Juvenile Drug Court: Enhancing Outcomes by Integrating
Scott W. Henggeler, Colleen A. Halliday-Boykins, Phillippe B. Cunningham, Jeff Randall, Steven B. Shapiro,
and Jason E. Chapman
Medical University of South Carolina
Evaluated the effectiveness of juvenile drug court for 161 juvenile offenders meeting diagnostic criteria
for substance abuse or dependence and determined whether the integration of evidence-based practices
enhanced the outcomes of juvenile drug court. Over a 1-year period, a four-condition randomized design
evaluated outcomes for family court with usual community services, drug court with usual community
services, drug court with multisystemic therapy, and drug court with multisystemic therapy enhanced
with contingency management for adolescent substance use, criminal behavior, symptomatology, and
days in out-of-home placement. In general, findings supported the view that drug court was more
effective than family court services in decreasing rates of adolescent substance use and criminal behavior.
Possibly due to the greatly increased surveillance of youths in drug court, however, these relative
reductions in antisocial behavior did not translate to corresponding decreases in rearrest or incarceration.
In addition, findings supported the view that the use of evidence-based treatments within the drug court
context improved youth substance-related outcomes. Clinical and policy implications of these findings
Keywords: juvenile drug court, multisystemic therapy, contingency management, delinquency, substance
Drug- and alcohol-abusing and dependent delinquents represent
a large and greatly underserved population that is at high risk of
presenting significant deleterious outcomes and long-term costs
for themselves, their families and communities, and society
(Belenko & Dembo, 2003). As described by recent reviewers
(Belenko & Logan, 2003; Cooper, 2002), juvenile drug courts
began emerging in the 1980s to address these problems, and 268
juvenile drug courts were in operation in December 2003
(Huddleston, Freeman-Wilson, & Boone, 2004). Although juve-
nile drug courts have continued to proliferate, evaluation of their
capacity to reduce offender substance use and criminal activity has
lagged. Indeed, only one modest (N ? 40) randomized trial of a
juvenile drug court has been completed, and that study is unpub-
lished (Dickie, 2000). The purpose of this article is to describe key
1-year outcomes from a randomized trial of juvenile drug court
that included four treatment conditions and addressed two impor-
tant gaps in the extant literature on juvenile drug courts.
The first gap pertains to the effectiveness of juvenile drug court,
per se. Although the effectiveness of juvenile drug courts has
received little empirical attention, several factors suggest that such
courts might be more effective than traditional justice services at
reducing adolescent substance use. First, the related literature on
adult drug courts (Belenko, 2001) suggests that close collaboration
of criminal justice professionals and alcohol and drug treatment
providers under a drug court rubric increases retention in treat-
ment, provides closer and more comprehensive supervision, re-
duces substance use, and produces short-term cost savings. Sec-
ond, the treatment services promoted by at least some juvenile
drug courts (see, e.g., Parnham & Wright, 1998) are intended to
address an array of the correlates of adolescent substance abuse at
the individual (e.g., development of drug refusal skills), family
(e.g., improving parental discipline), school (e.g., enhancing
school performance), and community (e.g., involvement in proso-
cial activities) levels. Addressing the known determinants of clin-
ical problems comprehensively likely increases the probability of
Scott W. Henggeler, Colleen A. Halliday-Boykins, Phillippe B. Cun-
ningham, Jeff Randall, Steven B. Shapiro, and Jason E. Chapman, Family
Services Research Center, Department of Psychiatry and Behavioral Sci-
ences, Medical University of South Carolina.
Scott W. Henggeler is a board member and stockholder of MST Ser-
vices, LLC, the Medical University of South Carolina-licensed organiza-
tion that provides training in multisystemic therapy.
This article was supported by Grants R01AA122202 from the National
Institute on Alcoholism and Alcohol Abuse and Substance Abuse and
Mental Health Services Administration/Center for Substance Abuse Treat-
ment, R01ZDA from the National Institute on Drug Abuse, and
P01HS10871 from the Agency for Healthcare Research & Quality.
We sincerely thank the clinical and research teams, including Anita
Gordon, Darlene Hines, Marcella Hamilton, Jennifer Turner, Jacqueline
Conyers, Shanta Barron, Jennifer Powers, Michael Alexander, and Lynn
West. We also thank professionals at the Charleston County Juvenile Drug
Court—Judge Charlie Segars Andrews, Judge Joyclin Cates, and Drug
Court Coordinator Julius Scott. Finally, we thank the Charleston office of
the South Carolina Department of Juvenile Justice, former Director Diana
Vaughan, and current Director Ashley Standafer for their support in facil-
itating the success of this project.
Correspondence concerning this article should be addressed to Scott W.
Henggeler, Family Services Research Center, Department of Psychiatry
and Behavioral Sciences, Medical University of South Carolina, 67 Pres-
ident Street, Suite CPP, P.O. Box 250861, Charleston, South Carolina,
29425. E-mail: email@example.com
Journal of Consulting and Clinical Psychology
2006, Vol. 74, No. 1, 42–54
Copyright 2006 by the American Psychological Association
attaining favorable outcomes. Third, several of the defining char-
acteristics of drug courts (Belenko, 2001; National Association of
Drug Court Professionals, Drug Court Standards Committee,
1997) are consistent with well-documented principles of behavior
change. For example, adolescent drug use is monitored closely
through frequent urine screens and sets of sanctions, and rewards
are applied by the judge consistently and swiftly based on the
results of the screens. Such monitoring and consistent application
of consequences might decrease juvenile drug use, regardless of
any treatment interventions conducted outside the court. This
possibility is addressed by including drug court with community
services (DC) and usual family court with community services
(FC) as two of the four intervention conditions.
The second gap addresses an area that has been largely ne-
glected in the drug court literature—the integration of evidence-
based clinical services into the treatment protocols for youths in
drug court (Belenko & Logan, 2003). As currently configured,
judges and court personnel generally assume that alcohol and drug
treatment available in the community is effective. Yet, as noted by
the Institute of Medicine (1998) and leading policy analysts (e.g.,
McLellan, Carise, & Kleber, 2003), evidence-based substance
abuse treatments are rarely used in community substance abuse
practice. Hence, this study also examined the effects of incorpo-
rating a well-validated family- and community-based treatment as
the community intervention component of the drug court process.
Specifically, drug court with multisystemic therapy (DC/MST;
Henggeler, Schoenwald, Borduin, Rowland, & Cunningham,
1998) was included as the third treatment condition. MST is one of
the few adolescent treatments cited by the National Institute on
Drug Abuse (NIDA; 1999) as evidence based and has achieved
favorable long-term reductions in drug use and criminal behavior
among substance using and abusing juvenile offenders (Henggeler,
Clingempeel, Brondino, & Pickrel, 2002; Schaeffer & Borduin,
2005). Thus, the third treatment condition integrated MST into
juvenile drug court to determine whether drug court outcomes
would be enhanced.
Finally, a third aim of this study was to evaluate whether the
integration of contingency management (CM; Petry, 2000) into the
more ecologically oriented MST treatment protocol would im-
prove substance use outcomes for MST. Although MST substance-
related outcomes with substance abusing juvenile offenders have
been generally favorable (Henggeler, Pickrel, & Brondino, 1999),
the bulk of MST therapeutic efforts have emphasized change in
youths’ family, peer, and school relations that were associated with
substance use, rather than focusing on the substance use, per se.
Perhaps alcohol and drug outcomes for MST would be enhanced if
substance use behavior was a specific and intensive focus of this
family intervention model. An examination of the outcome liter-
ature in the area of adolescent substance abuse (e.g., Liddle &
Dakof, 1995; Waldron, 1997) combined with knowledge gained
from the corresponding adult literature (e.g., Institute of Medicine,
1998; NIDA, 1999) suggests that a behavior therapy approach that
includes close monitoring of substance use can be reasonably
regarded as effective. Budney and Higgins (1998), for example,
developed an effective (NIDA, 1999) behavioral treatment proto-
col for adult drug abusers that includes a voucher system linked
with results from frequent urine analyses, functional analysis of
drug use, and development of drug refusal skills. A variation of
this approach developed by Donohue and Azrin (2001), CM, has
produced promising results for adolescents. Drug court with MST
enhanced with CM (DC/MST/CM), consequently, represented the
fourth treatment condition.
In summary, this study used a randomized design with intent-
to-treat analyses to evaluate key 1-year outcomes for substance
use, criminal behavior, incarceration, and symptomatology for 161
substance-abusing and dependent juvenile offenders assigned to
one of four treatment conditions (FC, DC, DC/MST, or DC/MST/
CM). In general, primary aims were to determine whether (a)
juvenile drug court improved outcomes in comparison with family
court, (b) integrating evidence-based interventions into juvenile
drug court improved juvenile drug court outcomes, and (c) inte-
grating CM into MST enhanced standard MST outcomes for
Design and Procedures
A 4 (treatment type: FC, DC, DC/MST, DC/MST/CM) ? 3 (time:
pretreatment, 4 months, 12 months) factorial design, with random assign-
ment of youths and families to treatment conditions, was used. Henggeler
and Randall (2000) discussed strategies used to gain the collaboration of
juvenile justice stakeholders in conducting randomized trials. Assessments
were conducted with each youth and his or her caregiver at three points in
time: within 72 hours of recruitment into the study (pretreatment; T1); 4
months postrecruitment (T2), corresponding to the average end of MST
treatment; and 12 months postrecruitment (T3), corresponding to the av-
erage end of drug court. Research assistants administered the assessment
battery to families in their homes or in detention facilities, for youths in
juvenile justice custody. Families were paid $75 for each completed
assessment as compensation for their time.
Participants were one hundred sixty-one 12- to 17-year-old adolescents
recruited from the Department of Juvenile Justice (DJJ) in Charleston
County, South Carolina who met Diagnostic and Statistical Manual of
Mental Disorders (4th ed.; DSM–IV; American Psychiatric Association,
1994) diagnostic criteria for alcohol or drug abuse or dependence. Inclu-
sion criteria were (a) age of 12–17 years, (b) diagnosis of psychoactive
substance abuse or dependence, (c) formal or informal probationary status,
(d) residence in Charleston County, and (e) residence with at least one
parent figure. Adolescents were excluded if they were already formally
involved in substance abuse treatment or if a family member had already
received MST treatment. To enhance the external validity of findings, no
youths were excluded based on preexisting mental health, physical health,
or intellectual difficulties.
Recruitment and Randomization
All cases entering the DJJ as new referrals or repeat offenders from
January 2000 to June 2003 (N ? 2,123) were screened by probation staff
for possible alcohol or drug abuse. If substance abuse was suspected, and
the youth and family met the other inclusion criteria, the Structured
Clinical Interview for DSM–IV (First, Spitzer, Gibbon, & Williams, 2001)
was administered to both the caregiver and the youth. All inclusion criteria
were met by 165 youths and their families. These families were recruited
for study participation by a researcher who obtained informed consent and
assent and opened a sealed envelope that informed the family of the
condition to which they were assigned. One hundred sixty-one families
agreed to participate (98% recruitment rate). All forms and consent pro-
JUVENILE DRUG COURT OUTCOMES
cedures were approved by the institutional review board at the Medical
University of South Carolina.
Please note that all youths were supervised via probation or parole,
which included a minimum of approximately 2 hr of juvenile justice
contact per month for approximately 1 year. The four intervention condi-
tions were defined as follows.
Youths in the FC condition appeared before a family court judge
on average once or twice per year. These youths were referred by their DJJ
intake representative to receive outpatient alcohol and drug abuse services
from the local state-funded alcohol and drug treatment facility. Youths
were directed to attend group treatment for 1.5 hours, 4 days a week for 12
weeks. The group focused on risk reduction, peer influence, conflict
resolution, and anger management. Youths concurrently received 6 weeks
of a group treatment pertaining to drug selling behavior; 12 weeks of
individual sessions; and 1.5 hr, 2 days a week, for 12 weeks of family
group therapy. The theoretical orientations of these groups were cognitive–
behavioral and systems theory. The interventions were not manual driven,
and selection of material was left to the therapist’s discretion. Services
were office based, with little community outreach. Less intensive services
were offered if needed following the completion of the 12-week program.
In this condition, the same aforementioned community services
were provided under the drug court rubric. Drug court was held once a
week, and procedures were typical of those provided in juvenile drug
courts nationally. Prior to court, urine drug screens were obtained, and the
youth, caregiver, and substance abuse counselor completed forms concern-
ing the youth’s behaviors in four domains: drug use, compliance with rules
at home, school behavior, and attendance and participation in community
services treatment groups. Youth and caregiver came before the judge and
read the forms. If the youth’s drug screen was dirty or if negative behavior
was reported in any of the other domains, sanctions could be imposed.
Sanctions varied in intensity and ranged from community service to de-
tention. Clean drug screens and positive behavior in the behavioral do-
mains resulted in the judge providing the youth with incentives that also
varied in level of intensity (e.g., meal at fast food restaurant to tickets for
a sporting event). Drug court consisted of three levels (i.e., weekly,
biweekly, and monthly required attendance in court), with graduation from
one level to the next dependent on clean drug screens and acceptable
behavior in other areas. The standard duration of juvenile drug court
was 12 months.
In this condition, MST was provided under the drug court
rubric. MST is a manualized (e.g., Henggeler et al., 1998) evidence-based
treatment that includes several key features. (a) A comprehensive set of
identified risk factors (e.g., across individual, family, peer, school, and
neighborhood domains) is targeted through individualized interventions.
(b) These interventions integrate empirically based clinical techniques
(e.g., from the cognitive–behavioral and behavioral therapies), which have
historically focused on a limited aspect of the youth’s social ecology (e.g.,
individual youth, family), into a broad-based ecological framework that
addresses relevant factors across family, peer, school, and community
contexts. (c) Interventions focus on promoting behavioral changes in the
youth’s natural ecology by empowering caregivers with skills and re-
sources to address difficulties that will arise in raising adolescents and
empowering youths to cope with family, peer, school, and neighborhood
problems. (d) Therapist efforts to achieve desired clinical outcomes are
supported through an intensive, sustained, and standardized quality assur-
ance system. (e) Services are delivered via a home-based model, which
facilitates high engagement and low dropout rates. In this study, therapists
carried low caseloads of four to five families per clinician, which allowed
intensive services to be provided to each family (2–15 hours per week,
depending upon the clinical need). Clinical services were delivered in
home, school, and/or neighborhood settings at times convenient to the
family. Therapists were available to respond to clinical problems 24 hours
a day, 7 days a week.
MST and CM, as used by Budney and Higgins (1998),
have several important similarities that support the viability of their inte-
gration: a strong commitment to empirical validation, the use of functional
analyses to identify the proximal determinants of identified problems and
guide initial intervention design, the use of pragmatic and goal-oriented
intervention strategies, a broad-based view of interventions that specifi-
cally address risk and protective factors in the patient’s social ecology, and
programmatic commitment to remove barriers to service access. As spec-
ified in a treatment manual (Cunningham et al., 2003), three components of
CM were specifically added to MST in this study: a voucher system that
rewarded clean substance screens, a detailed functional analysis of drug-
use behavior that served as the basis for self-management planning, and
protocols for self-management. Throughout treatment, and consistent with
MST treatment principles, the youth’s caregivers were closely involved in
the functional analyses and subsequent design of interventions as well as in
monitoring of the self-management plans and substance outcomes of the
Six full-time master’s level therapists with degrees in social work,
psychology, or education provided MST. Three therapists formed a team
that worked in the DC/MST condition, and three formed a team that
worked in the DC/MST/CM condition. Two of the six therapists had
previous MST experience (range ? 1–3 years), three were African Amer-
ican, three were European American, and all were female. Therapists
ranged in age from 25 to 50 years and had an average of 5 years of
postmaster’s degree experience (range ? 1–11 years). Only one therapist
left the project during the course of the study—during the last quarter of
Community-based treatment in the FC and DC conditions was provided
by 10 therapists employed by the local state-funded agency mandated to
provide substance abuse services in Charleston County. Eight of the
therapists had master’s degrees in social work, psychology, or education,
and the other two therapists had bachelor’s degrees. Five of the therapists
were African American, five were European American, and six were
female. Therapists ranged in age from 28 to 59 years and had an average
of 10 years experience in practice.
Sustaining Treatment Fidelity
Therapists from both MST conditions received standard MST training
and ongoing quality assurance. Included were an initial 5-day orientation,
weekly group supervision following a specified protocol (Henggeler &
Schoenwald, 1998), individual supervision as needed, and 1.5-day quar-
terly booster training. In addition to standard MST training, DC/MST/CM
therapists also received 1-day specialized training in the CM model. Group
supervision was conducted separately for therapists in the DC/MST and
DC/MST/CM conditions and occurred for 30–90 min, twice a week. This
task- and goal-oriented supervision focused on promoting adherence to
MST treatment principles, developing solutions to difficult clinical prob-
lems, and designing plans to overcome any barriers to obtaining strong
treatment adherence and favorable outcomes. DC/MST/CM therapists were
also required to specifically review their implementation of CM compo-
nents for each youth. Quarterly booster trainings were provided in areas
identified by therapists and the MST supervisor as presenting difficulties in
adherence or achieving clinical outcomes.
Evaluating Treatment Fidelity
Caregiver reports of therapist behavior were gathered through monthly
phone interviews. Questionnaire items assessed therapist behavior regard-
ing implementation of MST and CM.
HENGGELER ET AL.
al., 1998) was assessed with the 26-item Likert-format Therapist Adher-
ence Measure (Henggeler & Borduin, 1992), which has been validated in
several studies (e.g., Henggeler, Melton, Brondino, Scherer, & Hanley,
1997; Huey, Henggeler, Brondino, & Pickrel, 2000). More recently, a large
and diverse sample of 1,226 families and 255 clinicians allowed a thorough
evaluation of the properties of the Therapist Adherence Measure (Schoen-
wald, Sheidow, Letourneau, & Liao, 2003). Confirmatory factor analyses
supported a single-factor solution using 15 reliable items. The factor,
labeled Therapist Adherence to MST, reflects the interrelated principles
that operationalize MST. Cronbach’s alpha for the scale was .90. Adher-
ence scores were relatively stable over the treatment episode within fam-
ilies (i.e., intraclass correlation coefficient for time of .51). Hence, scores
were averaged across time to provide one adherence score per caregiver.
The predictive validity of this score has been demonstrated for several
youth outcome measures (Schoenwald et al., 2003).
Based on a procedure using expert consensus, nine Likert format
items were developed for use in this study. These items were rated as
essential for CM, though not necessarily incompatible with MST or stan-
dard community services. Items tapped the monitoring of substance use,
the provision of consequences based on that monitoring, and the develop-
ment of problem solving strategies to attenuate future substance use. Items
included, for example, “The therapist helped my child make a list of things
that might cause him/her to use drugs or alcohol” and “The therapist tested
my child for alcohol or drug use by Breathalyzer or drug screen.” Cron-
bach’s alpha for the scale was .88.
Standard community services.
As noted previously and in contrast with
the MST conditions, community services focused on providing group
treatment for the adolescents and did not follow standardized protocols.
Therapist adherence to MST treatment principles (Henggeler et
Key youth outcomes were assessed with a multimethod and multire-
spondent measurement battery.
Alcohol and drug use.
Substance use was measured through self-
reports and biological indices. Reported substance use by the adolescent
was assessed with the Form 90 (Miller, 1991), which is an interview based
on the time line follow-back methodology originally used to quantify
specific amounts of alcohol consumed on a daily basis by individuals, but
more recently adapted to assess other drug use using similar methods. A
calendar of the previous 90 days was first used to highlight important
events and then used to record specific quantities and types of substances
consumed on each day during the period. The numbers of days were
tabulated for alcohol use, heavy alcohol use (i.e., more than four standard
drinks), marijuana use, and polydrug use. Urine drug screens for cannabis,
cocaine, and amphetamines were collected before each drug court appear-
ance for the youths in the DC, DC/MST, and DC/MST/CM conditions
using the 3-Test Integrated Cup supplied by BioTechNostix (Markham,
Ohio). The minimum detectable level for cannabis is 50 ng/?l, and the
sensitivity is 50%. Following standard protocols for the juvenile drug
court, youths with unexcused absences (e.g., runaway, did not show) and
youths that missed court because they had been recently placed in detention
were counted as having positive urine screens for cannabis. Likewise,
youths with excused absences (e.g., a GED class) were counted as having
negative (i.e., clean) drug screens. Analyses are based on the percentage of
positive screens per youth.
Criminal activity and mental health symptoms.
assessed through self-reports and arrest records, and mental health symp-
toms were assessed through youth and caregiver reports. The 47-item
Self-Report Delinquency Scale (SRD; Elliott, Ageton, Huizinga, Knowles,
& Canter, 1983) is the best validated of the self-report delinquency scales
(Henggeler, 1989). The SRD taps a broad range of criminal behavior
perpetrated during the past 90 days and includes subscales that pertain to
status offenses, general theft, and crimes against the person (e.g., assault).
Criminal behavior was
Arrests were tracked through computerized records maintained by the
South Carolina DJJ. Because youths as young as 16 years could be charged
as adults, adult criminal records were also collected from the South
Carolina Law Enforcement Division for youths 16 years of age and older.
Externalizing and internalizing symptoms were assessed by adolescent and
caregiver ratings on the 113-item Child Behavior Checklist (CBCL;
Achenbach, 1990, 1991), one of the best-validated measures of child
behavioral functioning. T scores for the total problem scale from 60 to 63
are considered to be in the borderline clinical range, and scores above 63
are in the clinical range.
Services outcomes pertained to completion of drug
court for youths in the drug court conditions (i.e., DC, DC/MST, DC/MST/
CM) and out-of-home placements for all study participants. Those youths
who formally graduated from drug court were considered to have com-
pleted drug court. The primary criteria for graduation from the 12-month
drug court program included clean urine screens for at least the past 5
weeks and regular school attendance, employment, or serious efforts to
gain employment. Secondary graduation criteria pertained to high rates of
keeping curfew and attending treatment sessions. Days in out-of-home
placement were documented through criminal justice records in the juve-
nile justice and adult correctional systems as well as with an abbreviated
version of the Service Utilization Survey (Henggeler, Pickrel, Brondino,
Ward, & Rowland, 1997). This semistructured phone survey was admin-
istered to caregivers monthly. For present purposes, days in foster care,
group homes, residential treatment centers, juvenile justice facilities, and
mental health or substance abuse inpatient facilities were aggregated from
T1 to T3 to index days in out-of-home placements.
Characteristics of Participants
Youths averaged 15.2 years of age (SD ? 1.1); 83% were male,
and 17% were female. Racial representation reflected that of
youths in the juvenile justice system in Charleston, South Carolina,
with 67% African American, 31% White, and 2% biracial. Only
15% of the youths lived with both biological or adoptive parents,
whereas 21% lived with a biological parent and another adult
caregiver, 52% lived with a single biological or adoptive parent,
and 12% lived with other relatives. Socioeconomically, median
family income was in the $10,000–$15,000 range, 38% of families
were receiving financial assistance, and the median educational
status of the primary caregivers was 12th grade. The youths
averaged 3.6 arrests (SD ? 2.5) prior to study entry. Finally, in
spite of significant behavioral and substance use problems, only
35% of the youths had received mental health or substance abuse
Regarding the juvenile offenders’ clinical presentations, 19%
met diagnostic criteria for alcohol abuse (9%) or dependence
(10%), 98% met criteria for cannabis abuse (38%) or depen-
dence (60%), 4% met criteria for cocaine abuse (3%) or dependence
(1%), and 21% were abusing or dependent on two or more of these
substances. Fifty-seven percent of the youths met diagnostic cri-
teria for at least one co-occurring psychiatric disorder. The most
prevalent externalizing disorders were conduct disorder (36%),
oppositional defiant disorder (24%), and attention-deficit disorder
of any type (9%). The most prevalent internalizing disorders were
specific phobias (14%), major depression (6%), and obsessive–
compulsive disorder (4%).
JUVENILE DRUG COURT OUTCOMES
Comparability of Groups at Pretreatment
Between-groups comparisons for each of the aforementioned
demographic and psychosocial variables were examined using
chi-square tests and analyses of variance (ANOVAs). No between-
groups differences emerged for the demographic measures. For the
psychosocial measures, significant T1 between-groups differences
emerged for reported alcohol use, F(3, 157) ? 4.26, p ? .006, and
polydrug use, F(3, 157) ? 3.81, p ? .011, on the Form 90. Youths
in the DC/MST/CM condition reported significantly more alcohol
use than did their DC and DC/MST counterparts (ps ? .038) as
well as more polydrug use than youths in the DC condition (p ?
.035) at pretreatment. Examination of individual records showed
that, by chance, the five youths in the project reporting the most
frequent use of alcohol at T1 had been randomized to the DC/
Follow-Up Rates and Missing Data
Follow-up rates and missing data varied with the nature of the
measure. Follow-up was 100% (N ? 161) for archival measures of
arrest and criminal justice placements across all treatment condi-
tions as well as for urine drug screens for youths in the three drug
court treatment conditions. Regarding self-report measures, of the
original 161 families consenting to participate, 6 (3.7%) did not
participate in the T2 assessment, and 24 (14.9%) did not partici-
pate in the T3 assessment. One hundred thirty-five families
(83.9%) participated in all three assessments. Because families
could participate in an assessment without completing all mea-
sures, the amount of missing data varies slightly from outcome to
outcome. Attrition on these measures was evenly distributed across
treatment conditions. For example, 79% of FC, 82% of DC, 76%
of DC/MST, and 86% of DC/MST/CM youths completed the
Form 90 at all assessment points. The primary reasons for research
attrition included a lack of family responsiveness when the youth
was in detention and frequent family moves. In no cases were
available data excluded from the analyses (described subse-
quently) because a participant dropped out of treatment, failed to
complete a specified number of sessions, or did not otherwise
collaborate with the requirements of the treatment condition to
which he or she was assigned. As depicted in Figure 1 and detailed
in Table 1, the archival and drug screen analyses examined out-
comes for all participants, and the analyses of self-report measures
used listwise deletion of missing data.
sures were collected per caregiver, and scores were averaged
within caregivers for the following analyses. As anticipated, the
MST treatment conditions did not differ from each other on the
MST fidelity measure. Likewise, and as expected, FC and DC
conditions did not differ from each other on the MST fidelity
measure. As also anticipated, the MST conditions differed signif-
icantly from the DC, t(103) ? 6.44, p ? .001 (one-tailed),
d ? 1.26, and FC, t(99) ? 7.28, p ? .001 (one-tailed), d ? 1.45,
conditions on MST adherence. These findings support the view
that therapists in the MST conditions were more likely to be
delivering MST than were counterparts in the DC and FC
The second set of fidelity evaluations pertained to the CM
adherence measure, and results showed clear differentiation among
the treatment conditions. As expected in light of the drug court
emphasis on substance testing and consequences, CM implemen-
tation scores were significantly higher for youths in the DC con-
dition in comparison with counterparts in the FC condition,
t(99) ? 3.04, p ? .002 (one-tailed), d ? .83. Moreover, scores
were significantly higher in the MST conditions than in the DC
condition, t(103) ? 5.85, p ? .001 (one-tailed), d ? 1.18, and CM
scores were significantly higher for youths receiving DC/MST/CM
than for youths receiving DC/MST, t(74) ? 2.09, p ? .02 (one-
tailed), d ? .48. These findings support the view that drug court
focused significantly greater attention on monitoring and conse-
quenting adolescent drug use than did family court, that MST
significantly enhanced the capability of drug court to monitor and
provide consequences for drug court, and that the integration of the
CM protocol into MST (i.e., DC/MST/CM) further enhanced this
Intensity of treatment services.
contact hours supports the fidelity of services delivered in the MST
A mean of 2.8 (SD ? 1.4) adherence mea-
A review of mean number of
Participant flow diagram.
HENGGELER ET AL.
conditions. Youths in the DC/MST and DC/MST/CM conditions
averaged 66 hr (SD ? 32 hr) and 57 hr (SD ? 30 hr), respectively,
of direct (i.e., face-to-face meetings with any family member
present) or indirect (i.e., phone contacts, school visits without
youth or caregiver present) treatment over 4 months. Across the
MST conditions, approximately half the hours were with multiple
family members exclusively, 13% involved school personnel, 5%
focused directly on peer interactions, 14% were with the youth
individually, and 19% involved other community resources (e.g.,
church, neighbors, court personnel). The intensity and systemic
emphases of these contacts is consistent with that delivered in
other MST clinical trials with juvenile offenders. Additionally, in
Outcome Measures at Each Assessment
FC DC DC/MSTDC/MST/CM
M SDM SDM SDM SD
Self-Report Delinquency Scalec
Crimes against persons
Days in placemente
n ? 33
n ? 31
n ? 29
n ? 37 1.53d*
n ? 35, 27
n ? 29
n ? 38, 29
n ? 29
n ? 42, 38
n ? 37
n ? 32
n ? 42
n ? 38
n ? 38
n ? 430.86
1.001.15 1.451.35 1.401.521.281.44
n ? 29
n ? 29
n ? 28
n ? 34 0.292.31
n ? 42
n ? 38
n ? 38
n ? 430.73
differ at p ? .05 or better. Analyses on T2 and T3 have covaried the T1 scores. FC ? family court with community services; DC ? drug court with
community services; DC/MST ? drug court with multisystemic therapy; DC/MST/CM ? drug court with multisystemic therapy and contingency
management; T1 ? pretreatment assessment; T2 ? 4-month assessment; T3 ? 12-month assessment; CBCL ? Child Behavior Checklist.
aDays of use during past 90 days. n ? 130.
bPercentage of screens positive for cannibis. T1–T2 n ? 115 and T2–T3 n ? 94. For drug screens subsample sizes, the first value refers to T1–T2, and
the second value refers to T2–T3.
cCrimes committed during past 90 days. n ? 127.
dRepeated measures analysis of variance F.
en ? 161.
fT score for caregiver reports. n ? 120.
* p ? .05. ** p ? .01. *** p ? .001.
Within an assessment session, means with no subscripts and those that share the same subscript do not differ. Means with nonoverlapping subscripts
JUVENILE DRUG COURT OUTCOMES
support of the 12-month program requirements of the juvenile drug
court, the MST therapists averaged approximately 2 hr of family
contact per month for the remainder of the time the youth was in
drug court. Unfortunately, the unreliability of the record-keeping
systems for the community services conditions (FC and DC) did
not allow an accurate determination of the number of hours of
services received. Our impressions, however, are that (a) youths in
the FC condition received fewer hours of service than did their DC
counterparts, and (b) youths in the DC condition received fewer
hours of service than did counterparts in the MST conditions.
Overview of Outcome Analyses
treat strategy described previously and were guided by the layered
strategy used by Waldron, Slesnick, Turner, Brody, and Peterson
(2001) in their similarly designed (i.e., 4 treatment conditions, 3
time points) randomized trial of treatment for adolescent substance
abuse. First, when multiple related measures were tapped (e.g.,
Form 90, SRD), a 4 (treatment condition: FC, DC, DC/MST,
DC/MST/CM) ? 3 (time: pretreatment, 4 months, 12 months)
doubly multivariate repeated measures ANOVA (RMANOVA)
was conducted. Second, if a significant interaction was observed
for the RMANOVA, 4 (treatment condition) ? 3 (time) repeated
measures ANOVAs were conducted on each measure to examine
treatment group differences in both linear and quadratic effects of
time. Quadratic effects were examined because other MST studies
(Henggeler et al., 2003) and clinical trials in children’s mental
health (e.g., Weersing & Weisz, 2002) have found nonlinear
changes in symptom trajectories. Third, if a significant interaction
effect was observed, simple main effects analyses were conducted
to examine the effect of time in each of the four treatment condi-
tions. Fourth, if a simple main effect for time emerged, significant
differences between time points were identified by comparing each
pair of the estimated marginal means between T1, T2, and T3
follow-up for that treatment condition. Following Waldron et al.
(2001), univariate ANOVAs separate from the aforementioned
multilayered strategy were conducted for all measures on both T2
and T3 scores, with T1 scores covaried. Significant effects for
treatment conditions were identified by examining differences
between each pair of estimated marginal means. Results of these
analyses are presented in Table 1.
Effect sizes (ESs).
Treatment ESs for outcomes measured at
T2 and T3 were calculated according to the method described in
Morris and DeShon (2002). First, the within-group ES was calcu-
lated for each condition as the difference between the T2 (or T3)
mean and the T1 mean divided by the T1 standard deviation. The
resulting ESs for DC, DC/MST, and DC/MST/CM were then
subtracted from the ES for FC to yield time-adjusted treatment
ESs. For outcomes defined by change between time points (i.e.,
arrests, urine drug screens, and days in placement), ESs were
computed as the difference between conditions at T2 (or T3)
divided by the pooled standard deviation. The ESs are presented in
Table 2 and support the following text, which describes the treat-
The outcome analyses used the intent-to-
Alcohol and Drug Use Outcomes
Time interaction effect was significant, F(24, 870) ? 1.53, p ?
As shown in Table 1, the RMANOVA Group ?
.049, as were each of the subsequent univariate linear interaction
effects for alcohol use, heavy alcohol use, marijuana use, and
polydrug use, Fs(3, 126) ? 2.66, ps ? .051. The evaluation of
simple main effects and treatment effects within the two follow-up
periods revealed an interesting pattern that is discussed by
For alcohol use, significant time simple linear and quadratic
effects were observed only for youths in the DC/MST/CM condi-
tion (Fs ? 7.19, ps ? .011). A priori comparisons showed that
self-reported alcohol use decreased significantly from T1 to T2
(p ? .009) and that such decrease was sustained at the 12-month
follow-up (i.e., T1–T3; p ? .008). At T2, controlling for T1 scores,
youths in the DC and DC/MST/CM conditions reported signifi-
cantly less alcohol use than counterparts in the FC condition (ps ?
.035). At T3, controlling for T1 scores, youths in the DC/MST/CM
condition continued to report significantly less alcohol use than
counterparts in the FC condition (p ? .013). Together, these
findings suggest that DC/MST/CM was relatively effective at
Effect Sizes for Drug Court Conditions in Comparison With
Family Court Condition
Self-Reported Delinquency Scale
Crimes against persons
Days in placement
the drug screens, which are in comparison with the DC condition. FC ?
family court with community service; DC ? drug court with community
services; DC/MST ? drug court with multisystemic therapy; DC/MST/
CM ? drug court with multisystemic therapy and contingency manage-
ment; T1 ? pretreatment assessment; T2 ? 4-month assessment; T3 ?
12-month assessment; CBCL ? Child Behavior Checklist.
All effect sizes are in comparison with the FC condition, except for
HENGGELER ET AL.
decreasing youth alcohol use, especially when compared with
youths who were not enrolled in drug court.
For heavy alcohol use, despite a significant linear treatment
effect as noted earlier, none of the groups showed a significant
time simple linear or quadratic effect. At T3, however, controlling
for T1 scores, youths in the DC/MST and DC/MST/CM conditions
reported significantly less heavy alcohol use than did their coun-
terparts in the FC condition.
For marijuana use, significant simple linear time effects were
observed for each of the four treatment conditions (Fs ? 6.62,
ps ? .015), and significant simple quadratic time effects were
observed for the DC, DC/MST, and DC/MST/CM conditions
(Fs ? 10.49, ps ? .003). In general, the quadratic effects showed
that marijuana use decreased rapidly to T2 and then either leveled
off or increased slightly to T3. A priori comparisons showed that
self-reported marijuana use decreased significantly from T1 to T2
for youths in each treatment condition (ps ? .013) and that these
decreases were sustained for youths in each treatment condition
from T1 to T3 (ps ? .015). At T2, controlling for T1 scores,
significant between-groups differences did not emerge. At T3,
however, controlling for T1 scores, youths in the DC/MST and
DC/MST/CM conditions reported less marijuana use than FC
counterparts. Together, these results suggest that youth marijuana
use decreased considerably for youths across treatment conditions
from T1 to T2 but that use of the evidence-based interventions was
more likely to sustain the decrease in marijuana use at T3.
For polydrug use, significant simple quadratic time effects were
observed for each of the three interventions involved with juvenile
drug court (Fs ? 4.69, ps ? .039), and a simple linear time effect
was observed for youths in the DC/MST/CM condition, F(1,
36) ? 6.67, p ? .014. The groups did not differ significantly at T2,
controlling for T1 scores. At T3, however, controlling for T1
scores, youths in the DC/MST and DC/MST/CM conditions re-
ported significantly less polydrug use than did FC counterparts
(ps ? .012). Together, these results suggest that drug court was
more effective than community services at decreasing offender
polydrug use and that the evidence-based practices effectively
sustained these changes to T3.
To facilitate representation of the relatively complex findings
from the Form 90, Figure 2 presents an aggregate view of the
outcomes. Scores for each of the four measures were standardized
and then averaged for each of the treatment conditions at each time
point. This figure depicts the relative advantage of the drug court
conditions over FC as well as the capacity of the evidence-based
conditions to enhance drug court outcomes.
Consistent with the between-groups differences in days of sub-
stance use described previously, the ESs provided in Table 2 show
relatively strong and positive ESs for youths in the drug court
condition and assessment time point. T1 ? pretreatment assessment; T2 ? 4-month assessment; T3 ? 12-month
assessment; FC ? family court with community services; DC ? drug court with community services;
DC/MST ? drug court with multisystemic therapy; DC/MST/CM ? drug court with multisystemic therapy and
Composite summary indices of Form 90 measures of substance use as functions of treatment
JUVENILE DRUG COURT OUTCOMES
conditions (i.e., DC, DC/MST, DC/MST/CM) in comparison with
FC counterparts, especially at T3. ESs for alcohol, heavy alcohol,
and polydrug use at T3, for example, were almost all greater
than 1.00. Moreover, the evidence-based conditions showed larger
ESs than did the DC condition on each of the substance use
measures at T3.
Drug urine screens.
A mean of 22.3 drug urine screens
(SD ? 13.4) were collected for youths across the three drug court
conditions (i.e., DC, DC/MST, DC/MST/CM). In light of the low
base rate of positive screens for cocaine (e.g., 1.3% of screens
were positive from T1 to T3) and amphetamines, analyses of the
more than 2,000 drug urine screens focus solely on cannabis use.
As shown in Table 1, for the first 4 months of drug court,
equivalent to T1–T2, a significant between-groups difference
emerged for youths’ percentage of positive drug screens, F(2,
112) ? 40.47, p ? .001. Post hoc analyses showed that youths in
the DC/MST and DC/MST/CM conditions had significantly lower
percentages of positive drug screens than did their DC counterparts
(ps ? .001; DC ? 69%, DC/MST ? 28%, DC/MST/CM ? 18%).
Similar between-groups differences emerged for youths’ percent-
age of positive drug screens collected during the time from T2 to
T3, F(2, 91) ? 14.12, p ? .001. Again, post hoc analyses showed
that youths in the DC/MST and DC/MST/CM conditions had
significantly lower percentages of positive drug screens than did
their DC counterparts (ps ? .001; DC ? 45%, DC/MST ? 7%,
DC/MST/CM ? 17%). The ESs provided in Table 2 parallel these
between-groups differences. In comparison with DC youths, coun-
terparts in the DC/MST and DC/MST/CM conditions had very
large ESs at T2 (1.38 and 2.05, respectively), and these remained
large at T3 (1.27 and .82, respectively).
Criminal Behavior and Mental Health Symptom Outcomes
significant, F(18, 738) ? 1.84, p ? .019, as shown in Table 1. A
significant univariate linear interaction effect was observed for
status offenses, F(3, 123) ? 4.35, p ? .006, and a significant
univariate quadratic interaction effect was observed for crimes
against the person, F(3, 123) ? 4.12, p ? .008. A significant
interaction effect did not emerge for general theft.
For status offenses, significant time simple linear effects
emerged for each of the drug court treatments (i.e., DC, DC/MST,
DC/MST/CM; Fs ? 6.98, ps ? .012), and a significant time
simple quadratic effect was observed for the DC/MST condition,
F(1, 28) ? 6.68, p ? .015. A priori comparisons showed that
self-reported status offenses decreased significantly from T1 to T2
for youths in the DC/MST condition (p ? .009) and from T1 to T3
for youths in each of the treatments involving drug court (ps ?
.012). At T3, controlling for T1 scores, youths in the DC, DC/
MST, and DC/MST/CM conditions reported significantly fewer
status offenses than did their FC counterparts (ps ? .004).
For crimes against the person, significant time simple linear
effects emerged for the DC and DC/MST/CM conditions
(Fs ? 9.92, ps ? .004), and a significant time simple quadratic
effect was observed for the FC condition, F(1, 31) ? 6.95, p ?
.013. A priori comparisons showed that self-reported crimes
against the person decreased significantly from T1 to T3 for youths
in the DC and DC/MST/CM conditions (ps ? .004). Accounting
for the quadratic effect, youths in the FC condition showed a
The RMANOVA Group ? Time interaction effect was
significant decrease from T1 to T2 (p ? .025) and then showed a
substantial increase in self-reported crime. At T3, controlling for
T1 scores, youths in the DC and DC/MST/CM conditions reported
significantly fewer crimes against the person than did counterparts
in the FC condition (ps ? .029).
To facilitate representation of the SRD outcomes, the general
delinquency score of the SRD (this is a summary score) is graphed
by treatment condition in Figure 3. This figure depicts the relative
advantage of the drug court conditions over FC in decreasing
reported criminal activity by T3. For example, youths in the FC
condition reported an average of about 50 crimes during the
past 90 days, whereas youths in the drug court conditions averaged
fewer than 20.
ESs shown in Table 2 reflect the between-groups differences
described previously for status offenses and crimes against per-
sons. In comparison with counterparts in the FC condition, youths
in the drug court conditions had very large ESs for status offenses
at T3 (i.e., approximately 2.00) and medium ESs for crimes against
persons—averaging approximately .50.
As shown in Table 1, analyses revealed no between-
groups differences in the average number of arrests for youths
across the treatment conditions or for the percentage of youths
arrested within the conditions from T1 to T3. Across conditions,
youths averaged 1.3 arrests (SD ? 1.4), and 62% were arrested
between T1 and T3. Reflecting the nonsignificantly higher average
number of rearrests for youths in the drug court conditions com-
pared with FC youths, the ESs in Table 2 are negative for the
youths in the drug court conditions.
Mental health symptoms.
The analysis focused on the care-
giver CBCL reports, as the T score for youth reports for the total
problem scale was almost exactly at the mean for the normative
sample. As seen in Table 1, significant effects were not observed
for maternal reports on the CBCL total problem scale. At T2,
controlling for T1 scores, youths in the DC/MST/CM condition
had significantly fewer symptoms than counterparts in the FC and
DC conditions (ps ? .032), but group means were equivalent at
T3. These differences, or lack of differences, are reflected in the
near-zero ESs for youths in the DC and MST conditions (see Table
2), whereas DC/MST/CM counterparts had ESs of .79 at T2 and
.50 at T3 in comparison with youths in the FC condition.
Service System Outcomes
Drug court completion.
more likely to have graduated from drug court than were their DC
counterparts (45% vs. 28%), ?2(1, N ? 121) ? 3.16, p ? .038.
As shown in Table 1, across treat-
ment conditions, youths spent an average of more than 2 months
(i.e., 76 days) in out-of-home placements, and 64% of these days
were in correctional settings. Although the average number of days
in placement did not differ significantly for youths across the
treatment conditions (see near-zero ESs in Table 2), the percent-
ages of youths placed differed significantly, ?2(3, N ?
161) ? 10.33, p ? .016. Youths in the FC condition had the lowest
rate of placement (55%), and youths in the DC condition had the
highest rate (87%). FC youths were significantly less likely to get
placed than their DC, ?2(1, N ? 80) ? 7.61, p ? .006, and
DC/MST/CM, ?2(1, N ? 85) ? 4.45, p ? .035, counterparts.
Hence, being in drug court, even while receiving a family based
Youths in the MST conditions were
HENGGELER ET AL.
treatment, was associated with increased use of out-of-home place-
ment. Triggers for out-of-home placement for youths in the drug
court conditions included, in order of priority, dirty drug screens,
school related behavior problems (e.g., fighting, weapons, tru-
ancy), home behavior (e.g., curfew violations, incorrigibility, run-
away), and new DJJ charges. In light of the frequent court appear-
ances of youths in the juvenile drug court conditions, judges in the
juvenile drug court condition, compared with their family court
counterparts, had much greater awareness of the occurrence of
youths’ behaviors that triggered placement.
The overriding purpose of this study was to examine the effec-
tiveness of juvenile drug court for youths meeting diagnostic
criteria for substance abuse or dependence and to determine
whether the integration of evidence-based practices enhanced the
outcomes of juvenile drug court. In general, but certainly subject to
debate, the findings support the view that drug court was more
effective than family court services in decreasing rates of adoles-
cent criminal and substance use behavior. Possibly due to the
greatly increased surveillance of youths in drug court, however,
these relative reductions in antisocial behavior did not translate to
corresponding decreases in rearrest or incarceration. A second set
of findings pertains to the capacity of the evidence-based practices,
MST and MST/CM, to enhance drug court outcomes. Here, the
findings support the view that use of evidence-based treatments
within the drug court context improved youth substance-related
outcomes. When embedded within juvenile drug court, however,
the well-documented capacity of MST to reduce rearrest rates and
out-of-home placements (Curtis, Ronan, & Borduin, 2004) did not
emerge. Findings from the different outcome domains and their
corresponding interpretations and implications are discussed next.
Arguably, the number one priority of a juvenile drug court is to
reduce adolescent substance use. In this regard, the Form 90
analyses suggest that MST/CM, and to a lesser extent MST,
combined with drug court were effective at decreasing youth
alcohol, heavy alcohol, marijuana, and polydrug use in comparison
with FC. Evidence also emerged that DC was more effective than
FC at decreasing youth alcohol and polydrug use. Moreover,
although youths in the DC/MST and DC/MST/CM conditions
frequently reported better outcomes than did FC counterparts, in
no comparison did DC/MST or DC/MST/CM produce outcomes
superior to DC for the self-report measures. In consideration of
these outcomes and the corresponding ESs, it seems reasonable to
contend that DC was more effective than FC at reducing youth
substance use, DC/MST was slightly more effective than DC, and
condition and assessment time point. T1 ? pretreatment assessment; T2 ? 4-month assessment; T3 ? 12-month
assessment; FC ? family court with community services; DC ? drug court with community services;
DC/MST ? drug court with multisystemic therapy; DC/MST/CM ? drug court with multisystemic therapy and
The general delinquency score of the Self-Report Delinquency Scale as functions of treatment
JUVENILE DRUG COURT OUTCOMES
DC/MST/CM was slightly more effective than DC/MST. This
view is supported, in part, by the findings from the drug urine
screens. Here, the MST conditions were shown to be considerably
more effective than DC at decreasing cannabis use during both the
early and latter stages of drug court. In addition, though not
significant, youths in the DC/MST/CM condition averaged fewer
positive screens than did their DC/MST counterparts (18% vs.
28%) during the initial 4 months of drug court when these treat-
ments were being delivered.
Although outcome results for criminal behavior varied by the
nature of measurement, we contend that the findings support the
capacity of juvenile drug court to decrease the criminal activity of
the participants. Self-report measures showed, in general, that
youths in the drug court conditions (i.e., DC, DC/MST, DC/MST/
CM) engaged in significantly fewer status offenses and crimes
against the person during the course of the assessment period than
did their FC counterparts (see Figure 3). On the other hand, arrest
data showed no between-groups differences. In light of the fact
that the SRD is generally regarded as one of the best-validated
measures of delinquent behavior (Howell, 2003), and the validity
of arrest is apparent, how can these findings be reconciled? Al-
though certainly not definitive, we suggest that the arrest (and
placement) rates were inflated for youths in the drug court condi-
tions by the intensive surveillance provided to these youths
through frequent judicial review during drug court sessions in
which urine screens were collected and youth behavior at home, at
school, and in the community was closely tracked. In contrast,
youths in the FC condition had rare appearances before a judge as
well as much less surveillance and accountability for their behav-
ior. The hypothesis that increased surveillance resulted in an
increased probability of being arrested, in spite of the decreased
self-reported criminal behavior for youths in the drug court con-
ditions, is supported by the literature on intensive supervision
(Altschuler, Armstrong, & MacKenzie, 1999). For example, in a
well-controlled study, Land, McCall, and Williams (1990) found
that intensive supervision of youths with prior delinquent offenses
almost doubled their rate of rearrest. In the present sample, in
which youths averaged 3.6 prior arrests and reported an average
of 40 criminal acts in the 90 days preceding the pretreatment
assessment, the increased surveillance of drug court, therefore, was
more likely to detect subsequent criminal behavior even though
such behavior was occurring at a lower rate than for FC
Youths in the drug court conditions experienced very high rates
of out-of-home placement (i.e., 87% for DC, 71% for DC/MST,
and 74% for DC/MST/CM vs. 55% for FC), although treatment
conditions did not differ for average days in placement. The high
rates of out-of-home placement for youths in drug court were most
likely the product of two circumstances. First, the intensive and
continuous surveillance experienced by drug court participants
regarding drug use and any problem behavior in the community
(e.g., dirty drug screens, school problems, problems at home,
treatment attendance), as suggested previously, likely increased
the chance that difficulties would be identified. Second, although
drug court personnel emphasized a strength-focused approach,
clear sanctions (e.g., placement) for repeated youth problems were
readily available to the court. In contrast with FC youths, who
might present significant problems over an extended period of time
without detection by juvenile justice authorities, youths in drug
court had no such grace.
In addition, the high rates of out-of-home placement for youths
in the drug court conditions raises the possibility that the favorable
results for self-reported substance use and delinquency for these
youths in comparison with FC counterparts might have resulted
from their removal from the community (i.e., such removal might
have acted as an effective punishment). To test this hypothesis,
youths in the drug court conditions (DC, DC/MST, DC/MST/CM)
were divided into two groups, those that had been sent to place-
ments and those who had not, and t tests were conducted on the
self-reported substance use (Form 90) and criminal behavior
(SRD) measures at T2 and T3. Five of the 14 comparisons showed
significant results opposite of predictions based on the aforemen-
tioned possibility (i.e., youths sent to placements reported worse
outcomes), and the other comparisons showed no between-groups
differences. These results are consistent with the fact that problem
behavior is often a precipitant of placement as well as the lack of
evidence supporting the effectiveness of placements for juvenile
offenders (U.S. Public Health Service, 2001).
The CBCL findings did not reveal treatment effects (i.e., sig-
nificant Condition ? Time interaction effect) and showed that
caregiver reported symptomatology decreased to normative levels
for youths across the four treatment conditions. Evidence emerged,
however, that youths in the DC/MST/CM condition evidenced a
more rapid short-term decrease in symptoms than did FC and DC
counterparts but that symptom levels were decreased and similar
for all groups at T3. This differential trajectory of symptom change
is similar to that observed in an MST trial with youths presenting
psychiatric crises (Henggeler et al., 2003) and is consistent with
the pattern described by Weersing and Weisz (2002) in their study
comparing the effectiveness of community psychotherapy for de-
pressed youths with findings from cognitive–behavioral efficacy
trials. That is, the evidence-based treatments (i.e., MST and
cognitive–behavioral treatment) produced more rapid symptom
reduction, but symptom scores converged with comparison and
community samples by about 12 months posttreatment.
Clinical and Policy Implications
The findings have several clinical and policy implications. First,
the generally favorable outcomes for the DC condition support the
viability of the juvenile drug court model. Such support is heart-
ening in light of the considerable resources that federal, state, and
local entities have devoted to the development and implementation
of juvenile drug courts. Second, the findings support the capacity
of evidence-based treatments of adolescent substance abuse to
enhance juvenile drug court outcomes pertaining to substance use.
A key issue in this regard, however, is whether such enhanced
substance-related outcomes are cost-effective—worth the expense
of integrating an evidence-based treatment into drug court. An
economic evaluation of this project is currently being conducted to
address this issue. Third, across published MST outcome studies,
the average ESs for reductions in rearrest and incarceration have
been greater than .50 (Curtis et al., 2004). Yet, the MST conditions
showed no such effects in comparison with FC in the present
study. As this was the first controlled MST study to include an
intensive juvenile justice surveillance component, the findings
suggest that certain types of favorable MST outcomes might not be
HENGGELER ET AL.
realized within such a context, in spite of the best efforts of the
MST therapists. Fourth, although not statistically significant, the
substance-related findings, along with anecdotal reports from the
MST/CM therapists and supervisor, suggest that the integration of
CM into the more ecologically oriented MST treatment protocol
(DC/MST/CM) was a valuable addition.
Several limitations should be noted. First, implementation of the
guiding principles for juvenile drug courts likely varies from site to
site. Hence, the generalization (external validity) of the favorable
juvenile drug court findings to other juvenile drug courts cannot be
assumed. Second, the timing of the assessments might have fa-
vored certain treatment conditions. For example, the 4-month
assessment coincided with the approximate completion of MST.
The most favorable intervention effects for the MST conditions,
however, emerged at 12 months rather than at 4 months, which
argues against the outcomes being unfairly influenced by the
timing of the assessments in this case. A third and related concern,
however, is that the study did not include a follow-up. The 12-
month assessment coincided with the completion of drug court for
the youths in the three drug court conditions. Consequently, the
better performance on self-reported substance use for these drug
court youths in comparison with FC youths might have been due
to the fact that with a 12-month duration of drug court, the drug
court youths were continuing to be tested regularly for drugs.
Hence, the sustainability of the favorable drug court outcomes has
not been determined, and the timing of the 12-month assessment
might have favored the drug court conditions. Perhaps these out-
comes will attenuate as youths are no longer under drug court
supervision. It is also possible, however, that the drug court out-
comes for rearrest and incarceration will improve after intensive
surveillance of the youths has been removed. Fortunately, these
outcomes are being examined in a long-term (5-year) follow-up of
the youths and families participating in this project. Fourth, the
present study was part efficacy trial (e.g., therapists employed by
the research center, excellent supervision) and part effectiveness
trial (e.g., minimal exclusion criteria, implementation in commu-
nity settings). As such, one cannot assume that the favorable
results for DC/MST and DC/MST/CM will easily transport to
other community-based settings (Henggeler, 2004). Fifth, although
RMANOVA, followed by a multistep process, provided some
control for Type I error, more conservative procedures were not
used in light of the early stage of research in this area. Thus, Type
I error is likely inflated. Sixth, resources were not available to
conduct frequent drug urine screens for youths in the FC condition.
Hence, the corresponding outcomes for the FC youths could not be
compared with those of the youths in the drug court conditions.
The aforementioned 5-year follow-up study, however, is collecting
biological indices of substance use for youths across all the treat-
Achenbach, T. M. (1990). Manual for the Youth Self-Report and 1991
Profile. Burlington: University of Vermont, Department of Psychiatry.
Achenbach, T. M. (1991). Manual for the Child Behavior Checklist and
1991 Profile. Burlington: University of Vermont, Department of
Altschuler, D. M., Armstrong, T. L., & MacKenzie, D. L. (1999). Reinte-
gration, supervised release, and intensive aftercare. In S. Bilchik (Ad-
ministrator), Juvenile justice bulletin (July, pp. 1–23) Washington, DC:
Office of Juvenile Justice and Delinquency Prevention.
American Psychiatric Association. (1994). Diagnostic and statistical man-
ual of mental disorders (4th ed.). Washington, DC: Author.
Belenko, S. (2001). Research on drug courts: A critical review 2001
update. New York: The National Center on Addiction and Substance
Abuse, Columbia University.
Belenko, S., & Dembo, R. (2003). Treating adolescent substance abuse
problems in the juvenile drug court. International Journal of Law and
Psychiatry, 26, 87–110.
Belenko, S., & Logan, T. K. (2003). Delivering more effective treatment to
adolescents: Improving the juvenile drug court model. Journal of Sub-
stance Abuse Treatment, 25, 189–211.
Budney, A. J., & Higgins, S. T. (1998). A community reinforcement plus
vouchers approach: Treating cocaine addiction (NIH Publication No.
98–4309). Rockville, MD: National Institute on Drug Abuse.
Cooper, C. S. (2002). Juvenile drug treatment courts in the United States:
Initial lessons learned and issues being addressed. Substance Use &
Misuse, 37, 1689–1722.
Cunningham, P. B., Donohue, B., Randall, J., Swenson, C. C., Rowland,
M. D., Henggeler, S. W., & Schoenwald, S. K. (2003). Integrating
contingency management into multisystemic therapy. Charleston: Med-
ical University of South Carolina, Family Services Research Center.
Curtis, N. M., Ronan, K. R., & Borduin, C. M. (2004). Multisystemic
treatment: A meta-analysis of outcome studies. Journal of Family Psy-
chology, 18, 411–419.
Dickie, J. (2000). Summit County Juvenile Court Drug Court evaluation
report: July 1, 1999–June 30, 2000. Akron, OH: University of Akron,
Institute for Health and Social Policy.
Donohue, B., & Azrin, N. H. (2001). Family behavior therapy. In E. F.
Wagner and H. B. Waldron (Eds.). Innovations in adolescent substance
abuse (pp. 205–228). New York: Pergamon Press.
Elliott, D. S., Ageton, S. S., Huizinga, D., Knowles, B. A., & Canter, R. J.
(1983). The prevalence and incidence of delinquent behavior: 1976–80
(Report of the National Youth Survey, Project Rep. No. 26). Boulder,
CO: Behavioral Research Institute.
First, M. B. Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (2001).
Structured Clinical Interview for DSM–IV–TR Axis I Disorders, re-
search version, patient edition (SCID-I/P). New York: New York State
Psychiatric Institute, Biometrics Research.
Henggeler, S. W. (1989). Delinquency in adolescence. Newbury Park, CA:
Henggeler, S. W. (2004). Decreasing effect sizes for effectiveness studies:
Implications for the transport of evidence-based treatments: Comment
on Curtis, Ronan, and Borduin (2004). Journal of Family Psychol-
ogy, 18, 420–423.
Henggeler, S. W., & Borduin, C. M. (1992). Multisystemic Therapy
Adherence Scales. Unpublished instrument, Medical University of
South Carolina, Department of Psychiatry and Behavioral Sciences,
Henggeler, S. W., Clingempeel, W. G., Brondino, M. J., & Pickrel, S. G.
(2002). Four-year follow-up of multisystemic therapy with substance
abusing and dependent juvenile offenders. Journal of the American
Academy of Child & Adolescent Psychiatry, 41, 868–874.
Henggeler, S. W., Melton, G. B., Brondino, M. J., Scherer, D. G., &
Hanley, J. H. (1997). Multisystemic therapy with violent and chronic
juvenile offenders and their families: The role of treatment fidelity in
successful dissemination. Journal of Consulting and Clinical Psychol-
ogy, 65, 821–833.
Henggeler, S. W., Pickrel, S. G., & Brondino, M. J. (1999). Multisystemic
treatment of substance abusing and dependent delinquents: Outcomes,
JUVENILE DRUG COURT OUTCOMES
treatment fidelity, and transportability. Mental Health Services Re-
search, 1, 171–184.
Henggeler, S. W., Pickrel, S. G., Brondino, M. J., Ward, D. M., &
Rowland, M. D. (1997). Service Utilization Tracking Form II. Charles-
ton: Medical University of South Carolina, Family Services Research
Henggeler, S. W., & Randall, J. (2000). Conducting randomized treatment
studies in real world settings. In D. Drotar (Ed.), Handbook of research
in pediatric and clinical child psychology (pp. 447–461). New York:
Kluwer Academic/Plenum Press.
Henggeler, S. W., Rowland, M. D., Halliday-Boykins, C., Sheidow, A. J.,
Ward, D. M., Randall, J., et al. (2003). One-year follow-up of multisys-
temic therapy as an alternative to the hospitalization of youths in
psychiatric crisis. Journal of the American Academy of Child & Ado-
lescent Psychiatry, 42, 543–551.
Henggeler, S. W., & Schoenwald, S. K. (1998). The MST supervisory
manual: Promoting quality assurance at the clinical level. Charleston,
SC: MST Institute.
Henggeler, S. W., Schoenwald, S. K., Borduin, C. M., Rowland, M. D., &
Cunningham, P. B. (1998). Multisystemic treatment of antisocial behav-
ior in children and adolescents. New York: Guilford Press.
Howell, J. C. (2003). Preventing & reducing juvenile delinquency: A
comprehensive framework. Thousand Oaks, CA: Sage.
Huddleston, C. W., Freeman-Wilson, K., & Boone, D. L. (2004). 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.
Huey, S. J., Henggeler, S. W., Brondino, M. J., & Pickrel, S. G. (2000).
Mechanisms of change in multisystemic therapy: Reducing delinquent
behavior through therapist adherence and improved family and peer
functioning. Journal of Consulting and Clinical Psychology, 68, 451–
Institute of Medicine (1998). Bridging the gap between practice and
research: Forging partnerships with community-based drug and alcohol
treatment. Washington, DC: National Academy Press.
Land, K. C., McCall, P. L., & Williams, J. R. (1990). Something that works
in juvenile justice: An evaluation of the North Carolina court counselors’
intensive protective supervision randomized experiment project, 1987–
1989. Evaluation Review, 14, 574–606.
Liddle, H. A., & Dakof, G. A. (1995). Efficacy of family therapy for drug
abuse: Promising but not definitive. Journal of Marital and Family
Therapy, 21, 511–543.
McLellan, A. T., Carise, D., & Kleber, H. D. (2003). Can the national
addiction treatment infrastructure support the public’s demand for qual-
ity care? Journal of Substance Abuse Treatment, 25, 117–121.
Miller, W. R. (1991). Form 90: Structured Assessment for Drinking and
Related Behaviors. Washington, DC: National Institute on Alcohol
Abuse and Alcoholism.
Morris, S. B., & DeShon, R. P. (2002). Combining effect size estimates in
meta-analysis with repeated measures and independent-groups designs.
Psychological Methods, 7, 105–125.
National Association of Drug Court Professionals, Drug Court Standards
Committee. (1997). Defining drug courts: The key components. Wash-
ington, DC: U.S. Department of Justice, Office of Justice Programs,
Drug Courts Program Office.
National Institute on Drug Abuse. (1999). Principles of drug addiction
treatment: A research-based guide (NIH Publication No. 99–4180).
Washington, DC: Author.
Parnham, J., & Wright, R. (1998). Escambia County family-focused juve-
nile drug court. Pensacola, FL: Escambia County Juvenile Court.
Petry, N. M. (2000). A comprehensive guide to the application of contin-
gency management procedures in clinical settings. Drug and Alcohol
Dependence, 58, 9–25.
Schaeffer, C. M., & Borduin, C. M. (2005). Long-term follow-up to a
randomized clinical trial of multisystemic therapy with serious and
violent juvenile offenders. Journal of Consulting and Clinical Psychol-
ogy, 73, 445–453
Schoenwald, S. K., Sheidow, A. J., Letourneau, E. J., & Liao, J. G. (2003).
Transportability of multisystemic therapy: Evidence for multilevel in-
fluences. Mental Health Services Research, 5, 223–239.
U.S. Public Health Service. (2001). Youth violence: A report of the Sur-
geon General. Washington, DC: Author.
Waldron, H. B. (1997). Adolescent substance abuse and family therapy
outcome: A review of randomized trials. Advances in Clinical Child
Psychology, 19, 199–234.
Waldron, H. G., Slesnick, N., Turner, C. W., Brody, J. L., & Peterson, T. R.
(2001). Treatment outcomes for adolescent substance abuse at 4- and
7-month assessments. Journal of Consulting and Clinical Psychol-
ogy, 69, 802–813.
Weersing, V. R., & Weisz, J. R. (2002). Community clinic treatment of
depressed youth: Benchmarking usual care against CBT clinical trials.
Journal of Consulting and Clinical Psychology, 70, 299–310.
Received February 7, 2005
Revision received June 15, 2005
Accepted June 22, 2005 ?
HENGGELER ET AL.