Sleep consistency is a low-cost reliable indicator of nocturnal glycemic control: observations from 227,860 nights of real world, free-living smart ring and continuous glucose monitoring data

Nocturnal glucose regulation is modulated by autonomic and circadian mechanisms, yet their dynamic interplay in apparently healthy, free-living populations remains poorly studied. Here, we assessed 227,860 nights of concurrent sleep data from Ultrahuman AIR ring and M1 continuous glucose monitoring (CGM) system across 5849 adults globally to examine nocturnal cardio-metabolic coupling. We found that higher sleep consistency was inversely associated with glucose variability, and vice versa. Unsupervised clustering of metrics characterizing nightly sleep quality and demographic factors revealed phenotypes corresponding to better vs poorer metabolic management. Clustering on aggregated sleep scores differentiated users on metabolic metrics with larger effect sizes, rather than on base sleep metrics. A subgroup analysis of sleep sessions in the upper and lower quartiles of the sleep-metabolic spectrum, revealed an asymmetric coupling between metabolic and sleep factors in determining phenotype. Nights corresponding to poorer sleep-metabolic management displayed greater shape similarity between nightly heart rate (HR) and glucose curves, compared to sleep sessions with better sleep-metabolic management. These findings demonstrate that multi-sensor digital phenotyping can improve the profiling of sleep and metabolic alignment in largely healthy adults, with simple sleep/wake regularity emerging as a behaviorally tractable determinant of cardio-metabolic homeostasis.