Longer Sleep Linked to Lower Glycemic Variability, Reduced Deviation Risk in Adolescents

In free-living adolescents, longer nightly sleep was associated with lower next-day glycemic variability and reduced risk of extreme glucose excursions, with higher daytime glycemic variability in turn predicting shorter subsequent sleep, according to findings published in Sleep.¹ Glucose homeostasis is critically affected by sleep, with sleep disruptions impairing the body’s ability to regulate blood glucose.²

Researchers from the Copenhagen Prospective Studies on Asthma in Childhood (COPSAC) examined bidirectional associations between nightly sleep duration and continuous glucose monitoring (CGM)–derived glycemic outcomes in a real-world adolescent cohort.¹ Using wrist-worn accelerometry alongside CGM over a 14-day free-living monitoring period, the study captured high-resolution, temporally ordered data in 206 adolescents (mean age, 17.7 years), yielding 2245 person-days of overlapping sleep and glucose recordings (median 13 days per participant). Multivariable linear mixed-effects models adjusted for sociodemographic, behavioral, circadian, and cardiometabolic factors were used to examine associations between each prior night's sleep duration and CGM outcomes during the subsequent waking period.

On a day-to-day basis, each additional hour of sleep was associated with higher next-day median glucose (β = 0.39 mg/dL; 95% CI, 0.15-0.63; P = .002), lower glycemic variability as reflected by glucose standard deviation (β = −0.12 mg/dL; 95% CI, −0.23 to −0.01; P = .036), and a more favorable deviation risk profile, including lower Average Daily Risk Range (ADRR; β = −0.27; 95% CI, −0.43 to −0.10; P = .002). Time-varying models aligned to each participant's wake-up time revealed that the association between longer sleep and higher mean glucose was strongest in the first 7 hours after waking, attenuating toward zero by 10 to 11 hours post wake and turning negative thereafter. Reductions in glucose variability emerged later in the day, from approximately 10 to 16 hours after waking.

Within- vs Between-Person Effects

Decomposing sleep duration into within-person day-to-day fluctuations and between-person habitual differences revealed distinct patterns across glycemic domains. On days when participants slept longer than their average, median glucose was higher (β = 0.50 mg/dL; P = .002) and deviation risk was lower, as reflected by a reduced ADRR (β = −0.33; P = .003). In contrast, participants who habitually slept longer demonstrated markedly lower glycemic variability—a lower glucose standard deviation (β = −0.57 mg/dL; P = .002)—without corresponding differences in mean glucose levels. These findings suggest that day-to-day sleep variation is more closely tied to acute fluctuations in glucose concentration and risk, while habitual sleep duration is more strongly related to long-term glycemic stability.

Bidirectionality and Prewake Mediation

The study also identified a bidirectional relationship between sleep and glucose regulation. Higher daytime glycemic variability was associated with shorter subsequent sleep duration (β = −0.11 hours per SD increase in glucose SD; P < .001), as was greater time spent in hyperglycemia (β = −0.12 hours; P < .001). Glucose concentration measures were not associated with subsequent sleep duration.

Mediation analysis identified a steeper prewake glucose rise, averaging 9.7 mg/dL in the 2 hours before waking, as a partial mediator of the association between longer sleep and higher next-day median glucose (indirect effect 5.0%; P = .036). The investigators interpreted this preawakening glucose increase as consistent with a dawn-like physiological response, whereby longer sleep amplifies hypothalamic-pituitary-adrenal axis and sympathetic output near wake-up, transiently elevating glucose.

Sex differences were noted for glycemic concentration outcomes and glycemic risk assessment diabetes equation scores, with stronger associations observed in men than women. Weekend effects were also more pronounced, potentially reflecting recovery sleep and associated behavioral and circadian shifts common in adolescence. No significant interactions were observed with polygenic risk scores for sleep duration, fasting glucose, hemoglobin A_1c, or type 2 diabetes, suggesting that modifiable behavioral and circadian factors predominate in shaping these associations rather than genetic predisposition, according to the study.

Study Limitations and Implications

The authors acknowledged several limitations, including the observational design, which precludes causal inference, as well as the absence of sleep architecture data and unstandardized dietary intake. The cohort was also relatively socioeconomically advantaged, which may limit generalizability to populations facing greater environmental or social stressors.

Despite modest absolute effect sizes, the investigators underscored that glycemic variability is increasingly recognized as a driver of vascular and inflammatory stress, and that adolescence represents a developmentally sensitive period during which early behavioral patterns may begin to shape long-term cardiometabolic trajectories. The findings point to the awakening transition as a key physiological window through which sleep influences glucose dynamics and suggest that stabilizing sleep may support glycemic stability, and vice versa, through reciprocal coupling.

References

1. Horner D, Evensen K, Ye Z, et al. Night-to-night sleep duration and wake-anchored glycaemia: associations with continuous glucose monitoring in free-living adolescents. Sleep. Published online June 18, 2026. doi:10.1093/sleep/zsag158

2. Knutson KL. Impact of sleep and sleep loss on glucose homeostasis and appetite regulation. Sleep Med Clin. 2007;2(2):187-197. doi:10.1016/j.jsmc.2007.03.004