HIGHLIGHTS
  • Intervention: A treatment targeting multiple factors linked to anti-social juvenile behavior.
  • Key Findings: Randomized controlled trials show sizable decrease in the amount and severity of criminal behavior by juvenile offenders; however, effectiveness may depend critically on (i) close adherence to the intervention’s key features and (ii) population/setting in which it is implemented.

Description of the Intervention

Multisystemic Therapy is a treatment for juvenile offenders that uses a combination of empirically-based treatments (e.g. cognitive behavior therapy, behavioral parent training, functional family therapy) to address multiple variables (i.e. family, school, peer groups) that have been shown to be factors in juvenile behavior.

Masters-level therapists provide Multisystemic Therapy (thru the process below) at the youth’s home and community locations (e.g. school, recreation center). The therapists are available to the youth and his/her family 24 hours a day, 7 days a week. Each therapist has a small caseload (between one and five families). On average, the treatment lasts for four months, with the therapist spending several hours per week with the youth and his/her family.

Multisystemic Therapy’s overall goals are to improve the youth’s ability to make good decisions when choosing his/her peer group, and the family’s ability to monitor his/her behavior. To achieve these goals, the Multisystemic Therapist: (1) interviews the youth, his/her family and peers, and school officials to identify the youth’s problem behaviors and their causes; (2) identifies the youth’s personal strengths, and positive aspects of his or her family, peer group, and school, which can be used to address the problem behavior (e.g. an athletic youth might be encouraged to join a sports team to keep him or her occupied after school, or a family member who lives nearby could help supervise the youth); and (3) sets goals for the youth (e.g. regular school attendance, less contact with delinquent peers) and his/her parents (e.g. enforcement of curfew, more frequent communication with the youth’s teachers) to be achieved during treatment.

Progress towards these goals is carefully monitored, and if goals are not met, the therapist works with the youths, their families, friends, teachers and principals to remove obstacles to the goals’ achievement.

Because Multisystemic Therapy involves such a specific protocol, a well-defined supervisory system exists to ensure that treatment is provided with fidelity to the Multisystemic Therapy model (e.g. therapists report progress and setbacks to a supervisor who, in turn, reports to an Multisystemic Therapy expert consultant).

Multisystemic Therapy costs approximately $5,800 (in 2007 dollars) per youth treated.

Click here to go to the program’s website.

EVIDENCE OF EFFECTIVENESS

Study 1

Randomized controlled trial of 176 serious offenders (age 12-17) in Missouri. Prior to the intervention, these youths averaged four felony arrests, and 48% had been arrested for at least one violent offense. The youths were predominantly male (69%), white (76%), and of low socio-economic status (70%).

The youths were randomly assigned into an intervention group receiving Multisystemic Therapy, or a control group receiving individual therapy. Individual therapy is the normal treatment for youth offenders, and, unlike Multisystemic Therapy, does not typically address systemic factors (i.e. family, school, peer group) that contribute to the youth’s delinquency.

The average length of treatment in both groups was 25 weeks. Multisystemic Therapy was provided by trained graduate students who were closely supervised.

Results for the Multisystemic Therapy group 4 years after completion of the intervention (versus the control group):

  • 26% of the Multisystemic Therapy group had been arrested at least once, compared to 71% of the control group participants.
  • 88% reduction in the average number of arrests, versus the control group (0.45 vs. 3.88).
  • Multisystemic Therapy recidivists’ arrests were for less serious offenses.

Results for the Multisystemic Therapy group 13.7 years after completion of the intervention (versus the control group)– when participants’ average age was 29 years:

  • 50% of the Multisystemic Therapy group had been arrested at least once, compared to 81% of the control group.
  • 14% of the Multisystemic Therapy group had been arrested for a violent offense, compared to 30% of the control group.
  • 13% of the Multisystemic Therapy group had been arrested for a drug offense, compared to 33% of the control group.
  • 54% reduction in the average number of arrests, versus the control group (1.8 vs. 4.0).
  • 57% reduction in the average number of days incarcerated as an adult, versus the control group (582 vs. 1357).

Discussion of Study Quality (click here for glossary of terms)

  • This study had low attrition and a long-term follow-up: At the 4-year and 13.7-year follow-ups, outcome data on arrest rates were collected for 94-95% of the original sample.
  • The study reported outcomes using an intention-to-treat analysis: Although approximately 20% of both the Multisystemic Therapy and control groups dropped out of treatment, the study collected arrest data for almost all dropouts.
  • The study used official arrest data, rather than self-reports, as its primary outcomes measure.
  • Outcome data were collected by research assistants who were blind as to whether youths were in the Multisystemic or control groups.
  • There were no significant differences between the Multisystemic Therapy and control groups prior to the intervention.

Study 2

Randomized controlled trial of 118 substance abusing juvenile offenders (average age 15.7), mostly from low socioeconomic backgrounds. The sample averaged three prior arrests, and was 79% male, 50% African American, and 47% white.

The youths were randomly assigned to receive either Multisystemic Therapy or a control group that received the usual community services (namely, required attendance in a 12-step group therapy program).

Results for the Multisystemic Therapy group 4 years after the completion of the intervention (versus the control group):

  • The Multisystemic Therapy group had 75% fewer official adult convictions for aggressive crimes than the control group (and significantly fewer self-reported aggressive crimes).
  • The impact of Multisystemic Therapy on drug use is unclear: Drug tests suggested a decrease in drug use, but youths’ self-reports indicated no decrease.

Discussion of Study Quality (click here for glossary of terms)

  • The study had a 4-year follow-up.
  • The study appears to have measured outcomes using an intention-to-treat analysis (however, the study report is not explicit on this issue).
  • The study used official arrest data and biological testing to supplement self-reported survey data on criminal behavior and substance abuse.
  • Study Limitation: At the 4-year follow-up, the study could not obtain outcome data for 32% of the original sample, which represents a moderately high level of attrition. Statistical tests suggest that the attrition did not result in observable differences between the Multisystemic Therapy and control groups that might undermine the randomization. However, the study results should be treated with caution because it is possible that the attrition may have resulted in unobservable differences between the two groups, leading to inaccurate estimates of Multisystemic Therapy’s impact.

Study 3

Randomized controlled trial of 105 youth offenders (average age 15.1 years) who appeared before a county family court in Ohio between 1998 and 2001, and had a previous felony conviction. The sample averaged approximately seven prior offenses, with approximately two of them being felonies. The youths were predominantly male (78%) and white (78%).

The youths were randomly assigned either to an intervention group that received Multisystemic Therapy or to a control group that did not, but could receive any other services available in the community.

Results for the Multisystemic Therapy group 2 years after random assignment–i.e. about 18 months after completion of the intervention (versus the control group):

  • 23% less likely to be re-arrested (67% of the Multisystemic Therapy group had been re-arrested at least once, versus 87% of the control group).
  • 39% fewer arrests and arraignments per youth over the two years (1.4 vs. 2.3).
  • No significant difference in the percentage of felonies versus misdemeanors between the two groups.

Discussion of Study Quality (click here for glossary of terms)

  • The study had low attrition: At the 2-year follow-up, outcome data on arrest rates were collected for 89% of the original sample.
  • The study appears to have measured outcomes using an intention-to-treat analysis (however, the study report is not explicit on this issue).
  • The study was conducted in a community mental health setting under the conditions in which Multisystemic Therapy is normally implemented, thus providing evidence of its real-world effectiveness. Also, the study sample was representative of typical youth offenders appearing before the county court, as evidenced by the fact that almost 90% of youth meeting the study’s eligibility criteria agreed to participate in the study.
  • The study used official arrest data to measure criminal behavior.
  • The study was conducted by independent researchers (i.e. not the program’s developers).
  • There were no significant differences between the Multisystemic Therapy and control groups prior to the intervention.

Study 4

Randomized controlled trial of 409 juvenile offenders (average age 14.6) in Ontario , Canada, who were predominantly male (74%) and from low or middle-income backgrounds. 67% of the sample had been convicted for at least one offense, and 36% had been arrested for at least one violent offense.

The youths were randomly assigned to a group that received Multisystemic Therapy or a control group that received the government’s usual services.

In contrast to the three U.S. studies reported above, this study found that Multisystemic Therapy had no effect on participants’ criminal activity (see results below). Factors that may explain the differing results include: (1) Multisystemic Therapy implementation in Ontario did not adhere as closely to the Multisystemic Therapy model, according to key measures of adherence; (2) the Ontario sample was of higher average socioeconomic status, including more middle-income youths; and (3) the control group in the Ontario study had access to more extensive social services (e.g., mental health services) than the control groups in the U.S. studies. Thus, the Ontario results raise questions about Multisystemic Therapy’s effectiveness, particularly when implemented under conditions other than those in studies 1-3 above.

Results for the Multisystemic Therapy group 3 years after the completion of the intervention (versus the control group):

  • Multisystemic Therapy had no effect on participants’ conviction or incarceration rates, compared to the control group.

Discussion of Study Quality (click here for glossary of terms)

  • The study had no attrition: Conviction and incarceration data were collected for 100% of study participants at the 3-year follow-up.
  • The study measured outcomes using an intention-to-treat analysis.
  • The study measured outcomes using official conviction and incarceration data rather than self-reports.
  • The study was conducted by independent researchers (i.e. not the program’s developers).
  • There were no significant differences prior to the intervention between the Multisystemic Therapy and control groups.

REFERENCES

(Click on linked authors’ names for their contact information)

Study 1

  • Schaeffer, Cindy M. and Charles M. Borduin. “Long-Term Follow-Up to a Randomized Clinical Trial of Multisystemic Therapy With Serious and Violent Offenders.” Journal of Consulting and Clinical Psychology, 2005, Vol. 73, No. 3, pp. 445-453.
  • Borduin, Charles. “Disseminating Evidence Based Practices Four States Speak-out on MST.” Presentation at National Council for Community Behavioral Healthcare (March 30-April 1, 2003).

Study 2

  • Henggeler, Scott W.W. Glenn ClingempeelMichael J. Brondino and Susan G. Pickrel. “Four-Year Follow-up of Multisystemic Therapy With Substance Abusing and Substance-Dependent Juvenile Offenders.” Journal of the American Academy of Child and Adolescent Psychiatry, 41:7, July, 2002, pp. 868-874.

Study 3

  • Timmons-Mitchell, Jane, Monica B. Bender, Maureen A. Kishna and Clare C. Mitchell. “An Independent Effectiveness Trial of Multisystemic Therapy with Juvenile Justice Youth.“ Journal of Clinical Child and Adolescent Psychology, 2006, Vol. 35, No. 2, pp. 227-236.

Study 4