Importance: Mobile mental health applications (apps) for moderate to severe depression are proliferating, likely owing to their capacity to overcome the limitations of conventional psychotherapy, but research on the potential moderators of treatment efficacy is lacking.
Objective: To examine the treatment efficacy associated with mobile app interventions for moderate to severe depression and identify the potential moderators associated with better treatment outcomes.
Data sources: PubMed, Embase, and PsycINFO were searched from their inception to January 22, 2023.
Study selection: Only randomized clinical trials evaluating mobile app treatments in adults with moderate to severe depression that published their results in English were included in the analysis.
Data extraction and synthesis: Three independent researchers extracted and assessed relevant studies, their risk of bias, the characteristics of the population and study design, and the components of the intervention program following the Preferred Reporting Items for Systematic Reviews and Meta-analyses reporting guidelines. A fixed-effects model was used for data analysis, and exploratory post hoc meta-regression and subgroup analyses were also conducted. Data were analyzed from February 16 to March 25, 2023.
Main outcomes and measures: The main outcome was changes in depression symptom severity from before to after treatment, measured by standardized depression assessment instruments. Secondary outcomes included study-, intervention-, and patient-level factors associated with app efficacy.
Results: Of 2128 studies identified, 13 studies evaluating 16 intervention apps with 1470 participants with moderate to severe depression were included in the analysis. The overall pooled effect size of mobile app interventions vs both active and inactive control groups was 0.50 (95% CI, 0.40 to 0.61). Interventions with in-app notifications were associated with significantly lower treatment outcomes (standardized mean difference [SMD], 0.45; 95% CI, 0.29-0.60) than interventions without (SMD, 0.71; 95% CI, 0.54-0.87; P = .02). In addition, app interventions delivered for less than 8 weeks were associated with a significantly greater effect size (SMD, 0.77; 95% CI, 0.59-0.96) than interventions delivered for 8 weeks or longer (SMD, 0.43; 95% CI, 0.30-0.57; P = .004).
Conclusions and relevance: In this systematic review and meta-analysis, the feasibility and efficacy of mobile app interventions were supported in treating moderate and severe depression, and practical implications were also provided for developing effective app-based interventions in clinical practice.