What is already known on this topic

Major depressive disorder is a leading global cause of lifelong disability, with the greatest increase in incidence occurring in mid-to-late adolescence into young adulthood. Onset in childhood predicts poor long-term educational, social and health outcomes. Therefore, interventions to prevent depression in children and adolescents have considerable potential to reduce the global burden of depression. Prior research has suggested that selective and indicated prevention interventions targeted, respectively, at high risk or subsyndromal children and young people may outperform universal interventions that deliver to all individuals in the population.1–3

Methods of the study

The specialised register of the Cochrane Common Mental Disorders Group (up to September 2015, including the Cochrane Library, EMBASE, MEDLINE and PsycINFO databases), conference abstracts and reference lists were searched and experts contacted. Studies were included if they reported a randomised controlled trial of cognitive–behavioural therapy or interpersonal therapy relative to a comparison arm (treatment-as-usual, no treatment, wait list, attention placebo) for prevention of depression in participants aged between 5 and 19.9 years who did not currently meet diagnostic criteria for depression. Studies focusing on other disorders were excluded. Selection was conducted by at least two of the authors working independently. The primary outcomes were prevalence of depression diagnosis at follow-up (4–12 months) and depression symptoms at postintervention assessment. Results were meta-analysed including random-effects meta-regressions, with heterogeneity and publication bias assessed.

What this paper adds

• It updates a previous review3 adding 40 new independent trials and excluding 26 trials using more homogeneous inclusion criteria focused on depression. It confirms previous findings that psychological interventions have a small but statistically significant effect on onset of depression at up to 12 months follow-up (standardised mean difference −0.12, 95% CI −0.18 to −0.05) and that targeted trials have stronger effect sizes on depression diagnosis (RD −0.04, 95% CI −0.07 to −0.01) than universal prevention, which had no effect on depression diagnosis (RD −0.01, 95% CI −0.03 to 0.01).

• It highlights that universal preventions show no effect when compared with attention placebo controls and that few trials compare targeted interventions versus attention control. However, there was no significant effect of control condition on depression diagnosis for either targeted interventions (p=0.71) or universal interventions (p=0.67).

Limitations

• The majority of interventions (67 of 83) were delivered in school settings, raising questions as to the generalisability of the observed effects to other modes of delivery such as the internet or digital apps.

• Few trials had follow-ups >12 months, thereby lacking a long-term follow-up necessary to fully evaluate preventive effects: beneficial effects may not endure beyond 12 months.

• Studies that only examine change in depression symptoms postintervention assess a treatment effect more than a prevention effect and favour trials where participants have greater baseline symptoms and greater opportunity for symptom improvement, such as indicated prevention trials.

What next in research

Trials need to compare indicated or selective interventions against attention placebo, as well as usual care, using depression diagnosis as a primary outcome and with 2-year plus follow-ups. Furthermore, digital technology (internet, apps) provides a potential means to overcome practical difficulties in the implementation of scalable targeted prevention:4 a recent proof-of-principle trial found that internet CBT targeted at high worrying adolescents halved the 12-month onset of depression and anxiety relative to usual care.5 Such approaches need large-scale evaluation.

Do these results change your practices and why?

Yes and no. One clinical implication is that, as the authors conclude, we are not yet ready to generally implement public health depression prevention programmes for children and adolescents. Nonetheless, there may be merit in implementing prevention for the most-at-risk children and adolescents who meet selective and indicative criteria (eg, elevated symptoms of depression with personal and/or parental history of depression) where interventions have the strongest preventive benefit, although we cannot disentangle primary prevention from secondary (relapse) prevention.