A notification arrives telling you that your watch, or your ring, has seen breathing disturbances consistent with sleep apnoea. The useful question is not whether to believe it. It is what that notification is worth, and what silence from the same device is worth, because those two things are not equally informative.
Obstructive sleep apnoea is common and the majority of people who have it have never been diagnosed. Part of the reason is that the headline symptom belongs to someone else. Snoring is noticed by a partner, and people who sleep alone lose that signal entirely.
The symptoms that belong to you are quieter and easy to attribute to something else:
That blood pressure point is worth sitting with. Untreated sleep apnoea is one of the most common causes of hypertension that resists medication, which is why it comes up in what to do about a smartwatch blood pressure alert as well.
As of August 2026 there are three features cleared by the FDA to screen for sleep apnoea: Apple's sleep apnoea notification feature, Samsung's original feature, and Samsung's updated version that extends it to their smart ring, the first over the counter smart ring authorised to assess risk of obstructive sleep apnoea.
None of them measure airflow. They infer disturbed breathing from movement, pulse and oxygenation patterns, usually over a multi night evaluation window rather than from a single night.
The accuracy numbers are the part worth reading carefully, because they are asymmetric.
Apple's feature misses a lot but rarely cries wolf. Published performance for detecting moderate to severe apnoea sits around 66 per cent sensitivity with specificity above 98 per cent. In plain terms: if it tells you something, take it seriously. If it tells you nothing, that rules out very little.
Samsung's watch catches more. Reported sensitivity of around 91 per cent for moderate to severe cases.
Neither diagnoses anything. That is not a disclaimer, it is the regulatory status. These are screening prompts cleared to tell you to get tested.
The practical upshot: a notification is a reason to book a test. No notification, when you have the symptoms above, is not a reason to drop it. This is the same trap people fall into with every consumer screening feature, and it is the single most common way they do harm.
Two routes, and most people now start with the first.
A home sleep apnoea test. A small kit worn for one to three nights, measuring airflow, effort, oxygen saturation and pulse. Convenient, much cheaper, and good enough for straightforward cases in people with a reasonably high likelihood of the condition.
In laboratory polysomnography. An overnight study with full monitoring including brain activity. It is the right test when the home study is negative but suspicion remains, when central sleep apnoea or another sleep disorder is possible, or when there is significant heart or lung disease.
Either way the result is expressed as an apnoea hypopnoea index, the number of breathing events per hour of sleep. Five to fifteen is mild, fifteen to thirty is moderate, above thirty is severe. The number guides treatment, but symptoms and cardiovascular risk weigh just as heavily, which is why two people with the same index can be given different advice.
CPAP is still the most effective treatment, and adherence is the entire game. Most people who abandon it do so in the first fortnight, almost always because of mask fit, pressure settling or dryness, and almost all of those are fixable if you report them rather than giving up. Ask for the follow up appointment at the point you are given the machine.
Mandibular advancement devices hold the lower jaw forward and are a genuine alternative for mild to moderate disease, or for people who cannot tolerate CPAP. Fitted by a dentist with the relevant training, not bought online.
Positional therapy matters for the substantial group whose events happen almost entirely while lying on their back.
Weight loss changes the disease itself where excess weight is driving it, and this is where the landscape has moved. Tirzepatide is now approved specifically for moderate to severe obstructive sleep apnoea in adults with obesity, which makes it the first drug treatment aimed at the condition rather than at its symptoms. It is an addition to treatment, not a replacement for it, and the GLP-1 class has its own trade offs worth understanding first.
Hypoglossal nerve stimulation, an implanted device that moves the tongue forward with each breath, is an option for selected people with moderate to severe disease who cannot use CPAP.
Treating nasal obstruction, moving alcohol away from bedtime and reviewing sedating medication all help at the margins, and none of them substitute for the main treatment.
The appointment goes better when you arrive with evidence rather than a notification screenshot.
Keeping that in one place, rather than across three apps and a notes file, is the difference between a ten minute appointment and a useful one. The vitals and timeline in a Health Passport exist for exactly this, and if you want to organise the symptoms first, a structured symptom check will ask the questions a sleep clinic would.
Untreated moderate to severe sleep apnoea raises the risk of hypertension, atrial fibrillation, stroke, type 2 diabetes that is harder to control, and road traffic accidents. Treated, most people describe the change in daytime function as the most noticeable health improvement they have had. Few conditions offer that ratio of effort to payoff.
Consumer sleep apnoea features are good at pointing, bad at reassuring. Apple's misses roughly a third of moderate to severe cases, Samsung's catches more, and none of them can diagnose. If you were flagged, book a sleep test. If you were not flagged but you wake unrefreshed, nod off during the day, or have blood pressure that will not come down, book one anyway.
This article is for general education and is not a substitute for professional medical advice. Consumer wearable features are screening tools, not diagnostic devices. Excessive daytime sleepiness affects driving safety and should be discussed with a clinician promptly.
Smartwatch hypertension alerts are specific but miss roughly half of cases, and their meaning depends heavily on your age. What the notification is telling you and what to do next.
Normal blood tests don't always mean you're well. What standard panels miss, the tests to ask for by name, and how to advocate for a proper investigation.
CGMs and metabolic and hormone wearables are everywhere in 2026, but the marketing runs ahead of the evidence. An honest look at what they actually measure and when the data is useful.