Although cognitive behavioural therapy for insomnia (CBT-I) has been recommended the initial therapy for insomnia, its clinical usage remains limited due to the lack of therapists. Digital CBT-I (dCBT-I) can potentially circumvent this problem. This meta-analysis aims to evaluate the short-term and long-term efficacy of dCBT-I for adults with insomnia.
Systematic search of PubMed, EMBASE, PsycINFO, and CENTRAL from inception till 5 March 2020 was conducted. Randomised controlled trials (RCTs) comparing dCBT-I with controls (wait-list/treatment-as-usual/online education) in adults with insomnia were eligible. The primary outcome was insomnia severity index (ISI) at post-intervention, short-term follow-up (ranging 4 weeks to 6 months) and 1-year follow-up. Mean differences were pooled using the random-effects model.
94 articles were assessed full-text independently by two team members and 33 studies were included in this meta-analysis. 4719 and 4645 participants were randomised to dCBT-I and control respectively. dCBT-I significantly reduces ISI at post-intervention with mean difference -5.00 (95% CI -5.68 to -4.33, p < 0.0001) (I = 79%) compared to control. The improvements were sustained at short-term follow-up, -3.99 (95% CI -4.82 to -3.16, p < 0.0001) and 1-year follow-up, -3.48 (95% CI -4.21 to -2.76, p < 0.0001). Compared to dCBT-I, face-to-face CBT-I produced greater improvement in ISI, 3.07 (95% CI 1.18 to 4.95, p = 0.001). However, this was within the non-inferiority interval of 4 points on ISI.
This meta-analysis provides strong support for the effectiveness of dCBT-I in treating insomnia. dCBT-I has potential to revolutionise the delivery of CBT-I, improving the accessibility and availability of CBT-I content for insomnia patients worldwide.

Copyright © 2020 Elsevier B.V. All rights reserved.