Sleep and diabetes are connected across population prevalence, sleep duration, sleep quality, apnea, timing, and complications. The figures below distinguish U.S. surveys from international observational studies and meta-analyses, while keeping measurement periods and definitions explicit.
Key Sleep and Diabetes Statistics
These are the most broadly useful sleep and diabetes statistics in the available evidence:
- In August 2021–August 2023, 15.8% of U.S. adults aged 20 and older had total diabetes, including diagnosed and undiagnosed cases.
- In August 2021–August 2023, 11.3% of U.S. adults aged 20 and older had diagnosed diabetes, based on self-reported clinician diagnosis.
- In August 2021–August 2023, 4.5% of U.S. adults aged 20 and older had undiagnosed diabetes, based on fasting glucose or HbA1c thresholds.
- In 2024, 30.5% of U.S. adults reported sleeping less than 7 hours on average.
- In 2024, short sleep affected 33.4% of U.S. adults aged 35–49.
- In 2024, short sleep affected 40.2% of Black non-Hispanic U.S. adults.
- From 1999–2000 to August 2021–August 2023, age-adjusted total diabetes prevalence rose from 9.7% to 14.3% among U.S. adults aged 20 and older.
- A systematic review of prospective cohorts found short sleep associated with incident type 2 diabetes at RR 1.58 among clinically assessed studies.
- A 2024 systematic review found sleep below 7 hours associated with type 2 diabetes at OR 1.18 in international observational studies.
- The same review found sleep above 8 hours associated with type 2 diabetes at OR 1.13.
- In studies through October 1, 2024, poor sleep quality was associated with type 2 diabetes at OR 1.50.
- In the same evidence base, evening chronotype was associated with type 2 diabetes at OR 1.59.
- Across 41 articles identified through a 2019 search, pooled sleep-disorder prevalence among adults with diabetes was 52%.
- In that review, pooled obstructive sleep-apnea prevalence among adults with diabetes was 60%.
- In a Sub-Saharan Africa meta-analysis, poor sleep quality affected 53.22% of people with diabetes.
- In U.S. NHANES respondents from 2005–2008, more than 90% reported at least one examined sleep problem.
- In the 2005–2008 U.S. analysis, apnea was associated with diabetes at OR 3.52 among adults below age 60.
- ADA’s 2026 evidence summary describes a U-shaped association with the lowest type 2 diabetes risk typically around 7 hours per day.
Contents
- U.S. diabetes prevalence and short sleep
- Sleep duration and type 2 diabetes risk
- Sleep quality, insomnia, and diabetes outcomes
- Sleep apnea and diabetes prevalence
- Sleep timing and circadian disruption
- Sleep problems, complications, and pregnancy outcomes
U.S. Sleep and Diabetes Statistics: Prevalence and Short Sleep
CDC/NCHS estimated total diabetes in U.S. adults aged 20 and older at 15.8% in August 2021–August 2023. The estimate combines 11.3% diagnosed diabetes, based on self-reported clinician diagnosis, with 4.5% undiagnosed diabetes identified through fasting glucose or HbA1c thresholds among people reporting no diagnosis.
These are NHANES weighted estimates, not sleep-specific measures. CDC/NCHS, Prevalence of Total, Diagnosed, and Undiagnosed Diabetes in Adults: United States, August 2021–August 2023
| U.S. measure | Population or period | Result |
|---|---|---|
| Total diabetes | Adults aged 20+, Aug. 2021–Aug. 2023 | 15.8% |
| Diagnosed diabetes | Adults aged 20+, Aug. 2021–Aug. 2023 | 11.3% |
| Undiagnosed diabetes | Adults aged 20+, Aug. 2021–Aug. 2023 | 4.5% |
| Total diabetes, age-adjusted | 1999–2000 to Aug. 2021–Aug. 2023 | 9.7% to 14.3% |
The trend comparison uses age-adjustment to the 2000 Census. Diagnosed diabetes rose from 5.9% to 10.1% over the same 1999–2000 to August 2021–August 2023 span, while total age-adjusted diabetes changed from 14.8% in 2017–March 2020 to 14.3% in August 2021–August 2023 without a significant difference.
CDC/NCHS reported that 30.5% of U.S. adults slept less than 7 hours on average in 2024. The NHIS measure is self-reported average sleep duration and is a general-adult benchmark rather than a diabetes subgroup estimate. CDC/NCHS, Short Sleep Duration and Sleep Difficulties Among Adults: United States, 2024
Short sleep was reported by 30.6% of men and 30.4% of women. By age, the figures were 27.2% at ages 18–34, 33.4% at 35–49, 34.5% at 50–64, and 27.2% at 65 and older.
By race and Hispanic origin, short sleep affected 40.2% of Black non-Hispanic adults, 27.9% of Asian non-Hispanic adults, 28.9% of White non-Hispanic adults, and 29.0% of Hispanic adults. These are descriptive survey estimates and should not be treated as causal comparisons.
Sleep Duration Statistics and Type 2 Diabetes Risk
The evidence on duration describes associations, not proof that a specific number of hours causes or prevents diabetes. A systematic review included 10 studies representing 13 cohorts and 107,756 participants, providing an evidence base for prospective risk estimates. Quantity and Quality of Sleep and Incidence of Type 2 Diabetes: A systematic review and meta-analysis
- In clinically assessed prospective studies, short sleep was associated with incident type 2 diabetes at RR 1.58, with a 95% CI of 1.13–2.21.
- Difficulty maintaining sleep was associated with diabetes at RR 1.67, with a 95% CI of 1.30–2.14.
A newer systematic review searched four major databases through October 1, 2024 and pooled sleep duration, quality, chronotype, and napping evidence from international observational studies. Sleep features and the risk of type 2 diabetes mellitus: a systematic review and meta-analysis
| Sleep feature | Association with type 2 diabetes | Interpretation |
|---|---|---|
| Below 7 hours | OR 1.18 (95% CI 1.13–1.23) | Pooled observational association |
| Above 8 hours | OR 1.13 (95% CI 1.09–1.18) | May reflect illness or confounding |
| Poor sleep quality | OR 1.50 (95% CI 1.30–1.72) | Instruments varied |
| Evening chronotype | OR 1.59 (95% CI 1.18–2.13) | Chronotype measures varied |
The review found a lowest-risk reference pattern around 7 hours, with both shorter and longer sleep linked to higher type 2 diabetes odds. ADA’s 2026 evidence summary similarly says sleep under 6 hours or over 9 hours can carry up to a 50% higher type 2 diabetes risk, although the magnitude varies across the cited evidence.
Odds ratios and relative risks are not interchangeable, and neither pooled measure is an absolute probability. The duration definitions, instruments, populations, and adjustment sets differed across the observational studies.
Sleep Quality, Insomnia, and Diabetes Outcomes
Sleep quality is broader than duration: it can include difficulty initiating or maintaining sleep, fragmented sleep, and other sleep disorders. A systematic review covering studies found through a 2019 search estimated pooled sleep-disorder prevalence among adults with diabetes at 52% (95% CI 42%–63%) across 41 articles. The association between sleep and diabetes outcomes: A systematic review
The same review estimated unspecified sleep apnea at 69% (95% CI 59%–78%), obstructive sleep apnea at 60% (95% CI 39%–80%), and restless-leg syndrome at 27% (95% CI 20%–34%). The apnea figures are pooled estimates from heterogeneous samples; unspecified sleep apnea is not equivalent to objectively diagnosed obstructive sleep apnea.
A 2024 Sub-Saharan Africa review included 11 studies and 3,766 participants. It estimated poor sleep quality among people with diabetes at 53.22% (95% CI 36.26%–70.19%), a regional and heterogeneous result rather than a global prevalence. Quality of Sleep and Its Determinants Among People with Diabetes Mellitus in Sub-Saharan Africa: A Systematic Review and Meta-Analysis
Within that review, longer diabetes duration was associated with poor sleep quality at OR 2.94 (95% CI 1.85–4.67), poor glycemic control at OR 2.24 (95% CI 1.51–3.32), depression at OR 4.09 (95% CI 2.07–8.05), and comorbidity at OR 1.94 (95% CI 1.31–2.88). Direction of causality remains unresolved, and definitions varied.
An older CDC analysis of U.S. NHANES respondents from 2005–2008 found that more than 90% reported at least one examined sleep problem. Each individual problem in that analysis was reported by roughly 10% to 40% of respondents, but “any examined problem” is not the same as a diagnosed sleep disorder. CDC, Prevalence of Self-Reported Sleep Problems Among People With Diabetes in the United States, 2005-2008
Sleep Apnea and Diabetes Prevalence Statistics
Apnea showed different diabetes associations by age in the CDC’s 2005–2008 cross-sectional analysis. Among U.S. adults below age 60, apnea was associated with diabetes at OR 3.52, compared with OR 1.59 among adults aged 60 and older.
Other reported sleep-related symptoms also varied by age and sex:
- Leg symptoms were associated with diabetes at OR 1.94 below age 60 and OR 1.24 at age 60 and older.
- Nocturia was associated with diabetes at OR 3.09 below age 60 and OR 1.55 at age 60 and older.
- Leg symptoms had a stronger association among women than men, with OR 1.65 versus 1.28.
These are cross-sectional associations, so they do not establish timing or causation. Nocturia also has multiple possible causes, and symptom-based findings should not be read as equivalent to sleep-laboratory diagnoses.
The pooled evidence points in the same broad direction but uses different methods: the international review estimated obstructive sleep apnea prevalence at 60% among adults with diabetes, while the CDC analysis reported age-stratified associations between self-reported symptoms and diabetes. The measures answer different questions and should not be combined as one prevalence figure.
Sleep Timing, Chronotype, and Circadian Diabetes Data
Sleep timing adds a dimension beyond hours slept. ADA describes a U-shaped association between sleep duration and type 2 diabetes incidence, with the nadir typically at 7 hours per day, while the newer meta-analysis found evening chronotype associated with type 2 diabetes at OR 1.59.
That OR corresponds to 59% higher odds, not 59% higher absolute risk. ADA’s evidence summary reports evening-chronotype odds 2.5 times morning-chronotype odds, independent of sleep duration and sufficiency in the cited evidence, but this remains an observational estimate and chronotype categories vary.
The National Institute of Diabetes and Digestive and Kidney Diseases’ Diabetes in America chapter covers sleep and circadian disturbances in glucose metabolism, type 2 diabetes, and gestational diabetes or hyperglycemia research. It is a scope statement covering diverse evidence, not one pooled estimate. NIDDK, Impact of Sleep and Circadian Disturbances on Glucose Metabolism and Type 2 Diabetes
Because chronotype, timing, duration, and sleep sufficiency can overlap, the reported odds ratios should not be added together. The 2024 CDC short-sleep benchmark of 30.5% among general U.S. adults also cannot be substituted for a diabetes-specific timing estimate.
Sleep Problems, Diabetes Complications, and Pregnancy Outcomes
Diabetes prevalence itself varies substantially across age, BMI, and education categories in the U.S. August 2021–August 2023 NHANES estimates. These comparisons provide population context for why sleep-related findings should retain their underlying population definitions.
| Category | Total diabetes | Diagnosed | Undiagnosed |
|---|---|---|---|
| Age 20–39 | 3.6% | 2.2% | 1.3% |
| Age 40–59 | 17.7% | 12.1% | 5.6% |
| Age 60+ | 27.3% | 20.5% | 6.8% |
| BMI under 25 | 6.8% | 5.2% | 1.6% |
| BMI 25–29.9 | 12.3% | 9.5% | 2.8% |
| BMI 30+ | 24.2% | 16.3% | 7.9% |
The BMI categories are under 25, 25–29.9, and 30 or higher; the estimates are descriptive, and obesity is not a proxy for a sleep disorder. The same CDC report found total diabetes at 19.6% with high-school education or less, 17.2% with some college, and 10.7% with a bachelor’s degree or more.
Diagnosed diabetes across those education groups was 14.6%, 12.0%, and 7.3%, respectively. Undiagnosed diabetes was 5.0%, 5.2%, and 3.4%; the CDC reported that these education differences were not significant. CDC/NCHS, Prevalence of Total, Diagnosed, and Undiagnosed Diabetes in Adults: United States, August 2021–August 2023
NIDDK’s chapter identifies gestational diabetes or hyperglycemia as one area in which sleep and circadian disturbances have been studied, alongside glucose metabolism and type 2 diabetes. The available scope statement does not provide one universal pregnancy prevalence or outcome estimate, so pregnancy findings should remain tied to their original populations and definitions.
Across this evidence, sleep duration, sleep quality, apnea, timing, and diabetes outcomes are related but distinct measures. Prevalence, odds ratios, and relative risks should be read separately, and no figure here implies that a mattress or sleep product prevents or treats diabetes.