Apple announced a slew of new health features for its Apple Watch on Wednesday, including an FDA-cleared electrocardiogram sensor that is designed to alert users to potential problems with their heart so they get to a doctor in time.
But cardiologists aren't yet convinced that it will be a net positive for their patients. The device could save lives — and already is — but it also comes with a set of risks.
For starters, doctors could be overwhelmed by healthy patients with Apple Watches rushing to the emergency room with inaccurate results. Or worse still, some patients might not go to the doctor if they think their Apple Watch is monitoring them when it isn't.
"It can be game-changing for heart health if it is accurate," says Patricia K. Nguyen, assistant professor of medicine in the cardiovascular division at Stanford University.
But Nguyen told CNBC she needs to know a lot more about how the technology works before making that assessment. She told me there are four key things she'd need to see to get comfortable recommending it to patients.
First, Nguyen would want to take a closer look at the data that Apple presented to the FDA to get clearance for its new sensor, as well as the alert system that informs patients if there's a potential problem. That's pretty standard for clinicians, who will often take time to review data rather than simply relying on an FDA approval alone.
Second, she'd want to see researchers conduct a head-to-head comparison of the Apple Watch's electrocardiogram compared to the current gold standard, the Zio Patch, which doctors routinely recommend to patients at a higher risk of medical conditions such as atrial fibrillations. She'd need to see that Apple's product is as accurate and sensitive (or close to it) as the current alternatives on the market.
Apple will need to prove to the medical community through studies like these that it won't deliver an abnormally high rate of false positives (people who think they have a condition when they don't) and false negatives (people who think they're fine when they're not).
Thirdly, these tests would need to be performed using a diverse set of patients. If Apple Watch only studies relatively young and healthy people, it doesn't necessarily mean that it would work well for older patients with more serious medical problems.
Finally, and most importantly, it's well known in medical circles that sensors can often work well in a lab when the patient is sitting still. But would the Apple Watch provide accurate data if the patient is moving around?
"It would need to be accurate," says Nguyen, "in all types of patients when they are sitting, moving and exercising."