The Apple Watch can detect atrial fibrillation: so what now? | Hacker News Reader

The Apple Watch can detect atrial fibrillation: so what now?

nature.com

10 pointsvo2maxer13 comments

When you consider this came from research that was funded by Jawbone, Eight and that the author has equity in InCarda, it makes me question the authenticity of their claims. Can someone from this field corroborate what is claimed?

The review by Dr. Marcus, a cardiologist [1], refers to a study funded by Apple and published in The New England Journal of Medicine [2]. His main point is that “insufficient evidence exists to recommend atrial fibrillation (AF) screening in asymptomatic adults.” [3] There would be a significant number of alarmed users with false-positive results who would undergo costly additional evaluation and treatment where there may be of little benefit. It is troublesome, in his view, that “private industry has bypassed expert consensus to initiate screening for AF in the general population.” He adds that “this is an industry instigated inversion of the relationship between patients and physicians...” [4] Larry Husten at the Cardiobrief blog gives a good overview of some of the pitfalls [5].

[1] https://profiles.ucsf.edu/gregory.marcus

[2] https://www.nejm.org/doi/full/10.1056/NEJMoa1901183

[3] https://jamanetwork.com/journals/jama/fullarticle/2695678

[4] https://www.nature.com/articles/s41569-019-0330-y

[5] http://www.cardiobrief.org/2019/11/13/what-can-we-learn-from...

This is my humble opinion, but all doctor talk of "inversion of the relationship between patients and physicians" is complete quackery. We will see medicine change for the better over the next 10 years, but it will be a long road. First step is competition in medical education.

What you’re saying is that evidence should not drive decisions in health care but rather industry marketing. And if this what you mean, how is it quackery to suggest that more studies are needed or that users need to be given more information on devices with such a significant rate of false-positives?

You deformed his “humble opinion” into something completely different. There are other ways to expose your views, this one just looks like you have an axe to grind.

Please, explain to me how I deformed it when they label “all doctor talk” of the “inversion of the relationship between patients and physicians” as quackery without any reasoned argument to support such “humble opinion.” The position in the original submission is straight forward: The study was funded by Apple to drive sales by people who generally don’t suffer with atrial fibrillation. Consequently, this will create unnecessary alarm in the worried well, who will undergo needless testing and possibly harmful therapy. There is already plenty of superfluous medical evaluations and treatments at a huge cost. Are we supposed to encourage this when it is clearly industry driven with little evidence in how it advances healthcare. If this is my ax to grind, so be it, but as a physician, shouldn’t I be cynically happy that more business will be coming my way?

Mr Huston in the last link is complaining that there were too many subjects in the test and that the test was pretty expensive for Apple. How am I to take this seriously? It’s just another Apple hater.

In regards to your observation, Husten’s review notes that the study [1] included subjects where the incidence and prevalence of atrial fibrillation (AF) is low [2], and given the study’s design flaws, its findings turned out to be largely irrelevant except as a marketing tool with relatively negligible costs for Apple [3].

I don’t know if Mr. Husten is an Apple hater, but it would be more interesting if you challenged his actual arguments including those in the study’s accompanying opinion piece: “The uncomfortable fact is that our personal health data have considerable financial value to those who want to use them in the myriad marketplaces connected to our $3.7 trillion health economy.” [4]

[1] https://www.nejm.org/doi/full/10.1056/NEJMoa1901183

[2] https://www.ajconline.org/article/S0002-91491301288-5/fullte...

[3] http://www.cardiobrief.org/2019/11/13/what-can-we-learn-from...

[4] https://www.nejm.org/doi/full/10.1056/NEJMe1913980

I’m not sure how the fact that data has value is a comment on Apple Health studies. The health data collected in these studies is not connected on contact information, so the only thing a user can be profiled as is as a person who participates in the study. The author is quick to dismiss this data that first was worth billions as worthless though.

> Now the same sort of breathless optimism about technology is infecting medicine and healthcare.

Clearly the author made up his mind before he even started.

> In other words, nearly half a million people were required to identify a few hundred people with AF.

Or, the study identified a few hundred people with AF, using nothing but the watch they were going to wear anyway. Impressive!

I’m neither partial nor against Apple, but picked up a Series 4 watch recently with the ECG crown feature. It’s a pain because it only registers when the app is activated and the opposite arm touches the crown. I’d also consider wearing a chest strap for constant ECG and sleep apnea monitoring if such a QS device existed at a reasonable price-point. Lastly, it would be nice if there were a continuous non invasive blood pressure monitor (CNAP/cNIBP) similar to CareTaker but more compact and tasteful in appearance that could also gather data (I have tachycardia and high diastolic BP from unknown causes, in addition to obstructive sleep apnea because of a Marfan-ish CTD and I’m overweight).

What's not clear is the difference between owning a watch and getting your annual EKG with your primary doctor? Can the watch detect fibrillation not detected in your annual? I've never understood that.

The Apple Watch used a sensor to intermittently measure changes in blood flow as long as the user carried the device. The signals then generated pulse intervals over one minute which ran through an algorithm to infer the presence of an irregular ventricular rhythm. Further evaluation was conducted for the group with abnormal results. On the other hand, a 12-lead electrocardiogram measures electrical activity over a very short period of time. It’s just one quick snapshot of the heart, therefore, episodic atrial fibrillation can occur outside that quick peek.

Alas the article is behind a hard paywall and not available on Sci-Hub...

Reply on news.ycombinator.com