Apple Watch sleep apnoea notifications: what the alert means and what to do next

If your Apple Watch has told you it found signs of possible sleep apnoea, the short answer is: take it seriously, but it is not a diagnosis. The feature is cleared by the US FDA as a screening tool that flags a pattern of breathing disturbances over a month, and in Apple’s own validation study it caught about two thirds of people with moderate-to-severe sleep apnoea while almost never flagging people who did not have it. The next step is a conversation with a clinician and, if they agree, a proper sleep study.

What the feature actually measures

The watch does not measure your breathing directly. According to Apple’s support documentation, when you wear the watch to bed it “uses the accelerometer to look for breathing disturbances while you sleep”. In plain terms, it watches for tiny wrist movements that tend to accompany interruptions in normal breathing. Each night’s data is sorted into one of two bins, “Elevated” or “Not Elevated”.

The feature runs in the background over rolling 30-day windows. Apple states you “must wear your Apple Watch to sleep for a minimum of 10 nights over a 30 day period” and that your data will be “analyzed every 30 days”. If enough nights come back Elevated to match a pattern the algorithm associates with moderate-to-severe sleep apnoea, you get a notification. It is intended for people aged 18 or over who have not already been diagnosed with sleep apnoea, and it runs on Apple Watch Series 9 or later, Apple Watch Ultra 2, or Apple Watch SE 3 (accessed 11 September 2026).

Two lines from Apple’s own page deserve more attention than the notification itself. The first: “Not all people with sleep apnea receive a notification.” The second: the feature “is not intended to diagnose, treat, or aid in the management of sleep apnea”.

What the FDA clearance says about accuracy

The regulatory document is where the numbers live. The FDA cleared the Sleep Apnea Notification Feature on 13 September 2024 under 510(k) number K240929, in a category with the deliberately modest name “Over-the-counter device to assess risk of sleep apnea”.

The clearance summary describes a prospective study enrolling 1,499 subjects at several US sites, with a home sleep apnoea test (the Nox T3s) as the reference standard rather than in-lab polysomnography. Of those enrolled, 1,278 contributed to the notification analysis; the rest had too little watch or reference data. The population deliberately spanned the severity range, with 559 normal subjects, 362 mild, 216 moderate and 201 severe by the apnoea-hypopnoea index (AHI).

The headline results, as reported in the FDA summary:

• Sensitivity for moderate-to-severe sleep apnoea (AHI of 15 or more): 66.3%, with a 95% confidence interval of 62.2% to 70.3%. Put another way, roughly one in three people with clinically significant sleep apnoea did not receive a notification within the study window.
• Specificity for people with normal-to-mild results (AHI under 15): 98.5%, with a 95% confidence interval of 98.0% to 99.0%. The summary adds that the feature “did not falsely notify any subjects with normal” AHI.

That asymmetry is by design. The summary explains the performance was tuned for high specificity given that the feature is a passive, recurring detector. Apple’s press release for the Series 10 launch put the same point in marketing language: “every participant identified by the algorithm had at least mild sleep apnea.” Both statements are true, and both mean the same thing for you: a notification is a fairly strong signal, but silence is not reassurance. The FDA text is explicit that “The absence of a notification is not intended to indicate the absence of sleep apnea.”

A few other details from the summary are worth knowing. The underlying deep-learning model was trained on more than 11,000 nights of watch data paired with sleep-lab or home-test recordings. The feature’s predicate device (Samsung’s) uses blood-oxygen data, whereas Apple’s uses accelerometer data only. And the clearance includes a predetermined change-control plan, which means Apple can retrain and adjust the algorithm within agreed limits without a fresh clearance, so the performance figures above describe version 1.0 and may drift.

Why the sleep-medicine field is cautious

The American Academy of Sleep Medicine’s position statement on consumer sleep technology (published 2018, and still the society’s stated position) says that, given the lack of validation and FDA clearance, consumer sleep technologies “cannot be utilized for the diagnosis and/or treatment of sleep disorders at this time”. Apple’s feature is a partial exception to the “lack of FDA clearance” part, because it has one. It is not an exception to the diagnosis part: the clearance itself says it is not a diagnostic device.

The same statement sets the bar clearly: “It is the position of the AASM that CST must be FDA cleared and rigorously tested against current gold standards if it is intended to render a diagnosis and/or treatment.”

The AASM’s clinical practice guideline on diagnostic testing is equally direct about what does count. Its strong recommendations include that “polysomnography, or home sleep apnea testing with a technically adequate device, be used for the diagnosis of OSA in uncomplicated adult patients presenting with signs and symptoms that indicate an increased risk of moderate to severe OSA”, and that “clinical tools, questionnaires and prediction algorithms not be used to diagnose OSA in adults, in the absence of polysomnography or home sleep apnea testing”. A watch algorithm belongs in that second category. The guideline also recommends that if a single home test is negative, inconclusive or technically inadequate, polysomnography be performed. That matters because Apple’s validation itself used a home test as its reference, not the in-lab gold standard.

What to do when you get the notification

The honest answer is boring: book an appointment. The notification is a reason to see a GP or a sleep clinic, not a result to act on by yourself. Apple builds this into the feature: it generates a PDF you can export and share with a clinician, and the support page says plainly that if you receive a notification you should talk to your doctor.

For a desk worker, the conversation is easier if you bring context. Note whether you snore, whether a partner has seen you stop breathing, whether you wake unrefreshed, and whether you fight off sleep in the afternoon (persistent daytime sleepiness has more causes than a heavy lunch; see why do I get tired after lunch). Bring the exported PDF. Then let the clinician decide between a home sleep test and a full lab study.

What not to do is treat a month of “Not Elevated” nights as a clean bill of health. With sensitivity of around 66% in the validation study, the feature misses a meaningful fraction of real cases. If you have symptoms, the symptoms matter more than the watch.

The bigger picture

Apple’s launch materials cite an estimate that sleep apnoea affects more than 1 billion people worldwide and “goes undiagnosed in most cases”. If a wearable nudges a portion of those people towards a proper test, that is a public-health win, and the high specificity means it should do so without flooding clinics with false alarms. But the same design choice means it is a net with wide holes. Treat the notification as a screening prompt, keep the sleep study as the standard, and read the device’s own fine print, which says exactly that. For how we choose and weigh sources like these, see how we research.

Common questions

Can an Apple Watch diagnose sleep apnoea?

No. The FDA clearance classifies it as an over-the-counter device to assess risk, and the clearance text states it “is not intended to diagnose, treat, or aid in the management of sleep apnea”. Apple’s support page says the same. A diagnosis requires polysomnography or a technically adequate home sleep apnoea test ordered and interpreted by a clinician.

How accurate is the Apple Watch sleep apnoea notification?

In the FDA clearance study of 1,499 enrolled adults, the notification had 66.3% sensitivity for moderate-to-severe sleep apnoea and 98.5% specificity for normal-to-mild cases. It flagged no one with a normal AHI. So a notification is a strong signal, but around one in three people with moderate-to-severe disease were not notified.

What should I do if I get a sleep apnoea notification?

Export the PDF report from the Health app and book an appointment with a GP or sleep specialist. Do not self-treat. The clinician will decide whether a home sleep test or an in-lab study is appropriate, following guidance such as the AASM’s diagnostic testing guideline. If you have symptoms such as loud snoring or daytime sleepiness, mention them.

Does no notification mean I do not have sleep apnoea?

No. Apple’s page states that “Not all people with sleep apnea receive a notification”, and the FDA summary says the absence of a notification is not intended to indicate the absence of sleep apnoea. The algorithm was tuned for high specificity, which trades away sensitivity. If you have symptoms, seek assessment regardless of what the watch shows.

Which Apple Watch models support sleep apnoea notifications?

Apple’s support page lists Apple Watch Series 9 or later, Apple Watch Ultra 2, and Apple Watch SE 3 with the latest watchOS. You need Sleep tracking set up, at least 10 nights of wear in a 30-day period, and to be 18 or older without an existing sleep apnoea diagnosis. Availability varies by country.