Abstract
Long sleep duration has been consistently associated with increased cardiovascular disease (CVD) risk, yet the physiological mechanisms underlying this relationship remain insufficiently understood, particularly among middle-aged adults. Furthermore, the extent to which psycho-behavioral factors moderate these biological relationships has not been adequately examined. Therefore, the goal of this dissertation is to investigate the relationships among long sleep duration, psycho-behavioral factors, and CVD risk by examining two plausible biological mechanisms—autonomic dysfunction and systemic inflammation—in middle-aged adults in the United States. To achieve this goal, three manuscripts are presented.
The first manuscript presents a systematic review of eight observational studies synthesizing current evidence on the association between long sleep duration and autonomic nervous system (ANS) function across age groups. The synthesized findings demonstrated that long sleep duration was broadly associated with autonomic dysregulation, including reduced parasympathetic tone and/or diminished/heightened sympathetic function. Methodological heterogeneity was notable, with long sleep definitions ranging from ≥7 to ≥10 hours and various autonomic function measurements. This review supports autonomic dysregulation as a plausible mechanism linking long sleep duration to autonomic dysregulation in older age and clinical groups and identifies the need for standardized, age-specific sleep definitions.
The second and third manuscripts were derived from a cross-sectional analysis of data from the Midlife in the United States (MIDUS 3) study, which assessed sleep duration, psycho-behavioral factors, and systemic inflammation—indexed by C-reactive protein (CRP) and interleukin-6 (IL-6)—among middle-aged U.S. adults (ages 40–64). Both subjective sleep duration (Pittsburgh Sleep Quality Index) and objective sleep duration (7-day wrist actigraphy) were examined. The second manuscript examined the direct effects of sleep duration on inflammatory biomarkers and tested physical activity as a moderator. The third manuscript similarly examined the direct effects of sleep duration on CRP and tested depression as a moderator.
In the second manuscript, sleep duration was not independently associated with CRP or IL-6 after controlling for covariates; however, physical activity significantly moderated these relationships. Among physically active individuals, shorter sleep duration (< 6 hours) was associated with lower CRP compared to normal sleepers, suggesting a potential buffering effect of physical activity against sleep restriction-related inflammation. In contrast, physically active individuals with long sleep duration (> 7.3 hours) showed significantly higher IL-6 compared to normal sleepers, proposing that long sleep in this group may reflect underlying physiological vulnerability rather than restoration. In the third manuscript, sleep duration was not independently associated with CRP among middle-aged adults after controlling for covariates. However, depression significantly modified the relationship between longer sleep duration and CRP. Among non-depressed individuals, longer sleep duration was associated with higher CRP compared to normal sleepers; among depressed individuals, who exhibited significantly higher baseline CRP overall, no such pattern was observed—suggesting that chronic inflammatory activation driven by depression may override the incremental inflammatory effects of abnormal sleep duration.
This dissertation suggests that long sleep duration is defined differently across studies, which limits comparability. It is consistently associated with autonomic dysregulation, characterized by diminished parasympathetic function and diminished or heightened sympathetic function in old adults and individuals with clinical conditions. This suggests that autonomic dysregulation may be a potential mechanism linking long sleep duration to cardiovascular disease. Long sleep duration is not a uniform behavioral exposure but a contextually dependent signal whose cardiovascular implications are shaped by psycho-behavioral context. Providing appropriate clinical support to identify and address the underlying drivers of longer sleep—such as depression, sleep fragmentation, or subclinical comorbidities—should be incorporated into strategies for CVD risk reduction in this population.
