A Pragmatic SMART Study of Medication and CBT Sequencing in Pediatric Anxiety Disorders: A Randomized Clinical Trial

OBJECTIVE: Clinicians treating pediatric anxiety disorders have inadequate information to decide which evidence-based treatment should initiate care and what strategies to adopt when initial care proves to be insufficient. The authors sought to determine whether beginning treatment with fluoxetine or exposure-based CBT yielded better outcomes, and, if 3 months of treatment failed to produce remission, whether optimizing the initial treatment or adding the other treatment modality (combination therapy) yielded better outcomes. METHODS: anxiety disorder; age 8-17; at least 2nd-grade reading level; and not receiving anxiety treatments. The primary outcome was score on the youth-report 41-item Screen for Child Anxiety Related Emotional Disorders (SCARED). Secondary outcomes were score on the parent-report SCARED and scores on the youth- and parent-report Child Anxiety Impact Scale. The authors hypothesized better outcomes from initial treatment with fluoxetine followed by combination treatment. RESULTS: A total of 316 youths with severe anxiety who had high levels of sociodemographic disadvantage and co-occurring diagnoses were randomized. Overall, youth-reported SCARED scores declined 31.7% over the 24-week trial. Improvement did not differ significantly by initial treatment, although CBT afforded a nonsignificant advantage over medication (24-week difference in mean group change: 1.45, 95% CI=-2.25, 5.16). In week-12 nonremitters, combination treatment did not yield superior week-24 improvement over continuing monotherapy (group difference in mean change from baseline: -2.74, 95% CI=-6.53, 1.05). Initial treatment with CBT followed by combination therapy significantly separated from other sequences on a subset of measures over 24 weeks. Non-Hispanic White youths benefited more from initiating treatment with fluoxetine and continuing it after week 12, whereas racial and ethnic minority youths benefited more from transitioning to combination therapy after week 12. CONCLUSIONS: This pragmatic SMART study shows that symptom reduction is similar for fluoxetine, CBT, their combination, and variations in sequencing when delivered to youths with anxiety disorders. Patients, families, and clinicians have a range of treatment options and may optimize outcomes by stakeholder preference.