Table 1

Methods and results of the meta regression and meta synthesis

Systematic review and meta regression [7]

Previous reviews and electronic database searches were used to identify relevant studies. For inclusion, studies had to be randomised controlled trials with populations suffering from depression or depressive symptoms. The intervention had to assist patients in the treatment of their symptoms, using a health technology such as written information, audiotape, videotape or computer presentation. Interventions were designed to be conducted predominantly independent of professionals (3 hours or less). Data on moderators of treatment effect (i.e. the patient populations, study design, internal validity, and intervention content) and outcomes were analysed using meta regression.

Thirty four studies were identified with 39 relevant comparisons. Results found that the overall effect of self-help interventions was 'medium' according to the current convention, with the pooled standardised mean difference -0.43, 95% CI -0.57 to -0.30. The variation in effect size attributable to heterogeneity (I2) was 77.3%. When the analysis was restricted to studies using 'guided self-help' the pooled standardised mean difference was large (-0.80, 95% CI -1.01 to -0.58), and the variation in effect size attributable to heterogeneity was 68.3%. Effectiveness of guided self-help was related to population factors such as recruitment in non-clinical settings, and recruitment of patients with existing depression rather than those 'at risk'. Aspects of the intervention which moderated effects included contact with a therapist, and the use of CBT techniques. In the subset of 'guided self-help' studies using therapist contact, there were no significant associations between outcomes and the number of sessions, their content, delivery mode or the background of the therapist.

Meta-synthesis [8]

Qualitative work is ideally suited to capture the complexity of care processes, and as such has a key role to play in the development of complex interventions. This study used meta synthesis, which has some similarity to quantitative meta-analysis, involving the development of an overview of research, but based on qualitative papers.

The meta-synthesis involved 1) Identifying the literature and appraising the studies. 2) Data analysis and interpretation, including extraction of main findings, synthesis of main findings into an explanatory framework, and application of the explanatory framework to the guided self-help intervention.

Medline, Embase, Cinahl, and Web of Knowledge were searched from 2000 – 2005. The British Sociological Association criteria for the evaluation of qualitative research papers was adapted to appraise the studies. The researchers looked across the different papers for common and recurring concepts.

From the 9 papers included in the review 5 key themes were determined: personal experience of depression, ambivalent help seeking and the covert presentation of psychological problems, control and helplessness in engaging with treatment, stigma associated with treatment and patients' understanding of self-help interventions. This broad explanatory framework was then applied to the specific issue of the development of the guided self help intervention.


Lovell et al. BMC Psychiatry 2008 8:91   doi:10.1186/1471-244X-8-91

Open Data