SCIENCE / SUMMARY / INSOMNIA THERAPY

Does insomnia therapy work as well online or by phone as it does face to face?

NETWORK META ANALYSIS · 52 TRIALS ■■■□ moderate evidence: based on randomised trials, but inconsistency between trials is high and the phone arm is a single trial

Who

52 randomised trials in adults diagnosed with insomnia disorder, 12,544 people in total. Mean age was 43.4 and 70.6% were women. The main outcome, insomnia severity, was analysed in 44 trials with 4,662 people. The search covered English and German studies published from 1987 to November 2021.

What they did

Eight ways of delivering the same therapy were separated: individual face to face, group, remote session by video or chat, guided and unguided booklet, guided and unguided online programme, and phone app. A network meta analysis lets formats that were never tested against each other be compared through shared comparison groups. The main measure was self reported, mostly the Insomnia Severity Index, and the reference was a waiting list.

What they found

Against the waiting list, standardised mean differences were -1.35 for remote sessions, -1.30 for individual face to face, -1.05 for guided booklets, -1.04 for the phone app, -1.01 for groups and -1.01 for unguided online programmes (large effects); -0.73 for guided online programmes and -0.67 for unguided booklets (medium effects). No format was significantly better than another. On objective measures from actigraphy and sleep labs, sleep efficiency and night waking did not change; objective total sleep time fell in most formats, which the authors link to the sleep restriction component. Completion rates for group and both online formats lagged behind the waiting list; the authors note this data was reported unevenly.

Limits

As the authors note: the phone app is represented by one trial (compared only with sleep hygiene education), guided booklets and remote sessions by four trials each; consistency between trials is low (I² 77.7%) and the network shows inconsistency; risk of bias is considerable because of missing outcome data; some papers described support levels poorly, so misclassification is possible. 2025 author correction: in two trials, Sleep Condition Indicator scores had been entered without inverting them; after correction the phone arm became significant and I² fell from 95.5% to 77.7%. This summary uses the corrected version. The study had no dedicated funding; two authors report leading externally funded projects on digital insomnia programmes.

Link to Nefesly

Background for a digital insomnia programme Nefesly may consider in the future; no such programme is in the app today. It is not a health claim. How we built it (Turkish)

Reference

Simon L, Steinmetz L, Feige B, Benz F, Spiegelhalder K, Baumeister H. Comparative efficacy of onsite, digital, and other settings for cognitive behavioral therapy for insomnia: a systematic review and network meta-analysis. Scientific Reports. 2023;13(1):1929. doi:10.1038/s41598-023-28853-0 · Author correction 2025: doi:10.1038/s41598-025-99375-0 · Open access (CC BY 4.0).

Why chosen

One of 15 sources chosen from 430 candidates on the digital insomnia therapy axis; included because it compares face to face and digital formats in the same network.

Summarised by

Nefesly, from the full text, 11 October 2026. Awaiting clinician check.

Copyright

Summarised from the full text, not translated from the abstract.