New research: Sleep apnoea tests missing patients' 'full story'

GrumpyHere wrote: ↑

Fri Feb 18, 2022 7:03 am

The validation study for the pad was published in the AASM journal behind a paywall.
https://jcsm.aasm.org/doi/10.5664/jcsm.9168

Accurate? ¯\_(ツ)_/¯

Actually in the same abstract the authors draw the conclusion:

CONCLUSIONS:Withings Sleep Analyzers accurately detect moderate-severe sleep apnea syndrome in patients suspected of sleep apnea syndrome. This simple and automated approach could be of great clinical value given the high prevalence of sleep apnea syndrome in the general population.

Is it reasonable to be skeptical of their conclusion? Yes, always, that's what the scientific process is all about !

BTW, thanks for the link !

Highly unlikely that the authors would draw any other conclusion in a published validation study with that MEAN error of 9 events per hour

Nope, that doesn't bode well for buyers using it for diagnostic purposes (in my opinion as a internet rando )

Even "mild" severity (5-15 AHI) can lead people to suspect they have sleep apnea.
So likely wasted money for those folks.

"Mild" apnea (mine was 12) can kill--it just takes longer.
(Whoopee.)
I would be concerned about FALSE NEGATIVES, saving money insread of lives.

post removed by author.

GrumpyHere wrote: ↑

Fri Feb 18, 2022 11:02 am

Highly unlikely that the authors would draw any other conclusion in a published validation study with that MEAN error of 9 events per hour

Nope, that doesn't bode well for buyers using it for diagnostic purposes (in my opinion as a internet rando )

Even "mild" severity (5-15 AHI) can lead people to suspect they have sleep apnea.
So likely wasted money for those folks.

Perhaps I've misinterpreted the intended users of this device. My impression was that the device was NOT intended to be used by patients to self-diagnose, but rather to be used by sleep professionals as a screening device to prioritize access to a formal polysomnographic study.

(I don't know about your region of the world, but in my area the wait time for a sleep clinic study is 6-9 months. Even home studies have a 30-60 day wait.)

(think Holter monitor application - longer observation period, inexpensive, reusable, and fair accuracy, but certainly not the "final" word in recommendation for cardiac surgery !)

I'm still in favor of longer observation periods (suggested by Flinders U), but running a weeks-long polysomnographic study is just not practical. !
Whether the Withings device is reliable is still (to my mind) questionable, but should be studied.

Or HMO's to deny treatment to those who may not be critical, but still need treatment.

dataq1 wrote: ↑

Fri Feb 18, 2022 3:52 pm

I'm still in favor of longer observation periods (suggested by Flinders U), but running a weeks-long polysomnographic study is just not practical. !

And using a low-pressure CPAP device to attempt to diagnose/get the "big picture" is idiotic.

I've had the sleep pad for several years. I mentioned previously that the sleep pad doesn't work as well as the original version that worked in conjunction with the Aura alarm clock. It was a better product back then. I have the Nokia version now and it can't even tell if I'm asleep or awake. I certainly would not rely on the sleep stage data. The app was really dumbed down when Nokia took over. The app continues to be quite problematic. It used to track irregular breathing disturbances and show you a graph but that part of the app is not functional now. When you click the button to look at the details nothing happens. The stupidest thing is I keep looking at the app each morning.

Forgot to mention the original Withings pad sprung a leak after the 1st year and then Nokia took over so the that was the replacement pad that I purchased.

dataq1 wrote: ↑

Fri Feb 18, 2022 3:52 pm

Perhaps I've misinterpreted the intended users of this device. My impression was that the device was NOT intended to be used by patients to self-diagnose, but rather to be used by sleep professionals as a screening device to prioritize access to a formal polysomnographic study.

Even as a screening tool, it leaves a lot to be desired.
So it's marketed as a "Quantified Self" and "IoT Home Automation" toy.

The company does not market it as a diagnostic tool <wink, wink, nudge, nudge> for good reason.
But the conclusion drawn in the paper clearly shows their intent.

(I don't know about your region of the world, but in my area the wait time for a sleep clinic study is 6-9 months. Even home studies have a 30-60 day wait.)

Pre-COVID, two years wait (per media reports).

I got lucky after I subjected myself to an ApneaLink HST (used as a CPAP selling tool).
Someone cancelled and a space opened up within 2 months.

I'm still in favor of longer observation periods (suggested by Flinders U), but running a weeks-long polysomnographic study is just not practical. !
Whether the Withings device is reliable is still (to my mind) questionable, but should be studied.

There are already PAT based tests in the market that can do multi-night testing.
Testing which can diagnose mild apnea unlike this toy.
https://en.wikipedia.org/wiki/Peripheral_Arterial_Tone

It's a shame they aren't being used as such but ... profits (OTOH no profit, no tools)

The major strike against NightOwl HST is that it was validated against "MICHELE SLEEP SCORING SYSTEM" (https://www.cerebra.health/)
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6175785/

So it's 87% (correlation against MSSS) multiplied against whatever correlation MSSS had against the gold standard PSG.

GrumpyHere wrote: ↑

Fri Feb 18, 2022 7:15 pm

Even as a screening tool, it leaves a lot to be desired.
So it's marketed as a "Quantified Self" and "IoT Home Automation" toy.

Well that about completes the response to the part of the question raised by @Respirator99, (Original poster) about a device being studied by Flinders U. I'd concur with your thinking.

The only other issue/discussion raised is the value of multi-night sleep observations.
I'd still (intuitively) contend that multi-night observation to obtain a patient's overall sleep disfunction is preferable to "snapshot" observation.

That in mind, investigation of any system that is reliable and feasible AND can be used over multiple nights to observe untreated sleep disfunction without imposing therapy is worthwhile pursuing .