Cognitive Behavioral Therapy for Chronic Insomnia: A Systematic Review and Meta-analysis
PRISMA-registered systematic review and meta-analysis (20 RCTs, n=1,162 adults with chronic primary insomnia) of face-to-face multimodal CBT-I (≥3 components among cognitive therapy, stimulus control, sleep restriction, sleep hygiene, relaxation) versus inactive controls: post-treatment mean reductions in sleep-onset latency (~19 min), wake after sleep onset (~26 min), and gains in sleep efficiency (~10 percentage points); total sleep time change small and CI-inclusive of null; no adverse events reported; authors note narrow exclusion of comorbid insomnia.
Design
PROSPERO CRD42012002863; searches through 31 March 2015 across MEDLINE, EMBASE, PsycINFO, CINAHL, Cochrane Library, and PubMed Clinical Queries; 20 parallel RCTs (n = 1,162; ~64% female; mean age ~56 y) meeting criteria for face-to-face multimodal CBT-i vs inactive comparators on sleep-diary overnight measures; trials with primary medical, sleep, or psychiatric comorbidity were excluded—limits generalization to comorbid insomnia clinics.
Pooled post-treatment diary changes (random-effects meta-analysis; 95% CI from abstract)
| Outcome | Mean change vs inactive | 95% CI |
|---|---|---|
| Sleep-onset latency (SOL) | −19.03 min | −23.93 to −14.12 |
| Wake after sleep onset (WASO) | −26.00 min | −36.52 to −15.48 |
| Total sleep time (TST) | +7.61 min | −0.51 to +15.74 (CI spans null) |
| Sleep efficiency (SE%) | +9.91% | +8.09% to +11.73% |
Durability & safety
Abstract reports changes appeared sustained at later follow-up assessments where data existed, with caution on precision of long-term estimates; no adverse outcomes were reported across included trials.
Evidence hygiene
- Inactive comparator arms (waitlist, attention control, minimal education) — pooled effects are not directly comparable to hypnotic medication or digital/app CBT-i without reading component-level network meta-analyses.
- Multimodal CBT-i here means programs combining ≥3 of the listed components—single-component trials were outside inclusion rules.
Pairing in this wiki
sleep-optimization— behavioral first-line insomnia treatment context next to sleep extension and hygiene rows.progressive-muscle-relaxation— relaxation can appear inside CBT-i bundles but PMR as a standalone somatic protocol has a different evidence map.magnesium-supplementation— pharmacologic/mineral RCT stream; do not merge outcome tables with CBT-i.
Outcomes
- Sleep Onset Latency-19.03min (Minutes)
- Wake After Sleep Onset (WASO)-26min (Minutes)
- Total Sleep Time7.61min (Minutes)
- Sleep Efficiency9.91% (Percentage)
- Effect Size (Cohen's d / SMD)Systematic review: no adverse outcomes reported across included RCTs; long-term follow-up estimates less certain per authors