Mental health conditions are a significant burden for individuals, healthcare systems, and economies worldwide. Despite their high prevalence, there is still a lack of accessible treatments and support. In recent years, mobile apps have emerged as a potential solution to bridge this gap. Many mental health-focused mobile apps now incorporate artificial intelligence (AI) and machine learning techniques to provide personalized support. However, the effectiveness of these AI-enabled apps remains unclear.

The purpose of this scoping review is to provide an overview of the current research landscape regarding the use of AI in mobile health apps for mental health support. Our search yielded 17 studies that evaluated AI-powered mental health apps, revealing a wide range of uses and applications.

From risk prediction and classification to personalization and conversational support, these AI-enabled apps aim to address various mental health needs, including depression, stress, and suicide risk. However, the research is still in its early stages, with many studies featuring small sample sizes and methodological limitations.

Despite these challenges, our findings demonstrate the feasibility of using AI to support mental health apps. With the increasing availability of these apps to a large population, it is essential to continue researching their effectiveness and potential for long-term benefits.

Methodology

We conducted a systematic search of PubMed for randomized controlled trials and cohort studies published in English since 2014 that evaluate AI-enabled mobile apps for mental health support. Our search yielded 1,022 studies, from which we selected four for inclusion in our review.

Characteristics of AI-Enabled Apps

The included studies featured a range of AI techniques, including machine learning algorithms, natural language processing, and predictive modeling. These techniques were used to develop various app features, such as:

  • Risk prediction and classification
  • Personalization and conversational support
  • Mood tracking and assessment
  • Suicide risk assessment and intervention

These apps aimed to address diverse mental health needs, including depression, stress, anxiety, and post-traumatic stress disorder (PTSD).

Study Characteristics

The studies' characteristics varied in terms of:

  • Methods: Randomized controlled trials, cohort studies, and mixed-methods approaches were used.
  • Sample size: Small sample sizes dominated the included studies.
  • Study duration: Most studies had short durations, ranging from a few weeks to several months.

Limitations and Future Directions

The early stages of research and methodological limitations highlight the need for more high-quality studies evaluating AI-enabled mental health apps. Additionally, future studies should focus on addressing knowledge gaps in areas such as:

  • Long-term effectiveness
  • Generalizability to diverse populations
  • Integration with existing healthcare systems and services

Conclusion**

This scoping review provides an overview of the current state of research on AI-enabled mobile health apps for mental health support. Our findings demonstrate the feasibility of using AI to develop effective mental health interventions. However, we also highlight the need for more rigorous research and the importance of addressing knowledge gaps in this emerging field.

Author Summary

Mental health concerns are a significant burden for individuals, healthcare systems, and economies worldwide. Mobile health apps have emerged as a potential solution to bridge this gap. This scoping review provides an overview of AI-enabled mental health apps and highlights the need for more research into their effectiveness.

Funding Statement

This research was funded by the NIHR Artificial Intelligence in Health and Care Award (grant reference number: AI_AWARD02176).

Competing Interests

ES is an employee of, and BI is an advisor for, Wysa Ltd., a company that has designed and developed an AI-enabled mental health app.