Use of continuous glucose monitoring to stratify individuals without diabetes

Early detection of metabolic dysfunction before diabetes onset remains a critical challenge in preventive medicine. Although glucose dynamics provide high-dimensional insights into metabolic states, it remains unclear which combination of glucose dynamics-derived measures most effectively summarises interindividual differences in glucose regulation. We analysed continuous glucose monitoring (CGM) data from 8025 adults without diagnosed diabetes, using all recordings of at least seven days’ duration, to derive a low-dimensional representation of glucose dynamics. Using exploratory factor analysis, we identified a small set of latent features that accounted for variation between individuals in CGM-derived metrics. We then trained and validated machine-learning models to predict postprandial glucose trajectories from these features in 863 participants who underwent standardised meal tests. We further assessed the generalisability of this representation in an independent oral glucose tolerance test dataset. Finally, we examined associations between the derived features and markers of vascular and liver health in 1784 non-diabetic adults. Three features, “mean”, “variance”, and “autocorrelation”, explain more than 80% of the interindividual differences in CGM-derived measures. A three-dimensional representation based on these features reconstructs postprandial glucose trajectories with high accuracy and outperforms fasting, mean, and two-hour postprandial glucose values. Each feature shows independent associations with carotid artery intima–media thickness and with hepatic steatosis and stiffness. By compressing high-dimensional glucose dynamics into three interpretable features with minimal loss of information, this framework provides a simple yet physiologically meaningful representation of glucose regulation that may facilitate a more precise and interpretable assessment of diabetes-related risk. Sugimoto et al. investigate the potential of continuous glucose monitoring-derived indices in predicting diabetes-related measures. They show that inter-individual differences in glucose dynamics can be explained by three features: mean, variance, and autocorrelation, each independently associated with diabetes-related measures. This study aimed to develop a simple way to identify early issues with blood glucose control before a diabetes diagnosis. Using wearable sensors and standardised meal tests, we investigated whether it was possible to reduce many detailed glucose metrics to a few simple ones without losing important information. We showed that three simple measures, average level, extent of variation, and similarity to recent readings, were sufficient to describe most patterns of blood glucose change. These three measures were linked to early signs of artery damage, liver disease, and insulin sensitivity. Together, this may help to identify individuals at higher risk earlier and support more informed decisions about preventing diabetes and its complications.