Temporal regulation of sleep structure in real life

• Applicant: Eva Winnebeck, Ph.D.
• Subject Area: Epidemiology and Medical Biometry/Statistics ; Public Health, Healthcare Research, Social and Occupational Medicine
• Term: since 2020
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 450622422

Sleep is intricately linked with short and long term health. Studying this link in humans requires a combination of investigative approaches owing to the complex organisation and regulation of sleep on scales from milliseconds to hours and days to months. Health cohorts usually rely on questionnaires to assess sleep, which provide important information on sleep timing also longitudinally but miss important information on sleep structure, the dynamic physiological processes during sleep. In turn, experimental laboratory studies provide high-resolution insights into sleep physiology but are conducted isolated from real-life conditions and are limited in their sample size and number of sleep episodes monitored per individual.We recently established a method that can bridge the gap between coarse questionnaire and detailed laboratory measures by estimating sleep structure from wrist-activity recordings (actigraphy). Activity can be recorded longitudinally under ecologically meaningful settings and has been used for decades in human sleep and circadian research. Instead of just estimating sleep times from actigraphy, our method exploits the residual movement during sleep to quantify patterns in sleep physiology. Our measure of Locomotor Inactivity During Sleep (LIDS) captures both ultradian sleep cycles as well as gradual changes reminiscent of the dynamics of EEG slow wave activity.In this project, I propose to harness this advanced actigraphy analysis for use in large datasets to study the temporal regulation of sleep structure, in particular sleep cycles, in real life. From lab studies, we know that sleep structure is under strong regulation by homeostatic and circadian processes. In everyday life, homeostatic and circadian challenges are omnipresent - from work hours, social obligations and suboptimal light exposure patterns - resulting in chronic sleep loss and circadian disruption. In order to gauge the health impacts of such persistent temporal challenges, we need to understand their effects on sleep structure in real life and if these differ from those under acute challenges. For sleep cycles, even acute effects of homeostatic and circadian manipulations remain undetermined given the need for big-data.I therefore propose to study the temporal regulation of sleep structure based on actigraphy by combining analyses in two large observational datasets and two experimental studies of homeostatic and circadian manipulation. Identified associations in the Munich Activity Database (n>700; 20,000 sleep bouts) and the health cohort UK Biobank (n>85,000; 560,000 sleep bouts) will be compared to effects in sleep restriction and forced desynchrony experiments (and vice versa) to determine causality and differences between acute and chronic temporal challenges. A better understanding of real-life sleep structure and its temporal organisation is a prerequisite for studying its link with health and dissecting its mechanistic underpinnings.

• DFG Programme: Research Grants