Sleep duration statistics vary substantially by population and definition. U.S. time-use data measures sleeping as a daily activity, adolescent surveillance uses an average school-night threshold, and health studies compare reported durations with outcomes in different populations.
Key Sleep Duration Statistics
The most broadly useful sleep duration statistics include:
- In 2024, U.S. people age 15 and over averaged 9.04 hours of sleeping per day in the American Time Use Survey.
- In 2024, U.S. men age 15 and over averaged 8.96 hours of sleeping per day in the American Time Use Survey.
- In 2024, U.S. women age 15 and over averaged 9.11 hours of sleeping per day in the American Time Use Survey.
- In 2024, 99.9% of the U.S. population age 15 and over engaged in sleeping on an average day in the American Time Use Survey.
- In 2024, the U.S. population age 15 and over averaged 8.80 hours of sleeping on weekdays.
- In 2024, the U.S. population age 15 and over averaged 9.59 hours of sleeping on weekends and holidays.
- In 2025, 24% of U.S. high school students got at least 8 hours on an average school night.
- In 2025, 26% of U.S. male high school students got at least 8 hours on an average school night.
- In 2025, 22% of U.S. female high school students got at least 8 hours on an average school night.
- In 2025, 31% of U.S. ninth-graders got at least 8 hours on an average school night.
- In 2025, 20% of U.S. eleventh-graders got at least 8 hours on an average school night.
- From 2015 to 2025, the CDC classified the change in U.S. high school students getting at least 8 hours as worsened.
- In 2017 evidence synthesis, sleeping 7 hours was the reference relative risk of 1.00 for all-cause mortality.
- In 2017 evidence synthesis, sleeping 10 hours was associated with an all-cause mortality relative risk of 1.32 versus 7 hours.
- In 2017 evidence synthesis, each hour below 7 hours was associated with pooled mortality relative risk of 1.06.
- In 2017 evidence synthesis, sleeping 10 hours was associated with total cardiovascular disease relative risk of 1.37 versus 7 hours.
- In 2007–2015 YRBS data, more than 70% of U.S. high school students averaged less than 8 hours of sleep.
- In 2007–2015 YRBS data, students sleeping under 6 hours had mood/self-harm odds of 3.17 versus students sleeping at least 8 hours.
Contents
- Daily sleep and time-use statistics
- Healthy adolescent sleep duration data
- Sleep duration trends from 2015 to 2025
- Sleep duration and all-cause mortality
- Sleep duration and cardiovascular outcomes
- Short sleep and adolescent safety behaviors
Daily Sleep Duration Statistics in U.S. Time-Use Data
The 2024 American Time Use Survey (ATUS) describes sleeping as a primary activity for the U.S. civilian population age 15 and over. Its estimate includes naps and spells of sleeplessness, so it should not be treated as identical to a self-reported school-night sleep threshold.
The overall average was 9.04 hours per day. Men averaged 8.96 hours and women averaged 9.11 hours, while the participation rate was 99.9% for the total population, 99.9% for men, and 100.0% for women.
| 2024 ATUS measure | Total | Men | Women |
|---|---|---|---|
| Average sleeping time per day | 9.04 hours | 8.96 hours | 9.11 hours |
| Sleeping participation on an average day | 99.9% | 99.9% | 100.0% |
Source: American Time Use Survey – 2024 Results. The participation figures are rounded survey estimates, and the sleeping measure includes naps and spells of sleeplessness.
The same ATUS release shows a weekday and weekend pattern for the total population age 15 and over. Average sleeping time was 8.80 hours on weekdays and 9.59 hours on weekends and holidays; participation was 100.0% on weekdays and 99.9% on weekends and holidays.
| 2024 ATUS day type | Average sleeping time | Participation |
|---|---|---|
| Weekday | 8.80 hours | 100.0% |
| Weekend or holiday | 9.59 hours | 99.9% |
These are primary-activity time-use estimates rather than clinical measures of sleep quality, sleep opportunity, or sleep efficiency. They also do not establish that every respondent slept continuously for the reported duration.
Healthy Sleep Duration Statistics Among U.S. High School Students
The CDC Youth Risk Behavior Surveillance System (YRBSS) uses a different definition: the share of U.S. high school students reporting at least 8 hours on an average school night in 2025. The measure is self-reported and applies to school-night duration, not total daily sleeping time.
Only 24% of students met the threshold. The rate was 26% among male students and 22% among female students.
| 2025 student group | At least 8 hours on an average school night |
|---|---|
| All high school students | 24% |
| Male students | 26% |
| Female students | 22% |
| Ninth-graders | 31% |
| Tenth-graders | 25% |
| Eleventh-graders | 20% |
| Twelfth-graders | 20% |
Source: Sleep | Youth Risk Behavior Surveillance System (YRBSS). Values are self-reported average school-night durations.
The 2025 grade pattern ranges from 31% among ninth-graders to 20% among both eleventh- and twelfth-graders. The figures describe threshold attainment, so they do not show the average number of hours among students below or above 8 hours.
Reported percentages also differed across racial and ethnic groups. The supplied CDC values were 49% for American Indian or Alaska Native students, 18% for Asian students, 21% for Black students, 25% for Hispanic or Latino students, 33% for Middle Eastern or North African students, 25% for White students, and 20% for multiracial students.
- American Indian or Alaska Native students: 49% got at least 8 hours.
- Asian students: 18% got at least 8 hours.
- Black students: 21% got at least 8 hours.
- Hispanic or Latino students: 25% got at least 8 hours.
- Middle Eastern or North African students: 33% got at least 8 hours.
- White students: 25% got at least 8 hours.
- Multiracial students: 20% got at least 8 hours.
The subgroup estimates are self-reported, and the source notes that subgroup estimates may differ in precision. They should be read as surveillance results for U.S. high school students in 2025, not as universal sleep requirements.
Sleep Duration Trends Among Adolescents, 2015–2025
The YRBSS series provides a decade view of the percentage of U.S. high school students getting at least 8 hours on an average school night. The series fell from 27% in 2015 to 24% in 2025, with intermediate values of 25% in 2017, 22% in 2019, and 23% in both 2021 and 2023.
| Survey year | Students getting at least 8 hours |
|---|---|
| 2015 | 27% |
| 2017 | 25% |
| 2019 | 22% |
| 2021 | 23% |
| 2023 | 23% |
| 2025 | 24% |
Source: Sleep | Youth Risk Behavior Surveillance System (YRBSS). Each value refers to U.S. high school students reporting an average school-night duration.
The CDC described the 2015–2025 change as worsened, meaning fewer students met the at-least-8-hour threshold over the decade. It described the 2023–2025 change as no change, so the increase from 23% to 24% in the displayed series should not be interpreted as a statistically meaningful improvement.
Because this is a threshold series, it cannot be combined directly with ATUS average hours. The two sources measure different populations, recall contexts, and definitions: daily primary-activity sleeping for people age 15 and over versus an average school-night threshold for high school students.
Sleep Duration Statistics and All-Cause Mortality Risk
A 2017 systematic review and dose-response meta-analysis of prospective cohort studies used 7 hours as the reference category for all-cause mortality. Its relative risks are associations across cohort populations, not causal effects or universal prescriptions.
| Reported sleep duration | Pooled all-cause mortality relative risk vs. 7 hours |
|---|---|
| 3 hours | 1.12 (95% CI 1.10–1.14) |
| 4 hours | 1.08 (95% CI 1.06–1.09) |
| 5 hours | 1.04 (95% CI 1.03–1.05) |
| 6 hours | 1.01 (95% CI 1.00–1.01) |
| 7 hours | 1.00 reference |
| 8 hours | 1.04 (95% CI 1.04–1.05) |
| 9 hours | 1.15 (95% CI 1.14–1.16) |
| 10 hours | 1.32 (95% CI 1.29–1.35) |
| 11 hours | 1.53 (95% CI 1.47–1.59) |
Source: Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies. Estimates at extreme durations may be based on sparse data.
The dose-response analysis estimated that each 1-hour reduction below 7 hours was associated with mortality relative risk of 1.06, with a 95% confidence interval of 1.04–1.07. Each 1-hour increment above 7 hours was associated with relative risk of 1.13, with a 95% confidence interval of 1.11–1.15.
The short-sleep trend had moderate-to-high heterogeneity (I2 55.5%), while the long-sleep trend had high heterogeneity (I2 76.5%). Long sleep may reflect underlying illness, so the association cannot be interpreted as proof that longer sleep caused higher mortality.
Sleep Duration Statistics and Cardiovascular Outcomes
The same 2017 evidence synthesis examined total cardiovascular disease (CVD), coronary heart disease (CHD), and stroke. The reference and definitions varied by analysis, and the findings are associations from prospective cohort populations.
For total CVD, the pooled estimates compared with 7 hours included 1.09 at 4 hours, 1.05 at 5 hours, 1.02 at 6 hours, 1.03 at 8 hours, 1.16 at 9 hours, and 1.37 at 10 hours.
| Total CVD sleep duration | Relative risk vs. 7 hours |
|---|---|
| 4 hours | 1.09 (95% CI 1.06–1.13) |
| 5 hours | 1.05 (95% CI 1.03–1.08) |
| 6 hours | 1.02 (95% CI 1.00–1.03) |
| 8 hours | 1.03 (95% CI 1.02–1.05) |
| 9 hours | 1.16 (95% CI 1.13–1.19) |
| 10 hours | 1.37 (95% CI 1.29–1.45) |
For the dose-response models, each hour below 7 was associated with total CVD relative risk of 1.06 (95% CI 1.03–1.08), while each hour above 7 was associated with 1.12 (95% CI 1.08–1.16). The shortest-sleep category had total CVD relative risk of 1.14, and the longest-sleep category had 1.36.
The analysis also estimated CHD relative risk of 1.07 per hour below 7 and 1.05 per hour above 7. For stroke, the corresponding estimates were 1.05 below 7 hours and 1.18 above 7 hours; these figures are not causal effects.
The total-CVD short-sleep trend had moderate heterogeneity (I2 52.0%), and the long-sleep trend had high heterogeneity (I2 75.3%). Category definitions varied among cohorts, and the longest-category estimate had high heterogeneity (I2 71.2%).
Short Sleep Duration and Adolescent Safety Behaviors
A CDC-hosted JAMA Pediatrics research letter analyzed self-reported U.S. high school student data from the 2007–2015 Youth Risk Behavior Surveys. More than 70% of students averaged less than 8 hours of sleep, but this older prevalence estimate should not be treated as current 2025 prevalence.
Compared with students sleeping at least 8 hours, students sleeping under 6 hours had higher adjusted odds for several reported outcomes. The models adjusted for age, sex, race/ethnicity, and survey year where specified, but the findings remain cross-sectional associations.
| Reported outcome among students sleeping under 6 hours | Odds ratio vs. at least 8 hours |
|---|---|
| Other-drug use | 2.34 (95% CI 2.16–2.52) |
| Risky sexual activity | 1.65 (95% CI 1.53–1.78) |
| Aggressive behavior | 1.91 (95% CI 1.76–2.06) |
| Mood or self-harm measure | 3.17 (95% CI 2.92–3.44) |
| Considering suicide | 3.12 (95% CI 2.85–3.41) |
| Suicide plan | 3.17 (95% CI 2.87–3.51) |
| Suicide attempt | 3.39 (95% CI 3.00–3.82) |
| Suicide attempt resulting in treatment | 4.24 (95% CI 3.53–5.10) |
Source: Dose-Dependent Associations Between Sleep Duration and Unsafe Behaviors Among US High School Students. These are self-reported, cross-sectional associations; they do not show that sleep duration caused any outcome.
The largest reported association in this set was for a suicide attempt resulting in treatment, with an odds ratio of 4.24 for students sleeping under 6 hours compared with those sleeping at least 8 hours. Residual confounding, reverse causation, and the cross-sectional design limit interpretation.
Across these sources, the central measurement lesson is consistent: daily time-use averages, school-night thresholds, and pooled health-outcome estimates answer different questions. Sleep duration statistics are most useful when the population, period, threshold or reference category, and study limitations remain explicit.