Pilot Randomized Control Trial of Cognitive Behavioral Therapy for Insomnia Modified for Families with a Child with Autism Spectrum Disorder

Background: Chronic severe insomnia is one of the most common conditions in children with Autism Spectrum Disorder (ASD), affecting 60-80%. Insomnia in ASD have many possible biological, behavioral and cultural mechanisms. Scientists hypothesize that anomalies in the synaptic pathways of the brain may account for “arousal dysregulation” in ASD. Arousal dysregulation may produce a constellation of behavioral symptoms including anxiety, sensory sensitivities, hyperactivity and insomnia. Based on the idea that a subgroup of children with ASD are in a hyper-aroused state, we developed a calming module that included 12 relaxing and soothing activities that could decrease arousal levels. We modified Cognitive Behavioral Therapy for Insomnia (CBT-I) and included the calming module to address the internal factors and sensory sensitivities that threaten sleep. Currently, the standard care (SC) for behavioral insomnia in ASD is a 1-hour parent education session using a Sleep Tool Kit developed by the ATN and this may not be sufficient in all children with ASD to improve their sleep.

Objectives:

Therefore, the specific aims of this pilot Randomized Control Trial (RCT) were to:

1. determine feasibility and acceptability of implementing a CBT-I and SC vs. SC only RCT.

2. estimate the effects of CBT-I and SC vs SC only on sleep parameters and arousal/anxiety symptoms.

Methods:  Sample: 40 parent-child dyads, children ages 6-10 years with insomnia, Two-groups: n=20 SC only (control) vs n=20 CBT-I and SC. SC group received a 1hour education session.. The intervention group received SC plus CBT-I, which included 8 weekly home visits to teach and tailor positive evening routines, calming module, and faded bedtime protocol. Data was collected for both groups using actigraphy, sleep diary, CSHQ, PARS, SCARED, RBS and sensory profile at baseline, week 4 and week 8.

Results:  97 % of families reported that the randomization process and interventions were very acceptable. All children were able to tolerate the actigraph. Baseline mean sleep latency using actigraphy in both groups was over 30 minutes, meeting criteria for insomnia (36.73 min vs. 40.11 min). Both groups’ mean CSHQ score was above 42, indicating sleep problems. After intervention, CBT-I group had greater decrease in wake minutes (-65.68 min vs. -18.69 min, p=0.037), increase in sleep minutes (46.13 min vs. 8.24 min, p=0.073) and increase in percent sleep (10.61% vs. 2.26%, p=0.034) as compared to SC only. Sleep latency decreased for both groups (SC and CBT-I, -17.01 min, p=0.001; SC, -7.53min, p=0.083), but CBT-I group experienced a clinically significant decrease, with the average post-intervention sleep latency falling below the cut-off for insomnia (19.71 min).CBT-I parents reported a greater decrease in night wakings on the CSHQ than SC parents (-1.32 vs. -0.19, p=0.046).

Conclusions:  This pilot RCT was acceptable and feasible. The SC intervention was effective at decreasing sleep latency and is consistent with previously published data. CBT-I group had more consolidated sleep, shorter sleep latency and fewer wake minutes than SC only group. CBT-I was effective in significantly improving night wakings which has been a difficult problem to solve for families with ASD.