CGM-Led Multimodal Tracking with Chatbot Support: An Autoethnography in Sub-Health

2025-10-29

AI Summary

Metabolic dysregulation—particularly prediabetes and suboptimal health—poses a significant public health challenge, yet existing digital health interventions inadequately support non-clinical populations. This study introduces the first large language model (LLM)-driven health agent centered on continuous glucose monitoring (CGM) as its primary physiological signal, integrating multimodal physiological and behavioral data. Deployed in a six-week self-ethnographic study, it supported overweight, prediabetic, and high-stress individuals in daily reflection and intervention around diet, physical activity, and stress management. Its key contribution lies in repositioning CGM from a clinical monitoring tool to the central feedback modality for preventive human–computer interaction, thereby establishing a novel, data-informed, conversational paradigm for health intervention. Results demonstrate significant improvements in users’ understanding of glycemic dynamics, interoceptive awareness, and capacity for sustained health behavior change—offering a scalable methodological framework at the intersection of human–computer interaction (HCI) and digital health.

Technology Category

• Humans and AI: Intelligent User Interfaces
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Multiagent Systems: Agent Communication

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Metabolic disorders present a pressing global health challenge, with China carrying the world's largest burden. While continuous glucose monitoring (CGM) has transformed diabetes care, its potential for supporting sub-health populations -- such as individuals who are overweight, prediabetic, or anxious -- remains underexplored. At the same time, large language models (LLMs) are increasingly used in health coaching, yet CGM is rarely incorporated as a first-class signal. To address this gap, we conducted a six-week autoethnography, combining CGM with multimodal indicators captured via common digital devices and a chatbot that offered personalized reflections and explanations of glucose fluctuations. Our findings show how CGM-led, data-first multimodal tracking, coupled with conversational support, shaped everyday practices of diet, activity, stress, and wellbeing. This work contributes to HCI by extending CGM research beyond clinical diabetes and demonstrating how LLM-driven agents can support preventive health and reflection in at-risk populations.

Problem

• Addressing metabolic disorders in sub-health populations using CGM
• Integrating CGM with LLMs for personalized health coaching
• Exploring multimodal tracking for preventive health and wellbeing

Innovation

• Combining CGM with multimodal indicators from devices
• Using chatbot for personalized glucose fluctuation explanations
• Integrating CGM and LLM for preventive health tracking