The clearest current national estimate is that 30.5% of U.S. adults reported sleeping less than seven hours on average in 2024. In the same survey, 15.4% reported trouble falling asleep and 18.1% reported trouble staying asleep most days or every day during the past month 1.
Those figures come from interviews, not sleep studies. They describe the U.S. civilian noninstitutionalized adult population and do not diagnose insomnia or explain why someone slept poorly. Other widely quoted sleep statistics use different populations and methods. This guide keeps the population, year, measure, and limitation beside each number.
A current U.S. sleep snapshot
| Measure | Estimate and collection period | What the number represents |
|---|---|---|
| Adults sleeping less than 7 hours | 30.5% in 2024, 95% CI 29.8% to 31.1% 1 | Self-reported average sleep in a 24-hour period among U.S. adults |
| Adults waking well-rested | 54.8% in 2024, 95% CI 54.1% to 55.6% 1 | Self-reported feeling on most days or every day in the past 30 days |
| Adults with trouble falling asleep | 15.4% in 2024, 95% CI 14.9% to 15.9% 1 | A frequent symptom, not a clinical insomnia diagnosis |
| Adults with trouble staying asleep | 18.1% in 2024, 95% CI 17.6% to 18.6% 1 | A frequent symptom, not a clinical insomnia diagnosis |
| Adults using a sleep aid | 12.9% in 2024, 95% CI 12.4% to 13.4% 2 | Self-reported use most days or every day in the past 30 days |
| High school students getting at least 8 hours | About 1 in 4 in 2023 3 | Self-reported sleep on an average school night among U.S. high school students |
| Adults age 20 and older with obstructive sleep apnea | 83.7 million, or 32.4%, projected for 2024 4 | A literature-based model using a sleep-test threshold, not a diagnosed-case count |
| Deaths in crashes coded as drowsy-driving related | 644 in 2024 5 | An official police-report-based count that does not capture every drowsy-driving crash |
| Wage and salary workers usually on a non-daytime schedule | 16.4% in 2017 to 2018 6 | A national survey of workers' sole or main jobs, excluding self-employed workers |
CI means confidence interval. It describes sampling uncertainty around a survey estimate. It does not include every possible source of error, such as imperfect recall or differences in how people interpret a question.
How to read sleep statistics
Before comparing two figures, check five details:
- Who was studied? Adults, high school students, drivers, workers, and sleep-clinic patients are different populations.
- Where and when were data collected? A U.S. estimate from 2024 does not describe every country or establish the rate in 2026.
- What was measured? Hours slept, difficulty sleeping, a diagnosis, and a laboratory breathing index are not interchangeable.
- How was it measured? A questionnaire, time diary, sleep study, police report, and statistical model each have different strengths and blind spots.
- Is the figure a percentage, count, or rate? Prevalence is the share of a population with a condition or characteristic at a point or over a period. Incidence counts new cases over time. Most figures on this page are prevalence estimates, not incidence.
A difference between groups is also not proof that group membership caused the difference. Surveys can document patterns, but they usually cannot isolate the effects of work schedules, healthcare access, housing, stress, discrimination, health conditions, or other possible explanations.
Adult sleep duration and sleep difficulties
The 2024 National Health Interview Survey, or NHIS, is a nationally representative household survey of the U.S. civilian noninstitutionalized population. Adults reported their average hours of sleep in a 24-hour period, rounded to whole hours. Responses from zero through six hours were classified as short sleep duration 1.
Short sleep varied by age. It was reported by 27.2% of adults ages 18 to 34, 33.4% of those ages 35 to 49, 34.5% of those ages 50 to 64, and 27.2% of those age 65 and older. Men and women had nearly identical estimates, at 30.6% and 30.4% 1.
The report also found differences across its race and Hispanic-origin categories. Short sleep was reported by 40.2% of non-Hispanic Black adults, compared with 27.9% of non-Hispanic Asian adults, 28.9% of non-Hispanic White adults, and 29.0% of Hispanic adults. These are descriptive survey differences. The report does not show that race itself caused them or identify how much any social, environmental, occupational, or health factor contributed 1.
Sleep difficulty is not the same as insomnia
In the NHIS, 15.4% of adults said they had trouble falling asleep most days or every day in the past 30 days. The figure was 18.1% for trouble staying asleep. Women reported both symptoms more often than men: 18.5% versus 12.2% for falling asleep, and 21.4% versus 14.6% for staying asleep 1.
These percentages cannot be added together. One person may have reported both problems. They also should not be labeled the national prevalence of insomnia. An insomnia diagnosis considers the pattern, duration, daytime effects, sleep opportunity, and other possible causes. The survey questions captured two symptoms and their frequency over one month.
Feeling rested is another separate measure. In 2024, 54.8% of adults reported waking well-rested most days or every day. A person can meet a duration threshold and still have fragmented or poorly timed sleep, while a duration below seven hours does not by itself identify the reason 1.
Why another CDC survey gives different figures
CDC's 2022 Behavioral Risk Factor Surveillance System, or BRFSS, found that the age-adjusted percentage of adults reporting less than seven hours varied from 30% in Vermont to 46% in Hawaii. Within the BRFSS series, the national percentage did not change significantly from 2013 through 2022 7.
BRFSS is a telephone survey, while NHIS is a household interview survey. The surveys also cover different years and use their own sampling and weighting methods. The 2022 state estimates should not be compared directly with the 2024 NHIS figure to claim that adult sleep improved or worsened. Trends should be assessed within the same survey series whenever possible.
Sleep among U.S. high school students
In the 2023 national Youth Risk Behavior Survey, about one in four high school students reported getting at least eight hours of sleep on an average school night. Male students were more likely than female students to report at least eight hours, and the overall percentage decreased from 2013 to 2023 3.
This statistic applies to students attending U.S. high schools, not every adolescent. It reflects an average school night and a self-reported threshold, not sleep measured with a device. It also does not show why sleep was short. School schedules, homework, work, caregiving, health, social conditions, and personal habits may all matter, but this surveillance result alone cannot assign a cause.
Sleep-aid use
A separate analysis of the 2024 NHIS found that 12.9% of U.S. adults used at least one sleep aid most days or every day during the past 30 days to help them fall or stay asleep. The survey asked separately about prescription medication, over-the-counter medication or supplements, and marijuana or cannabidiol products 2.
The individual estimates were 5.2% for prescription medication, 5.7% for over-the-counter medication or supplements, and 3.7% for marijuana or cannabidiol. They sum to more than 12.9% because respondents could report more than one category. The survey did not establish whether a product was effective, safe for that person, or recommended by a clinician 2.
Use of any sleep aid increased across the age groups in the report, from 10.5% among adults ages 18 to 34 to 15.8% among adults age 65 and older. The mix changed with age: prescription and over-the-counter use increased, while marijuana or cannabidiol use decreased 2.
How common is obstructive sleep apnea?
A 2025 analysis projected that 83.7 million U.S. adults age 20 and older, or 32.4%, had obstructive sleep apnea in 2024. It defined the condition as at least five apneas or hypopneas with a 4% or greater oxygen desaturation per hour of sleep, known as an AHI4 of at least 5 4.
This is a modeled prevalence estimate, not the number of people diagnosed in medical records. The researchers reviewed eligible prevalence studies, estimated a 2004 baseline, and projected it to 2024 using an obesity population-attributable adjustment. Their model estimated prevalence of 39.1% among males and 26.0% among females, with 52% of modeled cases in the mild range 4.
The estimate depends on the included studies, the AHI4 definition, and the projection assumptions. A different breathing-event definition or population can produce a different result. Several authors worked for Apnimed, which is developing medication for obstructive sleep apnea, or for Clarivate Analytics, and the remaining author disclosed consulting relationships that included Apnimed. The paper is useful as a current model, but it should not be presented as an official census or a diagnosed-case count 4.
Drowsy-driving statistics
The National Highway Traffic Safety Administration reports that 644 people died in crashes coded as drowsy-driving related in the United States in 2024 5. This is an official count based on crash reporting, but drowsiness can be difficult to identify after a crash. There is no breath or blood test that confirms it, a driver may not survive or remember the event, and investigators may lack evidence of prior sleep 8.
A 2024 AAA Foundation for Traffic Safety analysis used in-depth crash investigations to build a model that inferred drowsiness when alertness was unknown. It estimated that 17.6% of fatal crashes from 2017 through 2021 involved a drowsy driver and that 29,834 people died in those crashes over the five-year period 8.
The model covered fatal crashes involving passenger vehicles and did not examine drivers of large trucks, buses, or motorcycles. Its estimate should not be compared directly with the 644 deaths recorded for 2024. The time periods, vehicle scope, and measurement methods differ. Together, the sources show why a police-coded count and a modeled estimate can be far apart without either figure serving as a precise total.
If you are struggling to keep your eyes open, drifting from your lane, missing exits, or forgetting the last part of a drive, stop driving as soon as it is safe. A population statistic cannot determine whether you are safe behind the wheel.
Shift-work statistics
In the 2017 to 2018 Leave and Job Flexibilities Module of the American Time Use Survey, 16.4% of U.S. wage and salary workers said they usually worked a non-daytime schedule at their sole or main job. This included 5.9% who usually worked evenings and 3.6% who usually worked nights. The remaining non-daytime schedules were rotating, irregular, split, or another arrangement 6.
The module included about 10,000 respondents and excluded self-employed workers. It defined a daytime schedule as one with most work between 6 a.m. and 6 p.m. The estimate describes work timing, not sleep duration, sleep quality, or shift work disorder. It is also a 2017 to 2018 estimate, so it should not be relabeled as the current share of all U.S. workers.
Non-daytime work was not distributed evenly. The survey estimated that 37% of wage and salary workers in leisure and hospitality, 26% in transportation and utilities, and 25% in wholesale and retail trade usually worked a non-daytime schedule 6. Those figures describe industries, not the probability that an individual worker has a sleep disorder.
What these numbers cannot tell you
Population statistics are useful for showing scale and disparities. They are poor substitutes for an individual assessment.
- Less than seven hours is a survey threshold, not proof of a sleep disorder.
- Trouble falling or staying asleep is a symptom, not automatically insomnia.
- A breathing-event prevalence model is not a count of diagnosed sleep apnea.
- A reported association does not establish which factor caused an outcome.
- A police-coded crash count and an imputed crash estimate should not be combined.
- Estimates from different surveys should not be used as a trend unless their methods are comparable.
Talk with a healthcare professional if sleep problems persist, interfere with daily life, or include loud snoring, gasping, pauses in breathing, repeated unintended sleep, or dangerous sleepiness. Seek urgent help rather than driving if you cannot stay awake.
The bottom line
The latest NHIS shows that short sleep and frequent sleep difficulties affect substantial shares of U.S. adults, while national youth surveillance shows that only about one in four high school students reported at least eight hours on an average school night. Other figures, including obstructive sleep apnea prevalence and drowsy-driving deaths, depend heavily on whether researchers used a model, a test threshold, or an official coded record.
A useful sleep statistic always comes with a population, place, collection period, measure, and method. Without those details, a precise-looking percentage may say much less than it appears to.
