Elisa Meth: From consequences of sleep loss to interventions

Abstract

Insufficient sleep and circadian misalignment impair physiological, cognitive, and emotional functioning, yet the mechanisms underlying individual vulnerability remain poorly understood. These gaps are especially evident in women, whose hormonal fluctuations across the menstrual cycle interact with sleep and stress systems but have been historically understudied. This thesis integrates experimental, physiological, psychological, and epidemiological approaches to examine how sleep, endogenous hormones, personality traits, and sensory-based interventions influence stress reactivity, sleep loss responses, and long-term metabolic risk.

Across four studies, mechanistic and translational questions related to sleep, stress, and neuroendocrine regulation are addressed. Paper I shows that bedtime use of a weighted blanket produces an increase in pre-sleep salivary melatonin concentrations compared with a light blanket, with no effects observed for cortisol, alpha-amylase, or oxytocin. These findings suggest that deep-pressure stimulation may influence circadian-related physiology and support further investigation of weighted blankets as a non-pharmacological sleep aid.

Paper II demonstrates that higher morning estradiol levels are associated with reduced autonomic arousal, indexed by smaller pupil dilations, during a socially evaluative karaoke task, indicating a buffering role of estradiol in stress reactivity. In addition, greater amounts of N3 sleep, particularly slow-wave sleep, were associated with lower subjective stress the following morning, highlighting complementary hormonal and sleep-related contributions to emotional resilience.

In Paper III, acute sleep deprivation in naturally cycling women led to increased circulating levels of p-Tau181, a biomarker linked to neurodegenerative processes, alongside impaired vigilance performance. Higher levels of neuroticism were associated with amplified biomarker responses, suggesting that psychological traits modulate vulnerability to sleep loss at both molecular and behavioral levels.

Finally, Paper IV analyzes data from over 200,000 adults in the UK Biobank, demonstrating that both short habitual sleep duration and unhealthy dietary patterns are independently associated with an increased risk of developing type 2 diabetes, with no evidence of interaction between these factors.

Together, these studies demonstrate that sleep, hormonal status, stress sensitivity, and lifestyle behaviors interact across multiple levels to shape both acute stress responses and long-term health outcomes. This thesis provides a multilevel framework for understanding sleep-related vulnerability and resilience in women and highlights the importance of integrating biological and behavioral factors in disease prevention.