Fischer · PloS one 2017 · Repeated cross-sectional survey analysis · n=53,689

Chronotypes in the US - Influence of age and sex.

Cited 514 times in the scientific literature.

Level 4 - case-series / case-control

Cross-sectional analysis of pooled national survey diary data

PubMed 28636610 · doi:10.1371/journal.pone.0178782 · record verified 2026-08-27

What was done

Researchers evaluated 12 years (2003–2014) of pooled time-use diary data from the American Time Use Survey (ATUS) to estimate population-level chronotypes across age and sex in the United States. Chronotype was calculated from the mid-point of sleep on weekends (MSFWe) across 53,689 participants.

What was found

Chronotype followed a near-normal distribution overall and within age subgroups. The mean shifted later during adolescence, peaking in 'lateness' at approximately 19 years of age before shifting earlier across older age groups. Males showed later chronotypes than females prior to age 40, but earlier chronotypes after age 40. The largest age-related differences occurred between ages 15 and 25 for both sexes, accounting for over 50% of the total chronotype variance across all age groups. Chronotype variability declined with increasing age and was generally higher in males than females.

Why it matters

This provides large-scale normative data on sleep timing and circadian patterns across the human lifespan in the US. The finding that adolescents exhibit the latest chronotypes provides empirical support for delaying school start times and adopting flexible work schedules.

Limits

The abstract reports relative patterns rather than exact numerical mid-point sleep times, confidence intervals, or standard deviations. Chronotype was estimated from self-reported weekend sleep timing rather than objective circadian physiological markers (such as dim-light melatonin onset), and the cross-sectional design cannot directly track individual longitudinal trajectories.

Cited by