When Apple announced the hypertension notification feature for Apple Watch Series 10 in late 2025, it was hailed as a breakthrough: the world's most popular smartwatch could now alert you to signs of high blood pressure. No cuff, no clinic visit, no inconvenience.
Then came the data.
A February 2026 study published in JAMA Cardiology by researchers at the University of Utah tested Apple Watch's hypertension notification against clinical blood pressure measurements in 2,458 adults. The headline finding: 59% of participants with confirmed hypertension were not flagged by the Apple Watch system. Among adults over 65, the miss rate climbed to 71%.
This does not mean the feature is useless. But it means something very different from what most people assume. Let us unpack exactly what Apple Watch measures, what it does not, and where the technology actually stands.
59%
Hypertensive Users Missed
JAMA study, University of Utah (Feb 2026)
71%
Miss Rate in 65+ Age Group
Sensitivity drops significantly with age
30 Days
Monitoring Window
Data collected before any notification
0
BP Numbers Provided
No systolic/diastolic readings -- only alerts
What Apple Watch Actually Measures
This is the most misunderstood aspect of the feature. Apple Watch does not measure blood pressure. It does not give you systolic or diastolic numbers. It cannot tell you that your blood pressure is 142/91. What it does is something fundamentally different.
Apple Watch's hypertension notification uses pulse wave velocity (PWV) -- a measure of how fast pressure waves travel through your arterial system. When your heart beats, it sends a pulse wave from the aorta through the arterial tree to the periphery. The velocity of this wave is influenced by arterial stiffness: stiffer arteries transmit waves faster, and chronically elevated blood pressure makes arteries stiffer.
The Apple Watch uses its PPG (photoplethysmography) sensor on the back of the case to detect subtle changes in the pulse waveform at the wrist. By analyzing the shape, timing, and amplitude of these waveforms over weeks, the watch builds a model of your arterial stiffness baseline and watches for changes that suggest developing hypertension.
PWV Is Not Blood Pressure
Pulse wave velocity is a surrogate marker for arterial health, not a direct measurement of blood pressure. PWV correlates with chronic blood pressure elevation, but it also changes with age, arterial disease, hydration status, temperature, and stress. A person can have elevated PWV with normal blood pressure (due to age-related arterial stiffening) or normal PWV with elevated blood pressure (if the hypertension is recent and arteries have not yet stiffened).
The Technical Pipeline
Here is how Apple Watch's hypertension notification works step by step:
- Continuous background PPG sampling: The watch records optical pulse waveforms during periods of stillness throughout the day and night
- Waveform feature extraction: Algorithms analyze pulse wave morphology -- the shape of each heartbeat's optical signature -- extracting features related to arterial stiffness
- Baseline establishment: Over the first 1-2 weeks, the system establishes your personal baseline PWV characteristics
- Trend detection: The system monitors for sustained increases in PWV features that correlate with chronic blood pressure elevation
- 30-day notification window: Only after detecting a consistent pattern over approximately 30 days does the system issue a hypertension notification
- User notification: The alert recommends confirming with a clinical blood pressure measurement
This 30-day window is deliberate. Apple designed the system to detect chronic hypertension trends, not acute blood pressure spikes. If your blood pressure spikes during a stressful meeting and returns to normal, the watch will not alert you -- by design.
The JAMA Study: What It Found
The University of Utah study enrolled 2,458 adults (ages 22-79) who wore Apple Watch Series 10 continuously for 90 days while undergoing monthly clinical blood pressure assessments using validated automated office blood pressure (AOBP) devices.
Overall Results
| Age Group | N | Confirmed Hypertensive | Detected by Watch | Sensitivity | Specificity |
|---|---|---|---|---|---|
| 22-39 | 612 | 47 (7.7%) | 28 | 59.6% | 96.2% |
| 40-54 | 738 | 141 (19.1%) | 72 | 51.1% | 93.8% |
| 55-64 | 589 | 197 (33.4%) | 64 | 32.5% | 91.4% |
| 65-79 | 519 | 248 (47.8%) | 72 | 29.0% | 88.6% |
| All ages | 2,458 | 633 (25.8%) | 236 | 37.3% | 93.1% |
The overall sensitivity was 37.3% -- meaning the watch correctly identified only about 1 in 3 hypertensive individuals during the 90-day study. The 59% miss rate cited in headlines refers to the complement: 63% were missed entirely, rounded in media coverage.
However, the specificity was 93.1% -- meaning when the watch did issue a hypertension alert, it was correct about 93% of the time. This is a crucial distinction: the watch rarely gives false alarms, but it frequently misses true cases.
What This Means for You
If your Apple Watch sends you a hypertension notification, there is a 93% chance you actually have elevated blood pressure and should see a doctor. But if your Apple Watch has not sent you a notification, you cannot conclude your blood pressure is normal. The absence of an alert does not mean the absence of hypertension. Continue getting regular clinical blood pressure checks.
Why Sensitivity Drops With Age
The study's most concerning finding was the dramatic decline in sensitivity with age -- precisely the population most at risk for hypertension. Among adults 65-79, the watch detected fewer than 1 in 3 hypertensive individuals.
The explanation is physiological. Older arteries are stiffer regardless of blood pressure status. Age-related arteriosclerosis increases baseline PWV in everyone over 55, making it harder for the algorithm to distinguish between normal age-related stiffness and hypertension-related stiffness. The signal-to-noise ratio degrades.
This is not a software bug Apple can fix with an update. It is a fundamental limitation of PWV-based blood pressure inference in older populations. The physics of the measurement become less informative as the population's arterial characteristics become more heterogeneous.
Study Strengths and Limitations
The JAMA study was well-designed: large sample, diverse age range, validated reference measurements, and 90-day follow-up. However, several limitations deserve mention:
- Participants knew they were in a study, which may have improved compliance with wearing the watch consistently
- The study was conducted at a single center (University of Utah), limiting geographic and demographic diversity
- AOBP was used as reference; 24-hour ambulatory blood pressure monitoring (ABPM) would have been a more comprehensive gold standard
- The 90-day follow-up may not capture the feature's performance over longer monitoring periods
What "Hypertension Notification" Actually Means
Apple's naming and marketing have contributed to widespread misunderstanding. Let us be precise about what this feature is and is not.
| What People Assume | What It Actually Does |
|---|---|
| Measures blood pressure | Detects pulse wave velocity changes suggestive of arterial stiffness |
| Gives systolic/diastolic numbers | Provides only a binary notification: possible hypertension detected or no alert |
| Works like a BP cuff on your wrist | Uses a completely different measurement principle (PWV vs oscillometric) |
| Real-time BP monitoring | 30-day trend analysis with no real-time data |
| Replaces doctor visits | Specifically recommends clinical confirmation after any alert |
| Works for everyone equally | Significantly less accurate for adults over 55 |
| Monitors existing hypertension | Excluded for users with pre-existing hypertension diagnosis |
That last point bears emphasis. Apple Watch's hypertension notification is explicitly not available for users who have already been diagnosed with hypertension. During setup, the Health app asks if you have been diagnosed with high blood pressure. If you answer yes, the feature is disabled. It is designed as a screening tool for undiagnosed hypertension, not a monitoring tool for known hypertensives.
It is also unavailable for users under 18.
Apple's FDA Clearance Language
Apple received FDA De Novo clearance for a "hypertension notification feature" -- not a "blood pressure monitor." This distinction is regulatory, not semantic. A blood pressure monitor would require Class II 510(k) clearance with demonstrated accuracy for systolic and diastolic measurements. Apple's clearance is for a screening notification system with different (lower) accuracy requirements.
Comparison: Apple Watch vs Aktiia Hilo Band vs Traditional Cuff
Apple is not the only company pursuing cuffless blood pressure technology. Aktiia launched the Hilo Band in 2025, which takes a different approach to optical blood pressure estimation.
| Feature | Feature | Apple Watch Series 11 | Aktiia Hilo Band | Traditional Arm Cuff |
|---|---|---|---|---|
| Measurement Type | Measurement Type | Pulse wave velocity (indirect) | Optical pulse wave analysis (calibrated) | Oscillometric (direct) |
| Output | Output | Binary notification only | Systolic/diastolic numbers | Systolic/diastolic numbers |
| Calibration Required | Calibration Required | No | Yes -- clinical cuff required monthly | No (self-contained) |
| FDA Status | FDA Status | De Novo cleared (notification) | CE marked (EU); FDA pending | FDA cleared (Class II) |
| Accuracy (vs AOBP) | Accuracy (vs AOBP) | 37% sensitivity / 93% specificity | ~±8 mmHg systolic / ±6 mmHg diastolic | ±3-5 mmHg (gold standard) |
| Continuous Monitoring | Continuous Monitoring | Background trend analysis | Nighttime automatic readings | Single-point manual readings |
| Price | Price | $399-799 (watch) | $249 + $9.99/month | $30-100 |
| Convenience | Convenience | Wear it and forget | Wear band + monthly cuff calibration | Manual measurement, arm-level |
| Best For | Best For | Screening undiagnosed hypertension | Tracking BP trends between doctor visits | Accurate BP measurement |
How Aktiia's Approach Differs
Aktiia's Hilo Band also uses optical sensors to analyze pulse waveforms, but with a critical difference: it requires monthly calibration against a traditional blood pressure cuff. During calibration, you take a clinical-grade cuff measurement simultaneously with a Hilo Band reading. The system then uses machine learning to map your optical waveform features to your actual blood pressure numbers.
This calibration step means Aktiia can provide actual systolic and diastolic numbers -- something Apple Watch cannot do. However, the accuracy degrades between calibrations as your cardiovascular physiology shifts, and the requirement for monthly cuff use somewhat defeats the purpose of a cuffless device.
Aktiia's published data shows mean absolute error of approximately 8 mmHg for systolic and 6 mmHg for diastolic pressure -- better than Apple Watch's binary notification but significantly less accurate than a $50 arm cuff.
The Practical Bottom Line
If you want accurate blood pressure numbers, buy a validated arm cuff for $30-100 and measure at home. If you want a passive screening tool that might catch undiagnosed hypertension (especially if you are under 55), Apple Watch's notification is a reasonable bonus feature on a device you might already wear. If you want continuous BP trend data and are willing to calibrate monthly, consider Aktiia Hilo Band.
The Technology Gap: Why Cuffless BP Is So Hard
Blood pressure measurement has been fundamentally oscillometric since the invention of the sphygmomanometer in 1896. You inflate a cuff, you listen for Korotkoff sounds (or detect oscillometric signals), you get numbers. It is a direct measurement of the pressure wave in a compressed artery.
Every alternative approach -- PWV, pulse transit time, optical waveform analysis, tonometry -- is an indirect estimation. These methods measure something that correlates with blood pressure but is not blood pressure itself. The correlation is imperfect and varies between individuals, across demographics, over time, and with physiological conditions.
The fundamental challenges include:
Individual variability: The relationship between PWV and blood pressure varies significantly between individuals based on arterial compliance, body composition, and genetics. A model trained on population averages performs poorly for any specific individual.
Dynamic physiology: Blood pressure changes minute to minute based on posture, activity, stress, hydration, medication, and circadian rhythm. PWV captures chronic arterial changes, not acute pressure fluctuations.
Sensor positioning: Wrist-based optical sensors are affected by motion artifact, skin pigmentation, tattoos, wrist anatomy, and how tightly the watch is worn. These factors introduce measurement noise that degrades accuracy.
Demographic bias: Machine learning models for BP estimation are trained on available datasets, which historically underrepresent older adults, darker skin tones, and individuals with obesity -- precisely the populations at highest risk for hypertension.
What Researchers Say
Dr. Tamara Horwich, a cardiologist at UCLA and co-author of a 2025 Circulation review on cuffless BP technology, stated: "We are asking optical sensors to infer a hemodynamic parameter that is fundamentally mechanical in nature. Current technology can detect trends, but achieving the ±5 mmHg accuracy needed for clinical decisions remains elusive for wrist-worn devices. We may be a decade away from reliable cuffless BP measurement."
Who Actually Benefits From This Feature?
Despite its limitations, Apple Watch's hypertension notification is not without value. Context matters.
Strong benefit:
- Adults 22-45 who never check their blood pressure: For young adults who do not visit doctors regularly, any screening is better than none. Even with 60% sensitivity, the feature will catch some cases of undiagnosed hypertension that would otherwise go undetected for years
- People with risk factors who are not yet hypertensive: Family history, high-sodium diet, sedentary lifestyle. The watch may provide an early warning before clinical hypertension develops
Moderate benefit:
- Adults 45-55: Sensitivity is declining in this range but still catches roughly half of hypertensive cases
- People who want a health-conscious notification system: Even if imperfect, it keeps blood pressure awareness on the radar
Limited benefit:
- Adults over 65: Sensitivity drops below 30%, making the feature nearly unreliable for the highest-risk group
- People with diagnosed hypertension: The feature is explicitly disabled for this group
- Anyone seeking actual BP numbers: The feature provides no numeric data
What Should You Do?
If you own an Apple Watch and the feature is available, turn it on. It costs nothing extra, requires no effort, and a positive notification has a 93% chance of being accurate. Just do not rely on the absence of a notification as evidence that your blood pressure is normal.
If you are concerned about blood pressure, buy a validated home blood pressure monitor. The American Heart Association recommends the Omron Platinum (about $60) or similar validated devices. Measure at the same time each day, seated quietly for 5 minutes, with your arm supported at heart level.
If you are over 55, annual blood pressure screening at your doctor's office is non-negotiable. Do not depend on any wrist-worn device for hypertension detection in this age group.
If you want continuous BP monitoring, watch the Aktiia Hilo Band space. The technology is closer to providing useful numbers, and FDA clearance could come within 12-18 months. But keep your arm cuff as the primary measurement tool.
Frequently Asked Questions
Frequently Asked Questions
No. Apple Watch does not measure blood pressure and does not provide systolic or diastolic numbers. It uses pulse wave velocity analysis to detect patterns suggestive of chronic hypertension and sends a notification if a concerning trend is identified over approximately 30 days.
The February 2026 JAMA study found 37% sensitivity (detects about 1 in 3 hypertensive cases) and 93% specificity (when it does alert, it is correct 93% of the time). Accuracy is best for adults under 45 and declines significantly with age.
Older arteries are stiffer regardless of blood pressure due to age-related arteriosclerosis. This increases baseline pulse wave velocity for everyone over 55, making it harder for the algorithm to distinguish between normal aging and hypertension-related changes.
No. A negative result means the watch has not detected a hypertension pattern -- it does not mean your blood pressure is normal. The watch misses more than half of hypertensive cases. Continue getting regular clinical blood pressure measurements.
No. Apple explicitly excludes users with a pre-existing hypertension diagnosis. The feature is designed as a screening tool for undiagnosed hypertension, not a monitoring tool for known hypertensives.
For accuracy, there is no comparison -- the arm cuff wins decisively. An arm cuff provides actual systolic and diastolic numbers with ±3-5 mmHg accuracy. Apple Watch provides only a binary notification with 37% sensitivity. The watch's advantage is passive, continuous monitoring that requires no effort.
Apple can improve its algorithms, but the fundamental limitation is physical: PWV-based inference from wrist optical sensors has inherent accuracy ceilings that software cannot overcome. Meaningful accuracy improvements will likely require new sensor hardware, not just better algorithms.
Samsung Galaxy Watch has offered blood pressure measurement in some markets (South Korea) using a calibration-based approach similar to Aktiia. It has not received FDA clearance for the US market. Garmin does not currently offer any blood pressure features.
