The Royal Melbourne hospital in Australia is using pulse oximeters to manage COVID positive patients in their own homes. The program was setup in March this year to manage a large number of patients remotely (cost/safety).
The patients take their own measurements with a pulse oximeter and digital thermometer (both off the shelf consumer items). The person is prompted via SMS and submit their vitals via website. A software system orchestrates all of this and alerts patients and clinicians to anyone with worrying numbers.
This keeps beds free at the hospital, but still gets the small percentage of patients back to hospital that get really sick.
It also does all the other boring monitoring and administrative work needed when you're checking up on lots of real people.
Full disclosure. I worked on this project. A version has been open sourced and if you're a hospital or other medical service you're welcome to use our work. We're publishing improvements as we go.
https://github.com/rmhcovid/txtmon
Using simple standalone devices and a low-coding platform already in use by hospitals (REDCap) the whole project was crash-built in a couple of weeks and is saving lives. That platform has many shortcomings (messier even than Excel), but there's various medical/privacy rules that make more traditional development unattractive for quickly prototyping. It's been a rewarding project to work on despite many frustrations.
Thank you for sharing this. In the program guide it says a temperature reading of > 42 degrees or an oxygen saturation of < 90 are the thresholds for a MET call. Is there a basis for these specific values or are they general approximations?
Chosen by the emergency doctors who headed the project. Sats being the more important. Those are default thresholds, but each they can be overridden per-patient.
As I understand it there is a lot of communication going between emergency teams and specialists around the world and the odd pre-publication paper. This work has come out of the RMH's emergency department. Full credit goes to Dr Martin Dutch who's idea the whole thing was. As best I know he came to the idea (tracking sats via cheap electronics to spot risky covid patients) quite early, and independently. He also built an initial prototype. I'm a developer who was brought in to build it out into a working system during the first phase of Australia's covid outbreak.
PS: If you're thinking about using the system, it's worth having a read through the redcap_design_overview.md documentation. Building something like this in a research-survey tool was not easy and there's plenty of things in the design that will be head-scratchers when seen for the first time. Now that the model is proven we'll hopefully build out a 'real' version of it at some point.
We're happy to assist where we can to see other hospitals (particularly resource-constrained ones) pick it up. The operating costs are very low (a server + sms gateway cost). The hospital has also open sourced some other covid tools developed this year.
And we have another project for operating drive-through testing clinics that will hopefully be published in the next few weeks. It has been shared through hospital channels but isn’t up on github yet.
bearsnowstorm, I just had a chance to read your background. If you or your department is interested in talking with our emergency specialists by all means say hello.
When internal body temperatures start nearing 40 degrees celcius, there starts to be a real risk of dehydration. Anything above that starts nearing deathly levels, generally a temperature of 42 celcius will start leading to permanent brain damage and death. So uh, if a body temp is already at 42 it starts to be a bit late going to the hospital :D
The figures are actually set per-patient. But that 42 as placeholder does seem odd. There are two lower thresholds that trigger before getting to that one. But I’ll flag it with the clinician regardless. Thank you.
Thanks everyone (below) who was concerned about whether the 42degrees figure was too high. A doctor on the team confirmed that figure was on purpose, but for peculiar reasons, and was kind enough to offer a few pages of explanation which I'll do my best to summarise. Any errors below are mine, not theirs.
The purpose of the program is to watch a large population of COVID patients who are not sick enough to warrant being in hospital. Miserable no doubt, but OK. The purpose of the monitoring is to identify those who are getting sicker before they have become very sick.
The goal is not to have the system calling ambulances. It can (and has) but we should be spotting degradation and reacting before it comes to that. To that end we measure three vitals (temperature, heart rate and oxygen saturation). They are not each equal predictors of the sickness we wish to spot. They are also not linear indicators of sickness.
We have three levels of response to patient vitals. There are some automated suggestions for patient comfort at the bottom end. There is the metcall at the top end. But the most important is the Clinical Review in the middle. Clinical Review connects the patient with a clinician by phone so that we can dig into their situation.
Temperature it turns out is not an especially good marker of the sickness we need to find in the group of people we are monitoring. "[with temperature] there is a U shape association with mortality, but the error bars at any point a super wide". It has some correlation, but high temperatures are not a good enough predictor of what we're trying to spot. Oxygen Saturation is good because of what COVID can cause in the lungs. That can happen without the patient being aware of it so it's critical to what we must spot.
Infection alone wont bring someone up to a temperature that high (we're monitoring people with covid, not amphetamines overdoses or desert marathons). So COVID monitoring is not dependant on spotting patient temperatures that high, and if it was that high it would not be on account of the virus. For those patients this isn't the only healthcare the patient is involved with.
So why do we bother to give the patient a thermometer? It's easy to measure. The devices are cheap and readily available. It's useful at the slightly-elevated end and helps us recommend paracetamol if/when appropriate. Tracking a patient temperature over their monitoring period feeds data back into the ongoing analysis of the virus. We ask a number of supplementary questions of the patients each day. Hopefully there will be trends.
So finally, why is there a temperature threshold at 42 degrees? More an accident of the project history. We'd built the multi-step thresholds for each of the measurements, but it turned out the highest temperature one really isn't applicable in our project. The patient hits Clinical Review well before they get up to those temperatures. It may come out completely in a newer version.
Also they added "Whilst not completely analogous, we draw your attention to the NEWS2 illness scoring system used by the National Health Service (UK)
(https://www.mdcalc.com/national-early-warning-score-news-2). The system uses various parameters to calculate a score of 'how sick they are'. Whilst many measurements have a 3+ (add 3 to the score) threshold values, the NHS does not have a temperature threshold value 'worthy' of contributing 3 points to your illness score."
Off-topic, but an inventor of the Pulse Oximeter recently passed away [1]:
"Takuo Aoyagi, a Japanese engineer whose pioneering work in the 1970s led to the modern pulse oximeter, a lifesaving device that clips on a finger and shows the level of oxygen in the blood and that has become a critical tool in the fight against the novel coronavirus, died on April 18 in Tokyo. He was 84."
I measured myself at home in Mexico City and got 93%. I got kind of worried, knowing people sick with COVID-19 can have very low saturation levels without realising it. Then I remembered I was standing at about 2200 m above sea level. Lots of unexpected weirdness happens at high altutude
I’ll be honest, a lot if the portable oxygen saturation monitors seem to report slightly out of range readings. It often picks up after a few minutes, I suspect they are calibrated poorly tbh.
94-98% is the desired range.
even holding your breath for couple seconds will tank saturation, so pretty likely high altitude will do the same until your breathing adjusts to compensate
It shouldn’t effect your oxygen saturation’s at all. Respiration is driven by raised CO2 not drops in oxygen in the healthy (COPD is the classic example of where low oxygen levels drive respiratory effort)
I know I replied to a different comment of yours, but this paper also illustrates how this point, while matching common medical wisdom, is not necessarily correct when accounting for altitude: https://www.thelancet.com/journals/langlo/article/PIIS2214-1...
(Summary: Healthy children at higher altitude breathe more rapidly.)
At the time they reported that there were mentions of it in an iOS 14 leak, but it wasn't clear if it would be a software update to existing watches (somehow able to detect it with existing heartrate hardware) or in a watch hardware revision. Since it wasn't mentioned in the watchOS news at WWDC, looks like it'll be new hardware.
Either that or the code they found in iOS 14 was related to the health app and support for 3rd party oximiters, but that's not how the 9to5 post frames it so best I can do is take their word for it.
Just bought a Huawei Honor 5 watch with pulse oximeter and heart rate functions: numbers seem correct and the device is fast: results in 5 seconds for O2 sat; almost instantaneous for HR.
Compare to Apple Watch HR function which takes at least 10 seconds to show result. Simultaneous side-by-side comparison of 2 devices: near identical numbers.
Garmin seems to be highly inaccurate at measuring oxygen and the devices take a very long time compared to a finger oximeter. I get 93% but a cheap Walmart finger oximeter shows me at 99%.
Wow they finally did it. After the hardware being there for _years_ they managed to get around to enabling the software. I already jettisoned them for Garmin after waiting so long.
Same frustrations here! Couldn't believe when I bought their smart device based on advertising that said they had the hardware to do it.... but it wasn't enabled.
Does anyone know, in general, if a watch device already has a pulse measurement LED system, whether that can also serve as a pulse ox measurement? or are they different hardware requirements?
The problem with pulse oximeters is the people who need them often can't use them. Yes I know it's not meant as a medical device.
My dad has COPD and IPF but he can't use the SPO2 meter on the Samsung phone I gave him (and Samsung disabled the SPO2 function for Canada). A Garmin VivoSmart 4 can't give a decent reading on him or me and it takes forever. Part of the problem is blood flow dad's fingers are ice cold due to his condition.
My dad uses a cheap oximeter from Walmart it works better than anything else. It's the type that clips over a finger it's fast and seems to be accurate. I'd like to get a wristband or watch that has a good SPO2 meter but nothing seems to exist.
The most painful thing I've experienced medically was an arterial blood draw to measure my o2 sat levels. (They numb your inner wrist with novocaine then go in with a giant needle to the artery deep in there. If it's that painful numb, I can't imagine what it'd be like otherwise.
And each time it was no more accurate than the pulse oximeter.
Most of the time we don’t numb patients to be honest. I’ve had it done to me, we practiced on each other in Med school. It is painful, but I suspect the Novocain you got didn’t work because it’s the artery that’s painful not the skin.
But it does give you different information than a pulse ox. For example, people can have respiratory failure from too low oxygen (captured on a pulse ox), but also from top high CO2 (not measured on pulse ox). It also gives a measure of acid-base status, which can be thrown off for a variety of reasons (usually renal or respiratory). It is used somewhat too much in my opinion but can be life saving in some situations.
An Pulse Ox and ABG are not interchangeable exams, with the ABG giving a lot more data about respiratory and metabolic derangements. A pulse ox can also be outright fooled such as in cases of carbon monoxide poisoning or essentially wrong in other poisoning. Also for a wrist ABG there’s no reason to use a giant needle. A 23g is usually fine, length is usually not an issue as the artery is actually quite superficial. For reference, an 18 gauge is the smallest that might be considered “large”. And the sizes are bigger as the number gets smaller.
Went to a party last weekend where the host used a pulse oximeter (and thermometer) on everyone. Blood O2 level is apparently a leading indicator of COVID-19. Pretty cool stuff. The inner workings of it are actually quite simple as well, pretty amazing that it works as well as it does.
It’s probably a leading indicator of COVID pneumonia but that probably still takes some days to develop during which time infected people are still contagious. Probably better to stay home or at least wear a mask...
COVID causes blood O2 to drop because it causes a viral pneumonia. Early stages of a viral pneumonia can be pre-symptomatic or entirely asymptomatic. Asymptomatic patients can still transmit COVID to others. Patients with low oxygen should seek medical attention because their pneumonia can get worse.
Your source says “Many patients with COVID-19 disease have low oxygen levels even when they are feeling well.” These patients have low oxygen because of COVID pneumonia.
Your (uncited) claim that "These patients have low oxygen because of COVID pneumonia." There are plenty of sources out there[1] that talk about a strange, again, leading indicator, known as "silent hypoxia" which has nothing to do with pneumonia. Further, there's a lot of confusion in your post: if COVID ends up giving you pneumonia, you are definitionally not asymptomatic.
You are mistaken. You can have pretty much any diagnosis and there is no “definitional” reason you must be symptomatic.
Pneumonia is a diagnosis (well, class of), not a symptom nor a sign, which are distinct terms of art. You can have pneumonia that shows up on a chest X-ray (a sign) without any symptoms.
A better example might be high blood pressure, which many people understand can be silent/asymptomatic for years but is easily measurable.
Point still stands, symptoms are noticeable and reportable by the patient. If covid gives you pneumonia you may still be asymptomatic.
> You can have pneumonia that shows up on a chest X-ray (a sign) without any symptoms.
Your post is a bit confusing, I'm not exactly sure what you're arguing. Walking pneumonia (which is what you're describing; the term of art is atypical pneumonia) is extremely rare, and it's not what we're talking about here. Non-atypical pneumonia has symptoms, ergo is by definition not asymptomatic.
What problem if you don't mind? I have a Garmin that says I'm 94ish, though it's accuracy is questionable. Haven't had a chance to test with a proper one.
This is something that is still not quite understood by doctors.
I take my kid (Who has a respiratory condition) to hospital at 90.
In my experience if you are in hospital with oxygen because of a respiratory illness, the oxygen would probably be given once you dip below 92-93 assuming they know you have a respiratory condition. Not because that’s the danger level but because they want to be able to tweak the O2 if you fall even lower. Doctors then usually instruct nurses to adjust the flow so saturation stays in the 90-95 range.
People at high altitude or in airplane cabins can be at 70% for hours without even noticing. Meanwhile someone having a dip from 90 to 80 with pneumonia might be gasping for air.
Pulse oximeters are not all that simple. The general idea is easy enough to get across and it's possible to build a fairly simple one that will work some of the time. But the top end, highly reliable ones that work well at low sats are pretty complicated things.
Interesting. You're probably not the right person to ask about this sort of thing and I should probably stick with a doctor's opinion, but you may have a better intuition than I do about the reliability of these things' readings.
Suppose I use a pulse oximeter and I get readings of 90%, 91%, 92%, and 99%, all taken ~one minute apart.
Is it most likely that I am fine and there were some issues either with the hardware or its administration? Or is it possible to get spurious high readings?
For what it's worth I had this same experience at the doctor's today for my physical.
Former research engineer (intern) at a pulse oximiter company.
This was about 18 years ago, so the details are shaky.
Yes, this is likely the probe calibrating itself. Recall that the SpO2 is based on the ratio of two measurements. From TFA, you can see that the absorption in the IR is pretty flat, so the IR LEDs don't need to be known with much accuracy - and the IR LEDs you buy are also generally in narrow tolerances because of the telecom industry using them for fiber too. But on the Red side you have a steep slope for one of the spectra, that means that you need to have a better understanding of the wavelength of red LED you're using.
But you know where you get red LEDs? Cheaply? From traffic light manufacturers. In bulk. Apparently it's much cheaper to buy LEDs in huge quantities and pay in intern to develop a measurement device to test every single LED to determine the wavelength. (And probably sell the rest back into the traffic light supply chain.)
Even after that, the red LEDs wavelength was very dependent on temperature (your finger warms it up, for example) and even the current coming in from the probe.
I never worked on the FDA approved code, but there was a bunch of pre-processing put into the sensor to determine the red LED wavelength and make calibration changes before the signal got downstream to the main instrument.
I'm sure any pulse oximiter you buy off of Amazon doesn't go through the calibration that the FDA approved ones do.
Fun fact. As an intern when you're poor, it sounds like a good idea to do calibration studies: they put pulse oximiters on every finger and your earlobes and then have a nurse administer air to you and slowly bring your SpO2 down to 70% while watching the response of the sensors. Talk about light headed! And then they take the mask off and you pop back up to about 95% (our office was a 5000 feet, not sea level) in a minute or two. Fun and you get a $50 gift certificate.
It often takes a while for the probes to stablise and get an accurate reading, so this is a very common pattern, but I haven't looked into why. I think part of it may be the machine calibrating to the flow of blood through the area of skin you have chosen. You often apply the probe and sit it there for 30-60s waiting for it to stabilise.
If you are worried, ask them to leave it on throughout your consult. You should see it sit at 99-100% for the rest of your stay.
> I should probably stick with a doctor's opinion, but you may have a better intuition.
Yes, you should trust your doctor over strangers over the internet. They should have much more context as well. It's one thing to comment on equipment, but it is another to comment on what that result means for you.
Not really. It was a very small gathering (< 10 people) mostly with people I already had spent a lot of quarantine with (doing camping, kayaking, etc.).
It's always interesting to me how much of a hivemind the internet has and how many coincidences there are. Just last night I dug out my cheapo pulse oximeter and checked everyone in my house to get a baseline.
I then googled how they work and read pretty much the same thing.
This is more a function of how many people read HN than anything else. On a population that large in a crowd of technically inclined people there's bound to be somebody who in the space of 24 hours researched how oximeters work. I read the same article a few months ago.
Does anybody have any suggestions for how to check whether a pulse oximeter actually works properly?
I bought one online from a not particularly trustworthy shop. I have never seen it show any value other than 99%. How can I figure out whether it actually does anything and doesn't just show 99% no matter what?
I purchased a relatively cheap one from Amazon a couple of months ago. Less than $50, I believe.
I tried some simple stuff when it arrived. I believe I was able to get it to swing within the 94-99% range by holding my breath and by deliberate, deep breathing, over the span of just a minute or two. It definitely was not pegged at 99%. The changes were surprisingly fast. My wife was able to get the same response.
If it matters, I'm at about 500 feet above sea level (not high-altitude at all).
1) don't know how much I trust the stuff from Amazon - see my earlier comment above about working at a pulse oximiter company and all the calibration tests we did.
2) while being an intern and sitting alone in my cube, I had dozen of test meters just sitting there and when I got bored I'd do the same thing and try to get it as low as I could by holding my breath and then snap back up. Our office is at 5000 feet so 95% is normal for healthy folks, I could get down to 85 or so and then hyperventilate back to 97% or so pretty quickly.
I bought one a few days ago (weird coincidence), from Amazon for around $35 at the advice of a friend. Skeptical also, my 97% level upped to 98-99 with deep breathing, and then I held my breath as long as I could possibly do it and it dropped to about 93-94.
That's obviously not a confirmation that it's calibrated or trustworthy, but at least it shows a relative measure that seems real in both directions of oxygen saturdation. Surely if I have a drop due to illness I should notice, even if the number isn't exact.
I have a cheap one so I am not sure if the absolute values are correct. But they change very reliably and consistently. With breathing exercises I can take the value from the usual 95-97 to 99. When I hold my breath long enough I can take the value down to 75 or less.
There is a delay of 20-30 seconds between changing my breathing to the oximeter changing its value. I suppose this is how long it takes for the blood to get around in the body.
To test it's accuracy or just to see if it ever shifts from 99% (or random fluctuations)? For the latter you you'd just use inorganic material of varying thickness/opaqueness.
I have CF, so I regularly check my o2 sat levels. I actually used mine the same day as I had a checkup; I recall it being spot on. (Homedics model from Walgreens I think)
In my experience you need to hold your breath for a while (like 2-3 minutes) for blood O2 to start decreasing. Certainly doable (and safe as long as you're lying down away from water) but might take some practice getting used to the discomfort. You can hyperventilate a bit before to make it easier (though never do this in any other breath holding scenario).
If you weren't doing this already, try holding your breath after exhaling to reduce the amount of oxygen stored in your lungs, then the wait wouldn't be as long - 2-3 minutes is a good breath hold!
In my experience this isn't that much better from a discomfort perspective, because (I would guess -- not an expert on physiology) the lower available oxygen in the lungs also leads to a more rapid increase in blood CO2. It might be a bit better depending on your particular body though, so yes it's worth trying both out.
Actually I just tried it and yeah for me at least the full-lungs version is obviously longer but takes less discomfort for my SpO2 to start to drop (like well before contractions start), whereas on exhale I get more discomfort (around 15s before contractions) before a similar drop.
SO is an internal medicine doctor. When covid was ramping up at first and we thought there would be beds in school gyms and stuff the first thing she did was get a pulse oximeter off the internet.
FWIW some Samsung phones (and probably others) have this functionality built in. I've tested mine and I consistently get readings within 1 point of my gold-standard Nonin pulse oximeter.
It has a weird 80s-90s aesthetic that I can watch for hours. He got me back into electronics when he was talking about the NA electrical system, and bashing ring circuits
Was going to send this link, but alas you already did. Alec is so good at this and I'm a proud Patron of his work. It's always really well done and thoughtful.
Does anyone know the degradation in blood oxygen levels under moderate exercise for a person who is fairly health (not athlete but slim, does regular exercise)?
I had covid, so bought one of these, and I'm now building my way back to being able to jog, after months of shortness of breath. I would like to be able to understand what level my blood oxygen %, for a normal, healthy person, would drop to when say jogging.
If I really push during my 5k run, pulse oxymeter shows around 85%. Not sure how much to trust it, as soon as I stop it climbs back up. Unclear if it's due to vibrations, or my lungs catching up.
Mine starts beeping at 93% (if I hold my breath for as long as I can it drops to around 92%). My level does rebound again after a few seconds. 85%, sounds a bit low, not totally incredulous though.
From what I understand, this is more because having that low of oxygen at rest is highly likely to be a symptom of a deeper problem, not because the 90% oxygen by itself is a huge imminent problem. Afaik you can go days/weeks at that level and be mostly fine if you are otherwise healthy. For example, this paper suggests that some people who live at high altitudes have perpetual SpO2 of <90%: https://www.thelancet.com/journals/langlo/article/PIIS2214-1....
For those of us living near sea level, temporary drops below 90% should be ok if you are otherwise healthy (up to a certain point of course), because as someone else in the thread suggested, they will rebound to >=95% as soon as you return to rest/normal breathing.
(Disclaimer: I'm not a medical expert, but have had to study breathing a fair amount. Happy to have corrections from anyone with more knowledge on this!)
It’s weird, when I have a pulse oximeter on my finger and a blood pressure cuff on and it’s rather inflated I get electrical feeling shocks up my arm. The both have to be on the same arm. No idea what it is, probably a pinched nerve.
I see that you were an anesthesiologist, now I can finally asked this question that I’ve always wanted too! When under general anesthesia why do they put you on opioids? I read that if not people would still react to the pain? Like would the scream and cry just not know they were doing that?
Opioids are a component of so-called "balanced" general anesthesia (GA).
Over my 38 years in practice (1977-2015) and some 25,000-30,000 cases, probably 95% were GA and the remainder regional/I.V. sedation.
Of the GA's, I estimate I used IV opioids in 95%. My drug of choice was fentanyl: fast and relatively short acting, easily titrated to response.
Along with fentanyl, I routinely used diazepam or Versed (I started in 1977 and Versed didn't become available until 1985), and skeletal muscle relaxants for endotracheal intubation and skeletal immobility.
Induction of GA was with sodium thiopental (1977-1989), thereafter Propofol.
Having said that, know that by the time I retired, I was in a small minority of anesthesiologists (most older like myself) who trained when balanced anesthesia was standard.
Today an overwhelming majority of anesthesiologists use only inhalation agents for GA and do not give opioids intraop. Inhalation GA often does not require skeletal muscle relaxants (depending on the surgical site), and if a patient is "light," they will react by moving and show increases in heart rate and blood pressure.
Diff'rent strokes.
FWIW, I am one of very few anesthesiologists with extended time in practice who have never been sued.
>Like would they scream and cry, just not know they were doing that?
Awareness under anesthesia is a whole different subject of considerable importance.
It is really grating that all the visualizations are wrong. Blue is shorter wavelength than red. Infrared is longer wavelength than red. Every single image is wrong. :(
One is a red light, which has a wavelength of approximately 650 nm. The other is an infrared light, which has a wavelength of 950 nm. (Throughout our description, we will show the infrared light in light blue. In reality, infrared light is invisible to the human eye.)
The fact that those curves don't even come close to resembling sine waves seems more wrong to me, and really helps drive home the fact that the diagram was made in MS Paint with little regard for accuracy.
The blue color is being used for infrared so the wavelength is right. The author could have used a '-- red' for infrared rather than blue to be less confusing.
Back in April, I bought this unit and sent it to my wife's family in New Jersey. I spent some time googling what to get; I found several discussions on Reddit where doctors and med students agreed that the "Zacurate" units reliably matched their hospital units.
Fun fact about pulse oximeters: in California EMTs are regulated so strictly that it is illegal for them to use a pulse oximeter as it is deemed "invasive". So if anyone is experiencing a medical emergency that could successfully be delayed following the algorithm involving checking blood O2, they better hope the EMT is experienced/lucky enough to suspect low O2 or that there's a paramedic or nurse on board. It's definitely one of those laws that aren't really enforced but it's weird that in California EMTs have to break the law to do basic screening.
Citation please? Page two of LA County Health Services's "EMT Scope of Practice" document lists pulse oximetry as part of the basic diagnostic process for EMTs. This would seem curious if state regulations forbid that.
This is a very weird law. In NM, basic EMTs can use a pulse/ox, insert blind airways, inject narcan and epinephrine, administer nitroglycerine, attach an ECG monitor, and do half a dozen other things more invasive than a pulse/ox.
(for those who can't access the original site due to it being hugged to death)