Digital Behavioural Therapy for Insomnia and its Effects on Depression and Anxiety: An Individual Participant Data Meta-Analysis
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Insomnia is a transdiagnostic risk factor for depression and anxiety and frequently co-occurs with both conditions. Sleep restriction therapy (SRT) is considered a key active component of cognitive behavioural therapy for insomnia (CBT-I), is now delivered without therapist involvement via digital platforms such as SleepFix. Existing meta-analytic evidence suggests that digital behavioural therapy for insomnia (dBT-I) may improve anxiety and depression, but participant-level evidence remains limited. This individual participant data meta-analysis pooled data from two Australian randomised controlled trials (dBT-I n=220; control n=270; 78.3% female; mean age 66.0 years) to examine whether dBT-I, with SRT as the central component and delivered through the SleepFix program, reduces depressive and anxiety symptoms in adults with insomnia disorder, who were not specifically selected for anxiety and depression. We measured anxiety using the Generalised Anxiety Disorder 7-item scale and depression using the Patient Health Questionnaire-9 or Geriatric Depression Scale-15, with depression scores standardised to a common scale assuming a shared standard deviation of 4. We fitted linear mixed-effects models with random intercepts for participants and trials at Weeks 8 and 16, including baseline GAD-7 (mean 6.1, SD 4.8) in the anxiety model. dBT-I significantly reduced anxiety at Week 8 (mean difference −0.94 GAD-7 points, 95% CI −1.80 to −0.09, p=.030) and Week 16 (−0.94 GAD-7 points, 95% CI −1.86 to −0.02, p=.044), and depression at Week 8 (−0.40 SDs, 95% CI −0.66 to −0.14, p=.003) and Week 16 (−0.44 SDs, 95% CI −0.72 to −0.17, p=.002), with no evidence effects diminished between timepoints. However, the reductions were less than the smallest detectable difference for these questionnaires (i.e., 1 point). These findings support dBT-I as a scalable intervention with modest mental health benefits extending beyond insomnia.