On December 6, 2018, Apple released the ECG app and irregular heart rhythm notification for the Apple Watch Series 4 through a watchOS update. Both features had received De Novo classification from the FDA, the pathway for a genuinely new kind of low-risk device, which let Apple offer them over the counter. The ECG app lets a user take a single-lead electrocardiogram from the wrist in about thirty seconds and classifies the rhythm as atrial fibrillation, sinus rhythm, or inconclusive; the background notification periodically checks for an irregular rhythm suggestive of AFib without the user doing anything.
Apple reported the ECG app classified AFib with 98.3 percent sensitivity and sinus rhythm with 99.6 percent specificity in its validation. The accompanying Apple Heart Study, with over 400,000 participants, was the largest atrial-fibrillation screening study ever run; among people who got an irregular-rhythm notification and then wore an ECG patch, 80 percent showed AFib on the patch. Apple was careful to note the limits: the features do not detect heart attacks, strokes, blood clots, or most other cardiac conditions.
The significance was less the algorithm than the scale and the regulatory framing. Atrial fibrillation is common, often silent, and a major stroke risk, and putting a cleared screening tool on tens of millions of wrists changed who gets flagged and when. It also raised the now-familiar questions of consumer health AI: false positives, worried-well visits, and the burden a screening signal places on the health system downstream.
Why business readers should care: this is the template for consumer health AI - a sensor plus a model plus FDA clearance reaching a mass market - and it shows that scale itself becomes a clinical and policy issue once a screening tool is in everyone’s pocket.