Study
Santisteban and colleagues (2003) used an experimental design to assess the efficacy of BSFT for Hispanic youth with behavior problems and drug use. A total of 126 Hispanic adolescents and their families participated in the study. To be eligible for inclusion, adolescents needed to exhibit externalizing behavior problems according to parents or school.
Adolescents ranged in age from 12 to 18. The majority of participants (87 percent) were male. The ethnic breakdown was 64 Cuban, 18 Nicaraguan, 12 Columbian, 8 Puerto Rican, 4 Peruvian, 2 Mexican, and 18 from other Hispanic nationalities. Youths were randomly assigned to either BSFT or a group treatment control (GC).
GC participants received a participatory-learning group intervention; facilitators led the group and encouraged participants to discuss and solve problems among themselves. The facilitator encouraged group cohesion, disseminated information regarding drug use, and maintained a problem-solving atmosphere. Groups consisted of four to eight adolescents; family members were not included. Participants received from 6 to 16 sessions of weekly therapy; sessions lasted on average 90 minutes.
Attrition for the BSFT condition was 30 percent and for the GC condition 37 percent. Adolescent behavior problems were assessed using the Revised Behavior Problem Checklist. Drug involvement was measured using the Addition Severity Index and urine toxicology screens. And family functioning was measured using the Family Environment Scale and the Structural Family Systems Rating. The researchers conducted subgroup analyses on family functioning and family cohesion at baseline.
Study
Coatsworth and colleagues (2001) used an experimental pretest–posttest design with 104 families of Hispanic (n = 79) or African American (n = 25) descent. Families were eligible for the study if they had a 12- to 14-year-old child who had significant academic problems, had initiated drug or alcohol use, or about whom the family or school reported a complaint of externalizing problems in the form of misconduct or internalizing problems in the form of anxiety/depression. Adolescents who had attempted suicide were excluded from the study. The sample was 75 percent male, with a mean age of 13.1.
Participants were randomized to the experimental condition or the community comparison condition. The two groups did not significantly differ. The experimental group received BSFT, while the comparison group received whatever therapy the particular community agency used. Researchers assessed the adolescents’ behavior problems as well as engagement and retention in treatment at baseline and at the completion of treatment. The authors conducted subgroup analyses on conduct disorder intake scores in relation to treatment engagement and retention.
Study
Robbins and colleagues (2011) used a random assignment design at eight community treatment provider sites to assess the impact of Brief Strategic Family Therapy (BSFT) on family functioning and adolescent substance abuse. Therapists were randomly assigned to the treatment condition or to treatment as usual (TAU). Therapists who volunteered to participate in the study did not know in advance to which group they would be assigned. Drug use was assessed at baseline and at 12 monthly follow-up assessments. All other measures were assessed at baseline, then at 4, 8, and 12 months postrandomization. Other measures included family functioning and retention and engagement.
TAU varied across the eight community treatment providers. It included individual therapy, group therapy, parent training groups, nonmanualized family therapy, and case management. Participants received 12 to 16 sessions over a 3- to 4-month period. Participation in ancillary services (e.g., case management, Alcoholic Anonymous) was typical.
BSFT included 12 to 16 sessions over a 4-month period. Other systems could be addressed during these sessions (e.g., parents could be coached on how to communicate with a probation officer or a school official). Most sessions were delivered in either the home (52.2 percent) or the clinic (45.3 percent) but could also be delivered elsewhere, such as at school or work (2.5 percent). Participation in ancillary services was permitted (e.g., case management, Alcoholic Anonymous), but most sessions were classified as family therapy. Booster sessions were allowed for participants in either condition.
Families were recruited from eight community treatment centers. To be included in the study, adolescents had to self-report illicit drug use (other than alcohol or tobacco) within the past 30 days. They had to live with a family (any parent or guardian). Adolescents with pending criminal offenses were excluded to eliminate the possibility of participant incarceration. The urn randomization process was used to assign families to either the control or treatment condition. A total of 480 adolescents and their family members participated in the study across eight sites. Participants were predominately male (n = 377). The racial/ethnic breakdown consisted of 213 Hispanics/Latinos, 148 non-Hispanic whites, 110 non-Hispanic blacks, 5 American Indians/Alaskans, 2 Japanese/whites, 1 Persian, and 1 Lebanese. Seventy-two percent were referred from the juvenile justice system.
The statistical model used for analysis included random effects for site and therapists. Various methods were used to assess the differences in engagement and retention, drug use, and family functioning, such as logistic regression, contingency table methods, generalized estimating equations, and the Wilcoxon rank–sum test. The authors conducted subgroup analyses on race and gender in relation to treatment engagement and retention.