|
Comparative meta-analysis of behavioral interventions for insomnia and their efficacy in middle-aged adults and in older adults 55+ years of age |
Irwin M R, Cole J C, Nicassio P M |
|
|
CRD summary This review appears to conclude that behavioural interventions for insomnia can have moderate to large effects on sleep outcomes, with similar effects for both middle-aged and older adults for sleep quality, sleep latency and wakening after sleep onset. However, the potential for publication bias and the absence of information about study validity may limit the reliability of these findings. Authors' objectives To determine the relative efficacy of different behavioural treatments for insomnia and the relationship between age and sleep outcomes. Searching The Cochrane Controlled Trials Register, PsycINFO, PubMed and Social Sciences Citation Index were searched from 1966 to 2004 for relevant studies; the search terms were reported. In addition, the reference lists of included studies and relevant meta-analyses were also checked. Only studies published in English in peer-reviewed journals were sought. Study selection Only randomised controlled trials (RCTs) including at least one cognitive-behavioural therapy (CBT) intervention, or a recognised variant (further details reported in the review), were eligible for inclusion. In addition, RCTs had to include adult participants with a diagnosis of primary insomnia, and report at least one of the following sleep outcomes: sleep quality, sleep latency, total sleep time (TST), sleep efficiency and wakenings after sleep onset (WASO). Studies were excluded if data were markedly non-normal (i.e. the mean was larger than the standard deviation). RCTs had to provide sufficient information to enable the calculation of an effect size (ES).
The included studies evaluated omnibus CBT, relaxation training and behavioural training alone.
Three reviewers selected studies for inclusion, with any disagreements resolved by consensus. Assessment of study quality The validity of the RCTs was assessed according to criteria recommended by the Cochrane Collaboration. These included items on concealment of allocation, blinding, use of intention-to-treat analyses, completeness of follow-up, measures of sleep outcome used, and psychometric validity of outcomes used.
Three reviewers assessed the validity of the included studies, with any disagreements resolved by consensus. Data extraction The included studies were grouped by age cohort, intervention type and outcome measure. There were two age cohorts: one for studies with a mean age of less than 55 years, and one for studies where all participants were at least 55 years old. The interventions were grouped as omnibus CBT, relaxation therapy and behavioural therapy alone. Sleep outcomes were quality, latency, TST, efficiency and WASO.
ESs with 95% confidence intervals (CIs) were calculated for comparisons between post-treatment outcomes, unadjusted for baseline differences. Only self-report measures were included in the meta-analysis.
The authors did not state how the data were extracted for the review. Methods of synthesis ESs based on Cohen's d (adjusted for small sample bias) were pooled across studies. The Q statistic was used to assess heterogeneity. Where studies were found to be heterogeneous, a random-effects model was used; where the Q statistic was not significant, both fixed-effect and random-effects models were used. Potential moderating effects of intervention type and age cohort were investigated separately.
An ES of 0.20 was interpreted as a small effect, 0.50 was considered medium, and 0.80 was considered large. Results of the review Twenty-three RCTs (number of participants not reported) were included in the review.
The quality of the studies was not reported.
Overall statistically significant effects were observed for sleep quality (7 studies; fixed-effect ES 0.76, 95% CI: 0.48, 1.03; non significant heterogeneity, p=0.244), sleep latency (21 studies; random-effects ES -0.50, 95% CI: -0.82, -0.19; significant heterogeneity, p<0.001), sleep efficiency (8 studies; random-effects ES 0.74, 95% CI: 0.11, 1.38; significant heterogeneity, p<0.001) and WASO (15 studies; fixed-effect ES -0.64, 95% CI: -0.82, -0.47; non significant heterogeneity, p=0.86), but not for TST.
No significant effect of the intervention type was found on sleep quality, sleep latency, TST or WASO. On sleep efficiency, significant effects were observed for CBT (fixed-effect ES 1.47, 95% CI: 1.00, 1.94, p<0.01) and behavioural only interventions (fixed-effect ES 0.67, 95% CI: 0.29, 1.05, p<0.01), but not for relaxation interventions.
Significant effects of age cohort were found on the outcomes of TST and sleep efficiency, with significantly smaller effects in adults aged over 55 years (p<0.001). No significant effects of age cohort were observed for sleep quality, sleep latency or WASO. Authors' conclusions The authors' conclusion appears to be that behavioural interventions can have moderate to large effects on subjective sleep outcomes, with similar effects for different types of intervention and between middle-aged and older adults on the outcomes of sleep quality, sleep latency and WASO. CRD commentary This review was based on a question that was defined in terms of the participants, interventions, outcomes and study designs of relevance. The search covered several sources, but excluded studies that were not published in English or in peer-reviewed journals, therefore the potential for publication bias cannot be excluded. The validity of the included studies was assessed using an established checklist, but the results were not reported in detail and it is not clear if or how this validity assessment was used in the synthesis. This makes it difficult to assess the reliability of the data, although the authors stated that many studies had small sample sizes and inadequate periods of follow-up. While the methods of synthesis appear broadly appropriate for studies measuring outcomes on a range of different scales, the characteristics of the included studies were not provided, making it impossible for the reader to establish whether the studies were categorised appropriately. These limitations may to some extent undermine the reliability of the authors' conclusions. Implications of the review for practice and research Practice: The authors stated that behavioural interventions should be considered viable alternatives to pharmacological approaches for the management of insomnia in older patients.
Research: The authors stated that future research should use standardised procedures for diagnosing insomnia, polysomnographic evaluation of sleep outcomes, and health status measures of disability and disease, and should consider broader populations in which poor sleep potentially compromises quality of life and health. They added that studies are needed to evaluate the efficacy of behavioural treatments in populations with chronic disease such as rheumatoid arthritis and cancer, given the high rates of insomnia in these groups. Funding General Clinical Research Centers Program, grant numbers AA13239, DA16541, MH55253, AG18367, T32MH19925, AR/AG41867, RO1 AF498840-01 and MO1-RR00865; Cousins Center for Psychoneuroimmunology. Bibliographic details Irwin M R, Cole J C, Nicassio P M. Comparative meta-analysis of behavioral interventions for insomnia and their efficacy in middle-aged adults and in older adults 55+ years of age. Health Psychology 2006; 25(1): 3-14 Other publications of related interest This additional published commentary may also be of interest.
McMillan DE. Review: behaviour therapy is effective for insomnia. Evid Based Nurs 2006;9:118. Indexing Status Subject indexing assigned by NLM MeSH Behavior Therapy; Female; Humans; Male; Middle Aged; Randomized Controlled Trials as Topic; Sleep Initiation and Maintenance Disorders /psychology /therapy; Treatment Outcome AccessionNumber 12006006000 Date bibliographic record published 03/08/2007 Date abstract record published 01/12/2008 Record Status This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn. |
|
|
|