Sleep apnea guide statistics: the key numbers
Sleep apnea guide statistics show that OSA is one of the most common sleep disorders in the world — yet most people who have it have never been diagnosed. A 2019 modeling study estimated 936 million adults aged 30–69 have mild-to-severe OSA globally. A 2025 US analysis put the figure at 83.7 million US adults aged 20 and older.
- 936 million adults aged 30–69 estimated with mild-to-severe OSA worldwide (Benjafield et al., The Lancet Respiratory Medicine, 2019)
- ~81% of adults with moderate-to-severe OSA had never been diagnosed in a 2025 US community study (Plante et al., Sleep Health, 2025)
- 2.94 points average reduction in daytime sleepiness score (Epworth scale) with CPAP versus placebo across 11 trials, 706 patients (Sassani et al., JAMA Internal Medicine, 2004)
17 figures from 11 sources, data years 2015–2026, primarily US, UK, Canada, global. The full explanation is in our sleep apnea guide.
These figures describe groups of people, not you personally. They are for information, not medical advice.
How common is sleep apnea?
Sleep apnea is far more common than most people realise. A 2019 modeling study in The Lancet Respiratory Medicine put the global count at 936 million adults with mild-to-severe OSA. A more recent 2025 US analysis estimated 83.7 million US adults aged 20 and older have OSA — about 1 in 3 American adults. Both figures are modeled estimates, not direct counts.
936 million adults aged 30–69 worldwide were estimated to have mild-to-severe OSA in 2019.
425 million of those adults were estimated to have moderate-to-severe OSA. This group is part of the 936 million total.
83.7 million US adults aged 20 or older were estimated to have OSA in 2024. This was about 32.4% of US adults.
What this means for you: These are population estimates, not a reading from any one test. Many people with OSA have not been identified yet. If you recognise the symptoms, a medical assessment is the right next step. The sleep apnea guide explains what OSA is and how it differs from central sleep apnea.
Who gets sleep apnea?
Sleep apnea affects adults of all ages and sexes, though it is more common in men than women and rises sharply with age. Men are about twice as likely as women to have OSA in the general population. OSA prevalence approaches 50% in adults aged 65 and older. A 2025 US study estimated that 59% of US adults with OSA are male and 41% female.
59% male, 41% female An estimated 59% of US adults with OSA were male and 41% were female in 2024.
~50% A 2026 review estimated that OSA affects about half of adults aged 65 and older.
What this means for you: OSA is not just a condition for heavy-set, middle-aged men who snore loudly. Women, older adults and people without obvious snoring are all affected. The symptoms section of the sleep apnea guide covers less obvious presentations, including those common in women.
What do AHI scores and severity categories mean in numbers?
The breathing-event score (apnea-hypopnea index, AHI) groups events as mild, moderate, or severe. Doctors use these levels with symptoms and other test results. A score near a boundary does not decide treatment by itself. A 2025 US model estimated that over half of OSA cases were mild.
5–14 events per hour is the AHI range used for the mild sleep apnea category in NHS guidance — meaning about 5 to 14 breathing interruptions recorded per hour of sleep.
15–30 events per hour corresponds to the moderate sleep apnea category in NHS guidance.
More than 30 events per hour corresponds to the severe sleep apnea category in NHS guidance.
52% mild, 30% moderate, 18% severe A 2025 model estimated that 52% of US adults with OSA had mild cases, 30% moderate cases, and 18% severe cases.
What this means for you: More than half of estimated cases are in the mild range, but a mild AHI does not mean symptoms feel mild. Clinicians look at the full picture — symptoms, oxygen readings and daytime impact — not the AHI alone. Our AHI section explains what the score does and does not tell you.
How many people with sleep apnea go undiagnosed?
Most people with OSA do not know they have it. Multiple studies estimate that 80% or more of people who meet criteria for moderate-to-severe OSA have never received a clinical diagnosis. A 2025 community study found that about 4 in 5 participants with an AHI of 15 or more were undiagnosed.
81.4% In a 2025 US community study, 81.4% of adults with an AHI of 15 or more had never been diagnosed.
What this means for you: An undiagnosed diagnosis is common, not unusual. A quiet night or the absence of loud snoring does not rule OSA out. If you have concerns, our sleep apnea answers page addresses common questions about getting tested.
Does sleep apnea treatment improve daytime sleepiness?
Several randomised trials found that CPAP reduces daytime sleepiness compared with placebo. A 2004 review combined 11 trials with 706 patients. It found an average 2.94-point drop on the Epworth Sleepiness Scale. People with more severe OSA and greater starting sleepiness often improve more.
2.94 points — the average drop in Epworth Sleepiness Scale scores with CPAP versus placebo. This result combined 11 randomised trials with 706 patients who had OSA.
4.75 points — the average score drop in trials involving people with severe OSA and marked sleepiness. Their AHI was ≥30 and Epworth score was ≥11. Other groups in the same review had an average drop of 1.10 points.
What this means for you: A 2-to-3-point drop on the Epworth scale is a meaningful group-level improvement, but it does not predict how any one person will feel. Some people stay tired even with good CPAP use — and that deserves its own follow-up. The CPAP problems section of our guide covers reasons why fatigue can persist.
How many people use CPAP as directed?
CPAP adherence is one of the most studied — and most challenging — areas in sleep medicine. A review of 82 studies spanning 20 years found that about 1 in 3 people do not meet a standard 7-hours-per-night benchmark. Adherence rates across published studies generally range from 30% to 60%, with no meaningful improvement over two decades of data.
34.1% — the share of people who did not meet the 7-hour nightly CPAP use standard. This overall rate came from 82 studies published over 20 years.
What this means for you: Struggling with CPAP is common — you are not alone if it feels hard to use every night. Education, mask fitting and follow-up support can improve adherence. The Epworth scale meta-analysis above used patients who were actually using their devices — so real-world benefits depend on consistent use. Our answers page has practical questions about CPAP difficulties and what to do next.
How often does sleep apnea occur with other health conditions?
OSA is common among people with heart and blood vessel disease. In heart clinics and hospitals, 40% to 80% of patients with certain conditions also have OSA. These conditions include high blood pressure, heart failure, coronary artery disease, atrial fibrillation, or stroke. This link does not prove OSA causes these conditions. The relationship is more complex.
40–80% of patients with high blood pressure, heart failure, coronary artery disease, atrial fibrillation, or stroke are estimated to have OSA. These estimates come from heart clinics and hospitals.
What this means for you: OSA overlapping with heart conditions is common — if you have either, a conversation with your clinician about the other is worthwhile. These are associations between groups, not a guarantee of any individual outcome. The guide’s health risks section explains these links in plain terms.
How is sleep apnea treatment covered and what does it cost?
CPAP coverage depends on the country, insurer, and person’s eligibility. In the US, Medicare Part B may cover a 12-week CPAP trial for eligible patients. During this trial, patients usually pay 20% of the approved equipment cost after the deductible. This applies when their supplier accepts assignment. In England, the NHS provides CPAP free to patients who need it.
12 weeks — the first CPAP trial period that US Medicare may cover for eligible patients.
20% — the share of the approved equipment cost that eligible patients usually pay after their deductible. This applies when their supplier accepts assignment.
What this means for you: Coverage rules differ by country, plan and clinical situation. These figures describe US Medicare only. For a full picture of where to start getting tested and treated where you live, the sleep apnea compass shows the right starting point for your situation.
Sleep apnea guide statistics table
| What it measures | Figure | Who / based on | Compared with · when | Source |
|---|---|---|---|---|
| Global mild-to-severe OSA prevalence | 936 million adults (95% CI 903–970 million) | Adults aged 30–69 globally · 17 source studies, 16 countries | Moderate-to-severe subset below · data year 2019 | Benjafield et al., The Lancet Respiratory Medicine, 2019 |
| Global moderate-to-severe OSA prevalence | 425 million adults (95% CI 399–450 million) | Adults aged 30–69 globally · same 17 studies | Full mild-to-severe count above · data year 2019 | Benjafield et al., The Lancet Respiratory Medicine, 2019 |
| US OSA prevalence count | 83.7 million adults (approx. 32.4% of US adults) | US adults aged ≥20 · obesity-adjusted modeled estimate | Prior US estimate of 54 million (ages 30–69) · data year 2024 | Wosu et al., Sleep Medicine, 2025 |
| US OSA by sex (estimated) | 59% male, 41% female | US adults aged ≥20 with OSA · modeled estimate | General adult population split · data year 2024 | Wosu et al., Sleep Medicine, 2025 |
| OSA severity distribution (US estimate) | 52% mild · 30% moderate · 18% severe | US adults with OSA · modeled estimate | No prior direct comparator · data year 2024 | Wosu et al., Sleep Medicine, 2025 |
| OSA prevalence in adults aged ≥65 | ~50% (approaches 50%) | Adults aged ≥65 · narrative review of population studies | Lower rates in younger adults · published 2026 | Fabbri et al., Sleep Medicine Reviews, 2026 |
| Undiagnosed moderate-to-severe OSA | 81.4% | 155 US community adults with AHI ≥15 · cross-sectional | Those with prior OSA diagnosis · published 2025 | Plante et al., Sleep Health, 2025 |
| AHI mild category threshold | 5–14 events per hour | Adults · UK clinical guidance | Moderate category (15–30) · NHS guidance | NHS, Sleep apnoea, n.d. |
| AHI moderate category threshold | 15–30 events per hour | Adults · UK clinical guidance | Mild (5–14) and severe (>30) · NHS guidance | NHS, Sleep apnoea, n.d. |
| AHI severe category threshold | >30 events per hour | Adults · UK clinical guidance | Moderate category (15–30) · NHS guidance | NHS, Sleep apnoea, n.d. |
| CPAP effect on daytime sleepiness (Epworth score) | Mean difference −2.94 points vs placebo (P <.001) | 706 adults with OSA · 11 randomised trials | Placebo/sham CPAP · random-effects meta-analysis, 2004 | Sassani et al., JAMA Internal Medicine, 2004 |
| CPAP effect — severe OSA subgroup | Mean difference −4.75 points vs placebo | Severe OSA + sleepiness subgroup · within the same 11 trials | Other OSA groups (−1.10 points) · same meta-analysis | Sassani et al., JAMA Internal Medicine, 2004 |
| CPAP non-adherence rate | 34.1% non-adherent (based on 7-hour sleep standard) | Studies over 20 years · 82 papers | No improvement over time · published 2016 | Weaver & Grunstein, Journal of Clinical Sleep Medicine, 2016 |
| OSA prevalence in cardiovascular patients | 40–80% (range across conditions) | Patients with hypertension, heart failure, CAD, AF or stroke · clinical settings | General population rates · narrative review, 2025 | Mishraki et al., PMC Cardiovascular Review, 2025 |
| US Medicare CPAP trial coverage duration | 12 weeks | Eligible US Medicare Part B beneficiaries | Continuing coverage requires meeting criteria · current guidance 2026 | Medicare, CPAP therapy coverage, n.d. |
| US Medicare CPAP cost-sharing | 20% of Medicare-approved amount after deductible | Eligible Medicare Part B beneficiaries, supplier accepting assignment | Full cost without Medicare · current guidance 2026 | Medicare, CPAP therapy coverage, n.d. |
| NHS CPAP provision (England) | Provided free of charge to eligible patients | Patients in England assessed as needing CPAP | Private provision costs · NHS guidance | NHS, Sleep apnoea, n.d. |
Why the numbers differ
The global estimate of 936 million and the US estimate of 83.7 million use different methods and age ranges. Benjafield et al. (2019) studied adults aged 30–69 worldwide. Wosu et al. (2025) studied US adults aged 20 or older and adjusted for obesity trends. Neither figure is wrong. Each describes a different group of people.
AHI category boundaries differ slightly between guidelines and countries. The NHS uses 5–14 (mild), 15–30 (moderate) and over 30 (severe). The AASM uses similar thresholds but with different hypopnea scoring rules, which can shift a person’s category. This is why a score near a boundary needs full clinical interpretation rather than self-assignment.
Studies estimate that 80% to 93% of people with OSA go undiagnosed. Rates vary by the group studied, the AHI cutoff, and how diagnosis was checked. Plante et al. (2025) found 81.4% undiagnosed in a community sample using home polysomnography and AHI ≥15. Studies using AHI ≥5 or other groups report different rates. All show a large gap between estimated cases and confirmed diagnoses.
CPAP adherence figures vary widely — from 30% to 83% across published studies — because definitions differ. Some studies define adherence as more than 4 hours per night; others use 7 hours. Real-world device data tend to show higher usage than research trial data.
What the numbers can’t tell you
Population estimates for OSA are modeled, not counted. They apply projection methods from study samples to the whole population. The true prevalence depends on how OSA is defined (what AHI cut-off, which hypopnea scoring rules) and which population was studied. Figures from one country may not apply directly to another.
The AHI categories describe frequency of breathing events — they do not describe how someone feels. Two people with an AHI of 20 can feel very differently. A score at the boundary between categories does not settle treatment decisions on its own.
The CPAP sleepiness improvement figure is a group average across trials. Individual responses vary. Some people improve significantly; others do not, sometimes because another sleep problem overlaps. No figure here tells you what CPAP will do for you personally.
No reliable published figure was found for how many people stop CPAP within their first year of routine care. Estimates exist, but they vary greatly. No reliable figure was found comparing sleepiness outcomes for CPAP and oral appliances in one large direct trial. Country-level OSA rates were also not found for most regions outside North America and Europe.
We gathered figures from the published studies and official sources listed below. We recorded each figure’s outcome, population, comparison, timepoint, and year as published. We did not run our own surveys. The original figures belong to their publishers. Our compilation and wording may be reused with credit under CC BY 4.0.
See where these numbers fit in the bigger picture
The numbers tell you how common OSA is and what treatment can do — but they don’t tell you where you fit. The sleep apnea mind map connects every concept, or if you’re ready to act, the sleep apnea compass finds your right starting point. More context in our sleep disorders pillar.
See how it all connects →Sources
- Benjafield et al. — Estimation of the global prevalence and burden of obstructive sleep apnoea: a literature-based analysis (2019) — Tier 1 prevalence modeling study; used for: global mild-to-severe OSA estimate (936 million) and moderate-to-severe estimate (425 million)
- Wosu et al. — Unmasking obstructive sleep apnea: Estimated prevalence and impact in the United States (2025) — Tier 1 modeling study; used for: US adult prevalence (83.7 million, 32.4%), sex distribution (59% male), severity distribution (52% mild, 30% moderate, 18% severe)
- Plante et al. — Obstructive sleep apnea underdiagnosis: Prevalence and correlates from the Sleep Health and Knowledge in US Hispanics study (2025) — Tier 1 community study; used for: undiagnosed moderate-to-severe OSA (81.4%)
- Sassani et al. — Continuous positive airway pressure therapy for treating sleepiness in a diverse population with OSA: results of a meta-analysis (2004) — Tier 1 meta-analysis; used for: CPAP Epworth score reduction (2.94 points overall, 4.75 points in severe subgroup)
- Weaver & Grunstein — Trends in CPAP adherence over twenty years of data collection: a flattened curve (2016) — Tier 1 systematic review; used for: CPAP non-adherence rate (34.1% over 20 years, 82 studies)
- Fabbri et al. — OSA in the aging population: Diagnostic and therapeutic considerations (2026) — Tier 1 review; used for: OSA prevalence approaching 50% in adults aged 65 and older
- Mishraki et al. — Sleep Apnea: The Slept-Upon Cardiovascular Risk Factor (2025) — Tier 1 narrative review; used for: OSA prevalence 40–80% in cardiovascular patients
- NHS — Sleep apnoea (n.d.) — Tier 2 government patient guidance; used for: AHI severity category boundaries (mild, moderate, severe), NHS CPAP provision
- Medicare — Continuous Positive Airway Pressure (CPAP) therapy (n.d.) — Tier 2 government coverage guidance; used for: 12-week trial period, 20% cost-sharing
- NHLBI — Sleep Apnea Treatment (2025) — Tier 2 government patient education; used for: CPAP side-effect context and treatment overview
- NICE — NG202 Obstructive sleep apnoea/hypopnoea syndrome (2021) — Tier 1 clinical guideline; used for: follow-up evaluation context
Figures checked · 17 figures from 11 sources · Next review
How we made this page: figures gathered and checked from the sources above, written up with AI assistance and edited by Aamir Bashir, Founder & Editor. Read our editorial policy.
