Library Home

SMART Study

A sequential, multiple-assignment randomized trial asking a deceptively simple question: which anxiety treatment should come first for a child — and what should happen when it isn't enough. Browse the study by audience below, or walk through it as a participant in Treatment pathways.

At a glance 316 enrolled 8–17 years old 9 clinical sites 24 treatment weeks 10 years follow-up

OverviewThe Study

Treating childhood anxiety

Anxiety disorders are among the most common mental-health conditions of childhood, affecting roughly one in eight children. Two gold-standard treatments exist — cognitive behavioral therapy and fluoxetine — but the best way to sequence them was unknown.

Led by Dr. Bradley S. Peterson at the Institute for the Developing Mind, the study enrolled 316 children across 9 clinical sites throughout greater Los Angeles, and was designed to serve underserved, ethnically diverse families. Funded by PCORI.

DesignThe Question

Which treatment first?

CBT and fluoxetine each help about 60% of children. The study doesn't crown a winner — it tests the sequences: what to start with, and what to do next when the first treatment falls short.

Participants are randomized twice: once at the start, and again at Week 12 if anxiety hasn't remitted. Four sequences emerge — CBT→CBT, CBT→Med, Med→Med, Med→CBT — and the study compares them all.

More +

Week 0Stage 1

Initial treatment

Children are randomly assigned 50/50 to either CBT or fluoxetine — the two gold standards for pediatric anxiety.

Week 12Assessment

Reassessment point

Each child is assessed for clinical remission. Remitters continue; non-remitters enter a second randomization.

Wks 12–24Stage 2

Treatment adaptation

Non-remitters either intensify their current treatment or add the other one as a combination approach.

Week 24+Outcomes

Findings & follow-up

Final assessments compare all four sequences. Annual follow-up continues for up to 10 years.

Structure · 4 cards

At a glance 1 in 8 kids 8–17 years old 2 ways to feel better

Hey thereFor Kids & Teens

Your feelings matter

Lots of kids feel worried or nervous sometimes — that's totally normal. But sometimes worry gets so big it makes school, friends, or new things feel really hard.

When that happens, doctors know ways to help you feel like yourself again. You don't have to figure it out alone.

?The Basics

What is anxiety?

Anxiety is your brain's alarm system going off a little too often — even when there's no real danger. It's not your fault, and it can get better.

1 in 8Fast Fact

You're not alone

About one in every eight kids deals with big anxiety. That's probably a few kids in your class — you just can't always tell.

At a glance ~60% CBT response ~60% SSRI response 24 wks active treatment Adaptive design

OverviewFor Parents

Understanding childhood anxiety

Clinical anxiety affects roughly one in eight children, shaping school performance, friendships, and family life. The good news: effective, evidence-based treatments exist — and this research clarifies how to use them well.

CBT · RxTreatment

Two gold standards

Cognitive behavioral therapy and SSRI medication each have strong research support, with response rates near 60%. Neither is universally "first."

Because both work about equally well on average, the right starting point depends on your family — access, cost, timing, and your child's own preferences.

More +

2×Design

An adaptive approach

Treatment starts with one approach, is reassessed at Week 12, and adjusts based on how your child actually responds — intensify, or add the other treatment.

This mirrors what good clinical care already does informally — but here the adjustment is planned, measured, and compared across every possible order.

More +

At a glance 3 effective options 3–10% differences between options Week 12 reassessment 24 wks of treatment

OverviewFor Providers

What the study means for your practice

Among clinically complex, demographically diverse youths treated in primary care and mental health clinics, fluoxetine, CBT, and their combination all substantially improved anxiety symptoms and functional impairment over 24 weeks.

Differences between the options were relatively small — on average 3–10% relative to baseline. Until specific patient characteristics are found to predict much greater benefit from a particular treatment, the authors suggest choosing the first treatment and its sequence by patient preference and practical considerations: availability, patient burden, potential side effects, and expense.

Week 12Next Step

If anxiety hasn't remitted

If anxiety had not remitted by week 12, adding the other treatment tended to improve anxiety and impairment more than optimizing the first one — though the effects were small across sequences.

Starting with CBT, then adding fluoxetine, gave the greatest improvements, particularly compared with continuing CBT alone. If CBT is available and preferred but hasn't produced remission by month 3, the authors advise considering medication — especially for racial and ethnic minority youths.

More +

24 wksTiming

Give treatment time

Improvement was roughly linear: the second 12 weeks brought gains comparable to the first. The 12–16 weeks commonly allotted to exposure-based CBT appear inadequate.

Judged at week 12, parent reports would have misleadingly suggested fluoxetine was superior to CBT. CBT needed more time to reach or surpass fluoxetine-level effectiveness.

More +

Findings

31.7%Finding 01

Anxiety reduction

Anxiety severity fell across every treatment pathway — CBT-first and medication-first alike.

Measured with the SCARED scale — a validated 41-item questionnaire completed by both children and parents at each major visit.

More +

AllFinding 02

Every pathway improved

Whether children began with therapy or medication, meaningful clinical improvement followed.

Consistent with prior evidence of ~60% response rates for each treatment alone. Parents reported earlier perceived improvement in the medication arm during the first six weeks.

More +

4 pathsFinding 03

Sequencing matters

The order of treatments — and how they adapt when the first falls short — meaningfully shapes long-term outcomes.

Minority youth showed different response patterns across sequencing strategies — a key equity-focused finding the study was designed to surface.

More +

24 wksFinding 04

Longer holds better

Children who continued through all 24 weeks showed more durable improvement than those assessed earlier.

CBT showed particular strength in long-term functional outcomes — school, social, and family functioning — beyond symptom reduction alone.

More +

Clinical implications

RxPrimary Care

Medication as a first step

Exposure-based CBT is difficult to access, costly, and demands considerable patient time and effort — making medication a reasonable first-line treatment in primary care settings.

Response was faster when treatment started with fluoxetine, especially on parent reports at weeks 6 and 12. All treatments were well tolerated; reduced appetite was more common with medication.

More +

CBTCommunity Care

CBT works in the community

Medication and CBT produced similar improvement, and CBT adherence was acceptable — even with therapists who had limited prior CBT expertise and a 2-day training.

The authors conclude that CBT is a viable option for community intervention, when available. Some data in adults suggest CBT may be more cost-effective over the long term.

More +

Youth · ParentMeasurement

Ask youth and parents

Youth and parent reports of anxiety often differ moderately. Both are valid, and they complement each other.

Parent reports better capture what is visible in behavior, such as avoidance; youth reports better capture the inner experience of fear and worry.

More +

EquityWho Benefits

Background can shape response

Average differences between treatments were small, but could be several times larger within race and ethnicity subgroups.

In general, non-Hispanic White youths may benefit more from medication at all stages of treatment; racial and ethnic minority youths may benefit more from combination treatment.

More +

SMART Study

Treating childhood anxiety

Which treatment first?

Initial treatment

Reassessment point

Treatment adaptation

Findings & follow-up

Your feelings matter

What is anxiety?

You're not alone

Understanding childhood anxiety

Two gold standards

An adaptive approach

What the study means for your practice

If anxiety hasn't remitted

Give treatment time

Supplemental materials

Findings

Anxiety reduction

Every pathway improved

Sequencing matters

Longer holds better

Talking helps

Medicine can help too

The timeline

Working with your clinician

Clinical implications

Medication as a first step

CBT works in the community

Ask youth and parents

Background can shape response