Skip to content
Blog

One Sleep Study vs 3 Months of Tracking

A 2026 study tracked 92 adults with an under-mattress sleep tracker for 3 months alongside a single-night sleep study. For 16 of 78, the two landed on different sides of the same cut-off.

PulseOxTracker Team
Infographic. One sleep study night vs weeks of under-mattress tracking. 16 of 78 landed on different sides of the line, the line being 15 breathing events per hour. 29 were above the line on both, 33 were below the line on both, 10 were above on the mat and below on the sleep study, and 6 were above on the sleep study and below on the mat. Who was studied: 92 adults referred for suspected sleep apnea, average age 53, 52% women, 78 were in the comparison. What the study found. Tracking kept up: 85 of 92 had 14 or more usable nights in month one. Nights collected: median 80 usable nights in the first 3 months. The 10 had bigger night-to-night swings: average swing 10.5 events per hour, against 4.8 in the 6 and 15.1 in the 29. The 10 had a short sleep study night: they slept 292 minutes on average, against 389 in the 6. The 10 were mostly on their side: median 90% of the sleep study night on their side, against 60% in the 6. The 6: 3 of the 6 were shift workers, none of the 10. What it cannot show: which of the two answers was right. Observational feasibility study. The groups that disagreed were 10 and 6 people. The mat does not measure oxygen or body position. Withings gave the devices and part-funded a research post. Source: Lechat et al. 2026, npj Digital Medicine.
A summary of the 2026 study: how often one sleep study night and weeks of under-mattress tracking disagreed, and what the people who disagreed had in common.

A sleep study measures one night. Breathing during sleep changes from night to night.

A 2026 Australian study put a sleep tracker under people's mattresses for 3 months and compared it with their single-night sleep study.

Who was in the study?

  • 92 adults referred to sleep clinics in South Australia for suspected sleep apnea.
  • Average age 53, 52% women.
  • Each got a Withings Sleep Analyzer, a mat under the mattress that estimates breathing events per hour.
  • Each also had one sleep study: 54 in a lab, 38 at home.

Did people keep tracking?

  • 85 of the 92 (92%) had at least 14 usable nights in the first month.
  • The median was 80 usable nights over 3 months.

How often did the two disagree?

The cut-off was 15 breathing events per hour. The sleep study was compared with the mat's average over the weeks around it, in 78 people. The other 14 were on therapy already or started it too early.

  • 29 were above the cut-off on both.
  • 33 were below it on both.
  • 10 were above it on the mat only.
  • 6 were above it on the sleep study only.

So 16 of 78 got a different answer.

What did those people have in common?

The 10 above the cut-off only on the mat:

  • Had bigger night-to-night swings: 10.5 events per hour on average, against 4.8 in the 6. The 29 above on both had the biggest swings, 15.1.
  • Slept less on the sleep study night: 292 minutes on average, against 389.
  • Spent a median 90% of that night on their side in the paper's table, against 60% in the 6.

Of the 6 above it only on the sleep study, 3 were shift workers, against none of the 10. The authors say irregular sleep may throw off the mat's estimate.

More: back vs side sleeping.

What can't it show?

  • Which answer was right. Neither method was taken as the reference.
  • This was an observational feasibility study, and comparing the two methods was a secondary aim.
  • Why they disagreed. The authors urge caution with groups this small.
  • The mat doesn't measure oxygen or body position.
  • Withings gave the devices and part-funded a research assistant.

What can you do with this?

The study used a mat, not an oximeter.

Not a medical device. Does not diagnose, treat or prevent any condition. Discuss results and changes with a clinician.

References

  1. Lechat B, Proctor S, Manners J, Loffler KA, Leviton A, Toson B, Vena D, Vakulin A, Fitton J, Adams RJ, Mukherjee S, Eckert DJ. Zero burden multi night monitoring with AI enabled technology reduces obstructive sleep apnea misdiagnosis. npj Digital Medicine. 2026;9(1):753. doi:10.1038/s41746-026-02914-w