Apple Watch takes countermeasures to prevent false Afib notifications

Posted:
in Apple Watch edited March 2019
It takes five incidents of detected atrial fibrillation -- Afib for short -- before an Apple Watch will actually alert its owner to the condition, Apple's VP of Health said on this weekend's publication of the Apple Heart Study, run in conjunction with Stanford Medicine.

Apple Heart Study


The threshold was created after talking with the medical community about how to prevent false positives, Dr. Sumbul Desai told Men's Health. While not necessarily a problem in itself, Afib can be a precursor to strokes, blood clots, or even heart failure.

Of the study's 419,093 participants, only 0.5 percent received an alert. That ratio fell to just 0.16 percent for those under 40, but climbed past 3 percent for people over 65.

"This tracks with physicians' understanding of Afib as being more common as you get older," Dr. Desai said. He cautioned though that even the Series 4 Watch -- which includes an electrocardiogram sensor -- is not meant to diagnose someone, but simply prompt wearers to see a physician.





Other facts gleaned from the study were that 38 percent of participants had an obese BMI, 21 percent had high blood pressure, 5 percent had diabetes, and 1 percent had experienced an earlier stroke. Many participants reportedly did follow up with their doctors if they saw an Afib warning.

The Apple Heart Study is said to be the largest such study in history, easily dwarfing the 5,209 of the Framingham experiment.

"It shows Apple has the operational capability to do large-scale virtual trials," Dr. Desai commented.

Apple is already embarking on another project with Johnson & Johnson, meant to determine whether the Watch's Afib detection can prevent strokes.

The Stanford results were shared during a presentation at the American College of Cardiology's annual conference on Saturday. Apple first announced the study during the launch of the Series 3 in 2017, but only closed enrollment in August 2018, finally ending data collection this January.

Comments

  • Reply 1 of 5
    sflocalsflocal Posts: 6,093member
    I can understand why.  I can easily imaging an army of ambulance-chasing attorneys just waiting to initiate a class-action lawsuit for the moment a false-positive happens that involves the wearer to incur hospital charges.  

    Then again, I'm sure this article will do the same thing since that same wearer will sue Apple for not promptly being notified of a possible AFIB.


  • Reply 2 of 5
    knowitallknowitall Posts: 1,648member
    Erroneous detection and not a real ECG “In a conventional 12-lead ECG, ten electrodes are placed on the patient's limbs and on the surface of the chest.” (https://en.m.wikipedia.org/wiki/Electrocardiography
    The correction is a bit laughable, maybe make a better sensor?
  • Reply 3 of 5
    knowitall said:
    Erroneous detection and not a real ECG “In a conventional 12-lead ECG, ten electrodes are placed on the patient's limbs and on the surface of the chest.” (https://en.m.wikipedia.org/wiki/Electrocardiography
    The correction is a bit laughable, maybe make a better sensor?
    If they are detecting the missed beats in a row, then it becomes more accurate as opposed to detecting and alerting every stray electrical signal the watch picks up. Anyway from what I understand it is a combination of the pulse diode and the ECG function that confirms it. Even the older Apple Watches will still pick up Afib because of the use of the pulse diode.
  • Reply 4 of 5
    StrangeDaysStrangeDays Posts: 12,877member
    knowitall said:
    Erroneous detection and not a real ECG
    As has been explained to you in another AW thread, it's a real ECG. It's a single-lead, but it's still an ECG. This is not up for debate or differing opinions, it's just a fact.
    forgot usernameGeorgeBMacpatchythepirate
  • Reply 5 of 5
    GeorgeBMacGeorgeBMac Posts: 11,421member
    knowitall said:
    Erroneous detection and not a real ECG “In a conventional 12-lead ECG, ten electrodes are placed on the patient's limbs and on the surface of the chest.” (https://en.m.wikipedia.org/wiki/Electrocardiography
    The correction is a bit laughable, maybe make a better sensor?
    Both are "real" ECGs.  The difference is not in accuracy but in the amount and variety of data provided.   A single lead can only determine heart rate and rhythm.   A 12 lead can determine far more. 
    edited March 2019
Sign In or Register to comment.