A novel CGM-based disposition index (mDI-CGM) demonstrated a strong positive correlation with mean CGM glucose levels (R = 0.80), serving as a quantitative surrogate of relative β-cell function.
Observational (n=5,927)
A novel model-derived metabolic index (mDI-CGM) from routine CGM data strongly correlates with mean glucose and may serve as a scalable surrogate for relative β-cell function.
Effect estimate: R = 0.80
Introduction and Objective: Recent advances in continuous glucose monitoring (CGM) enable high-resolution, real-time assessment of glucose dynamics in individuals with diabetes. Currently, most CGM-based metrics remain descriptive and are not explicitly linked to underlying physiological parameters of glucose regulation, particularly β-cell function. The objective of this study was to develop and validate a CGM-derived metabolic index that quantitatively reflects relative β-cell function by applying a mathematical modeling approach to large-scale CGM data. Methods: We extended a previously validated mathematical model of glucose regulation (Ha et al., Journal of Clinical Endocrinology & Metabolism, 2025) to analyze CGM data collected from a large cohort in South Korea. CGM datasets from 5,927 individuals (mean BMI 24.2 ± 4.2 kg/m²) obtained via wearable devices were analyzed. Using model-based estimation, we derived a CGM-based disposition index (mDI-CGM) incorporating mean glucose and dynamic glucose patterns to capture relative β-cell functional capacity. Results: The derived mDI-CGM demonstrated a strong positive correlation with mean CGM glucose levels (R = 0.80). The index captured inter-individual variability in glucose dynamics beyond conventional CGM summary metrics, suggesting that mDI-CGM reflects integrated aspects of glycemic variability and β-cell function across a wide metabolic spectrum. Conclusion: This study introduces mDI-CGM, a novel model-derived metabolic index obtained from routine CGM data, as a quantitative surrogate of relative β-cell function. By linking CGM glucose dynamics to physiological modeling, mDI-CGM represents a scalable tool for metabolic phenotyping and risk stratification in large populations using wearable CGM technologies. Disclosure J. Lee: None. J. Ha: None. Funding National Research Foundation of Korea (RS-2025-00555223), Korea Health Industry Development Institute (RS-2024-00408915), National Science Foundation (DMS 2401921), dkNET Pilot Program of the National Institute of Diabetes & Digestive & Kidney Disease, NIH
Lee et al. (Fri,) conducted a observational in Diabetes / Metabolic spectrum (n=5,927). CGM-based disposition index (mDI-CGM) was evaluated on Correlation with mean CGM glucose levels (R = 0.80). A novel CGM-based disposition index (mDI-CGM) demonstrated a strong positive correlation with mean CGM glucose levels (R = 0.80), serving as a quantitative surrogate of relative β-cell function.