Study
Celinska and colleagues (2013) used a quasi-experimental design to compare youths receiving FFT with youths who received individual therapy or mentoring. Data on at-risk youth was collected between 2005 and 2007 in the state of New Jersey. The treatment group included youths referred by New Jersey Probation, Family Crisis Intervention Unit, Family Court, and Divisions of Youth and Family Services. Youths in the control group received services through the Youth Case Management program. They were identified to participate in the study by case managers. Eligibility criteria for participation in the study included the following: being between the ages of 11–17; living with a parent or guardian; and having a history of aggressive behavior, destruction of property, or chronic truancy. Youths with alcohol and other drug use or mental health issues were not eligible
The sample consisted of 72 youths, of which 36 received FFT and 36 were in the control group. Of the entire study sample, the majority of youths were male (69 percent) and slightly older than15 years. The treatment group receiving FFT was 36 percent African American, 26 percent Latino, 19 percent white, and 8 percent other race/ethnicity. The majority of the treatment group was male (69 percent) and the average age was 15.5 years. The comparison group had similar percentages for race/ethnicity (44 percent African American, 33 percent Latino, 14 percent white, and 8 percent other). The majority of the comparison group was also male (61 percent) and the average age was 15.1 years. There were no significant differences between the groups on demographic characteristics.
Data was collected using the Strengths and Needs Assessment (SNA), which provided a standardized way to collect information on youth functioning across life domains. Primary outcome measures included the following scales: 1) Life Domain Scale, which measured dimensions of family, school, and vocational functioning; 2) Child Strengths Scale, which included family life, personal achievements, and community involvement; 3) Acculturation Scale, which included items related to language and culture; 4) Caregiver Strengths Scale, which was based on caregivers’ involvement with their child and the level of stability provided in the home; 5) Caregiver Needs Scale, which referred to the mental and physical health needs of caregivers; 6) Child Behavior and Emotional Needs Scale, which assessed impulsivity, depression, anxiety, anger control, and substance abuse; and 7) Child Risk Behavior, which included suicide risk, self-harm, dangerous behavior toward others, sexual aggression, running away, and fire setting.
The goal of the study was to measure effects of FFT relative to services received in the comparison group. A one-way ANOVA test was conducted to test for differences in the duration of treatment and the seven life domain areas the scales represent. The sample size was not large enough to employ other methods.
Study
The Sexton and Turner (2010) evaluation included a comparison of FFT with probation services. This community-based evaluation was conducted within a statewide juvenile justice system of a large western state. Data collection and group assignments were conducted by an independent state evaluation center. A total of 917 families in 14 counties in both rural and urban settings participated.
The participating adjudicated juveniles had been remanded for probation services and were stratified at the county level and randomly assigned to either FFT or a control group receiving usual probation services. Intervention youths received an average of 12 FFT family-based sessions in their homes over a 3- to 6-month period. FFT was provided by a community-based therapist. Control youth received traditional probation services in their local county with no additional treatment services. Each group included more than 400 adolescents. All participants were followed for 18 months when 1-year posttreatment assessments were collected.
Participants’ ages were evenly distributed from 13 to 17 years. Seventy-nine percent were male, and 21 percent were female. Seventy-eight percent of participants were white, 10 percent African American, 5 percent Asian, 3 percent Native American, and 4 percent were not identified. Most of the participants had committed felony crimes (56.2 percent), and many had committed misdemeanors (41.5 percent).
Measures included family-focused risk and protective factors sections of the Washington State Juvenile Court Assessment completed by a probation officer. Other measures included a treatment adherence measure and a measure of the youth’s adjudicated felony criminal behavior in the 12-month period following randomization to treatment.
The primary outcome measure was the youth’s adjudicated posttreatment felony criminal behavior in the 12-month period following randomization to treatment.
A four-step statistical analysis was used. First, preliminary multivariate analysis of variance (MANOVA) and analysis of variance (ANOVA) analyses were conducted to assess potential outside variables that might influence the hypothesis testing. Second, hierarchical linear modeling and logistic regression analyses were used to test the main hypothesis that the FFT condition was associated with a lower level of adjudicated felony recidivism compared with the control group. Third, a secondary hypothesis concerning effects of therapist model adherence (low versus high) was analyzed using logistic regression. Fourth, analyses examined possible interaction effects between pretreatment family and peer risk factors (low family risk, high family risk, and high peer risk) and therapist adherence as predictors of felony recidivism.
Study
Gordon and colleagues (1988) used a quasi-experimental design to evaluate Functional Family Therapy (FFT) with lower socioeconomic status juveniles, most of whom had multiple offenses. All 54 participants were white, court-referred juveniles from a rural Southeastern Ohio county who were adjudicated delinquents (guilty of misdemeanors or felonies) or status offenders. Youths who had two offenses before treatment (some of whom were placed outside the home) were assigned to the treatment group. The treatment group (n = 27) consisted of 15 males and 12 females who were court-ordered to a university counseling service as a condition of probation. The comparison group (n = 27) consisted of 23 male and 4 female juveniles who were randomly selected from the group of delinquents who were in court during the same period as the treatment group but were not referred for family therapy. These assignment procedures probably resulted in favor of the comparison group.
The participants were all white and were living in an economically depressed community with high rates of unemployment and single-parent households. Status offenses—habitual truancy, unruliness, and running away—accounted for 57 percent of all offenses committed. Misdemeanors accounted for 30 percent of offenses committed and consisted of petty theft, vandalism, criminal trespass, and menacing. Felonies, which accounted for 13 percent of offenses committed, included breaking and entering, grand theft, and rape. Participants in both groups continued to meet with their probation officer one or two times each month. Participants in the treatment group attended a median number of 16 family sessions (range: 7 to 38), lasting an average of 1½ hours each and extending over a mean of 5½ months.
The outcome—recidivism rate—was calculated for each group as the percentage of juveniles convicted of one offense or more. The mean follow-up period for measuring recidivism rates was 27.8 months for the treatment group and 31.5 for the comparison group. Since these periods differed slightly, the recidivism rate was annualized to reveal the rate for any 12-month period.
Adult recidivism at 5 to 6 years after placement on probation—when participants, generally, were 20 to 22 years of age—was reported by Gordon, Graves, and Arbuthnot (1995).
