We have no real data, but I want to speculate too. How many healthy people who get alarmed at their Apple Watch EKG results who seek a second opinion from their doctor, wasting a little time and money, does it take to equal the cost of one person who DOES have a problem and doesn’t go in for a checkup and later has a really expensive problem? When they could have caught it early with the Apple Watch EKG?
This is how I would use the Series 4 watch.
If I receive a notification of a possible AFib, I would immediately launch the EKG app and take a few consecutive EKG samples. If the EKG reports AFib, I would PANIC!
Then, I would probably drive myself (or be driven) to an emergency room. I would not call an ambulance because the price of an ambulance ride is insane. You may end up paying thousands of dollars just for an ambulance trip. Believe me, it happened to me even though I had insurance.
At the emergency room, I would be at the mercy of their triage team, depending on what their views are of this new feature in the Apple Watch. Perhaps, in some emergency rooms, I would be put in the front of the line and immediately taken to the back for an EKG. It's also very possible that in other emergency rooms, I would be scoffed at for coming with this silly reason and placed at the end of line.
If the latter happens, I would probably be mad and worried for hours while waiting in the reception room for my turn. Then, I would probably wait in one of the rooms in the back for a few more hours until they finally come to do my EKG. Then, about three or four hours later (totaling 7-8 hours of waiting), a doctor (or most likely a PA) would show up and tell me that there may be a little deviation from the norm in my EKG and that I need to see a cardiologist. Two thousand dollars later, I will come home in the middle of the night, worried, upset, etc.
The following day, I would call a cardiologist to make an appointment, and would go through the regular cardio checkup.
I'm a little ambivalent about these new features. I'm sure there will be people whose lives it will save, but I'm afraid there will be people who will have a heart attack from a self-diagnosis. So for now, I will stick with my Series 3 watch.
This. This is the point that AI seems to be intentionally missing. Arrhythmia becomes more common as we age, and a doctor practicing defensive medicine or a wearer insisting there’s a problem out of ignorance and based solely on a single lead EKG creates unnecessary tests, some of which can be invasive, costly, and carrying risks of their own.
Unfortunately, this is the "journalistic" world we live in today. Every website wants clicks and they get those clicks by turning everything into horror (much like local TV news does with weather). If that means misleading editing of interviews, so be it.
When "The Verge" publishes an article stating that people who have home blood pressure gauges also might have false positives and seek care when they don't need it, then I'll take them a little more seriously. There's nothing wrong with people monitoring their own health. It doesn't mean they rush to a doctor (if we have a problem in this country, it's that they don't). It means they have a data point to tell their doctor the next time they go.
It’s acually likely that the person will have lots of data points to share, rather than just one. While it’s just a single-lead ECG, the fact that it’s strapped to your wrist all the time means you’ll probably have lots of readings taken over time that you can share with your doctor, which is a benefit over a single reading taken just when the patient isn’t feeling well. There may be a lot of baseline information available that can help the doctor contextualize a concerning reading that prompts the visit.
You are not going to have a lot of EKG readings (samples) taken unless you repeatedly place your finger on the digital crown over the course of multiple days. The Apple Watch Series 4 doesn't take EKG samples by itself like it takes heart-rate samples and monitors for AFib.
Boy howdy, let me spell it out for you. If you have one of these things, you’re probably going to play with it and take some readings. Yes. By opening the ECG app and placing your finger on the crown. That is indeed how you do it. Some people will even go beyond just mindlessly playing with it and will think, “hey, I should take the occasional reading for future reference.” Either way, you end up with data you wouldn’t have if you had to go to a medical facility for no particular reason to get it done.
Smoke detectors are a bad idea because they could lead to false positives that would overwhelm the fire departments. In a worse case they might even cause harm as people needlessly run around looking for a fire.
This is how I would use the Series 4 watch.
If I receive a notification of a possible AFib, I would immediately launch the EKG app and take a few consecutive EKG samples. If the EKG reports AFib, I would PANIC!
Then, I would probably drive myself (or be driven) to an emergency room. I would not call an ambulance because the price of an ambulance ride is insane. You may end up paying thousands of dollars just for an ambulance trip. Believe me, it happened to me even though I had insurance.
At the emergency room, I would be at the mercy of their triage team, depending on what their views are of this new feature in the Apple Watch. Perhaps, in some emergency rooms, I would be put in the front of the line and immediately taken to the back for an EKG. It's also very possible that in other emergency rooms, I would be scoffed at for coming with this silly reason and placed at the end of line.
If the latter happens, I would probably be mad and worried for hours while waiting in the reception room for my turn. Then, I would probably wait in one of the rooms in the back for a few more hours until they finally come to do my EKG. Then, about three or four hours later (totaling 7-8 hours of waiting), a doctor (or most likely a PA) would show up and tell me that there may be a little deviation from the norm in my EKG and that I need to see a cardiologist. Two thousand dollars later, I will come home in the middle of the night, worried, upset, etc.
The following day, I would call a cardiologist to make an appointment, and would go through the regular cardio checkup.
I'm a little ambivalent about these new features. I'm sure there will be people whose lives it will save, but I'm afraid there will be people who will have a heart attack from a self-diagnosis. So for now, I will stick with my Series 3 watch.
This. This is the point that AI seems to be intentionally missing. Arrhythmia becomes more common as we age, and a doctor practicing defensive medicine or a wearer insisting there’s a problem out of ignorance and based solely on a single lead EKG creates unnecessary tests, some of which can be invasive, costly, and carrying risks of their own.
And if this was actually happening with any frequency, we’d be hearing the horror stories already. Oddly, we’re not. We’re just hearing predictive conjecture of horror stories.
Given that atrial fibrillation is something that can go undetected and untreated until it results in blood clots causing a stroke, chances are that in an overall cost-benefit analysis, this device will be a net positive, and by a significant margin. We’ve already seen that there haven’t been rampant cases of people panicking over false alarms and spending hours in the ER getting unnecessary tests.
Over time, however, it’s likely that this device will result in a good number of people with AFib getting checked out by their doctor earlier, and as a result getting proper treatment and managing not to have strokes at all.
This is how I would use the Series 4 watch.
If I receive a notification of a possible AFib, I would immediately launch the EKG app and take a few consecutive EKG samples. If the EKG reports AFib, I would PANIC!
Then, I would probably drive myself (or be driven) to an emergency room. I would not call an ambulance because the price of an ambulance ride is insane. You may end up paying thousands of dollars just for an ambulance trip. Believe me, it happened to me even though I had insurance.
At the emergency room, I would be at the mercy of their triage team, depending on what their views are of this new feature in the Apple Watch. Perhaps, in some emergency rooms, I would be put in the front of the line and immediately taken to the back for an EKG. It's also very possible that in other emergency rooms, I would be scoffed at for coming with this silly reason and placed at the end of line.
If the latter happens, I would probably be mad and worried for hours while waiting in the reception room for my turn. Then, I would probably wait in one of the rooms in the back for a few more hours until they finally come to do my EKG. Then, about three or four hours later (totaling 7-8 hours of waiting), a doctor (or most likely a PA) would show up and tell me that there may be a little deviation from the norm in my EKG and that I need to see a cardiologist. Two thousand dollars later, I will come home in the middle of the night, worried, upset, etc.
The following day, I would call a cardiologist to make an appointment, and would go through the regular cardio checkup.
I'm a little ambivalent about these new features. I'm sure there will be people whose lives it will save, but I'm afraid there will be people who will have a heart attack from a self-diagnosis. So for now, I will stick with my Series 3 watch.
You need to find a doctor or medical plan that has call-in service for at minimum a nurse. This would be the first step I’d take. Most doctor’s offices and hospitals also have websites with secure email that usually are monitored by professionals who would initiate a return call if they felt the results were an issue. Use all 21st century methods instead of racing to ER.
This is how I would use the Series 4 watch.
If I receive a notification of a possible AFib, I would immediately launch the EKG app and take a few consecutive EKG samples. If the EKG reports AFib, I would PANIC!
Then, I would probably drive myself (or be driven) to an emergency room. I would not call an ambulance because the price of an ambulance ride is insane. You may end up paying thousands of dollars just for an ambulance trip. Believe me, it happened to me even though I had insurance.
At the emergency room, I would be at the mercy of their triage team, depending on what their views are of this new feature in the Apple Watch. Perhaps, in some emergency rooms, I would be put in the front of the line and immediately taken to the back for an EKG. It's also very possible that in other emergency rooms, I would be scoffed at for coming with this silly reason and placed at the end of line.
If the latter happens, I would probably be mad and worried for hours while waiting in the reception room for my turn. Then, I would probably wait in one of the rooms in the back for a few more hours until they finally come to do my EKG. Then, about three or four hours later (totaling 7-8 hours of waiting), a doctor (or most likely a PA) would show up and tell me that there may be a little deviation from the norm in my EKG and that I need to see a cardiologist. Two thousand dollars later, I will come home in the middle of the night, worried, upset, etc.
The following day, I would call a cardiologist to make an appointment, and would go through the regular cardio checkup.
I'm a little ambivalent about these new features. I'm sure there will be people whose lives it will save, but I'm afraid there will be people who will have a heart attack from a self-diagnosis. So for now, I will stick with my Series 3 watch.
This. This is the point that AI seems to be intentionally missing. Arrhythmia becomes more common as we age, and a doctor practicing defensive medicine or a wearer insisting there’s a problem out of ignorance and based solely on a single lead EKG creates unnecessary tests, some of which can be invasive, costly, and carrying risks of their own.
I'm sorry, but your article is incorrect. The real problem is that false positive findings will overwhelm true positive findings in the generally low-risk population that wears the Apple Watch. Please read this tweetorial by Dr. Murthy to understand the rate and risk of a false positive:
The take home message is that for every person diagnosed with atrial fibrillation, 100 will be misdiagnosed with atrial fibrillation. You may or may not find that telling someone they may have a heart rhythm disturbance is a big deal, but, as a cardiologist, I can verify that the threat of having a "heart problem" makes most people very anxious.
I find this comment particularly misinformed, "What body count will it take from people ignoring the Watch's recommendation for it to be an okay technology? In what world is not checking out something you're worried about better than getting something checked out that's minor?" Atrial fibrillation is not minor; in fact by some estimates, it causes about 15% of all strokes in the United States (https://www.stroke.org/understand-stroke/preventing-a-stroke/afib-stroke-connection/). In addition, it can be associated with congestive heart failure.
The diagnosis of atrial fibrillation commonly leads to the use of blood thinners and medications to keep the heart rate slow. The former carry a major bleeding risk of 3-5% annually. It is not a small thing. The big problem is that you don't need persistent atrial fibrillation to require blood thinning, you need it documented once and the right set of risk factors. If a 60 year old man with high blood pressure and diabetes came to my office and had an Apple Watch atrial fibrillation reading, I would feel compelled to implant a 30 day monitor to feel comfortable about not anticoagulating him for his risk of a stroke with atrial fibrillation would be 3-4% per year. It isn't a simple diagnostic yes/no.
Finally, medical screening tests are commonly not beneficial. Please see the United States Preventive Services Task Force webpage for a list. As an example, we don't do chest x-ray screens in smokers because the false positive findings lead to more procedures rather than any reduction in cancer or death. Screening needs to be reserved for people at risk for an event, otherwise the anguish this will cause without any reason will be significant.
This is how I would use the Series 4 watch.
If I receive a notification of a possible AFib, I would immediately launch the EKG app and take a few consecutive EKG samples. If the EKG reports AFib, I would PANIC!
Then, I would probably drive myself (or be driven) to an emergency room. I would not call an ambulance because the price of an ambulance ride is insane. You may end up paying thousands of dollars just for an ambulance trip. Believe me, it happened to me even though I had insurance.
At the emergency room, I would be at the mercy of their triage team, depending on what their views are of this new feature in the Apple Watch. Perhaps, in some emergency rooms, I would be put in the front of the line and immediately taken to the back for an EKG. It's also very possible that in other emergency rooms, I would be scoffed at for coming with this silly reason and placed at the end of line.
If the latter happens, I would probably be mad and worried for hours while waiting in the reception room for my turn. Then, I would probably wait in one of the rooms in the back for a few more hours until they finally come to do my EKG. Then, about three or four hours later (totaling 7-8 hours of waiting), a doctor (or most likely a PA) would show up and tell me that there may be a little deviation from the norm in my EKG and that I need to see a cardiologist. Two thousand dollars later, I will come home in the middle of the night, worried, upset, etc.
The following day, I would call a cardiologist to make an appointment, and would go through the regular cardio checkup.
I'm a little ambivalent about these new features. I'm sure there will be people whose lives it will save, but I'm afraid there will be people who will have a heart attack from a self-diagnosis. So for now, I will stick with my Series 3 watch.
Back in the early 1990s my grandmother-in-law would occasionally call my father-in-law and tell him to expect the ambulance at her house (she lived next door to him). When the ambulance arrived she would have vague descriptions of not-feeling-well symptoms but insist that she needed to go to the hospital, despite what the EMTs would tell her. No joke, my FIL would show up to be supportive and see what was wrong, etc, but also found she had already packed a suitcase for her "trip". GIL would instruct FIL to bring the suitcase to her room after she was settled in. After a few days she would be discharged with nothing apparently wrong with her. That probably happened 5 or 6 times over the years.
According to a former volunteer-EMT friend of mine that kind of behavior wasn't particularly uncommon. All of that with no Apple Watch in sight!
Also, it seems odd you would "PANIC!" and go to the ER when, at the very least, after reading this article you should have a general idea of how the Apple Watch EKG function works, what to expect and how to react to the result.
The take home message is that for every person diagnosed with atrial fibrillation, 100 will be misdiagnosed with atrial fibrillation. You may or may not find that telling someone they may have a heart rhythm disturbance is a big deal, but, as a cardiologist, I can verify that the threat of having a "heart problem" makes most people very anxious.
I'm curious, is that 1 person diagnosed to 100 misdiagnosed only from "smart watches" or is that in general?
Indeed, if the watch were going to generate an overwhelming rash of doctor visits based on false positives, we’d have already heard about it. We haven’t.
Next, since the Apple Watch is not marketed primarily as a fitness device, I would assert that there are plenty of lard-butts wearing them, not just ultra marathoners. There are also a lot of older people who can use an iPhone, and therefore are also just as likely to get the Apple Watch as well. A lot of people over 50 are tech savvy enough for an Apple Watch.
Finally, if anyone has a doctor who would go from an Apple Watch ECG reading straight to treatment without any other tests or dramatically obvious symptoms, well, that doctor was going to end up killing them anyway.
Speaking as a 71 year old male, I can assure you a lot of seniors can and do use both an iPhone and Apple Watch. The thought that if one is older they are not tech savvy is very wrong. If you like anecdotal evidence, when I was last in my Verizon store I watched and listened to one of their reps explain to a 20 something how to get to their iPhone's home screen. Based on that observation can I assume one in their 20's is a technical moron? Nope, but it does demonstrate the fallacy of generalities based on ones own experience.
I'm sorry, but your article is incorrect. The real problem is that false positive findings will overwhelm true positive findings in the generally low-risk population that wears the Apple Watch. Please read this tweetorial by Dr. Murthy to understand the rate and risk of a false positive:
The take home message is that for every person diagnosed with atrial fibrillation, 100 will be misdiagnosed with atrial fibrillation. You may or may not find that telling someone they may have a heart rhythm disturbance is a big deal, but, as a cardiologist, I can verify that the threat of having a "heart problem" makes most people very anxious.
I find this comment particularly misinformed, "What body count will it take from people ignoring the Watch's recommendation for it to be an okay technology? In what world is not checking out something you're worried about better than getting something checked out that's minor?" Atrial fibrillation is not minor; in fact by some estimates, it causes about 15% of all strokes in the United States (https://www.stroke.org/understand-stroke/preventing-a-stroke/afib-stroke-connection/). In addition, it can be associated with congestive heart failure.
The diagnosis of atrial fibrillation commonly leads to the use of blood thinners and medications to keep the heart rate slow. The former carry a major bleeding risk of 3-5% annually. It is not a small thing. The big problem is that you don't need persistent atrial fibrillation to require blood thinning, you need it documented once and the right set of risk factors. If a 60 year old man with high blood pressure and diabetes came to my office and had an Apple Watch atrial fibrillation reading, I would feel compelled to implant a 30 day monitor to feel comfortable about not anticoagulating him for his risk of a stroke with atrial fibrillation would be 3-4% per year. It isn't a simple diagnostic yes/no.
Finally, medical screening tests are commonly not beneficial. Please see the United States Preventive Services Task Force webpage for a list. As an example, we don't do chest x-ray screens in smokers because the false positive findings lead to more procedures rather than any reduction in cancer or death. Screening needs to be reserved for people at risk for an event, otherwise the anguish this will cause without any reason will be significant.
Okay, so, to be clear, you don't want people seeking help if they get an alert from the Apple Watch because it might be wrong. Got it. We've got a pile of "Apple Watch saves..." stories and exactly none "I tried to get help from the Apple Watch, and it was wrong" stories. Which has a potentially more dire end?
Doctor or no, your take is an idiotic one. Maybe the watch is wrong -- but maybe it isn't. Then, it's up to you, the doctor, to find out which using better technologies to figure it out. And, you even get to bill for it. I am fully aware AFIB is serious given that's what those two lines you cited literally mean. Maybe a data point that suggests that it might be happening shouldn't be ignored, like you suggest.
I'm sorry, but your article is incorrect. The real problem is that false positive findings will overwhelm true positive findings in the generally low-risk population that wears the Apple Watch. Please read this tweetorial by Dr. Murthy to understand the rate and risk of a false positive:
The take home message is that for every person diagnosed with atrial fibrillation, 100 will be misdiagnosed with atrial fibrillation. You may or may not find that telling someone they may have a heart rhythm disturbance is a big deal, but, as a cardiologist, I can verify that the threat of having a "heart problem" makes most people very anxious.
I find this comment particularly misinformed, "What body count will it take from people ignoring the Watch's recommendation for it to be an okay technology? In what world is not checking out something you're worried about better than getting something checked out that's minor?" Atrial fibrillation is not minor; in fact by some estimates, it causes about 15% of all strokes in the United States (https://www.stroke.org/understand-stroke/preventing-a-stroke/afib-stroke-connection/). In addition, it can be associated with congestive heart failure.
The diagnosis of atrial fibrillation commonly leads to the use of blood thinners and medications to keep the heart rate slow. The former carry a major bleeding risk of 3-5% annually. It is not a small thing. The big problem is that you don't need persistent atrial fibrillation to require blood thinning, you need it documented once and the right set of risk factors. If a 60 year old man with high blood pressure and diabetes came to my office and had an Apple Watch atrial fibrillation reading, I would feel compelled to implant a 30 day monitor to feel comfortable about not anticoagulating him for his risk of a stroke with atrial fibrillation would be 3-4% per year. It isn't a simple diagnostic yes/no.
Finally, medical screening tests are commonly not beneficial. Please see the United States Preventive Services Task Force webpage for a list. As an example, we don't do chest x-ray screens in smokers because the false positive findings lead to more procedures rather than any reduction in cancer or death. Screening needs to be reserved for people at risk for an event, otherwise the anguish this will cause without any reason will be significant.
I looked at the “tweetorial” you recommended, and amongst the snark, the entirely irrelevant use of reader polling, and the fact that it was disjointedly presented via Twitter in the first place (seriously?), the take home message is that I would not want you as my cardiologist. I have no idea what this guy was trying to say. If there was any information of actual value there, it was indecipherable. If this represents in any way what you find as an acceptable means of communicating complicated statistical and medical information, I’d be out the door looking for someone else who can actually help me.
If you can source some actual statistical information to make your case, by all means, try again, but stay away from twitter.
P.S. I think you’re trying to make an argument that harm from a large number of false positives is greater than the benefit from a much smaller number of true positives. I’m a reasonably intelligent person and can understand basic statistical information. The problem is this. I don’t know if you’re really a “Heart Doc” or a bored 14 year old, or a bored 14 year old prodigy who actually is a heart doc. As a result, if you’re going to assert that you have authoritative information, you’re going to have to source it from something verifiable, and Twitter ain’t it. The truth also remains that if you really are a heart doc and you think that mess you linked to is how to communicate, then you need to do some serious continuing education on proper bedside manner. Be sure to invite Dr. Murthy to go with you.
Unfortunately, this is the "journalistic" world we live in today. Every website wants clicks and they get those clicks by turning everything into horror (much like local TV news does with weather). If that means misleading editing of interviews, so be it.
When "The Verge" publishes an article stating that people who have home blood pressure gauges also might have false positives and seek care when they don't need it, then I'll take them a little more seriously. There's nothing wrong with people monitoring their own health. It doesn't mean they rush to a doctor (if we have a problem in this country, it's that they don't). It means they have a data point to tell their doctor the next time they go.
It’s acually likely that the person will have lots of data points to share, rather than just one. While it’s just a single-lead ECG, the fact that it’s strapped to your wrist all the time means you’ll probably have lots of readings taken over time that you can share with your doctor, which is a benefit over a single reading taken just when the patient isn’t feeling well. There may be a lot of baseline information available that can help the doctor contextualize a concerning reading that prompts the visit.
You are not going to have a lot of EKG readings (samples) taken unless you repeatedly place your finger on the digital crown over the course of multiple days. The Apple Watch Series 4 doesn't take EKG samples by itself like it takes heart-rate samples and monitors for AFib.
I think many people will do this just as a matter of novelty, and those who have a condition, or believe they should monitor, will intentionally form the habit. There’s a lot to be said for a tool so easy and convenient to use. Sure, it’ll be spot data rather than continuous, but that’s still a lot more than most folks have to show their doctor upon an initial visit.
This is how I would use the Series 4 watch.
If I receive a notification of a possible AFib, I would immediately launch the EKG app and take a few consecutive EKG samples. If the EKG reports AFib, I would PANIC!
Then, I would probably drive myself (or be driven) to an emergency room. I would not call an ambulance because the price of an ambulance ride is insane. You may end up paying thousands of dollars just for an ambulance trip. Believe me, it happened to me even though I had insurance.
At the emergency room, I would be at the mercy of their triage team, depending on what their views are of this new feature in the Apple Watch. Perhaps, in some emergency rooms, I would be put in the front of the line and immediately taken to the back for an EKG. It's also very possible that in other emergency rooms, I would be scoffed at for coming with this silly reason and placed at the end of line.
If the latter happens, I would probably be mad and worried for hours while waiting in the reception room for my turn. Then, I would probably wait in one of the rooms in the back for a few more hours until they finally come to do my EKG. Then, about three or four hours later (totaling 7-8 hours of waiting), a doctor (or most likely a PA) would show up and tell me that there may be a little deviation from the norm in my EKG and that I need to see a cardiologist. Two thousand dollars later, I will come home in the middle of the night, worried, upset, etc.
The following day, I would call a cardiologist to make an appointment, and would go through the regular cardio checkup.
I'm a little ambivalent about these new features. I'm sure there will be people whose lives it will save, but I'm afraid there will be people who will have a heart attack from a self-diagnosis. So for now, I will stick with my Series 3 watch.
"If our doctor rushed us into heart bypass surgery solely because of what he or she saw on our Apple Watches, it's not Apple we or our families would be suing afterwards."
Seriously? If a doctor SOLELY used this information to rush a person into HBS that's probably not the doctor that person wants to begin with. Good grief. That's like saying my tire's low so go ahead and give me a new one.
"If our doctor rushed us into heart bypass surgery solely because of what he or she saw on our Apple Watches, it's not Apple we or our families would be suing afterwards."
Seriously? If a doctor SOLELY used this information to rush a person into HBS that's probably not the doctor that person wants to begin with. Good grief. That's like saying my tire's low so go ahead and give me a new one.
That was kinda the point. A Doctor relying solely on a reading from the AW is not a good doctor.
To clear up some of the issues raised in response to my post:
1) "I'm curious, is that 1 person diagnosed to 100 misdiagnosed only from "smart watches" or is that in general?" This would be in general based on the average age of people who own an Apple 4 watch. However, it is not Apple Watch specific. If you applied a screening test to this age population, the ratio would be the same because the number of people with atrial fibrillation in this age group is so low.
2) "Okay, so, to be clear, you don't want people seeking help if they get an alert from the Apple Watch because it might be wrong." Nope, that is the sarcastic interpretation of my post. I don't want 100 out of 101 people inconvenienced and worried for nothing. I don't want 100 out of 101 people spending their hard earned money at the primary care doctor's office or emergency room getting tests to confirm or exclude the diagnosis. If you are worried about atrial fibrillation, then there are many better portable methods to diagnose it that do not require a doctor's visit. Here is one, for example: https://store.alivecor.com/products/kardiamobile. Why are they better? Because you will have to be sitting still to get a good reading and avoid noise in the measurement.
3) "The truth also remains that if you really are a heart doc and you think that mess you linked to is how to communicate, then you need to do some serious continuing education on proper bedside manner. Be sure to invite Dr. Murthy to go with you." Thanks for the ad hominem attacks. I'm sorry you didn't enjoy the conversation amongst cardiologists on twitter right after this was released. If you would prefer a discussion on Bayes theorem and the value of a sensitive test in low risk populations, you can read here: First, the very low prevalence of atrial fibrillation in people under 50 (Worldwide epidemiology of atrial fibrillation: a Global Burden of Disease 2010 Study.Circulation. 2014;129:837–847), second, a screening study that shows that in patients above age 65 years with a risk factor for stroke, atrial fibrillation is diagnosed in 1.4% of those screened (Screening to identify unknown atrial fibrillation: a systematic review.Thromb Haemost. 2013;110:213–222. doi: 10.1160/TH13-02-0165.), and third a discussion on screening for cardiovascular disease in low-risk populations: Screening Strategies for Cardiovascular Disease in Asymptomatic Adults. Prim Care. 2014 Jun; 41(2): 371–397. or one on ECG specific screening; Screening for Cardiovascular Disease Risk with Electrocardiography. JAMA. 2018;319(22):2308-2314.
4) "If our doctor rushed us into heart bypass surgery solely because of what he or she saw on our Apple Watches, it's not Apple we or our families would be suing afterwards." This is silly. Atrial fibrillation does not equal coronary artery disease (something wherein blockages may require bypass surgery). This is all about a rhythm disturbance and the risk of stroke or heart failure.
I will point out the irresponsible reporting by the Verge to not mention this is a US-only feature right now. So many international readers may not realize that fact.
That Verge article definitely has an anti-Apple slant to it. They are turning speculation into "facts". They claim that the Apple Watch will put extra strain the health care system due to false positives, but the Apple Watch is designed with so many more features that help people live healthier, thereby actually reducing strain on the healthcare system. They seem to be creating FUD against even buying an Apple Watch, because somehow it is more harmful to a system than helpful to the people.
Comments
See how stupid that sounds? It's the same logic.
Given that atrial fibrillation is something that can go undetected and untreated until it results in blood clots causing a stroke, chances are that in an overall cost-benefit analysis, this device will be a net positive, and by a significant margin. We’ve already seen that there haven’t been rampant cases of people panicking over false alarms and spending hours in the ER getting unnecessary tests.
Over time, however, it’s likely that this device will result in a good number of people with AFib getting checked out by their doctor earlier, and as a result getting proper treatment and managing not to have strokes at all.
I find this comment particularly misinformed, "What body count will it take from people ignoring the Watch's recommendation for it to be an okay technology? In what world is not checking out something you're worried about better than getting something checked out that's minor?" Atrial fibrillation is not minor; in fact by some estimates, it causes about 15% of all strokes in the United States (https://www.stroke.org/understand-stroke/preventing-a-stroke/afib-stroke-connection/). In addition, it can be associated with congestive heart failure.
The diagnosis of atrial fibrillation commonly leads to the use of blood thinners and medications to keep the heart rate slow. The former carry a major bleeding risk of 3-5% annually. It is not a small thing. The big problem is that you don't need persistent atrial fibrillation to require blood thinning, you need it documented once and the right set of risk factors. If a 60 year old man with high blood pressure and diabetes came to my office and had an Apple Watch atrial fibrillation reading, I would feel compelled to implant a 30 day monitor to feel comfortable about not anticoagulating him for his risk of a stroke with atrial fibrillation would be 3-4% per year. It isn't a simple diagnostic yes/no.
Finally, medical screening tests are commonly not beneficial. Please see the United States Preventive Services Task Force webpage for a list. As an example, we don't do chest x-ray screens in smokers because the false positive findings lead to more procedures rather than any reduction in cancer or death. Screening needs to be reserved for people at risk for an event, otherwise the anguish this will cause without any reason will be significant.
According to a former volunteer-EMT friend of mine that kind of behavior wasn't particularly uncommon. All of that with no Apple Watch in sight!
Also, it seems odd you would "PANIC!" and go to the ER when, at the very least, after reading this article you should have a general idea of how the Apple Watch EKG function works, what to expect and how to react to the result.
Doctor or no, your take is an idiotic one. Maybe the watch is wrong -- but maybe it isn't. Then, it's up to you, the doctor, to find out which using better technologies to figure it out. And, you even get to bill for it. I am fully aware AFIB is serious given that's what those two lines you cited literally mean. Maybe a data point that suggests that it might be happening shouldn't be ignored, like you suggest.
If you can source some actual statistical information to make your case, by all means, try again, but stay away from twitter.
P.S. I think you’re trying to make an argument that harm from a large number of false positives is greater than the benefit from a much smaller number of true positives. I’m a reasonably intelligent person and can understand basic statistical information. The problem is this. I don’t know if you’re really a “Heart Doc” or a bored 14 year old, or a bored 14 year old prodigy who actually is a heart doc. As a result, if you’re going to assert that you have authoritative information, you’re going to have to source it from something verifiable, and Twitter ain’t it. The truth also remains that if you really are a heart doc and you think that mess you linked to is how to communicate, then you need to do some serious continuing education on proper bedside manner. Be sure to invite Dr. Murthy to go with you.
1) "I'm curious, is that 1 person diagnosed to 100 misdiagnosed only from "smart watches" or is that in general?" This would be in general based on the average age of people who own an Apple 4 watch. However, it is not Apple Watch specific. If you applied a screening test to this age population, the ratio would be the same because the number of people with atrial fibrillation in this age group is so low.
2) "Okay, so, to be clear, you don't want people seeking help if they get an alert from the Apple Watch because it might be wrong." Nope, that is the sarcastic interpretation of my post. I don't want 100 out of 101 people inconvenienced and worried for nothing. I don't want 100 out of 101 people spending their hard earned money at the primary care doctor's office or emergency room getting tests to confirm or exclude the diagnosis. If you are worried about atrial fibrillation, then there are many better portable methods to diagnose it that do not require a doctor's visit. Here is one, for example: https://store.alivecor.com/products/kardiamobile. Why are they better? Because you will have to be sitting still to get a good reading and avoid noise in the measurement.
3) "The truth also remains that if you really are a heart doc and you think that mess you linked to is how to communicate, then you need to do some serious continuing education on proper bedside manner. Be sure to invite Dr. Murthy to go with you." Thanks for the ad hominem attacks. I'm sorry you didn't enjoy the conversation amongst cardiologists on twitter right after this was released. If you would prefer a discussion on Bayes theorem and the value of a sensitive test in low risk populations, you can read here: First, the very low prevalence of atrial fibrillation in people under 50 (Worldwide epidemiology of atrial fibrillation: a Global Burden of Disease 2010 Study.Circulation. 2014;129:837–847), second, a screening study that shows that in patients above age 65 years with a risk factor for stroke, atrial fibrillation is diagnosed in 1.4% of those screened (Screening to identify unknown atrial fibrillation: a systematic review.Thromb Haemost. 2013;110:213–222. doi: 10.1160/TH13-02-0165.), and third a discussion on screening for cardiovascular disease in low-risk populations: Screening Strategies for Cardiovascular Disease in Asymptomatic Adults. Prim Care. 2014 Jun; 41(2): 371–397. or one on ECG specific screening; Screening for Cardiovascular Disease Risk with Electrocardiography. JAMA. 2018;319(22):2308-2314.
4) "If our doctor rushed us into heart bypass surgery solely because of what he or she saw on our Apple Watches, it's not Apple we or our families would be suing afterwards." This is silly. Atrial fibrillation does not equal coronary artery disease (something wherein blockages may require bypass surgery). This is all about a rhythm disturbance and the risk of stroke or heart failure.
As we can see from Appleinsider (https://appleinsider.com/articles/18/08/30/apple-watch-other-wearables-increasingly-used-to-manage-chronic-health-conditions-study-says), the average age of an Apple Watch use is 41 years old with ~10% over 60. The incidence of atrial fibrillation in this group would be well below 1% and the 1% threshold is crossed after age 55 years (Europace (2013) 15, 486–493).
That Verge article definitely has an anti-Apple slant to it. They are turning speculation into "facts". They claim that the Apple Watch will put extra strain the health care system due to false positives, but the Apple Watch is designed with so many more features that help people live healthier, thereby actually reducing strain on the healthcare system. They seem to be creating FUD against even buying an Apple Watch, because somehow it is more harmful to a system than helpful to the people.