Effects of a Nonwearable Digital Therapeutic Intervention on Preschoolers With Autism Spectrum Disorder in China: Open-Label Randomized Controlled Trial (Preprint)

Abstract

Autism spectrum disorder (ASD) is a neurodevelopmental disorder that can cause difficulty with communication and social interactions as well as complicated family dynamics. Digital health interventions can reduce treatment costs and promote healthy lifestyle changes. These therapies can be adjunctive or replace traditional treatments. However, issues with cooperation and compliance prevent preschool patients with ASD from applying these tools. In this open-label, randomized controlled trial, we developed a nonwearable digital therapy called virtual reality–incorporated cognitive behavioral therapy (VR-CBT).

The aim of this study was to assess the adjunctive function of VR-CBT by comparing the effects of VR-CBT plus learning style profile (LSP) intervention with those of LSP-only intervention in preschool children with ASD.

This trial was performed in China on 78 preschool children (age 3-6 years, IQ>70) diagnosed with ASD who were randomized to receive a 20-week VR-CBT plus LSP intervention (intervention group, 39/78, 50%) or LSP intervention only (control group, 39/78, 50%). The primary outcome was the change of scores from baseline to week 20, assessed by using the parent-rated Autism Behavior Checklist (ABC). Secondary outcomes included the Childhood Autism Rating Scale (CARS), Attention-Deficit/Hyperactivity Disorder Rating Scale-IV (ADHD-RS-IV), and behavioral performance data (accuracy and reaction time) in go/no-go tasks. All primary and secondary outcomes were analyzed in the intention-to-treat population.

After the intervention, there was an intervention effect on total ABC (β=–5.528; <i>P</i>&lt;.001) and CARS scores (β=–1.365; <i>P</i>=.02). A similar trend was observed in the ABC subscales: sensory (β=–1.133; <i>P</i>=.047), relating (β=–1.512; <i>P</i>=.03), body and object use (β=–1.211; <i>P</i>=.03), and social and self-help (β=–1.593; <i>P</i>=.03). The intervention also showed statistically significant effects in improving behavioral performance (go/no-go task, accuracy, β=2.923; <i>P</i>=.04). Moreover, a significant improvement of ADHD hyperactivity-impulsivity symptoms was observed in 53 children with comorbid ADHD based on ADHD-RS-IV (β=–1.269; <i>P</i>=.02). No statistically significant intervention effect was detected in the language subscale of ABC (β=–.080; <i>P</i>=.83). Intervention group girls had larger improvements in ABC subscales, that is, sensory and body and object use and in the CARS score and accuracy of go/no-go task (all <i>P</i>&lt;.05) than the control group girls. Statistically significant intervention effects could be observed in hyperactivity-impulsivity symptoms in the intervention group boys with comorbid ADHD compared with those in the control group boys (β=–1.333; <i>P</i>=.03).

We found potentially positive effects of nonwearable digital therapy plus LSP on core symptoms associated with ASD, leading to a modest improvement in the function of sensory, motor, and response inhibition, while reducing impulsivity and hyperactivity in preschoolers with both ASD and ADHD. VR-CBT was found to be an effective and feasible adjunctive digital tool.

Chinese Clinical Trial Registry ChiCTR2100053165; http://www.chictr.org.cn/showproj.aspx?proj=137016