CBT, medication and the combination are effective for childhood anxiety

Commentary on: Wang Z, Whiteside SPH, Sim L, et al. Comparative Effectiveness and Safety of Cognitive Behavioral Therapy and Pharmacotherapy for Childhood Anxiety Disorders. A Systematic Review and Meta-analysis. JAMA Pediatrics. 2017;171:1049–56.

What is already known on this topic?

Anxiety disorders are the most common mental health disorders among children¹ and adolescents,² characterised by excessive and persistent fear and anxiety that is difficult to control and negatively impacts daily life.³ Cognitive-behavioural therapy (CBT) and selective serotonin reuptake inhibitors (SSRIs) are often recommended as first-line interventions in reducing anxiety symptoms and improving function,⁴ however, comparative effectiveness has not been determined.

Methods of the study

The present study⁵ conducted a systematic review and meta-analysis on effectiveness and safety of CBT, pharmacotherapy, and the combination for childhood anxiety disorders. Studies were included if (1) They treated children or adolescents between the ages of 3 years and 18 years with one or more diagnosed anxiety disorder (excluding trials solely treating PTSD or OCD). (2) They received CBT or any medication alone or in combination. (3) They had at least one control group (CBT, medication, pill placebo, wait-list/no treatment, or attention control/treatment as usual). A literature search yielded 115 RCTs (randomized controlled trial) and non-randomised comparative studies (with 7719 participants) that were appropriate for inclusion. The primary outcome was primary anxiety symptoms defined by standardised measures completed by child, parent or clinician. Secondary outcomes included treatment response, remission and adverse events. SSRIs and CBT significantly improved primary anxiety symptoms, remission and response rates compared with pill placebo, and wait-list/no-treatment, respectively. The combination of sertraline and CBT significantly improved anxiety (symptoms and response) compared with either treatment alone. SNRIs  (serotonin norepinephrine reuptake inhibtors) significantly reduced clinician-rated anxiety symptoms based on limited support, but benzodiazepines and tricyclics had no significant effect on anxiety symptoms. Non-serious adverse events were common with medications but not CBT. Fewer dropouts were associated with CBT compared with pill placebo or medication.

What this paper adds?

• The meta-analysis provides an update on estimates of efficacy of existing anxiety treatments for children including CBT, SSRIs and the combination.

• The authors suggest a need for head-to-head comparisons between medications and in combination with CBT, and further evaluation of suicidality in medication trials.

• Fewer dropouts were associated with CBT than medication or pill placebo suggesting that CBT may be better tolerated than medications for anxious youth.

Limitations

• Reports of primary anxiety symptoms were collected from child, parent and clinician, and compared. Thus, results varied depending on the informant.

• Pooling non-randomised comparative studies with randomised controlled trials may affect the accuracy of the findings, given the level of bias and uncertainty surrounding treatment effects in observational studies.

• Although blinding and publication bias were acknowledged in the discussion, the authors may have been less conservative in their use of the GRADE criteria for assessing quality of evidence than contemporary standards.

• Combining various antidepressants into classes of SSRIs or SNRIs may hinder the ability to examine differential effects of these drugs individually.

• Few adverse events were reported for CBT, however, most CBT studies did not evaluate for the occurrence of adverse events like embarrassment about talking about anxiety, distress from exposures and worsening of symptoms.

What next in research?

In addition to addressing the above limitations and extending this research…

• There is a strong need for the development of adverse event reporting measures and monitoring systems related to CBT as adverse events of behavioural therapy have not been studied as closely as adverse events from pharmacological approaches.

• Future research should examine and refine existing treatments to maximise intervention effects. For example, using the most robust components of CBT while minimising those which provide little additional benefit.

• Intervention acceptability for CBT, SSRIs and the combination should be studied in more detail for this population.

Do these results change your practices and why?

• No. The results reiterate the effectiveness of CBT and SSRIs for reducing childhood anxiety symptoms (CBT alone may yield relatively greater tolerability compared with SSRIs) and provide initial evidence supporting the effectiveness of SNRIs, however the methodological concerns about the present meta-analysis limit considerably the validity of these findings. In clinical practice, clinicians should consider assessing adverse events and concerns during CBT and other psychotherapies.