A randomised controlled trial of acceptance and commitment therapy for anxious adolescents: effectiveness and mechanisms for change

Research Doctorate - Doctor of Philosophy (PhD)Anxiety disorders affect approximately 10-30% of children and adolescents. While traditional Cognitive Behaviour Therapy (CBT) is the first-line psychosocial treatment for children with anxiety, a significant proportion are nonresponsive or exhibit residual symptomatology at treatment cessation. Acceptance and Commitment Therapy (ACT) has been found to be effective among adults with anxiety disorders and children with other psychiatric conditions. ACT fosters psychological flexibility via putative mediators of change that form a “hexaflex” model: acceptance, defusion, mindfulness, self-as-context, committed action and valued living. This research examined ACT versus CBT in the treatment of anxiety disorders among children and adolescents. Among adolescents, an exploratory investigation of ACT mediators for change was undertaken. One-hundred-and-ninety-three children were block randomised to a manualised 10-week group format ACT or CBT program, or to waitlist control (WLC). Repeated clinical – clinician/self/parent-reported anxiety, depression and problem behaviours – and quality of life (QOL) measures – anxiety interference, psychosocial and physical health-related QOL – were taken pretreatment, posttreatment and 3-month follow-up (3MFU). Completers were 157 children, 58% female, with a mean age of 11 years (SD = 2.8). Completer and intention-to-treat (ITT) analyses revealed ACT and CBT were both superior to WLC across outcomes, reflecting statistically and clinically significant differences, with gains maintained at 3MFU. While WLC improved significantly on some outcomes at posttreatment, improvements were not clinically significant. Both completer and ITT analyses found ACT and CBT to produce similar outcomes. However, on ITT 3MFU results, CBT evidenced significantly lower scores on clinician-, but not self- or parent-reported outcomes. Mediation results were mixed. The hexaflex mediated the relationship between treatment and clinician-rated anxiety severity for ACT only; with treatment common effects observed for depression and self-reported anxiety. Acceptance and defusion emerged as specific mediators and evidenced the same pattern of effects, with clinician-rated anxiety effects treatment common. Hexaflex effects were accounted for by acceptance and defusion, as all other process measures were nonsignificant. Mediation analyses for parent-rated and QOL outcomes were nonsignificant. Few changes in process measures were observed post to 3MFU and mediation effects were nonsignificant. In conclusion, ACT and CBT are both effective in improving clinical and QOL outcomes among children with anxiety. Despite mixed results, there was some evidence for acceptance and defusion as treatment common change mediators. Limited support was obtained for the hexaflex model, the processes of valued action and mindfulness/self-as-context and the treatment specificity of mediation effects. ACT may be a viable alternative evidence based treatment option for clinicians working with children with anxiety disorders. Despite their differences ACT and CBT may be underpinned by analogous mechanisms