Only one type of sleep app meaningfully beats insomnia

Insomnia is one of the most prevalent sleep disorders in the world, and face-to-face treatment cannot scale to meet the demand. Digital therapeutics, treatments delivered through apps and online platforms, promise help at scale, and the market now offers everything from structured CBT programmes to virtual reality and neurofeedback. Most existing reviews validate one product type at a time, which cannot tell a clinician or a consumer which approach is actually best. This analysis compared six categories of digital sleep therapy within a single evidence network.

The authors searched PubMed, the Cochrane Library, Embase and Web of Science for randomised controlled trials and included 96 of them, covering 18,419 participants. Six digital therapeutics were compared: digital CBT, digital brief behavioural therapy, digital mindfulness-based therapy, circadian rhythm support, virtual reality and tele-neurofeedback. Outcomes spanned the Insomnia Severity Index, the Pittsburgh Sleep Quality Index, sleep diary measures and objective sleep parameters. Every result was also checked against the minimal clinically important difference, the threshold at which a change matters to a patient. Of the included studies, 2 were rated low risk of bias, 79 raised some concerns and 15 were high risk.

Digital CBT reduced Insomnia Severity Index scores by 4.24 points versus control on moderate-certainty evidence, and it was the only therapy to clear the clinically meaningful threshold. It also improved Pittsburgh sleep quality scores by 2.28 points, shortened self-reported time to fall asleep by 12.57 minutes, improved self-reported sleep efficiency by 7.27% and cut self-reported night-time wakefulness by 17.19 minutes, all on low-certainty evidence. Digital mindfulness therapy reduced insomnia severity by 2.28 points and shortened objective sleep onset by 7.96 minutes, and virtual reality reduced objective night-time wakefulness by 13.38 minutes. Neurofeedback, digital brief behavioural therapy and circadian rhythm support showed no significant effects on any outcome.

The result strengthens the position of CBT-based approaches, extending support for the therapy into digital delivery at scale. It also injects healthy scepticism into the wider sleep-tech market: several approaches showed statistically detectable effects that fell short of mattering clinically, and three categories showed nothing at all. Head-to-head comparisons between the digital options were mostly too imprecise to rank them with confidence, and the objective sleep findings rest on fewer, smaller studies. The authors call for large, high-quality trials to confirm and extend the picture.

If poor sleep has you browsing app stores, this is a reason to weight your choice towards programmes built on cognitive behavioural therapy for insomnia, the only category here that produced a change big enough to feel. It is equally a reason for scepticism about neurofeedback, circadian and brief behavioural apps, which showed no significant benefit in these trials. The convenience of digital treatment appears real; the effectiveness depends heavily on what is inside the app.