Generative AI-Enabled Therapy Support Tool for Improved Clinical Outcomes and Patient Engagement in Group Therapy:...

Area of Science:

• Mental Health Technology
• Digital Therapeutics
• Artificial Intelligence in Healthcare

Background:

• Cognitive Behavioral Therapy (CBT) is effective for depression and anxiety but suffers from low patient engagement.
• Lack of between-session engagement with therapeutic materials is a key factor in CBT treatment failure.
• A significant proportion of patients do not achieve successful outcomes with standard CBT.

Purpose of the Study:

• To evaluate if a generative AI-enabled therapy support tool improves treatment success and adherence.
• To compare AI-supported CBT engagement with traditional static workbook exercises.
• To assess the impact of AI on patient engagement with therapeutic materials between sessions.

Main Methods:

• A real-world observational study involving 244 patients in UK National Health Service Talking Therapies services.
• Comparison of 150 patients using an AI-enabled tool versus 94 patients using standard CBT workbooks.
• Groups were matched for CBT content and human-led sessions; the intervention group received AI support for exercises.

Main Results:

• Patients using the AI tool showed higher session attendance and lower dropout rates.
• AI tool users achieved better reliable improvement, recovery, and reliable recovery rates.
• Engagement with AI-supported CBT predicted better adherence and success, with personalization playing a key role.

Conclusions:

• AI-enabled, personalized therapy support tools enhance the efficacy of human-led group therapy.
• This approach shows promise for improving mental health care adherence and outcomes.
• AI tools can facilitate patient engagement and application of CBT techniques in daily life.