How on-the-wrist sleep apnea detection works
empirical.health
empirical.health
1. Yes, we should regulate manufacturers intensely.
2. No, we shouldn't regulate to whom companies should be allowed to sell them. The defaults are good enough for most. Provide an easy to use control for max pressure and people will figure it out. Perhaps a knob?
I claim: During the past decade, the risk of being maltreated by a personally set-up CPAP unit has been far, far smaller than the risk of not being treated at all. (Ways to access these menus have been been shared online for 10+ years, many use them.)
I wonder how many premature deaths happen every year because of this - including secondary victims from traffic accidents.
These things should be $500 mass-market consumer units that you pick up at Target, Walmart, etc.
Large companies e.g. Google/Apple are just inherently untrustworthy in this regards, smaller overseas ones are no better (little to no regulation).
So please, no, stay the ** out of my health, thanks.
In the U.S. you already have relatively minimal rights in healthcare. Your stability and legal guarantees are already compromised compared to your worldly neighbors. I too distrust tech companies in healthcare due to their differing motivations and organizational cultures.
I want the red tape to go away, but I also want strong regulations and safety nets to ensure the drugs and devices we are prescribing to people are effective and safe.
I'm not certain those goals can be both met. You can have innovation, but some red tape is necessary.
Of course you could get into arguing that "not all red tape is made equal", but deterring cheap "innovation" from a tech shop trying to make a quick buck is absolutely necessary and useful. The barriers exists for very good reasons, not only as you said they have a different motivation and organizational culture, but they have different qualifications.
Personally, I consider big pharma companies the absolute least trustworthy, and I would never choose to purchase anything from them if I would not literally otherwise die. Regulations are not fundamentally bad, but the current situation is obviously a case of regulatory capture.
Wider competition with less control/accuracy will often mean that fads get more profitable: for instance "gluten free" sells enough that it makes sense for random makers to slap the label on their product even if they don't really care to follow the most rigorous rules.
Letting the market decide will result in the minority (those actually affected by the products) to get shafted, as more and more companies will move to the lower standards they can get by, instead of keeping high priced products that don't sell as well.
That doesn't mean that the status quo is great, but at least we need to be pretty cautious on the direction we want to go. "Let's just do something and see how it goes" isn't a desirable strategy when people are at risk.
Competition does not mean getting rid of or even lowering standards, just that if those standards are misaligned with real people's needs we're not stuck with them.
Maintaining a known-flawed status quo is not being cautious. People suffer severely every day from inaccessible medical care. We can always return to the current system if we don't like a change, but until we actually try it we don't even know how much damage the status quo is doing.
It's been 4 years and Phillips has yet to replace them. They also don't answer any kind of contact about it. I'm glad the local authority gave me a resmed but I still want those Phillipses replaced.
There have been over 500 confirmed deaths due to the Philips CPAP. The number of unconfirmed deaths would be an order of magnitude higher, in my opinion.
Thing is, the CPAP issue could be argued for both sides:
"See how a small manufacturing defect could cause so many deaths? We definitely need to regulate this heavily!"
OR
"See how many people died due to Phillips because they didn't have alternatives?"
These were units prescribed by licensed people. The issue here was the manufacturer.
As such I'm not sure what your complaint is.
What made you think I was complaining?
Even if they are honest I'd expect near 100% - people who don't expect they have issues are not going to be getting the at home test in the first place.
(I'm honestly surprised someone doesn't make the vendors require a 1000 character passphrase that locks you out of the device after 2 failed attempts.)
My dad's friend died like that, not a flu, but was traveling overnight and didn't take his machine and died in his sleep. Although I almost wonder if this heavy reliance on cpap machines causes people to lose some innate ability to recognize when they stop breathing or at least delays the normal instinct to wake up and breathe.
What change did you see in your sleep routine?
I hope there's some insight beyond what I've got, but just from my experience: Sometimes having it blasting at the pressure you need while trying to get drowsy makes that process uncomfortable, but, after you've gained drowsiness it's not going to interrupt you and wake you (unless it starts leaking pressure against your face)
I'm iffy on the machines themselves for this, mostly because I'm just ignorant about the risks and potential issues. But this absolutely should be the case for the consumable/wear items that go along with it. There's been a lot of push-back on retailers being able to sell things like the masks, the hoses, filters, etc. to the public at large that can't be used without a machine anyway.
> These things should be $500 mass-market consumer units.
Fundamentally yes. I'd be fine with it being something as simple as the local pharmacy can keep them in the back and you bring in the prescription from the doctor and they file it away saying that you're allowed to buy/replace it if there's some reasonable evidence that they need some modicum of control for safety of people. It shouldn't be any harder than getting some antibiotics.
As it is now I'm using a 10+ year old machine because I've changed insurance companies multiple times since and my current one wants a full sleep study before they'll sign off on anything at all (like allow the doctor to prescribe or let me order one). And they wouldn't tell me if they'd cover the sleep study or not.
Just curious, why do you think a Dr. should be involved at all?
Most of what doctors do should be basic checklists - this is the same thing millions of other people have and should be treated by checklist in best practice. It is the exceptions where you are that 1 in a million case and so there is no checklist that you need doctors to read all the data and think.
What do you think your doctor will do based on this data? How do you think your doctor will filter out the signal from all the noise? (Is there even a signal?)
> but once in a while there is something more going on and I need to be forced into the ER.
Do you have a concrete example here?
as for telling the difference we need a lot of help there.
I often change the setting myself, it's just a hidden key combo anyway.
Really helps great against snoring too. And the max pressure is 20mm h2o which is really not a lot.
Some people obviously are capable of diagnosing and solving their own medical problems and are willing to learn and do the work but most people aren't.
I don't think that opening up CPAPs to the masses would suddenly "solve" sleep apnea, most people will still need to go through a similar process that exists now.
Regulation that streamlines the involvement of insurance providers would probably be enough IMO, they are the real reason it is difficult to get a CPAP most of the time.
The obvious solution is to bypass the insurance providers.
Also: when it comes to fitting masks to faces and figuring out the optimal pressure rates - these seem like pretty obvious application of machine learning algorithms, don't you think?
Anecdotally, my impression is that most CPAP users are using APAPs at this point, like the very popular Resmed AirSense 10. Titration amounts to giving it a pretty wide range and letting the machine itself figure out the right pressure. It's not a complex learning algorithm, but seems to do the job.
The algorithm is more about detecting that you aren't asleep and thus don't need the full pressure at the moment and less about altering the target pressure setting on the fly.
I could be wrong though, I use is a bipap which works a bit differently than CPAP or APAP.
To my understanding (no personal experience), Philips APAPs have a different algorithm to try and achieve the same goal, and are regarded as less aggressive about it than Resmed.
There's also a new Resmed Airsense 11 out but I don't know anything about it and what changed.
BiPAPs are definitely a different beast, for sure. My dad used one of those. Not automatic at all, right? Constant pressure, but different between inhale and exhale?
That is one solution but I think it would be better to make it easier for people to go through the current process rather than making them pay out of pocket (assuming that it what you meant by "bypass the insurance providers").
Assuming that we aren't going to also fix the entire American medical system, making it easier for people to use the insurance they already have to get a medical device they need seems like the best approach to me.
> Also: when it comes to fitting masks to faces and figuring out the optimal pressure rates - these seem like pretty obvious application of machine learning algorithms, don't you think?
Yes but I don't think that means we should take the human expert(s) out of the loop.
I rather quickly changed the pressure settings on my bipap in the hidden menu after the titration but many, maybe most, people aren't like me and need there to be someone they can just say "this isn't working for me" to instead of being left on their own.
A $200 CPAP with a knob that lets you control pressure absolutely would solve sleep apnea for much more people than gatekeeping it behind an insane price + having to make a doctor's appointment through insurance just to get them to turn the knob for you.
Because those who can't turn a knob themselves would still be able to schedule an appointment with a knob-turner without making the rest of us have to do it as well.
I got a hand me down CPAP from my dad, bought a mask off Amazon, played with the knob a bit, and within a week I cured my central sleep apnea. It's just not hard for anyone who doesn't think the doctor has divine capabilities over what they could do themselves (turning a knob).
But instead we infantilize everyone by requiring them to do the equivalent of taking their car in for a licensed professional to check their oil dipstick. And, of course, making them or their insurance pay $300 for the pleasure.
But gosh, I hope this is accurate across the Watch's wide user base.
I wouldn't say it's likely, but I do wonder if the pulse oximeter is being used surreptitiously. In any case, if you could use the pulse oximeter, this functionality would immediately become much more powerful.
I don't think it has to be as accurate as other things like blood glucose levels, because with those people might use the results to inject insulin.
Detecting sleep apnea might be more like afib detection. It just tells someone they should be checked out.
Anyone know why that isn’t the method?
But the reality is that everything that needs any kind of power is being aggressively idled down and time sliced and being made sample-based as can be gotten away with. You can easily get a sense of this with heart rate monitoring, if you dig into Apple Health and observe the pattern of samples it takes.
Almost certainly, running the sensor at such a high frequency that you could call it "constant" would drain the battery long before you woke up.
Apple instead prioritizes 3000 nit displays, laptop-class processors, and 60fps touch interactions. These are all characteristics which make for great keynote slideshows, and maybe even close a lot of "first time smartwatch" sales, but I simply don't encounter anyone who wouldn't trade much of that for an Apple-made device that lasts 3+ days and has higher resolution sensor collection. The sleep apnea detection alone is, while interesting, something of a farce when it was followed up with a "18 Hour Battery Life" slide; yes you can make it work, but its an active decision by Apple to ask you to, and it doesn't have to be this way.
Didn't see the slide by my apple watch definitely lasts way longer than 18 hours. But, I also have the screen default to off instead of always on (lets be real, who spends all day staring at their watch; motion activated turning on seems like not even a compromise).
Its not helpful, productive, or useful to state that you turn off headline features people paid to have in order to get greater battery life than even Apple advertises the device as having. Please adjust your discourse and do not repeat your behavior.
2. That would require more battery life than the Apple Watch offers.
There are third-party options that do constant monitoring that either directly or indirectly support Apple Health. You'll need some additional intelligence, in the form of an app or something, to analyze that data. WatchOS won't do it at present.
(My flimsy understanding is that blood oxygen measurement is how at-home sleep studies work?)
I did one of these and it was the worst night of sleep I ever had. It's somewhat invasive. Meanwhile, I sleep with my watch every night with no problems.
What was funny was that at-home sleep study was damned good sleep because it held my nose open!
I got some sleep.
Depending on the numbers, why not?
If it's sensor accuracy a single night might not be enough, but if it detects the condition averaging over a few weeks even unreliable measurements can be made more reliable (given they're just noisy).
That being said, your observation that a nightly watch user can collect lots of data is an interesting one. If Apple (or whoever) can come up a more heuristic analysis than just computing AHI, they may end up with something better.
Also, not that you asked, but my sleep situation is a bit complicated and I actually ended up with a CPAP anyway.
What do they mean by "wasted medical spend" here?
(1) Patients incorrectly spending money on a misdiagnosis, or (2) apnea solution providers not making money off people when they could?
Reserving it to Series 10 and 9 really exposes their lack of hardware innovation. I mean, we were talking about them measuring glucose levels for years and this is what we get as a new flagship feature?
I’m confident there’s more to it than just sensor availability. Compute power for starters, as well as the fact that a Series 1 accelerometer may not have the necessary resolution, etc.
Reserving it to Series 10 and 9 really exposes their lack of hardware innovation.
Ah, the ol’ “Apple didn’t build what I want, ergo, Apple doesn’t innovate” trope.
Without knowing more details about the implementation, it may rely on combining data from other sensors as well, it may need a certain frequency/resolution from a sensor that's not available in earlier generations of hardware, it might place too much demand on compute or power to maintain prior products' expected behavior, etc
Love it or hate it, this mentality of "the feature is technically possible but we won't enable it because we're not happy with the compromises it demands" is, by design, exactly the sort of logic that Apple is known for.
However, reading on PAT (Peripheral arterial tone) and existing FDA-approved devices using that, they indeed use oxygen but not temperature. See for example "WatchPAT ONE" device.
Although it's possible that Sleep Apnea detection uses the oxygen saturation sensor but doesn't show its results directly.
I hope it’s just still disabled. It would suck for someone to buy a new watch, the fix the dispute in 2 months, and you never get the feature until you buy another new watch.
They’re clearly not bending. I don’t know if they’ve given up or waiting on an appeal or maybe home a new administration will tell the ITC to go away or what.
They appear to have restricted it to 8 GB of RAM devices or more (when those appear).
During The Talk Show interview at WWDC with John Gruber an Apple exec (Joz?) said they did it because of user experience issues, seeming implying speed.
And given how hard AI hits hardware I think that’s relatively believable.
I think that growth is going to slow no matter what they do. Western and Japanese markets are saturated, and Chinese markets (once most promising) are bringing fierce competition both from local competition and the government.
There are also other issues like power consumption overnight. Older watches have smaller batteries and slow charging. People will update the software, use the new features, and complain that Apple software updates are killing the battery (waking up to low battery plus slow charing means not being able to charge up before leaving for work). This has the effect of Apple only backporting new features if they don't consume too much battery.
Do you mean per hour?
Additionaly increasing blood pressure on the head helps for sleeping more. I do that and sleep for an average of 10 hours a day. 12 hours or 14 hours/day of sleep for me, is not rare either.
Also increasing the blood pressure on the head helps fix myopia 100%, like that woman astronaut who returned to earth and her eyes worked perfectly again.
In space that's the default situation. The heart pulses maniacally to keep blood circulating all around, but in no gravity places the legs are a pretty small organ, they don't need that much blood. So in space, a lot of blood ends up on the head. We can simulate that on earth for half an hour/day and it has numerous benefits.
How is increased head pressure 30-60min a day going to change the fact that your soft tissue at the back of your throat closes the airway when you're sleeping?
Any other solution to clear out liquids from nasal airways doesn't make sense. Nasal sprays for example are total garbage. Measuring how much the airways are blocked doesn't make any sense either.
The goal is to clear out stuff from the airways and there are several ways to achieve it, all of them much worse than what i describe. They all hover in 5% efficacy to 20% or 30% efficacy.
The only other solution with such a high percentage of efficacy, is to chew 10.000 times a day, which is very inefficient.
Monkeys when climbing on trees, turn upside down hundred times a day. People who climb trees and turn upside down hundred times a day, have no sleep apnea problems.
Instead of climbing on trees, there is a way to simulate that effect on the industrial age.
No, but that was a popular torture fathers did to kids some decades back. I know several people, all men, whom their fathers tied the legs to a rope and hung them upside down for some hours, when they were boys. Pretty funny!
Much simpler actually. Just hung your head lower than your body, from a bed, a sofa, or even a motorcycle as much as you can, 10 minutes or half an hour.
It is a good idea to exercise at the same time, your strongest muscle on your body, your jaw muscle, by chewing imaginary leaves. That will help blood to flow on the head much quicker, and much more of it as well.
Back when we were monkeys, we chewed a lot of leaves. According to my estimations, we moved our jaw up and down, more than 10.000 times per day. That helped move blood up, towards the head, while nowadays we chew 100 times per day.
But getting fit enough for a regular headstand is a pretty solid investment from my experience.
I then used an at home sleep study that came back negative.
I have no idea how some people can charge their smartwatch 1 or more times a day.
Also did some bike packing this year and charging it with a power bank was also not an issue. At one point my power bank was empty and my apple watch ran out of power as well - this has been the only time I whined for more battery life.
A nice jump but not earth shattering.
Little unclear on macrumors.
Internet Archive has it: https://archive.org/details/b28114450
Also look into "mouth taping".
https://en.wikipedia.org/wiki/George_Catlin#Observations_on_...
I mouthtape and I got a full beard! Just use the 3M medical tape from Walgreens.
I don't do it every night, but I do it on occasion. Used to do it more. I find it's helped me just sleep mouth closed as normal now.
I have a beard and use GrifGrips sports tape to mouth tape. I've found the level of adhesiveness to be a perfect balance of strength and removability. Not the cheapest, but quality stuff.
This has all the hallmarks of quackery.
Why do you think how to live healthy is a modern invention? Don't you think people had to figure that out ages ago, or else we wouldn't be here?
Your argument is just a mix of the appeal to nature fallacy and noble savage trope that people use to justify all sorts of medical treatments with no proven benefits.
The Romans had great health. Proof: here we are thousands of years later alive and using their letters.
Not really, no. They had many ideas about health that we know are false now. And we still don't know a lot. Generally people died really young in those days.
But these things don't really matter for the population continuity if people get sick after they get kids, and they used to have them really early in those days (and many). They just didn't usually live until their 90s.
And of course knowing how to live healthy and actually doing it are different things too. It's something we really struggle with as a society and me too.
But I did “look into ‘mouth taping’” as you suggested, and I’ll bet half the page titles were “Mouth-taping: is it safe?” or “…does it work?” At what point does Betteridge’s Law of Headlines[0] kick in?
[0] https://en.wikipedia.org/wiki/Betteridge's_law_of_headlines
Have you read the book? Do you think I'd be recommending something that didn't have a very strong dataset backing it?
> At what point does Betteridge’s Law of Headlines[0] kick in?
Nothing is easier to manipulate than digital symbols. Thus, when talking about patterns one finds in digital symbols, one much be very very careful to build a trusthworthy dataset. Do you have a trusthworthy dataset on Betterridge's headlines?
— Buster Baxter (written by Matt Steinglass, 2005)
People make stuff up, especially online. If you expect others to believe you about your "very strong dataset", you should provide that very strong dataset (complete with provenance).
This author is one of the good 19th-century anthropologists (as judged by his scathing criticism of the institutions of 19th-century anthropology), but he's still a 19th-century anthropologist. The books' framing device is the idea that “the Tribes of North and South America” (including Sioux! with horses!!) lived in a “primitive condition”: while his observations are careful, and believable, his conclusions are incredible leaps of logic.
I can confidently assert that mouth-breathing does not deform the spine.
> That's what you call a "very strong dataset"‽
Yes!!!
Then, he had visited England which did not practice closed mouths and had poor health and poor teeth.
That's strong advice.
Then, he visited England which did speak English (as a primary language) and had poor health and poor teeth.
Frankly, the conclusion is irrefutable.
A working model of nature >> a large symbolic dataset >> a large personal dataset >> anecdotes.
This is a bit of #1 and #3.
Sure, #2 would be great, but #1 (mouth breathing bad) seems highly likely given both studying anatomy and modeling the physics of breathe and also given his dataset #3.
I agree that mouth breathing is probably bad. I don't agree that it causes spinal deformation or smallpox. I really don't think taping the mouth shut is a safe treatment for sleep apnea.
"During my Ethnographic labours amongst those wild people I have visited 150 Tribes, containing more than two millions of souls ; and therefore have had, in all probability, more extensive opportunities than any other man living, of examining their sanitary system ;"