The effects of cognitive-behavioral therapy for depression are not falling: A re-analysis of Johnsen and Friborg (2015).

Abstract

Cognitive-behavioral therapy (CBT) has a solid evidence base as an effective treatment for depression. However, a recent meta-analysis (Johnsen & Friborg, 2015) including 70 studies, showed that the effect sizes of CBT for depression have been falling between 1977 and 2014. A possible important limitation in the Johnsen and Friborg (2015) study was that they did not investigate a leveling off in the decline over time of the effectiveness of CBT for depression. We therefore reanalyzed the data reported by Johnsen and Friborg (2015) using meta-analytic regression models that allowed for a curvilinear effect of publication year and also modeled separate estimates of the decline of treatment effect before and after 1995. Our analyses showed that adding a quadratic effect of time to a linear effect of time significantly improved the meta-analytic regression models (p = .017-.027). Furthermore, significant declines were only observed between 1977 and 1995 (p = .001-.009) and not between 1995 to 2014 (p = .987-.785). We conclude that the declining effect of CBT for depression observed by Johnsen and Friborg (2015) was highly influenced by 22 studies published before 1995 and that the 48 studies published after 1995 did not demonstrate such a decline. Thus, there are no indications that CBT for depression is gradually losing its value. (PsycINFO Database Record