Anticipated next‐day demand affects the magnitude of the cortisol awakening response, but not subjective or objective sleep

    loading  Checking for direct PDF access through Ovid


A range of studies have indicated that subjective and objective sleep disturbances are associated with stress (Bastien et al., 2004; Reynolds et al., 1992, 1993); that stressful events can predict future sleep disturbances (Lallukka et al., 2012; Vahtera et al., 2007); and that pre‐sleep stress levels are predictive of subsequent subjective sleep quality (Åkerstedt et al., 2012). Whilst the link between stress and sleep is well established, the impact of an anticipated stressor upon sleep is less clear, with the exception of a limited number of studies. In one such study, engineering staff completed sleep diaries whilst on board a ship, which included a 24‐h ‘watch period’ whereby they were allowed to sleep, but would be awakened by an automatic alarm system in the event of machinery malfunction (Torsvall et al., 1987). During the watch period, staff self‐reported a reduced total sleep time (TST) and poorer sleep quality compared with free nights and, interestingly, these disruptions were evident even when no alarms occurred. In a similar study, where sleep was measured objectively using polysomnography (PSG), a shorter duration of sleep and reduced amounts of rapid eye movement (REM) and slow‐wave sleep were observed during watch periods, as compared with free nights (Torsvall and Akerstedt, 1988). Therefore, it could be concluded that even the anticipation of an upcoming stressor can affect subjective and objective sleep.
That said, two similar laboratory studies have investigated the effects of an anticipated stressor upon sleep, with very different outcomes. One study examined whether neuroticism, repression and coping style moderated the effects of an anticipated stressor upon REM sleep parameters (Germain et al., 2003). Healthy, good‐sleepers were either allocated to a control group or a next‐day stress condition, and were informed immediately before sleep that they were required to perform a speech, which would be evaluated, upon awakening. Interestingly, there were no group differences in terms of subjective sleep quality, or objective measures of sleep continuity or architecture. A second study from this group, using a similar anticipatory stress‐induction paradigm, measured heart rate variability (HRV) as a marker of autonomic nervous system arousal (Hall et al., 2004). Whilst again there were no between‐group differences in objective sleep continuity or architecture, more subtle differences were observed: participants in the stress condition demonstrated lower parasympathetic modulation during non‐ (N)REM and REM sleep, and a higher sympathovagal balance during NREM sleep compared with the control group, representing an anticipatory stress response during the night. Taken together, the naturalistic studies, where the stressor may have occurred at any point over an 8‐h overnight period (Torsvall and Akerstedt, 1988; Torsvall et al., 1987), and stress‐induction laboratory studies, which require action the subsequent morning, suggest that the timing of the anticipated stressor is potentially important.
Although Hall et al. (2004) measured HRV as a physiological marker of stress, to date, no studies have simultaneously examined the effects of anticipated stress upon sleep in tandem with a more robust, relevant and well‐established physiological marker of stress. One such marker is cortisol, which is the end‐product of the hypothalamic‐pituitary‐adrenal (HPA) axis, an endocrine system that allows adjustment and adaptation to bodily and environmental demands (Fries et al., 2009). Cortisol is responsive to stress; acute psychological demand increases cortisol levels in a dose–response manner (Dickerson and Kemeny, 2004). One specific aspect of cortisol output that appears to be sensitive to an anticipated stressor is the cortisol awakening response (CAR), where cortisol levels sharply increase in response to morning awakening (Clow et al., 2004; Fries et al., 2009). Several ambulatory studies have indicated that the CAR is associated with periods of increased demand (Brant et al.
    loading  Loading Related Articles