Is the Apple Watch a good screening tool for hypertension?

It's better than nothing, and that's probably useful for many people who don't regularly check their blood pressure.

Dec 22, 2025

"As the year comes to a close, I would like to thank you, my readers, for reading my posts and sharing them around. Together, we can improve clinical practice one idea at a time. While anesthesia is my primary practice, I spend a lot of time learning and thinking about preventive medicine. To that end, I am starting another Substack called My Morning Soapbox where I aim to share preventive medicine and healthcare topics unrelated to anesthesia. I would love it if you would check it out. I’ve included my latest post below."

- Tyler Jones, MD

Apple’s new Hypertension Notifications are available for those using the Apple Watch Series 9 or later or the Apple Watch Ultra 2 or later. The Apple Watch uses photoplethysmography (PPG) in combination with a machine-learning algorithm to identify users who may have hypertension. Importantly, it is not designed or approved to diagnose hypertension. By notifying users who may have hypertension, the idea is that more people will become aware of their condition and seek treatment.

Is the Apple Watch good at detecting users with hypertension?

The overall sensitivity is 41.2%. As screening tests go, this is poor. When looking at stage 1 hypertension (BP ≥ 130/80), the sensitivity drops to only 29.6%! When looking at stage 2 hypertension (which, according to Apple’s documentation, is defined as BP ≥ 135/85, not the AHA definition of BP ≥ 140/90), the sensitivity improves to 53.7%. The overall specificity is 92.3% (increasing to 95.3% among those with normal blood pressure).1

This means that if a user receives a notification of possible hypertension, they are fairly likely to have hypertension. However, most users with hypertension will not have it detected by the Apple Watch.

Is it a good screening test by the strict definition of screening? Absolutely not. Home blood pressure monitoring is far superior. However, for people who do not regularly have their blood pressure checked, this feature could help improve overall detection. Given that many people do not regularly seek medical care, I view this as a great tool to improve people’s health.

Notably, individuals aged < 60 years and those with BMI < 30 had significantly lower sensitivity in the subgroup analysis and significantly higher specificity. Asian participants had a significantly lower unadjusted sensitivity, however, after covariate adjustment, the results were not statistically significant (albeit with wide error bars).