Computer-Assisted Cognitive-Behavior Therapy for Depression: A Systematic Review and Meta-Analysis.

Objective: To evaluate the efficacy of computer-assisted forms of cognitive-behavior therapy for major depressive disorder (MDD) and examine the role of clinician support and other factors that might affect outcomes.

Data Sources: Ovid MEDLINE, PsycINFO, PubMed, and Scopus from their beginnings to July 18, 2016. Keywords were "randomized, controlled trials of computer-assisted cognitive-behavior therapy for depression" and "randomized, controlled trials of mobile apps for cognitive-behavior therapy of depression."

Study Selection: Of 223 studies identified in the search, 183 were excluded yielding a sample of 40 randomized, controlled investigations of computer-assisted cognitive-behavior therapy (CCBT) for depression.

Data Extraction: Data were abstracted independently by two authors, and consensus was reached by discussion with a third author.

Results: The overall mean effect size for CCBT compared to control conditions was g = 0.502, a moderately large effect. Studies that provided support from a clinician or other person yielded significantly larger effects (g = 0.673) than studies in which no support was provided (g = 0.239). Completion rate and study setting also influenced outcomes. Lower mean effect sizes were observed in studies with lower completion rates and in studies conducted in primary care practices.

Conclusions: CCBT with a modest amount of support from a clinician or other helping person was found to be efficacious with relatively large mean effect sizes on measures of depressive symptoms. Self-guided CCBT for depression was considerably less effective. Future research should focus on enhancing the implementation of CCBT, including evaluating the amount and type of support needed for effective delivery, methods to improve engagement with computer-assisted therapies, and ways to improve treatment outcome in primary care settings.