Apple announced last week the release of Apple Watch Series 6 (AW6) which features a “revolutionary Blood Oxygen feature”.
How accurate or useful (or revolutionary) is that blood oxygen feature?
Should those who can afford the $399 rush out and buy one?
Can the future of health really be on our wrist?
Although the skeptical cardiologist loves his Apple Watch 4 (especially the ECG feature) and thinks a home pulse oximeter can be useful during COVID-19 he hasn’t yet ordered the AW6. Let’s examine why.
What is the Accuracy of Apple’s Blood Oxygen Feature?
Apple’s marketing release tells us that their blood oxygen sensor “employs four clusters of green, red, and infrared LEDs, along with the four photodiodes on the back crystal of Apple Watch, to measure light reflected back from blood. Apple Watch then uses an advanced custom algorithm built into the Blood Oxygen app, which is designed to measure blood oxygen between 70 percent and 100 percent.”
Apple does not provide references to any studies or data that would support the accuracy of their new approach but does mention three health studies they are partnering in which will look at the AW6’s utility in asthma, heart failure, and COVID-19.
Although we don’t know anything about AW6’s accuracy we do have studies showing us that reflectance oximetry (used in AW6) is not as accurate or stable as transmittance oximetry (used in fingertip pulse oximeters.) It’s important to under this significant technological difference so please indulge me for two paragraphs.
Oximeters determine oxygen saturation (SpO2) in tissues by quantifying the concentration of hemoglobin bound to oxygen (oxyhemoglobin or HbO2) and hemoglobin (Hb) without oxygen. Pulse oximeters which are widely utilized in medical settings utilize photoplethysmography (PPG) performed on pulsatile arterial blood at two different wavelengths to obtain a ratio of HbO2 and Hb. Optoelectronic sensors composed of light-emitting diodes and photodiodes which operate at red and near-infrared wavelengths capitalize on the differing absorptivity of Hb and HbO2.
Transmission mode oximetry has been the gold standard but requires sensing at tissue locations that can be transilluminated with the light shining through the tissue such as the earlobes and fingers.
With reflective mode oximetry, the light sources and photodiodes are on the same side of the tissue and a thin measurement site is not required. Thus it can be used at measurement sites such as the forehead, chest, and most relevant for this discussion, the wrist, where a band or watch can be utilized.
The reflectance mode for oximetry, the mode utilized by Apple Watch has been shown to be inferior to the transmission mode used by fingertip pulse oximeters likely due to increased variability introduced by ambient light, the pressure applied, and positional variation which are inherent to the method.
Some smartphone apps have claimed an ability to measure SpO2 using reflectance but when two were studied they were found to be seriously inaccurate. Dr. Walter Schrading who co-authored the smartphone oximetry study has told me that he does not recommend any smartphone app as a reliable pulse oximeter.
Pending the release of any data or objective study of the AW6’s accuracy, it is difficult to recommend individuals purchase the device for the purpose of monitoring oxygen levels. Although Apple uses the adjective “revolutionary” its competitors (FitBit and Garmin) have offered various bands and watches with SpO2 capability for several years.
Should You Purchase Apple Watch 6?
As soon as Apple announced ECG capabilities on its AW4 in the fall of 2018 I signed up to get one and I love it and cannot stand to be without it. I’m an unabashed Apple fan and tech junkie.
The face of my AW4 reveals its amazing utility.
With a touch of the screen I can start a timer, check my heart rate or heart rate recovery, initiate a work-out, play music on my airpods, check the weather, read and respond to text messages or make a recording.
Hmm. The ECG icon is not on there. (But readily accessible should anyone near me suddenly pass out or should I feel my heart palpitating.)
So if you don’t have any Apple Watch or have AW3 or less, have the money and want all of the above nifty features seamlessly integrated with your iPhone plus handy ECG capabilities go for the AW6
Is it worth upgrading from AW4 or AW5 to AW 6?
Apple is really promoting the new Apple Watch (witness the feature image-THE FUTURE OF HEALTH) for wellness and health enthusiasts:
Apple Watch Series 6 delivers many notable hardware improvements, including a faster S6 System in Package (SiP) and next-generation always-on altimeter, along with its most colorful lineup yet, featuring a beautiful palette of new case finishes and bands. watchOS 7 brings Family Setup, sleep tracking, automatic handwashing detection, new workout types, and the ability to curate and share watch faces, encouraging customers to be more active, stay connected, and better manage their health in new ways.
Of those new features mentioned, the only one that I’m interested in is sleep tracking. I’ve assessed many different sleep tracking wearables, including the Oura ring, the Garmin VivoSmart (which offers nocturnal oxygen saturation BTW), and multiple iPhone apps and have found them all to be inaccurate. They primarily rely on actigraphy to determine sleep versus wake.
We don’t know any details about the AW6 sleep tracking. This shot from the Apple website suggests to me the information will be rudimentary.
Matt Walker, (Why We Sleep) believes that all wearable sleep trackers have an accuracy of about 70% in determining sleep versus wake and the stage of sleep (and he serves as scientific advisor to Oura.) He makes an interesting comment in a recent Peter Attia podcast that “stickiness” is as important as accuracy when it comes to sleep tracking. Stickiness refers to how likely we are to continue using and wearing the device during sleep.
The Oura ring is very sticky. I continued to wear it for months after I had concluded it was not helpful in tracking or optimizing my sleep. The Apple Watch due to its size and location is not something I want to wear to bed frequently and therefore is not sticky.
My Apple Watch has long been reporting to me that I have an unbelievably good VO2 max (54 ml/kg/min) (average for men aged 20-29 years is 48).) Apple does not reveal how it calculates this key exercise parameter and like the AW SpO2 and heart rate variability, the number cannot be relied on. My AW VO2 max jumped up 10 ml/kg/min for no apparent reason 2 months ago and has remained there since.
Apparently, AW6 tweaks the VO2 measurement and reporting. They need to be transparent about how it is measured (really estimated, there is no way they can measure it) before I would rely on it for anything.
Thus far, I haven’t felt the urge to upgrade to the AW6. If you are primarily excited by the blood oxygen sensing capabilities of AW6 I would recommend purchasing instead a $20 finger tip pulse oximeter. At least we have some data on these devices and they are using the more reliable transmission mode for measuring oxygen saturation.
If you are excited by the sleep tracking capabilities, consider whether you want to wear a bulky watch to bed which is likely inaccurately tellling you how much you slept.
Oxisomnometrically Yours,
-ACP
13 thoughts on “Apple Watch 6: Should You Buy It For The Revolutionary Blood Oxygen Feature?”
Interesting comments. Here’s a different perspective, having upgraded to Watch 6 from Watch 5 {passed Watch 5 to my wife and traded in the Watch 4 I previously gave her :)].
A new Apple watch comes with a new battery and, if you restore from a backup, you are quickly up and running again. New Watch 6, with its new hardware, always seems much snappier and battery life has been much better for me that what I was getting on my Watch 5 (after 1 year of use),
The Watch 6 introduced a new “Solo” Silicone band (fits older watches also). The band is sized for your specific wrists circumference, is elastic and stretchable, is comfortable, and keeps the watch sensors against your skin nicely as your wrist changes circumference during the day.
My Apple 6 watch EKG’s. with the Solo band, have been very easy to obtain whereas my much more expensive Apple 5 genuine link bracelet was either too lose or too tight and it was frustrating that getting a good EKG trace was difficult. These issues have disappeared with the Solo band. Basically slide it over wrist and you’re all set.
Maybe the Solo band is worth a try for more consistency in readings for the EKG and the Blood Oxygen Sensor. Could be a problem solver for you.
FYI. I purchased a Maximo Mighty Sat Rx Fingertip Pulse Oxiometer, after I reviewed Dr. P’s post on . Model I got was 9809. This unit claims to be a professional Medical device. iPhone Maximo Professional Health App is downloadable by anyone and works with no issues with the Rx. Results go into the Maximo app and then into the Apple Health App. This convenience was what I was hoping for and it works well.
https://www.masimo.com/products/monitors/spot-check/mightysatrx/
No exhaustive testing on my part, by any means, but the readings on the Rx vs Watch 6 are close (a difference of 0 – 3 in the displayed percent SpO2). I trust the Rx more but the watch is always on me so it was nice to see they were close. Anyway, maybe the solo band on my Watch 6 is helping here.
Jim Kirk
Thank you for these very informative articles. As a fan of Apple Watch myself, your insights on its health features are particularly helpful.
May I suggest a future topic? I want my husband to wear AW for the high / low heart rate notifications and to track CV fitness progress. However, he is a STEMI survivor (10/28/19-proximal LAD / PCI-DES) which renders the ECG feature useless, and he would be unlikely to meet the daily “Exercise Goal” due to BB / ARB-controlled heart rate. Thus your thoughts regarding the usefulness of AW (or any other wearable) for MI survivors, and those on BB, would be invaluable. (The info on the Apple website about use with BB is quite sketchy and geared to marketing the device.)
Jean,
Thanks for your comments. The topic of beta blocker effects on fitness parameters is close to my heart as I am taking nebivolol.
I’ll put that on list of great topics for the blog.
What exactly does Apple say about BB?
There are a few Apple Support discussions where this question is asked. The best they have is to recalibrate the AW. My husband tried this with little success.
https://discussions.apple.com/thread/250329143
https://discussions.apple.com/thread/251519449
Many thanks for your response! Greatly appreciated.
Jean,
I’ve looked further into this. Apple’s stock response to this issue is to copy the section of their support instructions that deals with recalibration. Unfortunately, the calibration just tells Apple what your stride length is so that if you are walking or running without GPS to indicate distance it can estimate the distance. based on looking at the support info Aple requires a intensity equivalent to a “brisk walk”. They obviously have a heart rate response or absolute number that they are using but they don’t tell you that. This number might be based on age. It should be. Clearly if you run a lot and are in good shape your heart rate response to any vigorous activity will be less. What is clearly a brisk walk by speed could result in heart rate that doesn’t exceed Apple’s “brisk walk” cut-off. Similarly those of us on beta blockers may never get credit for the moderate or high intensity exercise we engage in because our heart rates don’t reach predetermined cut-offs. I’ve seen this with an app called PAI which requires my heart to get >135 bpm to get vigorous intensity credit, a HR I very rarely go over when running.
I agree, Oura is useless for sleep tracking. It has been widely critiqued by sleep doctors. Apple has done extensive sleep studies as Apple’s Kevin Lynch pointed out. Nothing on your wrist or finger can measure REM sleep.
When it comes to SpO2, most of people don’t realize it’s moving number like you blood pressure. And you have to be absolutely still when measuring it with AW6. Just like with taking an ECG. Movement will cause errors.
Apple has a great attention to detail: in watchOS 7, there’s a setting if you use certain heart or blood pressure medication (beta blockers, amlodipin) that restrict heart rate, it will take that to account when calculating VO2Max.
By the way, according to DC Rainmaker, Fitbit Sense’s heart rate tracking is really inaccurate.
Tom,
Thanks for your comments.
If you have any links/references of sleep doctors critiquing oura send them to me.
I’ll check out the feature that you mention regarding beta blockers and Vo2.
Dr P
A couple of comments: That paper showing the flakiness of the reflectance method used a “home made” sensor — the high-tech wizards at Apple and Fitbit might well do better with the technology. But for sure, this is not going to be very reliable. During my brisk walks (I’m 73), my Apple Watch 4 will give a few minutes of spuriously high HR readings (like 175 instead of 80) unless I move it up my arm an inch or two above my bony wrist, and tighten the strap more than I like. Apparently the motion is what throws it off. The HR sensor also uses a reflectance methodology, so it is easy to see how the oxygen sensor could get messed up pretty easily. So I’m staying with the 4 for sure – I’ll wait for the next generation or two.
BTW, I wear my Watch 4 while sleeping to monitor my sleeping HR and watch for paroxysmal afib. The comfort is not a problem, and the “theater” mode keeps it from lighting up every time I move my arm. It recharges while I shower in the morning. There are some sleep monitoring apps that produce some useful information using the motion and heart rate. I use Autosleep. Mainly it shows movement and HR and guesses when you are sleeping deeply vs lightly vs not at all. It will display that information in a variety of ways, but it is still basically limited, as you point out.
What’s the evidence for the accuracy of Apple’s EKG?
Dr P: “Apple Watch has long been reporting to me that I have an unbelievably good VO2 max”
Mine as well; It was reporting my VO2max as excellent for a 67 year old (38), but dropped to high average (28) when I took 6 weeks off to recover from prostatectomy and has remained there.
But based on 5k times, my VO2max should always have been in the bottom 30-something percentile, (38 minute 5k now would be approx. 23% VO2max and bottom 34th percentile.)
When each Apple Watch health feature is released you’ve had a history of finding some other device more accurate. Then later you adopt it. Apple does the best job of merging health features into a “watch”. Eventually you’ll like the blood ? oxygen feature. I guess that’s why your the sceptical doctor.
I ordered mine.
Fitbit will soon release its new Sense device which has FDA approval for ECGs, sleep tracking and lots more. You might want to check this out also.
wondering if the AW6 ekg capability is on par with the Kardia watch band. My husband uses the Kardia watchband on his AW3, but the main limitation is that the Kardia watchband is not at all waterproof, whereas the AW3 and beyond by themselves are waterproof. We have the Kardia in stick form also, but it is very helpful to have one on his wrist most of the time…