Cuffless Blood Pressure Monitoring: How It Works

Nearly every cuffless device starts with photoplethysmography (PPG) — the green or infrared light your watch shines into your wrist. Blood absorbs that light, so as each heartbeat pushes a pulse of blood through, the reflected light dips and recovers. Plot that over time and you get a pulse waveform.

That waveform carries more than a heart rate. Its shape encodes how stiff your arteries are and how strongly they rebound, both of which relate to pressure. Two main approaches exploit it:

• Pulse wave analysis. Read features of the waveform itself — the steepness of the upstroke, the position of the reflected-wave notch, the area under the curve — and map them to pressure.
• Pulse transit time. Measure how long a pulse takes to travel between two points, for example from the heart's electrical signal to its arrival at the wrist. Pressure stiffens arteries, and stiffer arteries carry the wave faster, so travel time falls as pressure rises.

Both give a number that correlates with blood pressure. Neither measures it. That distinction is the whole story of this technology.

The relationship between waveform and pressure is different for every person. Your artery stiffness, wall thickness, wrist anatomy and skin all shift the mapping. So the device has to be told, at least once, what your real blood pressure is at a known waveform — using an actual cuff. That is calibration, and it is why no credible cuffless device ships without a cuff in the box.

The harder problem is that the relationship does not hold still. Arteries stiffen with age, change with temperature, respond to medication, hydration, exercise and time of day. The further your body drifts from the state it was calibrated in, the more the estimate degrades. Hence periodic recalibration:

• Hilo Band (Aktiia) — a four-day initial calibration with the supplied cuff, then recalibration about every 30 days.
• Samsung Galaxy Watch — cuff calibration at setup, then roughly every four weeks, and re-calibration if you switch wrists.

The practical consequence is the one people skip: an uncalibrated cuffless device is not a slightly worse monitor, it is not a monitor. If a device asks you to recalibrate and you postpone it for two months, you are reading numbers generated from a stale model of your arteries.

Validated upper-arm cuffs are expected to sit within roughly ±3–5 mmHg of reference measurements. Independent evaluations of cuffless devices typically find mean errors in the 5–10 mmHg range for systolic pressure, with wider spread in individual readings — and worse performance the further you move from the calibration point.

Ten mmHg sounds small until you place it on the blood pressure chart: it is the whole distance from comfortably normal to stage 1 hypertension. That is why hypertension societies have been consistent that cuffless devices are not yet substitutes for cuffs in diagnosis or treatment decisions, and why they keep asking for validation protocols designed specifically for this technology — the existing protocols were written for cuffs and do not test the things that break cuffless estimates, like drift over weeks or behavior during a genuine blood pressure change.

There is a real and encouraging exception at the regulatory level: the Hilo Band (from Aktiia) was cleared by the FDA in July 2025 as the first cuffless blood pressure monitor authorized for over-the-counter use. Clearance is meaningful — it means evidence was reviewed against a standard. It does not mean the device matches a cuff, and the calibration requirement is part of what was cleared.

Comparing a cuffless device to a cuff on single readings is asking it to do the one thing it is worst at. Its advantage is somewhere else entirely: it can measure when a cuff cannot.

• Night-time readings. Blood pressure normally dips 10–20% during sleep. People whose pressure does not dip carry higher cardiovascular risk, and this is invisible to any daytime measurement. A cuff that inflates while you sleep tends to wake you; an optical band does not.
• Frequency. Dozens of readings a day beats two, for spotting patterns — the effect of a stressful meeting, an afternoon slump, a change in routine.
• The reactivity problem. Being measured raises your blood pressure, which is the mechanism behind white coat hypertension. A reading you did not notice being taken avoids it.
• Adherence. The best monitor is the one you actually use. A band you forget about collects more data than a cuff in a drawer.

Read this way, cuffless and cuff are complements: the cuff establishes where you are, the band shows how you move.

• Does it come with a cuff, and does it require calibration? If not, walk away. A device that has never measured your real blood pressure cannot estimate it, whatever the listing says.
• Is there a named regulatory clearance or CE mark for blood pressure specifically? Not "for wellness," not for heart rate — for blood pressure.
• How often must you recalibrate, and how long does it take? This is the maintenance cost of the device, and it is the step that determines whether the numbers stay meaningful.
• Where does the data go? Check it exports — to Apple Health, or as a file — so your history is not locked to one manufacturer's app.
• What does it claim beyond blood pressure? A device that also promises non-invasive blood glucose from the same optical sensor is telling you something about how seriously to take the rest of its claims.