typically selected youth with more severe histories, while JDCs “B” and “C” targeted youth earlier in their delinquency trajectories.
Table 1. Sample Characteristics
|JDC A||JDC B||JDC C||JDC D||JDC
Age at Start
|16.7 (1.0)||16.4 (1.0)||16.1 (1.2)||16.6 (0.9)||16.5 (1.0)||16.5 (1.2)|
Age at First Offense*
|13.8 (2.0)||15.3 (1.7)||14.7 (2.0)||14.5 (2.0)||14.6 (2.0)||14.3 (1.9)|
Age at First AOD Offense*
|15.6 (1.5)||15.9 (1.1)||15.6 (1.2)||15.9 (1.1)||15.8 (1.2)||15.7 (1.3)|
AOD Prior Offenses*
|2.4 (2.0)||1.7 (0.9)||1.9 (1.0)||2.8 (1.6)||2.2 (1.5)||2.9 (1.7)|
Delinquency Prior Offenses*
|4.2 (4.1)||0.8 (1.4)||1.3 (2.4)||3.1 (3.9)||2.4 (3.4)||4.5 (4.1)|
Total Prior Offenses*
|6.6 (4.5)||2.5 (1.7)||3.2 (2.7)||5.8 (4.2)||4.6 (3.9)||7.3 (4.4)|
On Probation during JDC (%)
Table 2. JDC Characteristics
|JDC A||JDC B||JDC C||JDC D|
|Average Youth per Year||31||53||22||60|
|Referral Source||Probation officers||Probation officers||Judge and other sources||Probation officers|
|Has Participant Handbook?||Yes||Yes||Yes||No|
|Number of Phases||4||3||No phases||4|
|Treatment Modality||Individual & group||Individual & group||Primarily group||Individual & group|
|Treatment Intensity||30-day social detox followed by outpatient||Primarily outpatient w/ IOP* as needed||Outpatient||Outpatient & IOP*|
|Frequency of Random Drug Testing||Varies by phase: 3 x per week in phase 1 to 1 x per week in phase 4||Varies by priority assignment: low = 2 x per month to high = 5-6 times per month||3 x per month||Varies by priority assignment: low = 4-6 x per month to high = 10-12 x per month|
|Judicial Hearing Frequency||Every other week||Bi-monthly||Once a month
|4 courtrooms: 2 meet every other week, 2 meet once per month|
Intervention Participation Details and Exit Status
The four JDCs also varied significantly in regard to program structure, services, and length; non-compliance; and graduation rates. Table 2 presents the program characteristics provided to the researchers by each JDC. In accordance with their acceptance of more severe youth, JDCs “A” and “D” reported more phases, more frequent drug testing, and more intense treatment than the other two JDCs. For example, JDC “A,” which accepted youth with the youngest average age at first offense and the second highest average of AOD priors (see Table 1), reported the most intensive treatment. In JDC “A,” all youth were required to participate in a 30-day social detoxification program at the start of JDC, followed by outpatient treatment for the remainder of the program (see Table 2). JDC “D” reported that the majority of their youth participated in both outpatient and intensive outpatient (IOP) treatment. JDC “B” reported that IOP was available “as needed,” but was not utilized by the majority of participants. At JDC “C,” only outpatient treatment was available locally during the study period.
JDC “A” had the longest average length (M = 289 days), the greatest number of non-compliance events (M = 1.4, contempt and/or violations), and the lowest graduation rate (49%; see Table 3), reflecting their target population as the most severe of the four JDCs. Across all of the JDCs combined, the average time in the program was 249 days and the overall graduation rate was 61%. Average time on probation for the AOD comparison group was 268 days.
Table 3. Participation Details and Exit Status
|JDC A||JDC B||JDC C||JDC D||JDC
|Days in Probation||268 (164)|
|Days in JDC^||289 (145)||227 (130)||200 (123)||267 (113)||249 (130)|
For Graduated Youth^
|16.7 (1.0)||16.4 (1.0)||16.1 (1.2)||16.6 (0.9)||16.5 (1.0)||16.5 (1.2)|
For Terminated Youth
Contempt and Violations^*
|13.8 (2.0)||15.3 (1.7)||14.7 (2.0)||14.5 (2.0)||14.6 (2.0)||14.3 (1.9)|
|15.6 (1.5)||15.9 (1.1)||15.6 (1.2)||15.9 (1.1)||15.8 (1.2)||15.7 (1.3)|
Alcohol and Other Drug Recidivism
To test the first hypothesis (JDC youth will have significantly fewer post-program AOD offenses than similar youth who received probation), we compared JDC and AOD probationer youth on AOD offending at three through 30 months post-exit. As shown in Figure 1, the two groups did not differ statistically significantly from each other on AOD recidivism at any point during the follow-up period. At 12 months post-exit, 29% of both groups had a new AOD recidivism event (juvenile referral and/or adult arrest), while 42% of JDC and 39% of AOD probationers had AOD recidivism at 30 months. Recidivism rates at each follow-up period were calculated only for those who completed the full follow-up period. All participants in the study had 12 months post-exit follow-up, 99.8% had 18 months, 96% had 24 months, and 90.1% had 30 months follow-up.
Post-program recidivism varied significantly by JDC location. The two JDCs that served youth with more severe juvenile court histories had higher AOD recidivism rates (see Figure 2). JDC “D” had the highest AOD recidivism rate, at 38% at 12 months and 50% at 30 months. JDC “D” also had the highest average number of AOD offenses pre-entry (M = 2.8) of the four JDCs (see Table 1). JDC “C” had the lowest AOD recidivism rate, at 18% at 12 months and 32% at 30 months post-exit, and the second lowest pre-JDC AOD offense average (M = 1.9).
Figure 1. Juvenile and Adult AOD Recidivism Post-Exit – JDC and AOD Probationers
Figure 2. Juvenile and Adult AOD Recidivism Post-Exit – Four JDCs
Delinquency and Criminal Recidivism
To test the second hypothesis (JDC youth will have significantly fewer post-program delinquency and criminal offenses), JDC and AOD probationer youth were compared on delinquency/criminal offending at three through 30 months post-exit. At all follow-up points, JDC youth had statistically significantly fewer delinquency/criminal offenses than AOD probationers, with the difference between the groups getting larger with longer follow-up periods (see Figure 3). For example, at 12 months post-exit, 24% of JDC participants had a new delinquency or criminal recidivism event compared with 35% of AOD probationers. At 30 months post-exit, 34% of JDC participants had a new offense, versus 48% of AOD probationers. Again, recidivism was calculated for each time period only for those participants who had the full duration of follow-up.
The four JDCs differed statistically significantly on delinquency/criminal recidivism in the 30 months post-exit (see Figure 4). JDCs “A” and “D” had the highest delinquency/criminal recidivism rates, at over 40% at 30 months post-exit, while JDCs “B” and “C” had the lowest (25% and 19%, respectively). As shown previously in Table 1, JDCs “B” and “C” also had the fewest pre-JDC delinquency offenses on average.
Figure 3. Juvenile and Adult Delinquency/Criminal Recidivism Post-Exit – JDC and AOD Probationers
Figure 4. Juvenile and Adult Delinquency/Criminal Recidivism Post-Exit – Four JDCs
Factors Related to Recidivism
AOD recidivism. Three demographic factors (gender, minority status, age at intervention start), three court history factors (age at first offense, number of AOD prior offenses, number of delinquency prior offenses), and one program compliance factor (number of contempt and violations during intervention) were examined in relation to having an AOD recidivism event post-exit (JDC or probation). The four factors that were statistically significantly related to AOD recidivism in the bivariate analysis (gender, AOD prior offenses, delinquency prior offenses, contempt/violations) and group membership (JDC versus probation) were loaded into a logistic regression with AOD recidivism as the dependent variable. In the first logistic regression, the factor of delinquency prior offenses failed to reach statistical significance and was removed from the model. The final model was statistically significant (see Table 4) and did not depart significantly from the ideal (Hosmer & Lemeshow χ2 = 6.68, p > .05); however, the estimated explained variance in AOD recidivism was very low, at approximately 6% (Nagelkerke R2 = 0.06). In the final model (as shown in Table 4), males, youth with more AOD priors before JDC/probation, and youth with more contempt/violations during JDC/probation were more likely to have a new AOD offense after leaving their program. After controlling for these significant factors, group membership (JDC versus probation) was not statistically significantly related to AOD recidivism. Put another way, after controlling for other significant factors, there were no differences between JDC youth and probationers on post-program AOD recidivism.
Table 4. Logistic Regression of Predictors of AOD Recidivism – Final Model
|Gender (female = 0, male = 1)||.79**||2.2|
|AOD Prior Offenses||.09*||1.1|
|Contempt and Violations||.13**||1.1|
|Group (probation = 0, JDC = 1)||.23||--|
|Model chi square (DF)||53.78**||(4)|
Delinquency/criminal recidivism. The same seven factors that we examined in relation to AOD post-program recidivism, we also examined in relation to delinquency/criminal recidivism in bivariate tests. Six of the seven factors (all except AOD priors) were significantly related to delinquency/criminal recidivism post-exit from JDC/probation and were loaded into a logistic regression. After the first model, we removed three factors (minority status, age at intervention start, delinquency priors) from the model because they failed to reach statistical significance. The final model was statistically significant (see Table 5) and did not depart significantly from the ideal (Hosmer & Lemeshow χ2 = 7.93, p > .05); however, the estimated explained variance in delinquency/criminal recidivism was low, at approximately 10% (Nagelkerke R2 = 0.10). In the final model (see Table 5), males, youth who were younger at the time of their first offense, and youth with more contempt/violations during JDC/probation were more likely to have a new delinquency/criminal offense after leaving their program. After controlling for these three significant factors, JDC youth were statistically significantly less likely than probationers to have a delinquency/criminal recidivism event. JDC youth were about 30% less likely than probationers to have a new delinquency/criminal offense post-program, even after controlling for other significant factors.
Table 5. Logistic Regression of Predictors of Delinquency/Criminal Recidivism – Final Model
|Gender (female = 0, male = 1)||.75**||2.1|
|AOD Prior Offenses||-.12**||0.9|
|Contempt and Violations||.17**||1.2|
|Group (probation = 0, JDC = 1)||-.39**||0.7|
|Model chi square (DF)||82.91**||(4)|
The present study contributes to the literature on JDC effectiveness by demonstrating that JDCs can have a positive impact on delinquency/criminal offending, even when tracked into adulthood. However, AOD recidivism was not significantly different between JDC and comparison probationer youth. An additional key finding was the relationship between compliance during the program and long-term recidivism.
Hypothesis 1: AOD recidivism. The first hypothesis was not supported in the present study. JDC youth were found to be just as likely to recidivate on AOD as the comparison group of probationers. This finding may be attributed to the fact that the probationer youth also received substance abuse treatment, some from the same providers as JDC youth. Both groups received some level of AOD treatment, which may have reduced the effect of JDC on AOD recidivism. This finding is similar to some past studies of JDCs that have noted no effects (or negative effect) of JDCs on substance abuse behaviors (Gilmore et al., 2005; Rodriguez & Webb, 2004). The results from these studies may reveal how the drug court model still struggles to address adolescent developmental issues, such as juvenile independence, maturation, and peer pressure (Belenko & Dembo, 2003).
On the other hand, the present study’s null findings on AOD recidivism were contrary to the positive results of the study by Henggeler and colleagues (2006) in which JDC participants were found to have less AOD use than non-JDC participants. When comparing three different JDC groups, Henggeler et al. found that participants who received evidence-based substance abuse treatment were more likely to have sustainable treatment effects (i.e., lower AOD use). This is similar to what Gottfredson et al. (2003) found among adults in drug courts. The researchers found that drug court participants who attended substance abuse treatment in qualified organizations were more likely to have lower recidivism rates. These findings may lead to the conclusion that the treatment component of drug courts is not identical in its implementation, dosage, and/or effect among drug courts, including JDCs. The present study did not measure treatment dosage or fidelity of implementation. However, the information provided by the four JDCs on their program and treatment structure indicates that they varied widely in the modality and intensity of substance abuse treatment.
Hypothesis 2: Delinquency/criminal recidivism. The second hypothesis, that JDC youth would have less delinquency/criminal recidivism than probationers, was supported by the analyses. In both the bivariate analyses that compared JDC and probationers on post-program delinquency/criminal recidivism and the logistic regression that examined the effect of group membership on delinquency/criminal recidivism, after controlling for significant individual factors, JDC youth were significantly less likely than AOD probationers to recidivate. The logistic regression finding is important, as significant pre-existing differences were noted between the JDC and comparison probationer groups. These findings are consistent with the studies by Rodriguez and Webb (2004) and Henggeler et al. (2006), both of which found reduced delinquent behavior for JDC participants. However, the present study’s findings did reveal long-term reduction of re-arrests (30 months), contrary to Gilmore et al. (2005). Similar to Pitts (2006), the present study combined juvenile and adult re-arrest recidivism to reveal long-term positive effects of JDC. By combining juvenile and adult recidivism measures and allowing for a sufficient follow-up period, this study demonstrated the potential long-term impacts of juvenile drug courts.
Program compliance. Program compliance, measured as the number of contempt and violation events during participation, was significantly related to both AOD and delinquency/criminal offending following both JDC and probation participation. Each additional non-compliance event during JDC or probation was associated with a 10% increase in the likelihood of an AOD offense post-exit and a 20% increase in the likelihood of a delinquency/criminal offense post-exit. The relationship between during-program compliance and negative outcomes has been documented in previous JDC studies. Belenko (2001) noted that participants need to establish periods of abstinence during JDCs in order to be successful in the program, while Miller et al. (1998) found that treatment compliance may be one of the most important factors in determining JDC success.
Strengths and Limitations
The strengths of this study are the large sample size and its extensive follow-up period into adulthood. Few studies have combined juvenile and adult recidivism to reveal the long-term effects of JDC (Thompson, 2004; Pitts, 2006). The present study also included a comparative non-JDC group, rather than simply JDC non-graduates.
The first limitation of this study is that it did not use a randomized control design and, therefore, causal effects cannot be inferred. Individual JDC criteria for participant inclusion were not included in this study. In addition, there were significant pre-existing differences between the probation youth and JDC youth, which may have influenced the results. Multivariate statistical analyses were conducted to address those pre-existing differences.
A second, and important, limitation is that the present study did not include a process evaluation to test the fidelity of the substance abuse treatment that was provided to the JDC and probation youth. Therefore, different types of treatment could be eliciting different levels of treatment effect on youth.
Conclusion and Areas for Future Research
This study demonstrated that four varied JDC programs in Utah were effective in reducing delinquency/criminal recidivism compared with juvenile probationers, but that they did not reduce AOD recidivism. Because of a lack of process data, specifically regarding substance abuse treatment, this study could not answer the “why” or “how.”
Future research should include process evaluations to assess whether programs are implementing the JDC model faithfully and following evidence-based practices for adolescent treatment. We did not assess program implementation in the present study due to poor treatment record-keeping and lack of data-sharing between the treatment providers and individual JDCs. A better understanding of the treatment focus and fidelity may have helped to explain why AOD recidivism differences were not found between JDC and probation youth. Henggeler et al. (2006) found that the use of evidence-based practices in JDC is more likely to have sustainable positive treatment effects than interventions without such practices. Past research also found treatment quality is an important factor in reducing recidivism in adult drug courts (Gottfredson et al., 2003). Getting inside the “black box” of drug court treatment is critical to understanding whether evidence-based practices in treatment are being followed, and to reveal the parts of drug court models that are most effective (Bouffard & Taxman, 2004).
Other areas that should be investigated in the JDC process include: level of supervision; appropriate use of rewards and sanctions for compliance/noncompliance; frequency of drug testing; judicial monitoring (techniques, frequency); non-adversarial team approach; and the use of evidence-based practices in assessments and treatments (e.g., actuarial assessments, cognitive behavioral approaches, and motivational interviewing techniques). Heck (2006) provides a list of recommended data elements for JDC practitioners and researchers that covers many of these areas.
Combining the strengths of this study (long-term follow-up with both juvenile and adult recidivism) with the proposed process measures in future research would significantly increase our understanding of why, how, and what parts of the drug court model are productive. Continued research on JDCs is necessary to provide guidance in the refinement of a more effective and efficient model of JDC.
The importance of determining the effectiveness of JDCs and subsequently replicating the evidence-based treatment models cannot be overstated. Substance use and abuse among youth is common (Substance Abuse and Mental Health Services Administration (SAMHSA), 2010), and the relationship between substance use and criminal behavior has long been documented (Belenko, 2002; De Li, Priu, & MacKenzie, 2000; Harrell, 2001; Inciardi & Martin, 1997; Inciardi, Martin & Butzin, 2004; van Kammen & Loeber, 1994). In 2009, the most recent year for which data are available, 10% of youth aged 12–17 were current illicit drug users, while 13% of youth aged 14–15 and 26.3% youth aged 16–17 were underage alcohol consumers (SAMHSA, 2010). Rates of illegal drug use and alcohol consumption among youth have declined only slightly since the first survey in 2002 (SAMHSA, 2010). Substance abuse has been proven to intensify and sustain criminal activity (Inciardi et al., 2004). Specifically among juveniles, the rates of person offenses, carrying a concealed weapon, and overall offending increased with the initiation of illegal drug use (or drug dealing), while discontinuing drug use (or drug dealing) was associated with a decrease in delinquency (van Kammen & Loeber, 1994). It is because of this long-standing relationship between substance abuse and criminality that adult drug courts were developed to address the strain of drug users on the criminal justice system (National Institute of Justice (NIJ), 2006). Due to the documented success of adult drug courts, JDCs and other problem-solving courts were developed (American University, 1999; Applegate & Santana, 2000).
The body of literature supporting the effectiveness of JDCs, although lagging behind the adult drug court literature (Henggeler & Marlowe, 2010), is growing (Anspach & Ferguson, 2005; Dickie, 2000; Latessa et al., 2002; Thompson, 2001; Willard & Wright, 2005). Henggeler and Marlowe (2010) summarized the research on JDCs, noting that when evidence-based treatment is incorporated, JDCs can have a 15% to 40% reduction in substance abuse and delinquency. The current study contributes to this literature by documenting the effectiveness of four Utah JDCs in positively impacting delinquency/criminal offending into adulthood, particularly when compared with youth who are processed through the traditional juvenile probation system. Future research that can further explain the relationship between specific juvenile drug court treatment models and positive AOD and delinquency/criminal outcomes will benefit the juvenile justice system greatly. Because of the link between substance abuse and delinquency/criminal behavior, JDCs and other evidence-based treatment models that address both substance abuse and delinquent behaviors will remain important in the juvenile justice system.
About the Authors
Audrey O. Hickert, M.A., is a research analyst at the University of Utah. She has conducted program and policy evaluations in the adult and juvenile justice systems.
Erin Becker, M.C.J., is a research analyst at the Utah Criminal Justice Center, University of Utah. She has conducted evaluations in areas including adult and juvenile alternatives to incarceration, and substance abuse and mental health interventions.
Moisés Próspero, Ph.D., is a consultant providing evaluation and training services to non-profit organizations and government institutions working with youth and adults involved in the criminal and juvenile justice systems.
Kristina Moleni, M.A., is a social research assistant and Ph.D. candidate, University of Utah. She has worked as a juvenile probation officer and conducted research in the justice system.
American University (1999). Drug court activity update: Summary. Washington, DC: US Department of Justice, Office of Justice Programs, Drug Court Clearinghouse.
Anspach, D., & Ferguson, A. (2005). Part II: Outcome evaluation of Maine’s statewide juvenile drug treatment court program. Portland, ME: University of Southern Maine.
Applegate, B. K., & Santana, S. (2000). Intervening with youthful substance abusers: A preliminary analysis of a juvenile drug court. The Juvenile Justice System Journal, 21, 281-300.
Belenko, S. (2001). Research on drug courts: A critical review. 2001 update. New York: Columbia University, The National Center on Addiction and Substance Abuse.
Belenko, S. (2002). Trends in substance abuse and treatment needs among inmates, final reports (NCJ Publication No. 197073). Washington, DC: U.S. Department of Justice, National Institute of Justice.
Belenko, S., & Dembo, R. (2003). Treating adolescent substance abuse problems in the juvenile drug court. International Journal of Law and Psychiatry, 26, 87-110.
Bouffard, J., & Taxman, F. (2004). Looking inside the “black box” of drug court treatment services using direct observations. Journal of Drug Issues, 34(1), 195-218.
De Li, S., Priu, H. D., & MacKenzie, D. L. (2000). Drug involvement, lifestyles, and criminal activities among probationers. Journal of Drug Issues, 30(3), 593-620.
Dickie, J. L. (2000). Summit county juvenile court drug court evaluation report: July 1, 1999 - June 1, 2000. Akron, OH: University of Akron, Institute for Heath and Social Policy.
Fradella, H., Fischer, R., Hagan Kleinpeter, C., & Koob, J. (2009) Latino youth in the juvenile drug court of Orange County, California. Journal of Ethnicity in Criminal Justice, 7, 271-292.
Gilmore, A. S., Rodriguez, N., & Webb, V. J. (2005). Substance abuse and drug courts: The role of social bonds in juvenile drug courts. Youth Violence and Juvenile Justice, 3, 287-315.
Gottfredson, D. C., Najaka, S. S., & Kearley, B. (2003). Effectiveness of drug treatment courts: Evidence from a randomized trial. Criminology & Public Policy, 2, 171-196.
Harrell, A. (2001). Reducing drug use and crime among offenders: The impact of graduated sanctions. Journal of Drug Issues, 31(1), 207-233.
Hartmann, D. J., & Rhineberger, G. M. (2003). Evaluation of the Kalamazoo county juvenile drug treatment court program. Kalamazoo, MI: Kercher Center for Social Research, Western Michigan University.
Heck, C. (2006). Local drug court research: Navigating performance measures and process evaluations. Alexandria, VA: National Drug Court Institute. Retrieved from http://www.ndci.org/publications/publication-resources/monograph-series on January 29, 2010.
Henggeler, S. W., Halliday-Boykins, C. A., Cunningham, P. B., Randall, J., Shapiro, S. B., & Chapman, J. E. (2006). Juvenile drug court: Enhancing outcomes by integrating evidence-based treatments. Journal of Consulting & Clinical Psychology, 74(1), 42-54.
Henggeler, S. W., & Marlowe, D. B. (2010). Introduction to the special issue on juvenile drug treatment courts. Drug Court Review, 7(1), 1-9.
Hills, H., Shufelt, J., & Cocozza, J. (2009). Evidence-based practice recommendations for juvenile drug courts. Delmar, NY: National Center for Mental Health and Juvenile Justice.
Huddleston, III, C. W., Marlowe, D. B., & Casebolt, R. (2008). Painting the current picture: A national report card on drug courts and other problem-solving court programs in the United States. Alexandria, VA: National Drug Court Institute.
Inciardi, J. A., & Martin, S. S. (1997). An effective model of prison-based treatment for drug-involved offenders. Journal of Drug Issues, 27(2), 261-279.
Inciardi, J. A., Martin, S. S., & Butzin, C. A. (2004). Five-year outcomes of therapeutic community treatment of drug-involved offenders after release from prison. Crime & Delinquency, 50(1), 88-108.
Kaminer, Y. (2001). Adolescent substance abuse treatment: Where do we go from here? Psychiatric Services, 52(2), 147-149.
Latessa, E. J., Shaffer, D. K., & Lowenkamp, C. (2002). Outcome evaluations of Ohio’s drug court efforts. Cincinnati, OH: Center for Criminal Justice Research.
Marlowe, D. (2010). Research update on juvenile drug treatment courts. Alexandria, VA: National Association of Drug Court Professionals. Retrieved from http://www.ndci.org/research on May 3, 2011.
Miller, M. L., Scocas, E. A., & O’Connell, J. P. (1998). Evaluation of the juvenile drug court diversion program. Dover, DE: Statistical Analysis Center.
National Association of Drug Court Professionals (NADCP) (1997). Defining drug courts: The key components. Alexandria, VA.
National Institute of Justice (NIJ). (2006). Drug Courts: The Second Decade (NCJ Publication No. 211081). Washington, DC: U.S. Department of Justice, Office of Justice Programs.
O’Connell, J., Nestlerode, E., & Miller, M. (1999). Evaluation of the Delaware juvenile drug court diversion program. Dover, DE: State of Delaware Executive Department, Criminal Justice Council’s Statistical Analysis Center.
Office of Justice Programs, Drug Courts Program Office (1997). Defining drug courts: The key components. Washington, DC: ONDCP.
Pitts, W. J. (2006). Measuring recidivism in a juvenile drug court: Systematic outcome study of a juvenile drug court using historical information. Southwest Journal of Criminal Justice, 3(1), 17-34.
Rodriguez, N., & Webb, V. J. (2004). Multiple measures of juvenile drug court effectiveness: Results of a quasi-experimental design. Crime & Delinquency, 50, 292-314.
Rossman, S. B., Butts, J. A., Roman, J., DeStefano, C., & White, R. (2004). What juvenile drug courts do and how they do it. In J. A. Butts & J. Roman (Eds.), Juvenile drug courts and teen substance abuse (pp. 55-106). Washington DC: Urban Institute Press.
Ruiz, B., Stevens, S., Fuhriman, J., Bogart, J., & Korchmaros, J. (2009). A juvenile drug court model in southern Arizona: Substance abuse, delinquency, and sexual risk outcomes by gender and race/ethnicity. Journal of Offender Rehabilitation, 48, 416-438.
Sloan, J., Smykla, J., & Rush, J. (2004). Do juvenile drug courts reduce recidivism?: Outcomes of drug court and an adolescent substance abuse program. American Journal of Criminal Justice, 29, 95-115.
Substance Abuse and Mental Health Services Administration (SAMHSA). (2010). Results from the 2009 National Survey on Drug Use and Health: Volume I. Summary of National Findings (Office of Applied Studies, NSDUH Series H-38A, HHS Publication No. SMA 10-4856 Findings). Rockville, MD.
Thompson, K. (2001). A preliminary outcome evaluation of North Dakota’s juvenile drug court—Recidivism analysis. Fargo, ND: North Dakota State University.
Thompson, K. M. (2004). An adult recidivism outcome evaluation of North Dakota’s juvenile drug court. Fargo, ND: North Dakota State University.
U.S. Government Accountability Office. (2005). Adult drug courts: Evidence indicates recidivism reductions & mixed results for other outcomes. Washington, DC.
van Kammen, W. B., & Loeber, R. (1994). Are fluctuations in delinquent activities related to the onset and offset in juvenile illegal drug use and drug dealing? The Journal of Drug Issues, 24(1), 9-24.
Willard, T., & Wright, S. (2005). Virginia drug treatment court programs: Executive summary. Richmond, VA: Transformation System, Inc.