Cognitive–behavioural therapy aims to increase quality of life by changing cognitive and behavioural factors that maintain problematic symptoms. A previous overview of cognitive–behavioural therapy systematic reviews suggested that cognitive–behavioural therapy was effective for many conditions. However, few of the included reviews synthesised randomised controlled trials.
This project was undertaken to map the quality and gaps in the cognitive–behavioural therapy systematic review of randomised controlled trial evidence base. Panoramic meta-analyses were also conducted to identify any across-condition general effects of cognitive–behavioural therapy.
The overview was designed with cognitive–behavioural therapy patients, clinicians and researchers. The Cochrane Library, MEDLINE, EMBASE, PsycINFO, Cumulative Index to Nursing and Allied Health Literature, Child Development & Adolescent Studies, Database of Abstracts of Reviews of Effects and OpenGrey databases were searched from 1992 to January 2019.
Study inclusion criteria were as follows: (1) fulfil the Centre for Reviews and Dissemination criteria; (2) intervention reported as cognitive–behavioural therapy or including one cognitive and one behavioural element; (3) include a synthesis of cognitive–behavioural therapy trials; (4) include either health-related quality of life, depression, anxiety or pain outcome; and (5) available in English. Review quality was assessed with A MeaSurement Tool to Assess systematic Reviews (AMSTAR)-2. Reviews were quality assessed and data were extracted in duplicate by two independent researchers, and then mapped according to condition, population, context and quality. The effects from high-quality reviews were pooled within condition groups, using a random-effect panoramic meta-analysis. If the across-condition heterogeneity was I ² < 75%, we pooled across conditions. Subgroup analyses were conducted for age, delivery format, comparator type and length of follow-up, and a sensitivity analysis was performed for quality.
A total of 494 reviews were mapped, representing 68% (27/40) of the categories of the International Classification of Diseases, Eleventh Revision, Mortality and Morbidity Statistics. Most reviews (71%, 351/494) were of lower quality. Research on older adults, using cognitive–behavioural therapy preventatively, ethnic minorities and people living outside Europe, North America or Australasia was limited. Out of 494 reviews, 71 were included in the primary panoramic meta-analyses. A modest effect was found in favour of cognitive–behavioural therapy for health-related quality of life (standardised mean difference 0.23, 95% confidence interval 0.05 to 0.41, prediction interval –0.05 to 0.50, I ² = 32%), anxiety (standardised mean difference 0.30, 95% confidence interval 0.18 to 0.43, prediction interval –0.28 to 0.88, I ² = 62%) and pain (standardised mean difference 0.23, 95% confidence interval 0.05 to 0.41, prediction interval –0.28 to 0.74, I ² = 64%) outcomes. All condition, subgroup and sensitivity effect estimates remained consistent with the general effect. A statistically significant interaction effect was evident between the active and non-active comparator groups for the health-related quality-of-life outcome. A general effect for depression outcomes was not produced as a result of considerable heterogeneity across reviews and conditions.
Data extraction and analysis were conducted at the review level, rather than returning to the individual trial data. This meant that the risk of bias of the individual trials could not be accounted for, but only the quality of the systematic reviews that synthesised them.
Owing to the consistency and homogeneity of the highest-quality evidence, it is proposed that cognitive–behavioural therapy can produce a modest general, across-condition benefit in health-related quality-of-life, anxiety and pain outcomes.
Future research should focus on how the modest effect sizes seen with cognitive–behavioural therapy can be increased, for example identifying alternative delivery formats to increase adherence and reduce dropout, and pursuing novel methods to assess intervention fidelity and quality.
This study is registered as PROSPERO CRD42017078690.
This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment ; Vol. 25, No. 9. See the NIHR Journals Library website for further project information.