Study 1
Rigter and colleagues (2013) conducted a randomized controlled trial to evaluate the efficacy of Multidimensional Family Therapy (MDFT) on substance use and dependence. The study sample included Western European youths from five outpatient treatment sites located in Berlin, Germany; Brussels, Belgium; Geneva, Switzerland; The Hague, Netherlands; and Paris, France. Eligible youths were 13–18 years old; diagnosed with a cannabis use disorder (misuse or dependence, determined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (or the DSM | 4) guidelines (with misuse being diagnosed if at least one of four abuse criteria had been met, and dependence being if at least three of seven dependence criteria had been met); and had at least one parent willing to participate in the treatment. Youths suffering from a current mental disorder or condition (e.g., psychosis, advanced eating disorder, suicide ideation) that required inpatient treatment or had a substance disorder that required maintenance treatment with methadone or buprenorphine were ineligible for the study.
The total sample consisted of 450 youths who were assigned (using a 1:1 ratio; except in Paris, where it was a 2:1 ratio) to the MDFT intervention group (n = 212) or the treatment-as-usual (TAU; n = 238) comparison group. The youths were stratified into the treatment groups based on gender, age (13- to 14-year-olds versus 15- to 18-year-olds), and level of cannabis use in the past 90 days (74 or fewer days of cannabis consumption versus 75 or more). Specific to the study sites, 120 youths (MDFT = 59; TAU = 61) were from Berlin, 60 (MDFT = 30; TAU = 30) were from Brussels, 60 (MDFT = 30; TAU = 30) were from Geneva, 109 (MDFT = 55; TAU = 54) were from The Hague, and 101 (MDFT = 38; TAU = 63) were from Paris. Youths in the MDFT intervention group received weekly therapy sessions (at least two sessions per week), conducted by trained MDFT therapists. Youths in the TAU group received the standard treatment that was provided by individual treatment sites (i.e., individual psychotherapy).
Of the total sample, 85 percent of the youths were boys. Forty percent of the youths were of first- or second-generation foreign descent, 40 percent had an alcohol use disorder, 33 percent had been arrested in the past 3 months (mostly for drug offenses, property crimes, and violent crimes), less than 5 percent had a substance-use disorder for drugs other than cannabis, and 56 percent had parents who were divorced or separated. The average youth age was 16.3 years. There were no statistically significant differences in baseline characteristics between youths in the MDFT and TAU groups; however, in Brussels, there was a higher rate of foreign-descent youths in the MDFT group, compared with those in the TAU group.
Data were collected at baseline and at the 3-month, 6-month, 9-month, and 12-month follow-up periods. The CrimeSolutions review of the study focused on the 12-month follow-up period. The study examined the following outcomes: diagnosis of cannabis use disorder, symptoms of cannabis use dependence, and frequency of cannabis use. The outcome of interest for the CrimeSolutions review was the diagnosis of a cannabis use disorder. The Adolescent Diagnostic Interview Light (ADI | Light for cannabis) was used to measure a diagnosis of cannabis disorder (abuse and dependence). A latent growth curve model was used to estimate the intent-to-treat effects between youths in the MDFT group and youths in the TAU group, at the 12-month follow-up. The study authors did not conduct subgroup analyses.
Study 2
Van der Pol and colleagues (2018) conducted a randomized controlled trial, using the same study sample as in Study 1 (Rigter et al., 2013), to evaluate the effects of the MDFT intervention on reducing criminal offending. Of the five study sites, only two of the original sites (Geneva and The Hague) were included in the study analysis because of the other sites’ inability to report data on the number and types of criminal offenses committed by adolescents over the previous 90 days (using the Self-Report Delinquency [SRD] scale). Thus, the total sample for this study consisted of 169 youths (MDFT = 85; TAU = 84).
Of the 85 youths in the MDFT group, 83 percent were boys. Fifty-eight percent of youths were of foreign descent, 61 percent had parents who were divorced or separated. Thirty-one percent had a parent with mental health or substance use problems. The average youth age was 16.2 years. Of the 84 youths in the TAU group, 85 percent were boys. Fifty-five percent of youths were of foreign decent, 61 percent had parents who were divorced or separated. Twenty-nine percent had a parent with mental health or substance use problems. The average youth age was 16.2 years. There were no statistically significant differences in baseline characteristics between youths in MDFT and TAU groups. Across treatment sites, there were several differences at baseline. For example, Geneva had a higher percent of youths of foreign descent, The Hague had a higher proportion of youths living with their families, and cannabis dependence was more common among youths in Geneva than in The Hague; however, these differences were not statistically significant between the two treatment conditions.
Data were collected at baseline and at the 6-month and 12-month follow-up periods. The CrimeSolutions review of the study focused on the 12-month follow-up period. Outcomes of interest related to criminal offenses included offenses of any type, property crimes, and violent crimes. The SRD scale measured criminal offenses, asking youths how many and which type of criminal offenses they had committed in the past 90 days. A latent growth curve model with Mplus was used to determine differences between youths in the MDFT and TAU group, at the 12-month follow-up. The study authors did not conduct subgroup analyses.
Study 3
Liddle and colleagues (2018) conducted a randomized controlled trial to evaluate the efficacy of MDFT, using youths with co-occurring mental health disorders who were referred for residential substance use treatment in the state of Florida. In this study, youths were at high risk for being placed in a long-term juvenile justice or a residential substance-use treatment facility owing to the severity of their substance-use symptoms and delinquency, the number of their psychiatric diagnoses, and the number of their previous substance-use treatment placements. There were 113 youths who consented and met the following eligibility criteria: 1) between the ages of 13 and 18; 2) diagnosed with a substance use disorder and at least one comorbid psychiatric disorder; 3) referred and approved by the Florida Department of Children and Families (DCF) for state-subsidized residential, dual diagnosis substance use treatment; 4) known to have failed a previous treatment for a substance use disorder, or presenting with severe symptoms warranting a higher level of care either because of safety reasons or because this treatment was ordered by a judge; 5) living in the custody of a parent/caregiver (i.e., not in DCF custody); and 6) not currently suicidal, demonstrating psychotic symptoms, or diagnosed with autism spectrum or intellectual disability disorders.
Of the total sample of 113 youths, 57 were assigned to the MDFT group, and 56 were assigned to the TAU comparison group, using an urn randomization procedure. The TAU group consisted of youths who were in residential treatment, provided by a large community-based substance use treatment provider in Miami, Fla. Of the 57 youths in the MDFT group, 75 percent were boys. The majority (70 percent) of youths were Hispanic, 20 percent were African American, and 12 percent were white. The average age of the youths was 15.3 years, with 53 percent of first-time drug use at the ages of 12–14, 39 percent younger than age 12, and 8 percent at ages 15–18. More than one third of the youths (37 percent) had a parent who was involved in the criminal justice system, and 49 percent had a parent with substance use problems. Of the 56 TAU youths, 74 percent were boys. The majority (70 percent) of youths were Hispanic, 17 percent were African American, and 14 percent were white. The average age of the youths was 15.3 years, with 50 percent of first-time drug use at the ages of 12–14, 39 percent younger than age 12, and 11 percent at ages 15–18. One third of the youths (33 percent) had a parent who was involved in the criminal justice system, and 43 percent had a parent with substance use problems. There were no statistically significant differences in baseline characteristics between youths in the MDFT and TAU groups.
Data were collected at baseline (i.e., intake) and at the 2-month, 4-month, 12-month, and 18-month follow-ups. The CrimeSolutions review of the study focused on the 18-month follow-up period. Outcomes of interest included substance use problems, frequency of delinquent behaviors, symptoms of externalizing behaviors, and symptoms of internalizing behaviors. Substance use was measured using 1) the Personal Experience Inventory, a 29-item scale that focuses on psychological and behavioral aspects of substance use and related consequences in the previous 30 days. Delinquent behaviors were measured through self-report, using the National Youth Survey Self-Report Delinquency scale, which assessed criminal behavior and delinquent acts based on the Uniform Crime Report.
A latent growth curve model was used to estimate the intent-to-treat effects between MDFT and TAU youths, at the 18-month follow-up. A subgroup analysis determined whether child welfare outcomes differed for families with different baseline characteristics. The study authors conducted subgroup analyses, using a latent class pattern mixture model, to evaluate differences in posttreatment placement patterns between MDFT and TAU youths. For additional information on this analysis, see the Other Information section.
