13. Apple Heart Study - Revisited

heading

The Apple Heart Study - Revisited

ND320 C4 L2 09 Apple Heart Study Part II

recap

Summary

Wearables allow us to collect more data under more conditions from more people. Typically we’d only measure ECGs from people that are suspected of having heart conditions. Now Apple Watch can measure and look for atrial fibrillation from its large, mostly healthy population. More research needs to be done into AF itself to really know how useful detecting it in this population actually is. For example, what does it mean when someone under 40 years old has AF. Are they actually at an increased risk of stroke? The Framingham study -- which found the association between AF and stroke -- was conducted in the 40s, 50s, and 60s when AF was diagnosed after coming to the hospital and taking an ECG. If a subject’s AF was not persistent enough to show up at that time, then it would go undiagnosed. In fact, for a lot of the stroke victims, it was unknown that they even had AF until after they came to the hospital for the stroke itself.

Although the population of Framingham study was at a 4-5X increase in the risk of stroke from AF, it is unclear if the population that the Apple Watch is discovering AF for is also at this increased risk. What does it mean to the public health system when you're alerting all these people of AF just because you can monitor it more closely than you ever could before? And how should patients and their doctors respond when they discover AF? The Apple Heart Study paper mentions these concerns explicitly:

…further studies are needed to better understand the public health implications of identifying irregular pulse in persons younger than 40 years of age.

and

…uncertainty remains about the benefits of diagnosing and treating asymptomatic atrial fibrillation…

We have to address these concerns before we can really understand the value of using the Apple Watch as a screening or diagnostic tool.