Sharma et al 2021 «Sharma S, Hucker A, Matthews T ym. Cognitive behav...»1 aimed to examine the efficacy of CBT for reducing anxiety in autistic youth. Searches for their systematic review and meta-analysis were undertaken from January 1990 until December 2020. They included randomised controlled trials (RCTs) which used CBT to reduce anxiety in autistic youth. 965 studies were identified and 19 RCTs met the inclusion criteria. Sample sizes ranged from 12 to 66 in the intervention groups (mean n=23) and from 10 to 31 (mean n= 18) in the control groups.
Patients: There were total N= 487 participants in the intervention group and N= 346 participants in the control groups (total N= 833 participants) in the 19 studies. The age of participants was up to 18 years, they were diagnosed with ASD according to DSM or ICD criteria, and they had a co-occurring anxiety disorder.
Intervention: The intervention consisted of 5 to 20 CBT treatment sessions (modal number being 16 sessions) typically on a weekly basis. Group CBT was used in 9 studies, individual CBT in 8 studies and 2 had a combination of group and individual sessions. The CBT intervention programs were adapted for autistic youth. The most common CBT interventions were Behavioural Interventions for Anxiety in Children with Autism (BIACA) (k = 4), Exploring Feelings (k = 3), Coping Cat (k = 2) and Cool Kids (k = 2).
Comparisons: The effect of CBT intervention on anxiety was compared to a control group (wait list control, treatment as usual or active control, e.g., counselling). Follow-up data was provided in 6 studies but only one study had follow-up data for controls.
Outcomes: Anxiety was measured by a parental-rated measures in 19 studies, by child-rated measures in 13 studies and by blinded clinician-rated measures in 10 studies. The Spence Children's Anxiety Scale-Parent and Child versions (k = 9) and the Anxiety Disorders Interview Schedule (k = 8) or the Paediatric Anxiety Rating Scale (k = 5) for clinicians were the most commonly used measures.
For clinician rated symptoms random effects meta-analyses revealed a large effect size (g = 0.88, 95% CI 0.55, 1.12, k = 11). The effects sizes (one extreme outlier removed) were significant but smaller for parent (g = 0.40, 95% CI 0.24, 0.56; k = 18) and child-reported anxiety (g = 0.25, 95% CI 0.06, 0.43; k = 13). At the follow-up these benefits were not maintained. In moderator analyses CBT was more efficacious when delivered as individual therapy in clinician ratings and for younger children in clinician and parent ratings. The moderator analyses could not explain the large amount of heterogeneity in studies including clinician ratings.
Risk of bias was measured using the Cochrane Risk of Bias 2 tool. The risk of bias was low in 4/19 (21%) studies, 12/19 (63%) had some concerns and 3/19 (16%) were at high risk of bias. The reporting bias was the main area of concern. Only 7/19 (37%) studies reported registering their protocol. There was little or no evidence of publication bias.
Summary: CBT was efficacious for anxiety in autistic children in the immediate intervention period.
Comments: Our confidence in the results is moderate due to the inconsistency emerged in the magnitude of benefit depending upon the rater (clinician, parent or child). All the assessments used to profile symptoms of anxiety (e.g., SCAS-P) may not have support for their use in the ASD trials without modifications. Conclusions about the long-term benefits cannot be made due to the small number of studies which included this data. The sample sizes in most studies included were small. The participants in these studies had mostly of average to above average intelligence and had sentence-to-conversational-level speech skills (e.g., high-functioning ASD).
Former meta-analyses by Ung et al (2015) «Ung D, Selles R, Small BJ ym. A Systematic Review ...»2, Perihan et al (2020) «Perihan C, Burke M, Bowman-Perrott L ym. Effects o...»3 and Sukhodolsky et al (2013) «Sukhodolsky DG, Bloch MH, Panza KE ym. Cognitive-b...»4 have also aimed to study the effect of CBT on anxiety in autistic youth. However, all the RCTs included in those meta-analyses are included in this meta-analysis.