Meta-Analysis: 13-Year Follow-up of Psychotherapy Effects on Youth Depression

Youth depression is a debilitating condition that constitutes a major public health concern. A 2006 meta-analysis found modest benefits for psychotherapy versus control. Has 13 more years of research improved that picture? We sought to find out.

We searched PubMed, PsychINFO, and Dissertation Abstracts International for 1960 to 2017, identifying 655 randomized, English-language psychotherapy trials for individuals aged 4 to 18 years. Of these, 55 assessed psychotherapy versus control for youth depression with outcome measures administered to both treatment and control conditions at post (κ = 53) and/or follow-up (κ = 32). Twelve study and outcome characteristics were extracted, and effect sizes were calculated for all psychotherapy versus control comparisons. Using a three-level random-effects model, we obtained an overall estimate of the psychotherapy versus control difference while accounting for the dependency among effect sizes. We then fitted a three-level mixed-effects model to identify moderators that might explain variation in effect size within and between studies.

The overall effect size (g) was 0.36 at posttreatment and 0.21 at follow-up (averaging 42 weeks after posttreatment). Three moderator effects were identified: effects were significantly larger for interpersonal therapy than for cognitive behavioral therapy, for youth self-reported outcomes than parent-reports, and for comparisons with inactive control conditions (eg, waitlist) than active controls (eg, usual care). Effects showed specificity, with significantly smaller effects for anxiety and externalizing behavior outcomes than for depression measures.

Youth depression psychotherapy effects are modest, with no significant change over the past 13 years. The findings highlight the need for treatment development and research to improve both immediate and longer-term benefits.