Article Text

Download PDFPDF

CBT increases remission of child or adolescent anxiety disorders compared with wait list control
  1. Katharina Manassis
  1. Department of Psychiatry, Hospital for Sick Children, Toronto, Ontario, Canada

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Question: Is cognitive behavioural therapy (CBT) effective for the treatment of childhood and adolescent anxiety disorders?

Outcomes: Primary outcomes Remission of the anxiety disorder, as diagnosed by valid structured interviews for the Diagnostic and Statistical Manual (DSM) or the International Classification of Diseases (ICD) child and adolescent anxiety disorders (Anxiety Disorder Interview Schedule for Parents or Children (ADIS-P and ADIS-C), Diagnostic Interview Schedule for Children, Adolescents and Parents or Clinical Global Impression scale (CGI-I)); and acceptability of treatment as determined by dropout.


Design: Systematic review and meta-analysis.

Data sources: The Cochrane Depression, Anxiety and Neurosis Group Register (CCDANCTR), searched July 2012. This contains trials from the Cochrane Central Register of Controlled Trials (CENTRAL), EMBASE, MEDLINE and PsycINFO from the respective database inception to most recent search date of CCDANCTR (May 2012), supplemented by data from international trial registers, drug companies and handsearch of key journals, conference proceedings and other non-Cochrane systematic reviews and meta-analyses.

Study selection and analysis: RCTs (including cluster and crossover RCTs) were eligible if they examined the use of CBT (at least nine sessions) in children or adolescents (age 4–18 years) with DSM or ICD diagnosis of generalised anxiety disorder (GAD), over-anxious disorder, separation anxiety disorder, social phobia, specific phobia or panic disorder. Eligible comparators were treatment as usual; waiting list; drug placebo; or attention only or psychological treatment involving no CBT element. Analyses were by intention-to-treat. A random effects model was used to calculate standardised mean differences for continuous data and ORs for dichotomous data. Heterogeneity was assessed using χ2 and I2 statistic. Study quality and publication bias were assessed.

Main results

Forty-one RCTs (n=1806) met inclusion criteria and were included in the main analysis. The RCTs addressed separation anxiety disorder (21 RCTs), social phobia (20 RCTs), GAD (17 RCTs), over-anxious disorder (6 RCTs), panic disorder (4 RCTs) and social phobia (4 RCTs). Twenty-six RCTs (n=1350) compared CBT with waiting list control and found that CBT increased odds of remission as compared with waiting list (remission: 59.4% with CBT vs 17.5% with controls; OR 0.13, 95% CI 0.09 to 0.19; heterogeneity: I2=30%, p=0.04). The number needed to treat (NNT) to gain an extra remission with CBT was 6 (95% CI 7.5 to 4.6). Meta-regression suggested that only a small amount of the variability was accounted for by level of comorbidity (6.7%), initial level of anxiety (5.5%), gender (5.32%), participant age (1.06%) or number of CBT sessions (0.23%). There was no difference in outcome based on whether CBT was given on an individual, group or family basis (p=0.97; I2=0). Subgroup analyses according to type of CBT found significant effects for individual CBT, group CBT and family/parental CBT. There was no difference in dropout rates between CBT and waiting list (10.8% vs 10.3%; OR 0.93, 95% CI 0.58 to 1.51; p=0.78). There was no significant difference in remission between CBT and other active comparators (6 RCTs, n=426; OR 0.66, 95% CI 0.34 to 1.29; 6 RCTs) or treatment as usual (2 RCTs, n=88; OR 1.87, 95% CI 0.8 to 4.39; see notes).


CBT increases remission compared with waiting list control in child or adolescent anxiety disorders. A few studies are available comparing CBT with active treatments such as medication.


The OR reported in the results section of the Cochrane review (OR 1.56, 95% CI 0.67 to 3.65) differs from that reported in the analysis figure and summary of findings table (OR 1.87, 95% CI 0.80 to 4.39). Both results indicate no significant difference in remission between CBT and treatment as usual.

Abstracted from


As examined in James and colleagues’ review, cognitive behavioural therapy (CBT) is the most widely studied treatment for childhood anxiety disorders, a highly prevalent and potentially debilitating group of child psychiatric disorders. This meta-analysis builds on existing reviews by distinguishing comparisons of CBT with waitlist from comparisons with other active interventions. It also explores long-term effects, though the dearth of controlled follow-up studies suggests it may be premature to draw conclusions in this area. Moreover, it replicates previous findings regarding CBT response rates (about 59%).

The main conclusion, that anxiety-focused CBT is superior to waitlist, but not necessarily to other active interventions, must be interpreted with caution as there are few studies for some comparisons (eg, two comparisons with treatment as usual and only a single trial that included medication); research samples are often less severely affected than patients in the community (possibly overstating the response rate to non-intensive interventions in research studies) and the authors’ exclusion criteria eliminated about 43% of potentially relevant studies. The inclusion of studies of very young, cognitively immature children and combining parent and child outcome reports may also have been problematic.

In order to improve CBT response rates, the authors state that studies of the active components of CBT, of predictors of CBT response and response to various formats and of CBT adapted to specific disorders and cognitive abilities are needed. However, studies of promising new treatments (eg, attention bias modification1) and further studies of treatment combinations are also indicated to offer hope to those children who do not respond optimally to CBT alone. Access to CBT outside academic centres continues to be a challenge.

Clinicians should, therefore, continue recommending CBT for children with anxiety disorders, and advocate for improved community access to CBT. Perhaps they should feel less guilt about offering bibliotherapy or other active interventions while patients wait for CBT. Findings on other treatments and treatment combinations for CBT non-responders are eagerly anticipated.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • Sources of funding: No external sources of funding reported.


  • Competing interests None.