An estimated 936 million adults aged 30 to 69 worldwide have obstructive sleep apnea, and 425 million of them have moderate to severe disease (Benjafield et al., Lancet Respiratory Medicine 2019). In the United States, a 2025 modeling study projects OSA prevalence will climb from 34.3% to 46.2% of adults aged 30 to 69 by 2050, or 76.6 million people, while ResMed’s latest 10-K estimates that less than 20% of Americans with OSA have been diagnosed or treated. Among those who do start CPAP, 47.7% of new users in France had stopped within three years. The treatment market is also shifting: the FDA approved the first OSA drug in December 2024, and the leading hypoglossal nerve stimulation company saw revenue fall 7.6% year over year in Q2 2026. For the broader picture on sleep duration and disorders, see our sleep statistics. We aggregated data from The Lancet Respiratory Medicine, the New England Journal of Medicine, the US Food and Drug Administration, ResMed and Inspire Medical Systems SEC filings and results, Apnimed, Philips and more than a dozen peer-reviewed studies listed in the methodology.
TL;DR
- 936 million adults aged 30 to 69 have mild to severe OSA worldwide (Benjafield et al., Lancet Respiratory Medicine 2019).
- 425 million adults have moderate to severe OSA (Benjafield et al., Lancet Respiratory Medicine 2019).
- US OSA prevalence is projected to rise from 34.3% to 46.2% by 2050, reaching 76.6 million adults (Boers et al., Lancet Respiratory Medicine 2025).
- Less than 20% of Americans with OSA have been diagnosed or treated (ResMed, Form 10-K fiscal 2026).
- 47.7% of 480,000 new French CPAP users stopped therapy within 3 years (Pepin et al., Journal of Clinical Medicine 2021).
- 74.0% of 4.13 million US patients met Medicare CPAP adherence criteria in 90 days (Drager et al., Journal of Clinical Sleep Medicine 2021).
- Telemonitoring plus an engagement tool cut long-term PAP termination from 27% to 11% (Woehrle et al., ERJ Open Research 2024).
- Tirzepatide reduced AHI by up to 23.8 events per hour more than placebo (Malhotra et al., NEJM 2024).
- AD109, an oral OSA pill, cut AHI by 55.6% in a 646-patient Phase 3 trial (Apnimed, 2026).
- The 2021 Philips recall affected three to four million CPAP devices globally (Internal Medicine Journal 2026).
- ResMed revenue reached 5.653 billion dollars in fiscal 2026, up 10% (ResMed, Q4 FY2026 results).
- Inspire Medical Systems guides 2026 revenue to 835 to 875 million dollars, down from 912.0 million in 2025 (Inspire, Q2 2026 results).
1. Prevalence: Close to a Billion Adults, and Rising
OSA is far more common than its diagnosis rates suggest. Benjafield and colleagues estimated that 936 million adults aged 30 to 69 have mild to severe OSA (AHI of 5 or more) and 425 million have moderate to severe OSA (AHI of 15 or more), with prevalence exceeding 50% in some countries (Benjafield et al., Lancet Respiratory Medicine 2019). The estimate rests on thin ground: only 17 articles gave reliable prevalence figures, covering 16 of 193 countries, and the rest were extrapolated by BMI, race and geography.
The US trend line points up. A 2025 dynamic population model from the same research group projects that OSA prevalence among US adults aged 30 to 69 will rise from 34.3% to 46.2% by 2050, or 76.6 million people, with women driving the growth: female prevalence rises 65.4% in relative terms, from 22.8% to 37.7%, versus 19.3% for men (Boers et al., Lancet Respiratory Medicine 2025). That is a demographic shift as much as a medical one: a disease long screened for in middle-aged, overweight men is becoming a condition of women and older adults too. Shift workers, whose sleep is already disrupted, are a separate risk pool covered in our shift work health statistics.
| Metric | Value | Source |
|---|---|---|
| Adults 30 to 69 with mild to severe OSA, worldwide | 936 million (95% CI 903 to 970 million) | Benjafield et al., Lancet Respiratory Medicine 2019 |
| Adults 30 to 69 with moderate to severe OSA, worldwide | 425 million (95% CI 399 to 450 million) | Benjafield et al., Lancet Respiratory Medicine 2019 |
| Largest national counts, mild to severe OSA | China 175.7M, US 54.1M, India 51.6M, Brazil 48.7M | Benjafield et al., Lancet Respiratory Medicine 2019 |
| US OSA prevalence, adults 30 to 69, now vs 2050 | 34.3% to 46.2% (76.6 million cases) | Boers et al., Lancet Respiratory Medicine 2025 |
| US women, projected 2050 | 22.8% to 37.7% (+65.4% relative), 30.4 million cases | Boers et al., Lancet Respiratory Medicine 2025 |
| US men, projected 2050 | 45.6% to 54.4% (+19.3% relative), 45.9 million cases | Boers et al., Lancet Respiratory Medicine 2025 |
| Moderate to severe sleep-disordered breathing, 2007-2010 | Men 30-49: 10%, men 50-70: 17%, women 30-49: 3%, women 50-70: 9% | Peppard et al., American Journal of Epidemiology 2013 |
| Americans with mild to severe OSA, industry estimate | 61 million | ResMed, Form 10-K fiscal 2026 |
Context note: Apnimed’s 2026 press releases cite “an estimated 80 million people in the United States and one billion people worldwide” with OSA, a higher US figure than ResMed’s 61 million. The gap reflects different age ranges and model years, not a measured change. The Peppard estimates (Wisconsin Sleep Cohort extrapolated to NHANES BMI data) are the most recent available direct US measurement and showed relative increases of 14% to 55% over two decades.
2. The Diagnosis Gap: Most Cases Never Reach a Sleep Lab
The defining number in OSA is not prevalence but detection. ResMed estimates that less than 20% of people with OSA in the US have been diagnosed or treated, and 10% or less in other markets (ResMed, Form 10-K fiscal 2026). That is a company estimate with an obvious commercial interest, but claims studies in Medicare point the same way: in a 2022 analysis of 142,893 older beneficiaries with cardiovascular disease, only 19,390 carried an OSA diagnosis, and untreated OSA was associated with 1.82 times the odds of hospitalization (Innovation in Aging, 2022).
Undiagnosed OSA is expensive before anyone names it. Among 41,500 Medicare beneficiaries with depression, 11% had occult OSA, and they used more inpatient care (rate ratio 1.53) and emergency care (rate ratio 1.48); mean total costs were significantly higher, with a reported estimate of 44,390 dollars (95% CI 32,076 to 56,703) (Wickwire and Albrecht, Journal of Clinical Sleep Medicine 2024). An earlier study by the same authors found occult OSA carried 3.19 times the risk of incident depression in the year before diagnosis. Consumer sleep trackers are one route people now take toward a test; see our wearable sleep tracker statistics.
| Metric | Value | Source |
|---|---|---|
| Share of US OSA cases diagnosed or treated | Less than 20% | ResMed, Form 10-K fiscal 2026 |
| Share diagnosed or treated outside the US | 10% or less | ResMed, Form 10-K fiscal 2026 |
| Medicare beneficiaries with CVD who had an OSA diagnosis | 19,390 of 142,893 | Innovation in Aging, 2022 (conference abstract) |
| Odds of hospitalization with untreated OSA, Medicare CVD patients | 1.82 (95% CI 1.77 to 1.87) | Innovation in Aging, 2022 |
| Occult OSA among Medicare beneficiaries with depression | 4,573 of 41,500 (11%) | Wickwire and Albrecht, Journal of Clinical Sleep Medicine 2024 |
| Inpatient and ED use with occult OSA (rate ratios) | 1.53 inpatient, 1.48 emergency department | Wickwire and Albrecht, Journal of Clinical Sleep Medicine 2024 |
| Higher mean total costs with occult OSA | 44,390 dollars (95% CI 32,076 to 56,703) | Wickwire and Albrecht, Journal of Clinical Sleep Medicine 2024 |
| Risk of incident depression with occult OSA | Risk ratio 3.19 (21,116 beneficiaries) | Wickwire and Albrecht, Sleep Health 2023 |
Context note: the Medicare studies use 2006-2013 claims (Sleep Health 2023) and are the most recent available national estimates of undiagnosed burden; “occult” is defined as the 12 months before a first OSA diagnosis code, so it captures late diagnosis, not cases never found.
3. CPAP Adherence: The 90-Day Bar and the Long Tail of Quitting
CPAP works when it is worn, and that is the problem. A systematic review of 82 studies covering 1994 to 2015 put CPAP non-adherence at 34.1%, with no significant improvement over twenty years, and behavioral interventions added only about one hour of use per night (Rotenberg et al., Journal of Otolaryngology - Head and Neck Surgery 2016). A 2026 JAMA Otolaryngology review still describes adherence rates of only 30% to 60%.
Connected devices changed what can be measured. In ResMed AirView data on 4,181,490 patients, 74.0% of US patients met the Medicare standard of at least 4 hours on at least 70% of nights in the first 90 days, and patients who used an engagement app were twice as likely to hit it (Drager et al., Journal of Clinical Sleep Medicine 2021). Long-term data are harsher: in French national insurance claims for 480,000 new users, 23.1% had stopped after one year and 47.7% after three (Pepin et al., Journal of Clinical Medicine 2021). In German data on 104,612 patients, telemonitoring-guided care cut termination from 27% to 20%, and to 11% with a patient engagement tool (Woehrle et al., ERJ Open Research 2024). Remote follow-up is a slice of the wider shift tracked in our telehealth statistics.
| Metric | Value | Source |
|---|---|---|
| CPAP non-adherence, 82 studies, 1994-2015 | 34.1%, no improvement over 20 years | Rotenberg et al., J Otolaryngol Head Neck Surg 2016 |
| 90-day CMS adherence, US / Brazil / Mexico | 74.0% / 71.7% / 66.4% (4,181,490 patients) | Drager et al., Journal of Clinical Sleep Medicine 2021 |
| Patients using an engagement app, US / Brazil / Mexico | 26% / 8.1% / 2.8% | Drager et al., Journal of Clinical Sleep Medicine 2021 |
| Adherent at 3 months still using PAP at 1 year | More than 80% | Drager et al., Journal of Clinical Sleep Medicine 2021 |
| CPAP termination after 1, 2 and 3 years, France | 23.1%, 37.1%, 47.7% (480,000 patients) | Pepin et al., Journal of Clinical Medicine 2021 |
| PAP termination: standard care vs telemonitoring vs telemonitoring plus app | 27% vs 20% vs 11% (104,612 patients) | Woehrle et al., ERJ Open Research 2024 |
| 1-year adherence, integrated sleep practice vs other suppliers | 52% vs 33% (3,884 patients aged 18-64) | Andry et al., Journal of Clinical Sleep Medicine 2021 |
| Non-compliant patients reaching CMS compliance after switching to bilevel | 56.8% (1,496 patients), +0.9 hours per day | Benjafield et al., BMJ Open Respiratory Research 2019 |
Context note: the large adherence datasets come from ResMed’s own cloud platform and several co-authors are ResMed employees or consultants; they cover only patients on connected devices. The 7-hour reference sleep time in the Rotenberg review is stricter than the CMS 4-hour rule, so the two rates are not directly comparable. Most recent available data for long-term termination is 2021 (France) and 2024 (Germany).
4. The Philips Recall: A Supply Shock That Showed Up in Usage Data
The 2021 Philips recall is the largest natural experiment in what happens when CPAP supply breaks. On June 14, 2021 Philips announced a voluntary recall of certain CPAP, BiPAP and ventilator devices because the polyester-based polyurethane sound abatement foam could degrade into particles or off-gas chemicals, citing a complaint rate of 0.03% in 2020 and booking an initial 250 million euro provision (Philips, recall notification, June 2021). A 2026 Australian study puts the recalled Philips CPAP fleet at three to four million devices globally (Internal Medicine Journal 2026).
The behavioral effect was measurable. In that Australian cohort of 368 patients, mean nightly CPAP use fell from 5.56 to 4.23 hours after a regulator notice that actually advised continuing therapy. In a survey of 136 US sleep clinicians, 91.5% said the recall affected patients’ health and well-being and 83.3% said it affected their trust in medicine (Journal of Clinical Sleep Medicine 2023). The Veterans Health Administration alone had to manage more than 700,000 affected Veterans between June 2021 and August 2022 (Frontiers in Sleep 2023).
| Metric | Value | Source |
|---|---|---|
| Recall announcement date | June 14, 2021 | Philips, 2021 |
| Initial cost provision | 250 million euros | Philips, 2021 |
| Foam complaint rate cited by Philips | 0.03% in 2020 | Philips, 2021 |
| Recalled Philips CPAP devices globally | Three to four million | Internal Medicine Journal 2026 |
| Mean nightly CPAP use before vs after regulator notice | 5.56 to 4.23 hours (368 patients) | Internal Medicine Journal 2026 |
| Clinicians who recommended continuing recalled devices | 62.4% (136 clinicians) | Journal of Clinical Sleep Medicine 2023 |
| Clinicians reporting impact on patient health / patient trust | 91.5% / 83.3% | Journal of Clinical Sleep Medicine 2023 |
| Veterans affected | More than 700,000 | Frontiers in Sleep 2023 |
Context note: the clinician survey also reported 59.9% of patients continued treatment and 26.5% stopped. Philips’ later consent decree and remediation totals could not be read from a primary page during research and are excluded here.
5. Drug Therapy Arrives: Tirzepatide, AD109 and Oral GLP-1s
For the first time, OSA has a pharmacy route. On December 20, 2024 the FDA approved Zepbound (tirzepatide) for moderate to severe OSA in adults with obesity, based on two 52-week trials in 469 adults without type 2 diabetes (FDA, December 2024). In the SURMOUNT-OSA trials, AHI fell by 25.3 events per hour on tirzepatide versus 5.3 on placebo in patients not using PAP, and by 29.3 versus 5.5 in patients on PAP (Malhotra et al., NEJM 2024).
The drugs are a complement, not yet a replacement. A 2026 JAMA Otolaryngology review notes that 42% to 50% of tirzepatide patients reached disease remission, but GLP-1 drugs did not match CPAP efficacy (roughly 22 versus 31 events per hour of AHI reduction), and weight regain is common after stopping. Obesity is present in 60% to 70% of OSA patients, which caps the eligible pool. Non-weight drugs are next: Apnimed’s AD109 pill cut AHI by 55.6% in the 646-patient SynAIRgy trial, with a potential FDA action date in the first quarter of 2027 (Apnimed, May 2026). Uptake of the same drug class for weight loss is covered in our GLP-1 weight loss statistics.
| Metric | Value | Source |
|---|---|---|
| FDA approval of tirzepatide for OSA | December 20, 2024; 469 adults, 52 weeks, 10 or 15 mg weekly | FDA, 2024 |
| Baseline AHI, SURMOUNT-OSA trials 1 and 2 | 51.5 and 49.5 events per hour | Malhotra et al., NEJM 2024 |
| AHI change, tirzepatide vs placebo (trial 1 / trial 2) | -25.3 vs -5.3 / -29.3 vs -5.5 events per hour | Malhotra et al., NEJM 2024 |
| Patients reaching remission on tirzepatide | 42% to 50% | Harris and Kaffenberger, JAMA Otolaryngology 2026 |
| Patients improving AHI severity category (post hoc) | 68% to 79% | Falcon et al., Journal of Clinical Sleep Medicine 2026 |
| AD109 AHI reduction from baseline at week 26 | 55.6% (646 patients, 69 centers) | Apnimed, SynAIRgy publication release 2026 |
| AD109 patients with AHI under 5 / with 50%+ AHI reduction | 22.3% / 39.6% | Apnimed, 2026 |
| Orforglipron Phase 3 OSA trial size (ATTAIN-OSA) | 712 participants | Malhotra et al., Contemporary Clinical Trials Communications 2026 |
Context note: AD109’s 55.6% figure uses the 4% desaturation hypopnea rule and is a within-group change from baseline; the press release does not state the placebo-adjusted difference. Apnimed raised up to 150 million dollars in debt financing in April 2026 to prepare a launch, contingent on approval.
6. Surgery and Nerve Stimulation: From UPPP to Implants
The surgical side of OSA has rotated away from tissue removal. Medicare OSA soft palate surgeries fell 65.7% between 2000 and 2021, from 4,208 to 1,443, and uvulopalatopharyngoplasty (UPPP) fell 87.6% (Torabi et al., Annals of Otology, Rhinology and Laryngology 2025), while palatoplasty grew 245.4%, from 185 to 639. The growth went to hypoglossal nerve stimulation: Inspire Medical Systems reports that physicians have treated more than 125,000 patients with its implant across the US, Europe and Asia (Inspire, Form 10-K 2025).
That business hit a wall in 2026, and reimbursement, not drugs, is the stated cause. Inspire grew revenue 14% to 912.0 million dollars in 2025, then guided 2026 to a decline after the CPT code used for approximately 10,000 Inspire V cases in 2025 was no longer available for Medicare reimbursement (Inspire, Q1 2026 results). Q2 2026 revenue was 200.6 million dollars, down 7.6% from 217.1 million (Inspire, Q2 2026 results). The company’s 10-K lists GLP-1 demand as a risk factor without quantifying it.
| Metric | Value | Source |
|---|---|---|
| Medicare OSA soft palate surgeries, 2000 vs 2021 | 4,208 to 1,443 (-65.7%) | Torabi et al., Ann Otol Rhinol Laryngol 2025 |
| UPPP procedures, 2000 vs 2021 | 3,455 to 428 (-87.6%) | Torabi et al., Ann Otol Rhinol Laryngol 2025 |
| Patients treated with Inspire therapy to date | More than 125,000 | Inspire, Form 10-K 2025 |
| Inspire US / international sales territories, end 2025 | 295 / 21 | Inspire, Form 10-K 2025 |
| Covered lives under positive US commercial policies | Over 300 million | Inspire, Form 10-K 2025 |
| Inspire revenue, 2025 | 912.0 million dollars (+14%), gross margin 85.4% | Inspire, Q4 and FY2025 results |
| Inspire V cases in 2025 billed under a code no longer available for Medicare | Approximately 10,000 | Inspire, Q1 2026 results |
| Inspire Q2 2026 revenue | 200.6 million dollars (-7.6% vs 217.1 million) | Inspire, Q2 2026 results |
Context note: Inspire’s 2026 guidance moved from 950 million to 1.0 billion dollars in February to 825 to 875 million in May (a 4% to 10% decline vs 2025) and 835 to 875 million in August. On October 1, 2026, after the latest results, Inspire announced that new CPT codes for hypoglossal nerve stimulation had been approved.
7. The CPAP Business: ResMed at Scale and Conflicting Market Sizes
CPAP remains the volume engine, and its leader is still growing despite GLP-1 drugs. ResMed reported fiscal 2026 revenue of 5.653 billion dollars, up 10% (8% in constant currency), with fourth-quarter revenue of 1.464 billion, up 9% (ResMed, Q4 FY2026 results). Masks are the recurring stream that rewards adherence: Q4 masks and accessories revenue was 542 million dollars against 750 million for devices, and masks outside the Americas grew 16%.
Scale is the moat. ResMed reports more than 35 million patients using cloud-connected devices on AirView and over 12 million registered on its myAir app, sells in more than 140 countries and employs about 11,370 people (ResMed, Form 10-K fiscal 2026). Third-party market sizing is less consistent: MarketsandMarkets puts the global sleep apnea devices market at 6.9 billion dollars in 2025, rising to 11.6 billion by 2032 at a 7.6% CAGR (MarketsandMarkets), while Precedence Research estimates 7.62 billion in 2026 and 14.00 billion by 2035 at 7.01%, with North America holding 49% in 2025 (Precedence Research).
| Metric | Value | Source |
|---|---|---|
| ResMed fiscal 2026 revenue | 5.653 billion dollars (+10%, +8% constant currency) | ResMed, Q4 FY2026 results |
| ResMed Q4 FY2026 revenue | 1.464 billion dollars (+9%) | ResMed, Q4 FY2026 results |
| ResMed Q4 devices / masks and accessories revenue | 750 million / 542 million dollars | ResMed, Q4 FY2026 results |
| ResMed revenue mix, fiscal 2026 | Devices about 51%, masks and accessories about 37%, software about 12% | ResMed, Form 10-K fiscal 2026 |
| Patients on cloud-connected devices / myAir users | More than 35 million / over 12 million | ResMed, Form 10-K fiscal 2026 |
| Global sleep apnea devices market | 6.9 billion dollars (2025) to 11.6 billion (2032), 7.6% CAGR | MarketsandMarkets |
| Global sleep apnea devices market, alternative estimate | 7.62 billion dollars (2026) to 14.00 billion (2035), 7.01% CAGR | Precedence Research |
| North America share of the devices market, 2025 | 49% | Precedence Research |
Context note: ResMed’s own fiscal 2026 revenue (which includes residential care software) is close to both firms’ total market estimates, which suggests the third-party figures undercount the category or define it narrowly. Treat them as directional.
Summary: Obstructive Sleep Apnea and CPAP by the Numbers
| Metric | Value | Source |
|---|---|---|
| Adults 30 to 69 with mild to severe OSA, worldwide | 936 million | Benjafield et al., Lancet Respiratory Medicine 2019 |
| Adults with moderate to severe OSA, worldwide | 425 million | Benjafield et al., Lancet Respiratory Medicine 2019 |
| US adults with OSA by 2050 | 76.6 million (46.2%) | Boers et al., Lancet Respiratory Medicine 2025 |
| Projected relative rise in female OSA prevalence by 2050 | +65.4% | Boers et al., Lancet Respiratory Medicine 2025 |
| US OSA cases diagnosed or treated | Less than 20% | ResMed, Form 10-K fiscal 2026 |
| Occult OSA among Medicare beneficiaries with depression | 11% | Wickwire and Albrecht, JCSM 2024 |
| CPAP non-adherence, 20-year review | 34.1% | Rotenberg et al., 2016 |
| 90-day Medicare adherence, US connected-device patients | 74.0% | Drager et al., JCSM 2021 |
| CPAP termination within 3 years, France | 47.7% | Pepin et al., Journal of Clinical Medicine 2021 |
| PAP termination with telemonitoring plus app | 11% vs 27% standard care | Woehrle et al., ERJ Open Research 2024 |
| Recalled Philips CPAP devices | Three to four million | Internal Medicine Journal 2026 |
| Veterans affected by the Philips recall | More than 700,000 | Frontiers in Sleep 2023 |
| FDA approval of first OSA drug | December 20, 2024 (tirzepatide) | FDA, 2024 |
| Tirzepatide AHI reduction vs placebo | -20.0 and -23.8 events per hour | Malhotra et al., NEJM 2024 |
| AD109 AHI reduction from baseline | 55.6% | Apnimed, 2026 |
| Medicare UPPP procedures, 2000 to 2021 | -87.6% | Torabi et al., 2025 |
| Patients treated with Inspire to date | More than 125,000 | Inspire, Form 10-K 2025 |
| Inspire Q2 2026 revenue change | -7.6% | Inspire, Q2 2026 results |
| ResMed fiscal 2026 revenue | 5.653 billion dollars (+10%) | ResMed, Q4 FY2026 results |
| ResMed patients on cloud-connected devices | More than 35 million | ResMed, Form 10-K fiscal 2026 |
Methodology and Sources
Every figure above was read during research for this article in a peer-reviewed abstract or full text (via the Europe PMC API and PubMed Central), a regulator press release, an SEC filing or a company results release. Web search was unavailable for this round, so primary documents were fetched directly. Statistics that appear on sleep clinic and supplier blogs without a primary document were excluded.
- Benjafield AV et al.: Estimation of the global prevalence and burden of obstructive sleep apnoea, Lancet Respiratory Medicine 2019
- Boers E et al.: Projecting the 30-year burden of obstructive sleep apnoea in the USA, Lancet Respiratory Medicine 2025
- Peppard PE et al.: Increased prevalence of sleep-disordered breathing in adults, American Journal of Epidemiology 2013 (Wisconsin Sleep Cohort)
- ResMed: Form 10-K for fiscal year ended June 30, 2026 and Q4 fiscal 2026 results, August 2026
- Medicare claims studies: Wickwire and Albrecht, Journal of Clinical Sleep Medicine 2024; Wickwire and Albrecht, Sleep Health 2023; Innovation in Aging 2022 conference abstract on untreated OSA in beneficiaries with heart disease
- Adherence: Drager et al., Journal of Clinical Sleep Medicine 2021; Pepin et al., Journal of Clinical Medicine 2021; Woehrle et al., ERJ Open Research 2024; Rotenberg, Murariu and Pang, Journal of Otolaryngology - Head and Neck Surgery 2016; Andry et al., Journal of Clinical Sleep Medicine 2021; Benjafield et al., BMJ Open Respiratory Research 2019
- Philips recall: Philips recall notification, June 14, 2021; Internal Medicine Journal 2026; Journal of Clinical Sleep Medicine 2023 clinician survey; Frontiers in Sleep 2023, VHA case study
- Drugs: FDA approval of Zepbound for OSA, December 2024; Malhotra et al., SURMOUNT-OSA, NEJM 2024; Harris and Kaffenberger, JAMA Otolaryngology - Head and Neck Surgery 2026; Falcon et al., Journal of Clinical Sleep Medicine 2026; Malhotra et al., ATTAIN-OSA design paper, Contemporary Clinical Trials Communications 2026; Apnimed, SynAIRgy publication release, May 2026 and Apnimed financing release, April 2026
- Surgery and neurostimulation: Torabi et al., Annals of Otology, Rhinology and Laryngology 2025; Inspire Form 10-K 2025; Inspire Q4 and FY2025 results; Inspire Q1 2026 results; Inspire Q2 2026 results
- Market sizing: MarketsandMarkets, sleep apnea devices market; Precedence Research, sleep apnea devices market
- Data watch: The global prevalence estimate (2019) and the Peppard US measurement (2007-2010 data) are the most recent available direct estimates; the 2025 Boers model and the industry counts (ResMed 61 million, Apnimed 80 million US and one billion worldwide) disagree because they use different ages, years and thresholds. Several adherence and recall studies use ResMed cloud data or have ResMed-affiliated authors, and connected-device data exclude non-connected users. The Medicare undiagnosed-OSA studies use 2006-2013 claims, and the heart disease figure is a 2022 conference abstract. Market size estimates conflict (6.9 billion dollars for 2025 vs 7.62 billion for 2026) and both sit near ResMed’s own revenue, so they likely define the market narrowly. The Philips consent decree and remediation totals, FDA recall pages and Grand View Research and Fortune Business Insights estimates could not be retrieved (404 or 403) and are excluded. The AASM economic burden PDF could not be parsed, and older AASM 2014 figures were left out as outdated. Apnimed’s AD109 NDA review status and an FDA decision expected in 2027 will change Section 5.
Last updated: October 3, 2026. We update this roundup quarterly, and the next refresh is expected when ResMed reports fiscal 2027 first-quarter results and Inspire Medical Systems reports Q3 2026 results, along with any FDA action on AD109.