NEWS & EVENTS 
SPECIALTY CARE
IHI 2025 Poster: Piloting an AI ChatBot Tool for Diabetes Nutrition Counseling
Updated: 12/08/25
Citation: Brannon E, Hisamatsu, R, Young L, Oshman L. Piloting an AI ChatBot Tool for Diabetes Nutrition Counseling. Institute for Healthcare Improvement Forum 2025. December 8-11, 2025, Anaheim, CA.
Presented at the Institute for Healthcare Improvement (IHI) Forum conference in December 2025, Anaheim, CA.
Poster Title: Piloting an AI ChatBot Tool for Diabetes Nutrition Counseling.
DOI: Not Available
Authors: Elliott Brannon, MD, PhD, MPH, Rina Hisamatsu, MPH, RDN, Larrea Young, MDes, Lauren Oshman, MD, MPH.
Introduction and Objective: The Standards of Care published by the American Diabetes Association recommends that all people with T2D receive individualized medical nutrition therapy and endorses low carbohydrate diets (LCDs) as a viable approach in T2D management.
However, clinicians report limited training and confidence in discussing LCDs with patients contributing to it being an underutilized treatment approach.
Methods: A conversational artificial intelligence (AI) chatbot called “LCD Patient Simulator” was developed with OpenAI for clinicians to practice counseling patients on LCDs for T2D management. The chatbot includes six simulated patients with different levels of health literacy and cultural backgrounds, and provides counseling feedback to the user.
MCT2D piloted the chatbot as a take-home self assessment tool during their Spring 2025 Low Carbohydrate Immersive Training Program. This weeklong training program was designed to provide clinicians with foundational knowledge of how to effectively counsel patients on LCDs. Participants were asked to use the chatbot to assess their understanding of LCDs and their counseling skills.
The pilot sought to establish feasibility of using a chatbot in LCD clinician education. Participants were asked to complete a post-workshop survey with questions aimed at evaluating the chatbot's usefulness as a tool for clinicians to practice their LCD counseling skills.
Conclusion:
- Clinicians had a positive experience using the AI chatbot for nutrition education.
- Clinicians preferred challenging AI patients to practice their skills.
- During the training, clinicians had the opportunity to work through multiple AI patients but most only chose to work through one.
- The chatbot will be improved and used for future iterations of the MCT2D LCD training.
Subscribe to the MCT2D Newsletter
Once per month we deliver news, updates, events, and more directly to you!
ccteam@mct2d.org
Support for The Michigan Collaborative for Type 2 Diabetes (MCT2D) is provided by Blue Cross Blue Shield of Michigan as part of the BCBSM Value Partnerships program.
Although Blue Cross Blue Shield of Michigan and MCT2D work collaboratively, the opinions, beliefs and viewpoints expressed by the author do not necessarily reflect the opinions, beliefs and viewpoints of BCBSM or any of its employees.