Deblinger, Lippman, and Steer (1996) conducted a randomized trial to evaluate the impact of Trauma-Focused Cognitive Behavioral Therapy (TF-CBT) on 100 sexually abused children. These children had experienced sexual abuse that had been substantiated by an investigation conducted by the Division of Youth and Family Services or the Prosecutor’s Office. Child participants exhibited a minimum of three posttraumatic stress disorder (PTSD) symptoms. Their ages ranged from 7 to 13 years, with a mean age of 9.9 years. Eighty-three percent were female; 70 percent were white, and 21 percent were African American. The biological father or stepfather was described as the perpetrator in 31 percent of cases.
Over the course of roughly 4 years, subjects completed an initial assessment and were then randomly assigned to one of four treatment conditions: community control condition, child-only intervention, parent-only intervention, or parent and child intervention. The community control condition consisted of providing parents with information about symptoms and encouraging them to seek therapy for their children. The child-only intervention included several cognitive behavioral therapy methods such as gradual exposure, education, coping, and body safety skills. The parent-only intervention consisted of teaching mothers skills for responding therapeutically to their children. Subjects assigned to the three intervention groups participated in 12 weekly treatment sessions provided by a trained therapist who followed a detailed treatment manual. Treatment sessions for child-only and parent-only interventions lasted 45 minutes each; treatment sessions for the parent and child intervention lasted 80 to 90 minutes.
Measures used in the study included a structured background interview to collect demographic and abuse-related data and information about parent and child coping responses and support resources. PTSD was assessed using the epidemiological version of the Schedule for Affective Disorders and Schizophrenia for School-Aged Children. Anxiety was assessed using the 20-item State Trait Anxiety Inventory for Children; depression was assessed using the Child Depression Index; and child behavior problems were assessed using the Child Behavior Checklist. Parents’ interaction with children was evaluated using the Parenting Practice Questionnaire.
Results were reported for children 3 months after the intervention. Follow-up results at 6 months, 1 year, and 2 years posttest were reported by Deblinger, Steer, and Lippman (1999), who addressed missing 2-year follow-up data using a multivariate analysis of covariance with imputed end-point data in which the last obtained scores on outcome measures were carried forward for remaining follow-up analyses. Data from only those participants who had completed all six evaluations was used in the follow-up analyses.
Cohen and Mannarino (1996) used a randomized experimental design with 67 sexually abused preschool children ages 3–6 and their parents. Participants were randomly assigned to either the treatment group, which provided cognitive behavioral therapy adapted for sexually abused preschool children or the control group, which provided nondirective supportive therapy. Subjects were referred from rape crisis centers, child protective services, pediatricians, psychologists, mental health agencies, police departments, and judicial systems. Of the 86 subjects recruited, 67 completed the study. The mean age of treatment completers was 4.7 years; 75 percent lived with both biological parents; 58 percent were female; 54 percent were white, and 42 percent were African American.
Treatment consisted of 12 individual sessions for both child and parent monitored for integrity through intensive training and supervision, use of treatment manuals, and audio-taped sessions. Outcomes were evaluated at posttreatment, which varied from 12 to 16 weeks after baseline.
Instruments used in this study included the Child Behavior Checklist form for 4- to 11-year-olds; the Parenting Practices Questionnaire; the 42-item Child Sexual Behavior Inventory, completed by parents regarding normative and inappropriate sexual behavior; the Weekly Behavior Report, a 21-item instrument for documenting problematic behaviors in preschool children completed by parents; and the Preschool Symptom Self-Report, a pictorial instrument used to obtain multiple sources of information.
Cohen and colleagues (2004) evaluated TF–CBT using a sample of 229 consecutively referred children who had experienced contact sexual abuse confirmed by Child Protective Services, law enforcement, or an independent forensic professional. Children were recruited from two outpatient treatment sites—both academically affiliated, with one in a large metropolitan area and one in a suburban area. Children had to meet at least five criteria for sexual abuse–related PTSD, as defined by the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition. Further, a parent or caretaker had to be willing to participate in the parent component of the treatment. The final sample included the 203 (89 percent) children who attended at least three therapy sessions. Of these children, 180 (89 percent) met full criteria for PTSD.
Since the sample included 14 sibling pairs, there were 189 caretakers in the study. An equal number of sibling pairs were assigned to treatment and control conditions. All children in the study exhibited multiple symptoms of PTSD. They all had at least one responsible, nonabusive parent or guardian willing to participate in the parental component of the study. About half of the children and their parents were randomized to treatment (receiving 12 weeks of TF–CBT treatment) and the other half to a comparison (receiving comparable levels of conventional child-centered therapy) group.
The sample was 79 percent female. Children in the sample ranged from 8 years to 14 years and 11 months, (mean=10.76 years). Sixty-percent were white, and 28 percent were African American. Participating parents included 78 percent biological mothers, 4 percent biological fathers, and the remainder adoptive mothers, foster mothers, grandmothers, and other female relatives.
Instruments used in this study included the PTSD, psychosis, and substance use disorders scales of the Schedule for Affective Disorders and Schizophrenia for School-Aged Children—Present and Lifetime Version; the Child Depression Inventory; the State Trait Anxiety Inventory for Children; the Children’s Attributions Perception Scale; the Child Behavior Checklist; the Child Sexual Behavior Inventory; the Beck Depression Inventory; the Parent Emotional Reaction Questionnaire; the Parent Support Questionnaire; and the Parenting Practices Questionnaire modified for this population.