| ||Literature Coverage Dates||Number of Studies||Number of Study Participants|
|Meta-Analysis 1||1973 - 2006||4||210|
|Meta-Analysis 2||1973 - 1997||8||749|
|Meta-Analysis 3||1990 - 2013||28||5564|
Baldwin and colleagues (2012) conducted a meta-analysis to examine the effectiveness of family-based treatment intervention programs on substance abuse, delinquency, and conduct problems in adolescents (ages 11–19). Specifically, the authors looked at randomized trials of Brief Strategic Family Therapy (BSFT), Functional Family Therapy (FFT), Multidimensional Family Therapy (MFT), or Multisystemic Therapy (MST). Studies were identified using an electronic search of PsycINFO, Medline, and Dissertation Abstracts International, through February 2009, including published and unpublished work. The treatment had to be delivered on its own (i.e., not in combination with other interventions), and prevention studies were excluded.
A total of 24 randomized controlled designed studies were included. The authors did not report an overall aggregated effect size because varying comparison types made interpreting results difficult. Instead they segregated the sample by types of comparison groups: 1) treatment as usual (i.e., referral to juvenile delinquency treatment center), 2) alternative therapy (i.e., any therapy besides family therapy), and 3) no-treatment control. The comparisons used in this CrimeSolutions.gov review comprised studies of BSFT (n
= 3) and FFT (n
= 1) compared with no-treatment control groups. The researchers did not provide any descriptive information regarding the study characteristics or sample demographics.
A weighted random effects model was used in the analysis. A mean effect size was calculated, using a standardized mean difference statistic (d
). Additional moderator, influence, and publication bias analyses were conducted to account for the heterogeneity across the small number of studies included.Meta-Analysis 2
Woolfenden, Williams, and Peat (2002) conducted a meta-analysis to examine the effectiveness of family-based treatment intervention programs on delinquency and and other problem behaviors in adolescents (ages 10–17). Studies were eligible for review if they were randomized controlled trials of family and/or parenting interventions for children and adolescents aged 10–17 with conduct disorder and/or delinquency. Conduct disorder was defined by a standardized psychological assessment (such as the Child Behavior Checklist) or a psychiatric diagnosis, and delinquency was defined by a referral from a juvenile justice or other legal system for a child/adolescent who committed a serious crime or offended on at least two occasions. Studies that only included sex or drug offenses were excluded. Literature was electronically searched (including Cochrane Controlled Trial Register, Medline, EMBASE, CINAHL, PsycINFO, SocioFile, ERIC, and HealthStar) through July 1999; published and unpublished work was included.
Eight randomized controlled designed studies were included in the review. The studies together included 749 youth participants. The studies were conducted in the United States, with the exception of one study from Australia. The study samples were predominantly male, with one study sample including a slightly greater number of girls than boys. Seven of the studies included youth referred to treatment by the juvenile justice system; the other focused on youth referred to treatment by schools, parents, or other youth agencies. The included studies used a number of different family and parenting interventions, including short-term family therapy, parent training, MST, multidimensional intervention foster care (MTFC), and an adolescent diversion project that included a family condition and a multi-focus condition. Researchers used a weighted mean difference to calculate the time in institutions, a relative risk ratio to measure adolescents’ risk of rearrests, and a standardized mean difference to calculate their rate of rearrest at 1 to 3 years later.Meta-Analysis 3
Dopp and colleagues (2017) conducted a meta-analysis on family-based treatments for serious and violent juvenile offenders (up to age 18). Studies were eligible if they 1) focused on one or more family-based treatments; 2) included a sample of serious juvenile offenders; 3) used a prospective research design, including at least one comparison condition); 4) included at least one outcome measure of antisocial behavior of targeted youth; and 5) included a published or unpublished English-language report of the study by January 1, 2014. Electronic databases searched included PsycINFO and Medline (via PubMed) to identity eligible studies. Reference lists from published reviews of youth psychosocial treatments was also conducted to identify relevant studies. In addition, websites of certain organizations were consulted and a hand search of the tables of contents of several journals that regularly published studies of psychosocial treatments was also conducted. Specific searches were also conducted to identify unpublished studies.
A total of 28 studies (which included 324 effect sizes) were included in the meta-analysis. The studies together included 5,565 youth participants. Of these studies, 19 were randomized controlled trials, and 9 were non-randomized. The majority of studies (n
= 21) were conducted in the United States, and the others were conducted in Europe (n
= 6) or Canada (n
= 1). The average age for participants at the start of treatment was 15.0 years. The majority of the sample was male (72.0 percent) and identified as white (58.3 percent). The sample also included a significant number of black (30.6 percent), Hispanic (9.5 percent); and Asian, American Indian, or multiracial (8.0 percent) youth. Of the total sample, an average of 4.34 had committed pretreatment offenses, and others had committed one or more felony (72.5 percent), violent offense (50.0 percent), or sexual offense (10.7 percent).
Of the 28 studies, 19 examined the effectiveness of MST, 4 examined Treatment Foster Care Oregon (formerly MTFC), 3 examined FFT, 1 examined BSFT, and 1 looked at intensive home-based services. Across the studies, these programs lasted for an average of 25.9 weeks and involved 50.8 hours of contact with the youth and/or family members.
Effect sizes were calculated using Cohen’s d
, so that a positive number represented a beneficial effect for the family-based treatment group relative to the comparison group. Potential moderators (e.g., characteristics of samples, treatments, methods, and measures) were entered as fixed effects in the meta-analytic model.