SCIENCE / SUMMARY / ACOUSTIC APNEA

How close do phones come to a sleep lab when screening for sleep apnea?

META ANALYSIS · DIAGNOSTIC ACCURACY ■■□□ low evidence: 11 studies, but methods and definitions differ widely and participants were already suspected of apnea

Who: 1,644 people in 11 English language studies published up to May 2021. Participants were people referred for polysomnography (a full night lab sleep test) because of suspected sleep apnea, not the general population.
What they did: Six databases were searched. Phone measurements based on movement, sound, finger oximetry or combinations were judged against moderate and severe apnea (15 or more breathing events per hour in the lab, AHI ≥15). Study quality was rated with the QUADAS-2 tool and results were split into subgroups by method.
What they found: All methods together: sensitivity 0.91 (95% CI 0.88 to 0.93), specificity 0.88 (0.82 to 0.92), area under the curve 0.917. In the 5 studies using sound alone, sensitivity was 0.88 and specificity 0.81; in the 3 studies using movement alone, specificity rose to 0.96. The mean difference between phone and lab AHI was -0.68 events per hour, but the authors report wide limits of agreement (around 6.2) and very high variation between studies for this comparison (I² 92.9%). The authors conclude that phones may help with screening but cannot replace the lab.
Limits: As the authors note: few studies; different detection methods and apnea definitions caused marked variation between results; three studies did not state the type of lab test; agreement on event counts rests on only four studies. In addition: only English papers were included, and since most participants were already suspected patients, the same figures should not be expected at home or in the general population. No funding received and no conflicts of interest reported.

What does this mean for you?

A phone recording at night can give you an idea of what happens while you sleep. But sound based measures carry a higher share of false alarms, and a single night's count can differ quite a lot from a lab result. If you suspect sleep apnea, the sleep test a doctor orders is what decides. This is not advice.

Link to Nefesly: One of the reasons Nefesly presents its night sound indicators in the language of "possible" rather than diagnosis, and does not show apnea severity. It led us to keep the false alarm share of sound alone in mind when reading high values recorded at home.
Reference: Kim DH, Kim SW, Hwang SH. Diagnostic value of smartphone in obstructive sleep apnea syndrome: A systematic review and meta-analysis. PLOS ONE. 2022;17(5):e0268585. doi:10.1371/journal.pone.0268585 · Open access (Creative Commons Attribution License).
Why chosen: One of 15 sources selected from 534 candidates on the acoustic snoring and apnea axis; a meta analysis that splits phone screening into subgroups by method.
Summarised by: Nefesly, from the full text, 11 October 2026. Awaiting clinician check.
Copyright: Summarised from the full text, not translated from the abstract.