Home sleep-apnea tests and under-mattress sensors: how close to a sleep lab?

There are now three tiers of sleep-apnea testing at home, and they are not interchangeable: a prescribed home sleep apnea test (a medical device your clinician orders), a validated consumer sensor such as an under-the-mattress mat, and the sleep score on a smartwatch. The first can diagnose clear-cut cases and is endorsed for exactly that; the second can flag severe apnea reasonably well; all of them share a blind spot for milder disease and for the events an in-lab study is designed to catch [s1][s2].

What the sleep-medicine guideline endorses

The American Academy of Sleep Medicine's clinical practice guideline is explicit about where home testing belongs. For uncomplicated adults with a high pre-test probability of moderate-to-severe obstructive sleep apnea, it recommends that diagnosis can be made with either in-laboratory polysomnography or a home sleep apnea test — the two are treated as acceptable alternatives in that specific group [s1].

The conditions attached to that endorsement are the whole point. The guideline restricts home testing to uncomplicated cases and recommends that if a home test is negative, inconclusive or technically inadequate, the patient should proceed to full in-laboratory polysomnography [s1]. It also recommends against using a home test in patients with significant heart, lung or neuromuscular disease, where the simpler recording is more likely to mislead [s1]. And it draws a hard line between medical home sleep apnea tests and consumer gadgets: the guideline advises against relying on a consumer device alone to diagnose the condition [s1]. A home test is a way to confirm a suspected obvious case, not a way to rule the condition out.

What a validated consumer sensor can do

Contactless and under-mattress sensors are the newer, tidier promise — no wires, no finger probe, just a mat under the mattress reading breathing and movement. The best-validated example is the Withings Sleep Analyzer, tested against simultaneous in-lab polysomnography in 118 patients with suspected apnea [s2]. It performed respectably at the severe end: for an apnea-hypopnea index of 30 or more events an hour, sensitivity was 86.0% and specificity 91.2%, with an area under the ROC curve of 0.954; at a threshold of 15 events an hour the area under the curve was 0.926 [s2].

But the same study exposes the ceiling. The average apnea-hypopnea index it estimated (32.8 events an hour) tracked the lab value (31.2) on average, yet the mean absolute error was 9.5 events an hour [s2]. An error of that size straddles the very cut-points — 5, 15, 30 — that separate no apnea from mild, moderate and severe. Averaged across a group the device looks accurate; for the individual whose true index is near a boundary, a nine-point swing can move them a whole severity category. That is why these devices are useful for flagging likely severe disease and prompting a proper work-up, and unreliable for grading milder cases or declaring someone clear.

The consumer tier below that

Below the validated medical and consumer sensors sit the wrist wearables, where the evidence is thinner still. As covered separately, smartwatch apnea-detection features have been validated against sleep studies with real but limited accuracy, and an Apple Watch apnea notification is a screening prompt, not a diagnosis. What every tier shares is the direction of its errors: they are far better at raising a suspicion than at putting one to rest.

Why access still matters

None of this argues against home testing — the opposite. Access to any sleep testing is the binding constraint in much of the world; a diagnostic gap documented across Chile and Peru leaves most apnea undiagnosed, and a home test that reaches people a sleep lab never will is a genuine gain. The point is to match the tool to the question. A prescribed home test, ordered when the suspicion is already strong, is guideline-endorsed and efficient. A consumer sensor is a reasonable prompt to seek that testing. Neither is a substitute for the in-lab study when the case is complicated, the home test is negative, or the diagnosis will drive long-term therapy whose adherence is itself hard to sustain.

What this means for a reader

If a home sensor or watch flags possible apnea, treat it as a reason to ask a clinician for proper testing, not as an answer. A prescribed home sleep apnea test can confirm a clear-cut case at a fraction of the cost and hassle of a lab night [s1]. But a normal reading from any at-home device — medical or consumer — does not exclude sleep apnea, especially the milder disease the numbers here show these tools miss most often [s2]. The measurement is getting cheaper and more comfortable. What it cannot yet do is safely tell you that nothing is wrong.

Sources

• [s1] Journal of Clinical Sleep Medicine — Clinical Practice Guideline for Diagnostic Testing for Adult Obstructive Sleep Apnea: An American Academy of Sleep Medicine Clinical Practice Guideline (2017-03-15)
• [s2] Journal of Clinical Sleep Medicine — Validation of the Withings Sleep Analyzer, an under-the-mattress device for the detection of moderate-severe sleep apnea syndrome (2021-02-16)