As mental health apps for depression and anxiety grow in popularity, understanding how users engage with these tools is essential.1 A new meta-analysis of 79 randomized trials reveals that while most users activate these apps, sustaining meaningful engagement remains challenging.
The study identifies strategies—including reminders and human support—that can reduce dropout and improve adherence, providing evidence-based guidance for designing more effective digital mental health interventions.
This meta-analysis is published in JAMA Psychiatry.
Mental health care for depression and anxiety has increasingly embraced digital tools, with smartphone apps emerging as scalable, accessible interventions.2 Apps have been found to produce small but significant improvements in depressive and generalized anxiety symptoms, with stronger effects when apps include cognitive behavioral therapy, mood monitoring, or chatbot features.
“From a research perspective, estimates of uptake, attrition, and adherence provide a valuable benchmark that can inform trial planning, including a priori power analyses and recruitment targets,” wrote the researchers of the study.1 “Clinically, these findings offer insight into how participants engage with mental health apps and highlight modifiable design features that could be critical for enhancing adherence and reducing dropout.”
In the study, the researchers examined patterns of uptake, adherence, and attrition in randomized clinical trials of mental health apps for depression and anxiety. Eligible trials, identified from a 2024 review and an updated Medline and PsycINFO search through May 2025, included participants with depression or anxiety and reported rates of app engagement or outcome completion.
Included in the meta-analysis were 79 trials encompassing 92 app conditions and 78 control conditions. Across 56 app conditions, the pooled uptake rate was high at 92.4% (95% prediction interval [PI], 48.4%-99.4%), indicating that most participants initially activated the apps.
Among 21 conditions reporting adherence, the pooled rate was 61.8% (95% PI, 53.2%-69.7%), reflecting moderate engagement over the course of the trials. Posttest attrition averaged 18.6% (95% PI, 2.4%-68.3%) and increased to 28.4% (95% PI, 3.0%-83.7%) at follow-up. Attrition was lower in trials incorporating reminders or human support and in apps without gamification features.
Although app conditions showed higher attrition compared with waitlist controls (risk ratio, 1.49; 95% PI, 0.34-6.48), attrition did not significantly differ compared with other active control groups.
However, the researchers acknowledged several limitations, which include limited adherence data and a focus on depression and anxiety, and no other conditions. Additionally, high heterogeneity in pooled estimates remained, likely influenced by unmeasured factors such as recruitment, motivation, cultural context, and symptom severity. Therefore, the researchers believe that more consistent reporting is needed to support future moderator analyses.
Despite these limitations, the researchers believe these findings highlight both the potential and the challenges of maintaining engagement in digital mental health interventions.
“Such results may help inform the development of more engaging and potent apps by identifying which features warrant prioritization to maximize retention and treatment fidelity,” wrote the researchers. “Future research may benefit from exploring how emerging technologies, such as generative artificial intelligence, can support the development of app features that show early promise for improving engagement. The digital mental health field is uniquely positioned for rapid iteration and innovation, with novel technology-enabled solutions increasingly within reach.”
References
1. Liu C, Torous J, Fuller-Tyszkiewicz M, et al. Uptake, adherence, and attrition in clinical trials of depression and anxiety apps. JAMA Psychiatry. doi:10.1001/jamapsychiatry.2025.3439
2. Linardon J, Torous J, Firth J, et al. Current evidence on the efficacy of mental health smartphone apps for symptoms of depression and anxiety. A meta-analysis of 176 randomized controlled trials. World Psychiatry. 2024;23(1):139-149. doi:10.1002/wps.21183