Apple Watch Afib data discrepancy?

Posted by nevets @nevets, Aug 4, 2025

Here's one for any techn minded folks out there...My Apple Watch is set up to continuously monitor Afib. When setting it up, I entered that I have been diagnosed with Afib, which I believe means it does not send an alert each time it detects Afib.

Here's my question: How is it that even though it indicates that throughout a 24 hr period over the past six weeks I had Afib 2% or less of the time (meaning I may have not had any Afib, but it's accuracy is not absolute), yet it also indicates that over the same six week period I had Afib 3% of the time on Saturdays and Sundays? How is that possible?

Interested in more discussions like this? Go to the Heart Rhythm Conditions Support Group.

Profile picture for gloaming @gloaming

@stryker66 I hope I don't come across as callous or dismissive....not my intent. We don't know each other, I don't know your history, and I'm no health expert...not by any means. Just taking what you have 'said' at face value, and mixing it with what I have learned (mostly for and about myself), humans tend to be biased toward information that appears to support or to confirm what they WISH TO believe....what they HOPE is true. This is widely known as 'confirmation bias'. I think, without knowing much about the two devices, that there is some danger here for you in assuming the 'better news' is correct. It may very well be that you are correct, and that the Apple watch is better, but remember that the FDA or whomever has approved BOTH devices for sale and with active apps that measure health passably accurately, enough that they are meant to inform. But different devices sit on the wrist differently due to differences in engineering and in form factor. Maybe one of their bands feels better for you when it's a bit tighter, and when it therefore gets better information, than the other. But which is it....which one....and how can you tell?

Not trying to rain on your parade. I would hope, for your sake, that the Apple IS the better of the two. But you can't be sure without some other measure that confirms it. Maybe, if you're up to the experiment and the relatively modest outlay of cash, a Kardia mobile or some other worn device would allow you to do that....confirm.

Jump to this post

@gloaming I think in the absence of conducting a reliable experiment, it is reasonable to rely on data from the internet and personal experience regarding the relative efficacy of the two devices, which is the same way many of us on this forum have become "knowledgeable" about all things Afib.

REPLY
Profile picture for nevets @nevets

@gloaming I think in the absence of conducting a reliable experiment, it is reasonable to rely on data from the internet and personal experience regarding the relative efficacy of the two devices, which is the same way many of us on this forum have become "knowledgeable" about all things Afib.

Jump to this post

@nevets My point was not that it is unreasonable, but that it might not be salutary...or wise. In fact, it might be dangerous to rely on confirmatory, or hopeful, information that is not confirmed using other sources of information, especially when your two available sources are contradictory. This is a textbook example of when a patient should seek confirmatory information to rule out a potentially harmful disorder.

REPLY

I have similar experiences with AF tech and the posts here confirm my observations. I started with Cardia, then GAlaxy 8 then Iwatch 10 and I even tried credit card Cardia. (I'm a touch nerdy) I've shared traces with my Ep and also paid QALY $15 for on line interpretations. I now pretty well use the Iwatch simply because i have Iphone. All are good at detecting sinus rhythms plus pure AF. Where they lack as these posts affirm is interpreting ectopic beats, ie PACs and PVCs. I find that if have only a couple of PACs during a 30 sec recording i get a Sinus rhythm diagnosis whereas if the PACs are more numerous then the watch returns a diagnosis of AF even though my EP and QALY dispute that. Apple has a measure called Afib burden which tends to fluctuate over time but I read that it's being generated by measuring the interval between beats from the HR algorithm which is pretty continuous in the background (unlike the ECG which requires activation and a finger on the crown) and if there's enough variation then it's considered AFIB which seems consistent with the false AF diagnoses for PACs. AS others here have written, It's notable that the AF burden tends to increase on weekends which doesn't make much sense in that I'm retired. We need a course in how to interpret the electronics. Gloaming's recommendation is where I land - the trace is a good indicator of whether I'm in AF, PACs or Sinus Rhythm, but don't trust the diagnosis..

REPLY
Please sign in or register to post a reply.