Chronic sleep restriction is prevalent in the U.S. population and associated with increased morbidity and mortality. The primary reasons for reduced sleep are unknown. Using population data on time use, we sought to identify individual characteristics and behaviors associated with short sleep that could be targeted for intervention programs.
Analysis of the American Time Use Survey (ATUS).
Cross-sectional annual survey conducted by the U.S. Bureau of Labor Statistics.
Representative cohort (N = 124,517) of Americans 15 years and older surveyed between 2003 and 2011.
Measurements and Results:
Telephone survey of activities over 24 hours. Relative to all other waking activities, paid work time was the primary waking activity exchanged for sleep. Time spent traveling, which included commuting to/from work, and immediate pre- and post-sleep activities (socializing, grooming, watching TV) were also reciprocally related to sleep duration. With every hour that work or educational training started later in the morning, sleep time increased by approximately 20 minutes. Working multiple jobs was associated with the highest odds for sleeping ≤ 6 hours on weekdays (adjusted OR 1.61, 95% CI 1.44; 1.81). Self-employed respondents were less likely to be short sleepers compared to private sector employees (OR 0.83, 95% CI 0.72; 0.95). Sociodemographic characteristics associated with paid work (age 25-64, male sex, high income, and employment per se) were consistently associated with short sleep.
U.S. population time use survey findings suggest that interventions to increase sleep time should concentrate on delaying the morning start time of work and educational activities (or making them more flexible), increasing sleep opportunities, and shortening morning and evening commute times. Reducing the need for multiple jobs may increase sleep time, but economic disincentives from working fewer hours will need to be offset. Raising awareness of the importance of sufficient sleep for health and safety may be necessary to positively influence discretionary behaviors that reduce sleep time, including television viewing and morning grooming.
Basner M, Spaeth AM, Dinges DF. Sociodemographic characteristics and waking activities and their role in the timing and duration of sleep. SLEEP 2014;37(12):1889-1906.