CBT and fluoxetine show comparable improvement for pediatric anxiety

Children and adolescents with anxiety disorders experienced substantial improvement with either cognitive behavioral therapy (CBT) or fluoxetine, with no clinically meaningful difference between the 2 as initial treatment, according to findings from a randomized clinical trial published in the American Journal of Psychiatry.¹

For Bradley S. Peterson, MD, chief of child and adolescent psychiatry at the University of Southern California and Children's Hospital Los Angeles, the findings provide reassurance that clinicians have more than 1 effective approach.²

"The good news is that kids get better. Kids who have anxiety disorders, they get better with treatment," Peterson said.

Is CBT or medication better for initial treatment?

The 24-week sequential multiple assignment randomized trial included 316 youths aged 8 to 17 years with severe anxiety. Participants were initially randomized to 12 weeks of exposure-based CBT or fluoxetine. Those who did not achieve remission were then randomized to continue and optimize the initial treatment or add the other therapy.

The population reflected patients commonly encountered in clinical practice. Approximately 64% were Hispanic or Latino, 56% had Medicaid coverage, 67% had co-occurring depression, and 46% had ADHD.

Overall, youth-reported anxiety scores declined 31.7% over 24 weeks. Initial CBT and fluoxetine did not differ significantly on the primary outcome, with CBT providing a small numerical advantage.

"They came out pretty comparable head to head. I mean, there was a little bit of advantage for one over the other, but not really meaningful clinically," Peterson said. "But kids got substantially better over the six months of treatment."

What should clinicians do if initial treatment is not enough?

Few participants achieved the study's stringent remission criteria after 12 weeks, and most required additional treatment. For these patients, combination treatment did not significantly outperform optimized monotherapy on the primary outcome at week 24.

"You can either stick with what you initially started with, or you can go to combination treatment," Peterson said. "There was some slight advantage to going to combination treatment, however."

The sequence of treatment provided additional information. Among the 4 approaches evaluated, starting with CBT and adding fluoxetine after 12 weeks produced the greatest improvement across most primary and secondary measures, although differences on the primary youth-reported anxiety outcome were not statistically significant.

"It was better if you started with CBT and then went to combination treatment," Peterson said. "And that is a sizable difference, a clinically meaningful difference."

Peterson also noted that analyses conducted after the published study suggested differences between treatment sequences became more pronounced during subsequent follow-up; those findings were not part of the published trial results.

How can treatment decisions incorporate family preferences?

Because CBT, fluoxetine, and combination treatment all produced improvement, the findings support considering patient preference, treatment availability, burden, potential adverse effects, and cost when choosing initial therapy. Exposure-based CBT can be difficult to access in some communities, making medication a reasonable option when CBT is unavailable.

"Some kids and some families may prefer one over the other. Some prefer medication over CBT. Some prefer CBT over medication," Peterson said. "And really, I think families should be supported in that decision because the findings suggest that both are comparably helpful on average."

Could patient characteristics help guide treatment?

The study also identified differences by race and ethnicity on secondary outcomes. Non-Hispanic White youths experienced greater benefit from initiating and continuing fluoxetine, whereas racial and ethnic minority youths showed greater benefit from transitioning to combination treatment.

Peterson said the finding warrants additional investigation rather than establishing a treatment rule based on race or ethnicity.

"It suggests that there may be some characteristics of some kids that we're going to learn about, like this, like race and ethnicity, that may guide us toward one of the treatments rather than another to provide the best chances of improving with anxiety," he said.

For pediatricians, the broader finding is that both established treatments can substantially reduce anxiety in clinically complex children, while reassessment over time can help determine whether continuing treatment or adding another modality is appropriate.

References

1. Peterson BS, West AE, Weersing VR, et al. A pragmatic SMART study of medication and CBT sequencing in pediatric anxiety disorders: a randomized clinical trial. Am J Psychiatry. Published online 2026. doi:10.1176/appi.ajp.20251037.

2. Children’s Hospital of Los Angeles. Children’s Hospital Los Angeles-Led Study Finds Multiple Effective Options for Pediatric Anxiety. Children’s Hospital of Los Angeles. June 25, 2026. Accessed August 26, 2026. https://www.newswise.com/articles/children-s-hospital-los-angeles-led-study-finds-multiple-effective-options-for-pediatric-anxiety/?sc=dwrecomm&xy=10069478&wt=dw_r_25062026