Abstract

Background: Because psychological approaches are likely to produce sustained benefits without the risk for tolerance or adverse effects associated with pharmacologic approaches, cognitive behavioral therapy for insomnia (CBT-i) is now commonly recommended as first-line treatment for chronic insomnia. Purpose: To determine the efficacy of CBT-i on diary measures of overnight sleep in adults with chronic insomnia. Data Sources: Searches of MEDLINE, EMBASE, PsycINFO, CINAHL, the Cochrane Library, and PubMed Clinical Queries from inception to 31 March 2015, supplemented with manual screening. Study Selection: Randomized, controlled trials assessing the efficacy of face-to-face, multimodal CBT-i compared with inactive comparators on overnight sleep in adults with chronic insomnia, with studies of insomnia comorbid with medical, sleep, or psychiatric disorders excluded. Data Extraction: Study characteristics, quality, and data were assessed independently by 2 reviewers. Main outcome measures were sleep onset latency (SOL), wake after sleep onset (WASO), total sleep time (TST), and sleep efficiency (SE%). Data Synthesis: Among 292 citations and 91 full-text articles reviewed, 20 studies (1162 participants [64% female; mean age, 56 years]) were included. Approaches to CBT-i incorporated at least 3 of the following: cognitive therapy, stimulus control, sleep restriction, sleep hygiene, and relaxation. At the posttreatment time point, SOL improved by 19.03 (95% CI, 14.12 to 23.93) minutes, WASO improved by 26.00 (CI, 15.48 to 36.52) minutes, TST improved by 7.61 (CI, -0.51 to 15.74) minutes, and SE% improved by 9.91% (CI, 8.09% to 11.73%), and changes seemed to be sustained at later time points. No adverse outcomes were reported. Limitation: Our narrow inclusion criteria limited applicability to patients with comorbid insomnia and other sleep problems, and the accuracy of estimates at later time points was less clear. Conclusion: CBT-i is an effective treatment for adults with chronic insomnia, with clinically meaningful effect sizes. Primary Funding Source: None. (PROSPERO: CRD42012002863).