Exploring AI-Driven Mobile Applications to Support Child Mental Health: A Systematic Review and Analysis

Abstract

Background: The mental health needs of children have become increasingly urgent, and Artificial Intelligence (AI)-powered mobile apps offer innovative approaches to meet these needs. This review investigates the landscape of AI-based apps specifically targeting child mental health, analyzing their availability, quality, readability, core features, and functional focus.

Methods: A systematic review was conducted to identify mobile applications that integrate AI tools for child mental health support. The Mobile Application Rating Scale (MARS) was applied to assess the overall quality, including domains such as engagement, functionality, aesthetics, information accuracy, and subjective quality. Readability was measured using the Automatic Readability Index, providing an estimate of the reading level required to understand app content and descriptions. A qualitative content analysis further explored app characteristics and operational models.

Results: From an initial pool of 369 apps, 27 met inclusion criteria. MARS evaluations revealed moderate app quality, with an average score of 3.47 out of 5. Readability results suggested that in-app content (mean grade level: 6.62) and app store text (mean grade level: 9.93) may be difficult for younger users to fully comprehend. Most apps lacked age-appropriate interface designs, limiting their appeal and usability for children. Functionally, the apps were grouped into three categories: chatbot interfaces (15 apps), mood and behavior tracking tools (9 apps), and therapeutic modules (3 apps). Although 74.1% of the apps claimed to be based on validated psychological frameworks, only two had undergone formal clinical trials. In terms of cost, only 7 were freely accessible; the remaining 20 required paid subscriptions, averaging $20.16 per month—an economic barrier for many families, particularly those in underserved communities.

Conclusions: While AI-driven apps present promising tools for advancing child mental health support, the current offerings are limited by design inefficiencies, financial barriers, and insufficient clinical validation. Prioritizing child-friendly interfaces, reducing cost barriers, and enhancing research-backed development are vital next steps to ensure these tools fulfill their potential in improving mental health outcomes for children.