Over the past decade, mobile apps have been widely adopted as a digital intervention method for mental health support, offering scalable and accessible solutions to address the growing global mental health challenges. However, sustaining user engagement in real-world settings remains a major challenge in the development of these applications. This study systematically examines factors that hinder user engagement in existing mobile mental health support systems through a scoping review of the literature. After an initial identification of 1,267 papers, we conducted a final analysis of 111 empirical studies using mobile app-based mental health support systems. The study investigates the main factors that negatively affect user engagement from user and system perspectives. Based on these findings, we propose guidelines for enhancing user engagement and structuring personalized emotional interaction design along three dimensions: adaptive, continuous, and multimodal interactions. Furthermore, we discuss the potential for integration with advanced AI methods (e.g., LLM-based AI agents) as a way to achieve these design implications and suggestions. Our results provide critical insights for enhancing long-term user engagement in the development of future mental health support systems.
https://dl.acm.org/doi/10.1145/3706598.3713732
The ACM CHI Conference on Human Factors in Computing Systems (https://chi2025.acm.org/)