Online CBT for insomnia adapted for autistic adults

Insomnia is not a minor inconvenience for autistic adults: an estimated 60% report insomnia symptoms, and the abstract notes this significantly impairs daily functioning and wellbeing. Cognitive behavioural therapy for insomnia is the usual first-line non-drug approach, but the standard programmes were not built with autistic users in mind, and face-to-face therapy is not always accessible. That raises a practical question — can a guided self-help internet programme, adapted with autistic input, shift insomnia in this group? This trial set out to test exactly that, and it is described as the first randomised controlled trial of autism-adapted internet CBT for insomnia.

The researchers adapted an existing non-autism-specific programme, i-Sleep, in co-creation with seven autistic adults who had sleeping problems, producing a 5-week guided self-help internet CBT-based intervention called i-Sleep Autism. They randomly allocated 163 adults with autism and insomnia (Insomnia Severity Index of 10 or above) to the intervention (n=82) or a waitlist control (n=81) that received psychoeducation and sleep hygiene. Assessments were taken at baseline, mid-intervention at 3 weeks, post-test at 6 weeks, and follow-up at 6 months. Insomnia severity was the primary outcome, with pre-sleep arousal, anxiety, depression, daily functioning and quality of life as secondary measures.

Intention-to-treat analyses showed a significant reduction in insomnia severity at both mid-intervention and post-test, with an effect size of d = 1.02 (95% CI 0.65 to 1.40) favouring the intervention. Both cognitive and somatic pre-sleep arousal fell at post-test, and the intervention group reported significantly higher quality of life at post-test. At post-test, 37.8% of the intervention group showed clinically meaningful improvement in insomnia versus 11.1% of controls, and the insomnia benefit was maintained at 6-month follow-up. Notably, there were no significant effects on anxiety, depression or daily functioning. Adherence was partial: 56.1% completed four modules and 50% completed all five.

This extends the well-established idea that CBT-based approaches can shift insomnia into a population that is rarely studied in sleep trials, and it does so through a low-intensity, remotely delivered format. The null results for anxiety, depression and daily functioning complicate any assumption that improving sleep automatically cascades into broader mental health or functional gains, at least over this timeframe. The comparator was a waitlist receiving psychoeducation and sleep hygiene rather than an active therapy, so how the adapted programme compares with unadapted CBT for insomnia remains open. The authors themselves flag that additional strategies are needed for dropouts and non-responders.

For autistic adults struggling with sleep, this is a reason to think a structured, guided online insomnia programme can meaningfully reduce insomnia severity, and that the gain may persist for months rather than fading immediately. It is also a reminder that roughly half of participants did not finish the whole course, so format and support matter as much as content. Better sleep here did not translate into measurable improvements in mood or day-to-day functioning, so expectations should be set on the sleep outcome itself. As always with one trial, this is a promising signal rather than a settled conclusion, and sleep problems with other causes need proper assessment.