JAMA study casts doubt on Apple Watch blood pressure alerts
Apple Watch hypertension alerts fail to flag nearly 60 percent of users with undiagnosed high blood pressure, and still generate false alerts in healthy people. A new JAMA study shows the feature works better in older and high-risk groups, but often provides little reassurance when silent.
Hypertension alerts were cleared by the FDA in 2025 as a tool to raise awareness of potential hypertension. It uses optical sensors to look for patterns in blood flow, then runs the data through an algorithm to decide whether to show an alert. There’s no cuff involved. The idea is to flag risk, not diagnose anything.
So how well does it work in the real world? That’s what researchers from the University of Utah and the University of Pennsylvania wanted to find out.
It misses a lot of people
Using publicly available accuracy data from Apple and national health survey data from the U.S., the researchers simulated how this alert system would perform at scale.
What they found was pretty straightforward. If someone actually has undiagnosed hypertension, there’s a good chance the Apple Watch won’t pick it up. In fact, about 59 percent of those people wouldn’t get an alert. On the other side, around 8 percent of users who don’t have hypertension would still get flagged anyway.
So the feature isn’t very accurate if you think of it like a medical test. That’s because it isn’t one. It’s not built to diagnose or replace a cuff. It’s built to nudge you into paying attention.
The researchers focused on how meaningful the alerts actually are, depending on who you are. And that’s where things get more interesting.
Age and risk level change what the alert means
For younger users under 30, the chance of having undiagnosed hypertension is low to begin with. So when the watch sends an alert, the chance you actually have high blood pressure jumps from 14 percent to 47 percent. That’s a decent signal, but still leaves a lot of uncertainty. If there’s no alert, your risk drops slightly to 10 percent.
Older adults, on the other hand, have a higher baseline risk. For people over 60, the probability of having hypertension without knowing it is already 45 percent. If they get an alert, that jumps to 81 percent. But even if they don’t get an alert, their risk is still sitting at 34 percent.
That’s the point the researchers keep coming back to. An alert might help confirm risk, but the lack of an alert is not all that reassuring for older or higher-risk users.
Group | Base prevalence (NHANES) | Risk with alert | Risk with no alert |
|---|---|---|---|
Adults under 30 | 14% | 47% | 10% |
Adults 60 and over | 45% | 81% | 34% |
Non-Hispanic Black adults | 36% | 75% | 26% |
Hispanic adults | 24% | 63% | 17% |
The analysis also broke things down by race and ethnicity.
For non-Hispanic Black adults, getting an alert bumps the chance of having hypertension from 36 percent to 75 percent. Without an alert, it drops to 26 percent. For Hispanic adults, the same alert moves the probability from 24 percent to 63 percent, and silence brings it down to 17 percent.
Should you trust the alert?
Not entirely. It can be a helpful early warning, especially for people who wouldn’t otherwise think to check their blood pressure. But it doesn’t offer the kind of accuracy you get from a cuff-based device.
And that’s where this feature could backfire a bit. If someone sees the alert and follows up, that’s good. But if someone assumes they’re fine just because the watch is quiet, they could miss a diagnosis that needs attention. That’s a real concern for older users and people in higher-risk categories.
Current guidelines still recommend cuff-based screening every few years for adults under 40, and yearly after that. Apple’s alerts aren’t changing that advice.
Source: Jama Network
Subscribe to our monthly newsletter! Check out our YouTube channel.
And of course, you can follow Gadgets & Wearables on Google News and add us as your preferred source to get our expert news, reviews, and opinion in your feeds.
