Group CBT Best Psychotherapy for Acute Anxiety in Kids?

Batya Swift Yasgur, MA, LSW

November 07, 2018

Group cognitive behavioral therapy (CBT) is the most appropriate psychotherapy for children and adolescents with anxiety disorders, new research suggests.

Using network meta-analysis methodology, investigators conducted an analysis of 101 studies encompassing over 6600 participants that compared 11 different psychotherapies with four specific control conditions.

Although the certainty of evidence was rated as "low to very low," most psychotherapies were still found to be significantly more effective than waitlist conditions.

Moreover, only group CBT was significantly more effective than other psychotherapies and all the control conditions.

"Group cognitive therapy is probably the best initial choice of psychotherapy for anxiety disorders in children and adolescents," author Andrea Cipriani, MD, PhD, professor of psychiatry, Oxford University, United Kingdom, told Medscape Medical News.

"These results should be replicated in future research with focus on disorder-specific psychotherapies and identification of moderators of treatment effect," he said.

The study was published online October 31 in JAMA Psychiatry.

Addressing Uncertainty

Psychological treatments, especially CBT, are commonly used to treat anxiety disorders in children and adolescents, the authors write.

Other psychotherapies including behavioral therapy (BT) and bibliotherapy are also used, but there is ongoing debate regarding the different components and format of psychotherapy; specifically, whether cognitive maturity is required for successful engagement in CBT for young children and whether differences exist in efficacy between psychotherapy delivered individually or in group settings.

"These issues lead to uncertainty in the decision making for healthcare professionals and patients," which have not been adequately addressed by previous pairwise meta-analyses, the investigators note.

To address previous gaps, the researchers used network meta-analysis, which "allows for better data synthesis because indirect comparisons can be made."

Using this approach, they "aimed to comprehensively compare and rank psychological interventions for the acute treatment of anxiety disorders in children and adolescents."

The researchers began with a comprehensive literature search of studies through November 30, 2017.

A study was required to include any structured psychotherapy for the acute treatment of children and adolescents (≤ 18 years old) with a primary diagnosis of anxiety disorders.

Psychotherapy was considered "structured" when it was accompanied by an explicit manual for therapists to follow and/or laid out in a manual for participants.

The anxiety disorders included generalized anxiety disorder, social anxiety disorders, specific phobias, panic disorder, separation anxiety disorder, and selective mutism.

Trials were required to be ≥ 6 weeks in duration and have a sample size of ≥ 10 patients.

As there are a variety of different delivery approaches in the various modalities (face-to-face or Internet-assisted), different modalities (childhood psychotherapy, parental involvement, or parent-only therapy), and formats (group, individual, or both) leading to different treatment effects, the researchers decided to consider them as independent nodes in the network meta-analysis.

"Parental involvement" was defined as parent attendance ≥ 40% of total sessions of children and parents, and at ≥ 40% involvement in each session.

Control conditions included no treatment; psychological placebo (defined as "a control condition that was regarded as inactive by the researchers but was presented to the participants as being an active therapy); treatment as usual (which included any nonstructured psychotherapy that might have some treatment effects); and waitlist conditions.

Efficacy post-treatment and at follow-up was assessed as mean change scores in anxiety symptoms from baseline to endpoint and from baseline to the end of follow-up (≤ 12 months).

Anxiety symptoms were measured using a variety of psychometrically continuous scales (for example, the Revised Children's Manifest Anxiety Scale, Spence Children's Anxiety Scale, and Pediatric Anxiety Rating Scale).

High Risk of Bias

The analysis included a total of 101 unique randomized clinical trials involving 6625 unique patients and encompassing 11 different psychotherapies: group BT, individual and group BT, individual BT with parental involvement, group CBT, group CBT with parental involvement, individual CBT, individual and group CBT, individual CBT with parental involvement, Internet-assisted CBT, parent-only CBT, and bibliotherapy CBT.

The studies, which were conducted in 14 countries, were published from 1994 to 2017.

Of the studies, 75 studies (74.3%) included patients with mixed anxiety disorders.

The median study sample size was 54 patients (range, 11 - 267 patients), with a mean (SD) age of 10.8 (3.0) years.

Approximately half of total participants (3350, 50.6%) were female, with a median proportion of female participants of 52% (range, 8%-100%).

Twenty trials enrolled only children, 49 enrolled only adolescents, and the remainder enrolled both children and adolescents.

The median duration of acute treatment was 12 weeks (range, 6-32), with a median number of sessions of 12 (range, 6 - 32), and a median number of sessions with family involvement of 4 (range, 0 - 20).

The median duration of the longest follow-up was 6 months (range, 1 - 12 months).

Most trials were of moderate or high risk of bias (72 [71.3%] and 21 [20.8%], respectively), with only 8 (7.9%) at low risk of bias.

Low-Quality Evidence

Pairwise meta-analysis revealed that for efficacy, group CBT, individual CBT, and parental involvement CBT were significantly more efficacious than the waitlist condition, both post-treatment and at follow-up.

Bibliotherapy CBT was found to be less acceptable than group CBT with parental involvement and the waitlist condition.

Group CBT with parental involvement, individual CBT, individual and group BT, internet-assisted CBT, and parent-only CBT were found to be significantly more beneficial than either the waitlist condition or psychological placebo in quality of life (QOL)/functioning.

The network meta-analysis revealed that, in terms of efficacy post-treatment, all psychotherapies were more beneficial than the waitlist condition, but only group CBT was significantly more effective than all neutral control conditions (standardized mean differences [SMD] range, −1.43 to −0.76) and most other psychotherapies (SMD range, −0.82 to −0.43).

At the end of follow-up, almost all of the investigated psychotherapies were significantly more effective than the waitlist condition and no treatment (SMD range, −2.80 to −1.64), but only group CBT was significantly more effective than group CBT with parental involvement and all control conditions at short-term follow-up (SMD range, −0.43 to −0.82).

Psychological placebo was significantly more efficacious than the waitlist condition, both at post-treatment and follow-up.

Only bibliotherapy CBT had significantly more all-cause discontinuations than some of the other psychotherapies and control conditions (range of odds ratios, 2.48-9.32).

Almost all CBT interventions, with the exception of BT, showed significantly more benefit in QOL/functioning compared with psychological placebo and the waitlist condition (SMD range, 0.73-1.99).

Although the common heterogeneity SDs were "relatively high," they were still "within the empirically estimated distributions," with a common heterogeneity SD of 0.65 (95% credible interval [CrI], 0.54 - 0.77) for efficacy post-treatment; 0.63 (95% CrI, 0.43 - 0.89) for efficacy at follow-up; 0.49 (95% CrI, 0.20 - 0.75) for acceptability; and 0.51 (95% CrI, 0.33 - 0.76) for QOL/functioning.

The test of global inconsistency did not show a significant difference between the consistency and inconsistency models for efficacy post-treatment (P = .50) but did show a significant difference for efficacy at follow-up (P < .001).

Group CBT Best Option

Sensitivity analyses revealed that, in terms of efficacy post-treatment, the most effective treatments were group CBT (93.4%) and group BT (86.1%), while the least effective was the waitlist condition (2.4%).

In terms of efficacy at follow-up, the most effective treatments were parent-only CBT, individual BT with parental involvement, and Internet-assisted CBT (67.9%, 66.1%, and 65.6% respectively), whereas no treatment was least effective (1.5%).

"According to the GRADE [Grading of Recommendations Assessment, Development and Evaluation] framework, the certainty of the evidence for efficacy was low for most comparisons and very low for some comparisons," the authors write.

"The delivery formats of psychotherapy for anxiety disorders in children versus adolescents in still under wide debate," they add.

"In our subgroup analyses we found different point estimates for group CBT for adolescents (mean age, ≥ 13 years; SMD, −0.82) versus younger patients (mean age, < 13 years; SMD, −0.50); however, the corresponding test for subgroup difference was nonsignificant (P = .45)," they add.

"We found that only group CBT was significantly more effective in reducing anxiety symptoms than other psychotherapies and all control conditions post-treatment and at short-term follow-up," said Cipriani.

More Research Needed

Commenting on the study for Medscape Medical News, John T. Walkup, MD, professor of psychiatry at the Feinberg School of Medicine, Northwestern University, and chair, Department of Child and Adolescent Psychiatry at Lurie Children's Hospital of Chicago, Illinois, who was not involved with the study, expressed concern that "although there were a lot of studies, the quality of the studies was poor in general and it is very difficult for the investigators to come to firm conclusions about much, especially that group forms of treatment are superior to other forms of CBT."

The process of looking in-depth at high-quality studies is "likely to be much more informative than looking at a large number of studies of variable quality," said Walkup.

As CBT has been proven to be effective, "the right conclusion, from a review of the CBT literature, is to match the young person and their family to an effective CBT intervention delivered by a skilled therapist," Walkup said.

He noted that CBT is "a treatment that all evidence-based practitioners understand to be effective and is considered to be the treatment of choice for the childhood anxiety disorders," especially social anxiety disorder, "as the group format preferentially works to reduce anxiety associated with social interactions."

The authors acknowledge that more research is needed to confirm their conclusions and that healthcare professionals, patients, and families "should carefully interpret these findings, bearing in mind the limited amount of information and the low quality of available evidence."

This study was supported by the National Institute for Health Research (NIHR) Oxford Cognitive Health Clinical Research Facility, NIHR Oxford Health Biomedical Research Centre, National Key Research and Development Program of China, and National Natural Science Foundation of China. Cipriani reports being an NIHR research professor. Disclosures for the other authors are listed in the article. Walkup reports receiving royalties from Wolters Kluwer for his contributions to UpToDate and funding from the National Institute of Mental Health and Pfizer for previous research.

JAMA Psychiatry. Published online October 31, 2018. Full text

For more Medscape Psychiatry news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: