Mental health apps for adolescents and young adults: A systematic review of randomised controlled trials

Objective: Smartphone applications ('apps') have the potential to improve the scalability of mental health interventions for young people, however, the effectiveness of stand-alone apps in mental health management remains unclear. This systematic review, with meta-analysis, provides an up-to-date summary of the available high-quality evidence. Methods: Eleven randomized controlled trials, involving a pooled sample of 1706 adolescents and young adults (age range 10 to 35 years), were identified from the Cochrane Library, Embase, Google Scholar, PsycINFO and PubMed databases. The reporting quality of studies was evaluated using the Cochrane Risk of Bias Tool 2.0 (RoB 2.0). Hedges' g effect sizes with 95% confidence intervals, p values and heterogeneity statistics were additionally calculated using a random effects model. Results: Study reporting quality was sound, with no trials characterized as 'high' risk. App interventions produced significant symptom improvement across multiple outcomes, compared to wait-list or attention control conditions (depression gw = 0.52 [CI: 0.18-0.84], p = .01, k = 8; stress gw = 0.30 [CI: 0.06-0.53], p = .02, k = 2). Longer-term benefits could not be established (k = 4), although individual studies reported positive trends up to 6 months post. Age was not identified as a significant moderator. Conclusions: Smartphone apps hold promise as a stand-alone self-management tool in mental health service delivery. Further controlled trials with follow-up data are needed to confirm these findings as well as determine treatment engagement and effectiveness across diverse groups of participants.

Automated summary

Smartphone apps show promise as a stand-alone self-management tool in mental health service delivery, producing significant symptom improvement. While longer-term benefits are still uncertain, some individual studies show positive trends up to 6 months post. Further controlled trials with follow-up data are needed to confirm these findings.