Evidence is growing that trauma-focused cognitive-behavioral therapy (TF-CBT) is an effective treatment for sexually abused children, including those who have experienced multiple other traumatic events. This article reviews the research that has examined treatments for sexually abused children and suggests future research priorities in this regard.
Most of the studies that have evaluated TF-CBT have been well designed. This treatment model represents a synthesis of trauma-sensitive interventions and well-established CBT principles (Cohen et al., 2001; Deblinger and Heflin, 1996). Trauma-focused cognitive-behavioral therapy was jointly developed by two groups of researchers who have recently worked together to conduct multisite, treatment-outcome studies for sexually abused and otherwise traumatized children. The therapy was developed to resolve posttraumatic stress disorder, and depressive and anxiety symptoms, as well as to address underlying distortions about self-blame, safety, the trustworthiness of others, and the world. The treatment also fits sexual abuse and other traumatic experiences into a broader context of children's lives so that their primary identity is not that of a victim.
Core components of TF-CBT are psychoeducation about child sexual abuse and PTSD; affective modulation skills; individualized stress-management skills; an introduction to the cognitive triad (relationships between thoughts, feeling and behaviors); creating a trauma narrative (a gradual exposure intervention wherein children describe increasingly distressing details of their sexual abuse); cognitive processing; safety skills and education about healthy sexuality; and a parental treatment component. Parents are seen separately from their children for most of the treatment and receive interventions that parallel those provided to the child, along with parenting skills. Several joint parent-child sessions are also included to enhance family communication about sexual abuse and other issues. Most of the TF-CBT treatment studies have consisted of 12 treatment sessions.
Cohen and Mannarino conducted two parallel, randomized, controlled trials for 67 sexually abused preschoolers (3 to 6 years old) and 82 children and young adolescents (7 to 14 years old), comparing TF-CBT to nondirective supportive therapy (NST) (Cohen and Mannarino, 2000, 1998a, 1998b, 1997, 1996a, 1996b; Cohen et al., in press). The nondirective supportive therapy consisted of play for younger children and child- or parent-directed supportive therapy for older children. The preschool study demonstrated the superiority of TF-CBT in improving PTSD symptoms (including sexualized behaviors) and externalizing and internalizing behaviors. These differences were maintained over a one-year follow-up (Cohen and Mannarino, 1997, 1996a).
The strongest mediator of treatment response other than type of treatment was parental emotional distress. At one-year follow-up, the strongest predictor of positive response was parental support of the child (Cohen and Mannarino, 1998a, 1996b). Among treatment completers in the young adolescent study, TF-CBT was superior to NST in improving depression and social competence at the end of treatment, and in improving PTSD and dissociation at one-year follow-up. (Cohen and Mannarino, 1998b; Cohen et al., in press). The strongest mediators of treatment response were parental support of the child and the child's sexual abuse-related attributions (Cohen and Mannarino, 2000).
Deblinger et al. (1996) randomly assigned 100 sexually abused children to standard community care or TF-CBT provided to the child only, the parent only or both. This study documented that TF-CBT provided directly to the child (in either the child-only or parent plus child condition) was superior in improving PTSD symptoms, while TF-CBT provided directly to the parent (in either the parent-only or the parent plus child condition) was superior in improving the child's depressive and behavior problems, as well as in improving positive parenting practices. This was the first study to directly demonstrate the benefit of including a parental treatment component for sexually abused children. These differences were maintained during a two-year follow-up (Deblinger et al., 1999).
Deblinger and colleagues (2001) conducted a group randomized, controlled trial for preschool children, comparing TF-CBT to supportive therapy. Each treatment was provided in parallel child and parent groups. Because of the group setting and the young age of these children, gradual exposure was not included. This study demonstrated that TF-CBT was superior to supportive therapy in improving children's body safety skills and parental distress related to their children's sexual abuse.