Factors to consider in the interpretation of trial results in psychotherapy for depression in adults, with reference to suicidal behaviour

Abstract

The results of randomised controlled trials and meta-analyses of trials in medicine have led to major improvements in treatments, and thereby significant reductions in mortality and morbidity worldwide. The implications of specific aspects of trial methodology for evaluating the effectiveness of psychotherapy for depression are explored, with reference to trial size, quality, bias, eligibility criteria and the use of non-inferiority design. In contrast to the treatment of systemic disease, pill placebo control in trials of psychotherapy for depression is not ‘no treatment’, and the use of such can reduce measured treatment effect. Trials have entry criteria, and thus use a selective sample, findings may therefore not be generalizable. For example, in psychotherapy, specific trial exclusion criteria (suicidal behaviour) appear to raise a concern that a therapy may be widely adopted and include the treatment of a category of individual excluded from a practice-informing trial. A diagnosis of depression is associated with a higher mortality, notably suicide, and the inter-relationship between depression and suicidal behaviour is explored. Aside from the direct utilisation of specific psychotherapies for suicidal behaviour; the effective treatment of depression, notably major depression, by psychotherapy, may lead to a reduction in the prevalence of suicidal behaviour.