Sleep During the COVID-19 Pandemic: Longitudinal Observational Study Combining Multisensor Data With Questionnaires

Nguyen Luong,Gloria Mark,Juhi Kulshrestha,Talayeh Aledavood

DOI: https://doi.org/10.2196/53389

2024-09-03

Abstract:Background: The COVID-19 pandemic prompted various containment strategies, such as work-from-home policies and reduced social contact, which significantly altered people's sleep routines. While previous studies have highlighted the negative impacts of these restrictions on sleep, they often lack a comprehensive perspective that considers other factors, such as seasonal variations and physical activity (PA), which can also influence sleep. Objective: This study aims to longitudinally examine the detailed changes in sleep patterns among working adults during the COVID-19 pandemic using a combination of repeated questionnaires and high-resolution passive measurements from wearable sensors. We investigate the association between sleep and 5 sets of variables: (1) demographics; (2) sleep-related habits; (3) PA behaviors; and external factors, including (4) pandemic-specific constraints and (5) seasonal variations during the study period. Methods: We recruited working adults in Finland for a 1-year study (June 2021-June 2022) conducted during the late stage of the COVID-19 pandemic. We collected multisensor data from fitness trackers worn by participants, as well as work and sleep-related measures through monthly questionnaires. Additionally, we used the Stringency Index for Finland at various points in time to estimate the degree of pandemic-related lockdown restrictions during the study period. We applied linear mixed models to examine changes in sleep patterns during this late stage of the pandemic and their association with the 5 sets of variables. Results: The sleep patterns of 27,350 nights from 112 working adults were analyzed. Stricter pandemic measures were associated with an increase in total sleep time (TST) (β=.003, 95% CI 0.001-0.005; P<.001) and a delay in midsleep (MS) (β=.02, 95% CI 0.02-0.03; P<.001). Individuals who tend to snooze exhibited greater variability in both TST (β=.15, 95% CI 0.05-0.27; P=.006) and MS (β=.17, 95% CI 0.03-0.31; P=.01). Occupational differences in sleep pattern were observed, with service staff experiencing longer TST (β=.37, 95% CI 0.14-0.61; P=.004) and lower variability in TST (β=-.15, 95% CI -0.27 to -0.05; P<.001). Engaging in PA later in the day was associated with longer TST (β=.03, 95% CI 0.02-0.04; P<.001) and less variability in TST (β=-.01, 95% CI -0.02 to 0.00; P=.02). Higher intradaily variability in rest activity rhythm was associated with shorter TST (β=-.26, 95% CI -0.29 to -0.23; P<.001), earlier MS (β=-.29, 95% CI -0.33 to -0.26; P<.001), and reduced variability in TST (β=-.16, 95% CI -0.23 to -0.09; P<.001). Conclusions: Our study provided a comprehensive view of the factors affecting sleep patterns during the late stage of the pandemic. As we navigate the future of work after the pandemic, understanding how work arrangements, lifestyle choices, and sleep quality interact will be crucial for optimizing well-being and performance in the workforce.