Daily Relations Between Stress and Electroencephalography-Assessed Sleep: A 15-Day Intensive Longitudinal Design With Ecological Momentary Assessments

Abstract

Background Recent studies have found bi-directional relations between stress and sleep. However, few studies have examined the daily associations between stress and electroencephalography (EEG) measured sleep. Purpose This study examined the temporal associations between repeated ecological momentary assessments of stress and EEG-estimated sleep. Methods Ninety-eight international or interstate undergraduate students (Mage = 20.54 ± 1.64, 76.5% female, 84.7% Asian) reported their stress levels four times daily at morning awakening, afternoon, evening, and pre-bedtime across 15 consecutive days (>4,000 total observations). Next-day stress was coded as an average of morning, afternoon, and evening stress. Z-Machine Insight+ recorded over 1,000 nights EEG total sleep time (TST), sleep onset latency, wake after sleep onset, sleep efficiency (SE), slow-wave sleep (SWS), and rapid eye movement (REM) sleep duration. Multilevel models, adjusted for covariates (i.e., sociodemographic, health factors, and daily covariates) and lagged outcomes, tested the daily within- and between-level stress-sleep associations. Results After adjusting for covariates, within-person shorter TST (b = −0.11 [−0.21, −0.01], p = .04), lower SE (b = −0.02 [−0.03, 0.00], p = .04), less SWS (b = −0.38 [−0.66, −0.10], p = .008), and less REM sleep (b = −0.32 [−0.53, −0.10], p = .004) predicted higher next-day stress. Pre-bedtime stress did not predict same-night sleep. No significant results emerged at the between-person level. Conclusions These findings demonstrate that poor or short sleep, measured by EEG, is predictive of higher next-day stress. Results for sleep architecture support the role of SWS and REM sleep in regulating the perception of stress. Given that only within-person effects were significant, these findings highlight the importance of examining night-to-night fluctuations in sleep affecting next-day stress and its impact on daytime functioning.

Protocol of the Budapest sleep, experiences, and traits study: An accessible resource for understanding associations between daily experiences, individual differences, and objectively measured sleep

“…Simply put, if participants experience very similar experiences and very similar sleep on all days, we can learn little about the relationship between these two variables in a daily observational study like BSETS, which effectively uses participants as their own controls. However, previous studies [ 20 , 22 , 24 ] reported moderate intraclass correlation coefficients, that is, indicators of within-individual similarity, of daily experiences and sleep variables, suggesting that in practice, we will see sufficient within-individual variance in BSETS to enable meaningful analyses.…”

“…However, some outstanding issues remain. First, while some studies [ 22 , 24 ] used EEG to provide an objective measurement of sleep, this is not standard procedure and this may be problematic as self-reports of sleep are only moderately accurate [ 25 ]. To our knowledge, no daily observational study used quantitative EEG to establish the link between daily experience and sleep, although based on experimental studies [ 81 , 82 ] it is highly expected that daily experiences will affect waveforms during the subsequent nights, for example because richer experiences require additional synaptic downscaling through increased slow-wave activity [ 4 ].…”

“…The main limitation of current daily observational studies is the reliance on self-reports in sleep timing and quality assessments. While a handful of recent daily observational studies [ 22 , 24 ] employed mobile EEG devices to establish the link between sleep and DAEs, this is not the case with most studies. While self-reports may provide valuable information about sleep, the accuracy of self-reports are at best moderate [ 25 ], and the estimation of sleep macrostructure (such as the proportion of NREM and REM) is not possible based on self-reports.…”

“…A possible general limitation of both experimental and daily observational studies is ignoring the role of between-individual variation. Sleep features across multiple nights tend to be highly similar within the same individual [ 20 , 22 , 26 – 28 ], in a large part due to genetic reasons [ 29 – 32 ]. The same is true for daily experiences ([ 20 , 22 , 24 ]), which are not random but in part follow the trait-like predispositions of individuals and tend to be themselves highly heritable [ 33 – 36 ].…”

Sleep and day-to-day PTSD symptom variability: an ecological momentary assessment and actigraphy monitored study in trauma-exposed young adults

“…However, a recent publication used a similar device to test temporal relationships between EEG measured sleep and daytime stress over two weeks in a non-clinical sample. They found that several sleep markers predicted next-day stress level (Yap et al, 2022 ) indicating feasibility of such methods.…”