A Scoping Review of Artificial Intelligence-Based Health Education Interventions for Patients with Type 2 Diabetes

Area of Science:

• Digital Health
• Medical Informatics
• Artificial Intelligence in Medicine

Background:

• Type 2 diabetes mellitus (T2DM) presents a significant global health challenge, necessitating effective patient education for self-management.
• Artificial intelligence (AI) offers scalable, personalized solutions for health education, yet its application in T2DM is not comprehensively synthesized.

Purpose of the Study:

• To map AI-based health education interventions for T2DM.
• To evaluate the technologies, effectiveness, and challenges of these AI interventions.

Main Methods:

• A scoping review guided by the Arksey and O'Malley framework.
• Searched seven academic databases for studies from 2008 to March 2025.
• Included 14 interventional studies involving 32,478 adult T2DM patients.

Main Results:

• AI interventions utilized diverse technologies like mobile apps, chatbots, and intelligent platforms for diet, glucose monitoring, and lifestyle management.
• Significant improvements were observed in glycemic control (HbA1c reduction up to 2.59%), self-management adherence (60-85%), and psychological outcomes.
• Key challenges included technical complexity, low long-term engagement, digital literacy gaps, and data privacy concerns.

Conclusions:

• AI demonstrates substantial potential for personalized and accessible T2DM health education.
• Future research should focus on overcoming technological hurdles, user-centered design, and healthcare system integration for sustainability and equity.