2.1 Background
Anxiety is a common problem in school‐aged children with ASD. CBT and other psychosocial interventions have been developed as alternatives to pharmacological intervention to treat anxiety symptoms in students with ASD without co‐occurring intellectual disability. This present synthesis of evidence is a systematic review and meta‐analysis examining the efficacy of interventions for reducing anxiety among school‐aged children with ASD.
2.2 Objectives
This review aims to address the question of what the relative effectiveness of interventions is for managing anxiety of school‐aged children with ASD in school, family, and clinical settings.
2.3 Search methods
The following databases were searched for references from 1996 up to 31 December 2018: EBSCO (including Academic Search Complete, British Education Index, CINHAHL, Education Research Complete, ERIC, PsychINFO, and SocINDEX), Informit (A + Education), Elsevier (including EMBASE and SCOPUS), PubMed and Proquest (CBCA Complete). We also searched the reference lists of published and unpublished literature papers, as well as gray literature sources, selected websites, trial registries, and experts in the field of autism to inquire about studies.
2.4 Selection criteria
Studies were included in the review if they met the following criteria.
1. The patient/client population was school‐aged children (5 to 18 years old) diagnosed with ASD (inclusive of autism, ASD, Autistic Disorder, Asperger's Disorder, Asperger Syndrome, atypical autism, and PDD‐NOS) by a professional eligible to diagnose these conditions, and also experiencing anxiety symptoms or a diagnosis of an anxiety disorder provided by a professional eligible to diagnose such conditions.
2. The intervention was focused on reducing anxiety symptoms and included at least one of the following seven elements: (a) psychoeducation, (b) exposure, (c) cognitive restructuring, (d) parent training or parent psychoeducation, (e) relaxation, (f) modeling, and (g) self‐monitoring.
3. At least one outcome measure was a standardized continuous measure of anxiety (parent, clinician or self‐reported).
4. The study was published between the years 1996 and 2018.
2.5 Data collection and analysis
Four authors independently selected and appraised studies for inclusion, while two authors evaluated the risk of bias in each subsequently included study. All outcome data were continuous, from which standardized mean difference effect sizes were calculated. We conducted random effects meta‐analysis, which means we assumed individual studies would provide different estimates of treatment effects. Where outlier studies were identified, analyses were repeated after the outlier had been removed from the list of studies. Analyses were conducted separately according to the respondent on the outcome measure of anxiety: clinician, parent or subject (child or youth). Moderator analyses were undertaken to examine differences in effect sizes depending on whether or not the family was involved and whether treatment occurred in groups or individually.
2.6 Results
Eighteen randomized controlled trials (RCTs) and six quasi‐experimental studies met the inclusion criteria. These studies evaluated the effects of interventions targeting anxiety in 931 (764 male and 167 females) participants aged 3–19 years. Overall, the effects of interventions on anxiety were statistically significant and of moderate to high effectiveness, compared to waitlist and treatment‐as‐usual control conditions at posttreatment (standardized mean difference after removal of outliers SMD = −0.71, 95% confidence interval [CI]: −0.97, −0.46; z = −5.42, p < .01), where SMDs of 0.05, 0.19, 0.45, and 0.70 were taken to be indicative of low, moderate, high, and very high effects, respectively. Results also suggested the reported effectiveness of treatment varied as a function of the informant on outcome measures—clinician reports indicate a very high statistically significant effect (SMD = −0.84, 95% CI: −1.15, −0.54; z = −5.43, p < .01), while parent reports indicate a high significant effect (SMD = −0.53, 95% CI: −0.76, −0.31; z = −4.73, p < .01). Results based on the subjects’ self‐reports indicated a moderate significant effect on the reduction of anxiety (SMD = −0.35, 95% CI: −0.55, −0.15; z = −3.41, p = .001).
Moderators indicated larger effects for treatments that involved parents (SMD = −0.74, 95% CI −1.06, −0.42; z = −4.55, p < .01) than for student‐only interventions (SMD = −0.60, 95% CI −1.03, −0.17; z = −2.73, p < .01). Treatments that were administered individually one‐on‐one (SMD = −1.24, 95% CI −1.75, −0.74; z = −4.87, p < .01), indicated larger effects than for treatments delivered in a group context with peers (SMD = −0.37, 95% CI −0.54, −0.19; z = −4.10, p < .01).
No adverse events were reported. Given the nature of the interventions and the selected outcome measures, the risk of performance and detection bias are generally high, particularly for those studies that used outcome measures based on parent and self‐reports.
2.7 Authors’ conclusions
There is evidence that CBT is an effective behavioral treatment for anxiety in some children and youth with ASD without co‐occurring intellectual disability. Evidence for other psychoeducational interventions is more limited, not just due to the popularity of CBT but also due to the quality of the smaller number of non‐CBT studies available.
While there is evidence that CBT is an effective behavioral treatment for anxiety in some children and youth with ASD, work remains to be done in terms of identifying the characteristics of these interventions that contribute to their effectiveness and identifying the characteristics of participants who are more likely to respond to such interventions.
Oversett med Google Translate
