Abstract

     Prior research suggests that sleep is associated with increased subjective stress and aggression, but important questions remain about the typical magnitude of these relationships, as well as their potential moderators. We therefore conducted the first meta-analysis of this literature. Across 340 associational and experimental studies, significant associations were identified between sleep with both subjective stress (r = 0.307, p < .001) and aggression (r = 0.258, p < .001) in individuals from the general population, as well as between sleep with subjective stress (r = 0.425, p < .001) in individuals with sleep disorders. Experimental sleep restriction also led to increased subjective stress (g = 0.403, p = .017) and aggression (g = 0.330, p = .042). These findings suggest that poorer sleep is associated with – and leads to – heightened levels of subjective stress and aggression. These findings, and their implications, are discussed in relation to neurobiological literature, which highlights the complex interplay between metabolic activity in the brain, hormonal changes, and behavior.

Introduction

     Sleep is a fundamental biological process that is critical for a range of key brain functions including cognition and mental health (Lo et al., 2015; Lovato and Gradisar, 2014; Pires et al., 2016). It is therefore of concern that, in recent years, the average nightly sleep duration per individual has decreased, with sleep disorders such as insomnia continuing to increase in prevalence (Adams et al., 2016; Ford et al., 2015; Knutson et al., 2010; Kronholm et al., 2008). This is largely due to both population aging (as sleep disturbances are more common in older age) and engagement in lifestyle behaviors that negatively impact sleep quality (such as use of internet devices and consumption of energy drinks; Adams et al., 2016). Sleep disturbances are also a common feature of many neurological and psychiatric disorders (Adams et al., 2016; Petit et al., 2004; Spiegelhalder et al., 2013; Videnovic et al., 2014), and chronic sleep disturbances or disorders have been linked to poorer health outcomes such as diabetes and cardiovascular disease (see Sigurdson and Ayas, 2007 for a review).

     The present meta-analysis was designed to investigate the relationship between sleep with both subjective stress and aggression. Although these are distinct constructs, both reflect important behavioral outcomes that have been investigated frequently in relation to sleep disturbances, and can negatively affect health and psychosocial wellbeing (Bosch et al., 2013; DeLongis et al., 1988; Jennings et al., 2017; Meyrueix et al., 2015; Paterson et al., 2011; Wu et al., 2015; Quick et al., 1987). For instance, stress can have detrimental effects on everyday cognitive functioning and mood (Bosch et al., 2013; DeLongis et al., 1988; Quick et al., 1987). Similarly, aggression is associated with generalized anxiety and depression symptoms (Meyrueix et al., 2015; Paterson et al., 2011; Wu et al., 2015), as well as increased cardiovascular risk and mortality (see Jennings et al., 2017). This study will therefore investigate the relationship between sleep with both subjective stress and aggression.

Results

Literature search

     In total, 340 studies met the inclusion criteria. This included 278 studies assessing the association between sleep and subjective stress in individuals from the general population, nine assessing the association between sleep and subjective stress in individuals with sleep disorders, 50 assessing the association between sleep and aggression in individuals from the general population, six assessing the causal impact of sleep restriction on subjective stress, and 10 assessing the causal impact of sleep restriction on aggression. The contributing studies comprised papers and theses published between 1977 and 2021. Overall, the studies included data from a total of 388,934 participants with an average age of 32.19 (SD = 13.46). Data was also gathered from countries across Asia, Africa, North America, South America, Europe, and Australia (see Supplementary Document for key characteristics of all studies contributing to the meta-analyses).