Two years ago, the World Health Organisation (WHO) released an important report: approximately four out of every five people with hypertension are not adequately treated. If countries scaled up coverage, an estimated 76 million deaths could have been averted between 2023 and 2050. One of the greatest challenges with hypertension is that it often flies under the radar. Many people show no symptoms, may not visit a doctor regularly, and, even during clinical appointments, a single reading can easily miss the condition. A new technology from Apple — hypertension notifications on the Apple Watch — may help shift that reality.

Announced in September, the feature is now available in India. Affecting around 1.3 billion adults globally, hypertension is the leading modifiable risk factor for heart attack, stroke and kidney disease. It is frequently undiagnosed for years. To be clear, hypertension notifications are not the same as blood pressure readings. Instead, the Apple Watch analyses 30-day periods of heart-rate data, beginning the day the feature is set up.

“When we think about any of our health features, there are really three key things we think about. One is making sure that they are grounded in science. One is making sure that they provide actionable insights for our users. And then the last is that they have privacy at their core… We know that 1.3 billion people have hypertension, 40 per cent are undiagnosed, and this can happen for a variety of reasons,” said Dr Adam Phillips, a cardiologist at Apple, speaking to this newspaper over a video call.

How the technology works

Hypertension notifications use data from the optical heart sensor to analyse how a user’s blood vessels respond to each heartbeat. The algorithm works passively in the background, reviewing data over rolling 30-day periods. If it detects signs consistent with hypertension, the Watch will notify the user. These alerts are designed to help people make potentially lifesaving lifestyle changes or begin treatment to reduce the risk of serious long-term illnesses.

“We are really rigorous in how we develop and validate features. For hypertension, we developed it with over 100,000 participants. We then validated it in a big clinical study of over 2,000 participants… We think it will notify over a million people in the first year alone,” said Dr Phillips.

To be precise, the feature was developed using advanced machine-learning techniques and training data from multiple studies totalling more than 100,000 participants. Its performance was then validated in a dedicated clinical study involving over 2,000 participants.

Accessing the feature

Setting it up is straightforward. Hypertension notifications are available on Apple Watch Series 9 and later, and on Apple Watch Ultra 2 and later, running watchOS 26. You do not need the newest model. On your iPhone, ensure you have the latest iOS version installed, then open the Health app. Under Health Checklist, scroll to Hypertension Notifications. The feature may have appeared earlier with a message that it was not yet available in this region.

The app will ask a few screening questions, such as whether you are aged 22 or above and whether you have previously been diagnosed with hypertension. It also clearly explains the feature: “Hypertension is a condition that can cause damage to your heart, blood vessels and other organs. Rather than measuring blood pressure, Apple Watch looks at your heart data and searches for patterns associated with hypertension.”

Apple emphasises that the Watch cannot identify every instance of hypertension, and that “regular measurements with a blood pressure cuff are the most effective way to look for hypertension”. You cannot take on-demand blood pressure readings on the device.

Hoping for a ‘positive impact’

“We’re really excited about the big impact it’s going to make. As a clinician, I’m really excited about trying to put a dent in this huge problem of hypertension in the world… We know exactly how to treat it. We just don’t know who has it all the time. With this feature, we’re going to detect a lot more people and get them on the proper treatment and management pathways. We think it’ll have a really positive impact,” said Dr Phillips.

According to WHO, hypertension — defined as blood pressure of 140/90 mmHg or higher — is common but can be dangerous if not treated. Risk factors include older age, genetics, obesity, physical inactivity, high-salt diets and excessive alcohol consumption.

“Lifestyle changes like eating a healthier diet, quitting tobacco and being more active can help lower blood pressure. Some people may still need to take medicines,” WHO noted in a September document.

Blood pressure is expressed as two numbers: systolic pressure (when the heart contracts) and diastolic pressure (when the heart rests). Hypertension is diagnosed if readings on two different days show systolic pressure ≥140 mmHg and/or diastolic pressure ≥90 mmHg.

“We like to share as much as possible with users and physicians so they understand exactly what our feature is doing and how it performs. We have a clinical validation paper on our website. We also developed a blood pressure logging experience… Users will log their readings twice daily for seven days and then take that data to their physician. We worked with clinicians to create a PDF that displays the information clearly with graphs and tables to support conversation and care pathways,” Dr Phillips explained.

Deep dive into the feature

Apple has put out a detailed “hypertension validation” paper. Hypertension Notifications is powered by a machine learning–based algorithm to identify key PPG patterns that may indicate hypertension. The algorithm uses 60-second segments of PPG signals as inputs, collected approximately every two hours throughout nonoverlapping 30-day evaluation windows.

Before samples are processed, Apple Watch filters PPG data using accelerometer data to determine whether the user is sitting still, which is required for the algorithm.

The ML-based algorithm comprises three key steps.

Step 1: The PPG Feature Extractor transforms raw PPG data into generalisable characteristics that can be used to assess hypertension risk through layered models. The extraction uses a deep learning (DL) algorithm developed with a self-supervised learning method based on large-scale unlabeled data from the Apple Heart and Movement Study. Training data did not include any blood pressure or hypertension data, focusing solely on identifying PPG signal features.

Step 2: For Hypertension Risk Scoring, a set of linear machine learning models analyse distinctive features produced by the PPG Feature Extractor to score hypertension risk for each PPG segment. The dataset used to develop these models included PPG data and reference data collected from at-home blood pressure cuffs.

Step 3: In Notification Assessment, an algorithm determines whether a notification should be surfaced. First, hypertension risk scores are filtered to exclude samples collected during sleep, as the model was trained on PPG data from waking hours to reach the most users. Remaining scores are aggregated and averaged over each 30-day evaluation period, then compared against a threshold to decide if a notification should be sent.

It may be time to strap on the Apple Watch and make some lifestyle changes, even if hypertension does not currently affect you. A healthy, low-salt diet, maintaining a healthy weight, regular physical activity and quitting tobacco all help lower blood pressure. Technology cannot replace a cuff, a doctor or medical treatment, but it may ensure more people discover a condition that is far too often missed.

“I think we are really excited about how successful this feature will be on a population level, trying to combat the prevalence of hypertension, and then we’re also really excited about how effective it will be on an individual user level, with giving them alerts at the right time to actually get diagnosed and manage their hypertension.”