Most Blood Pressure Readings Are Wrong

#

Why this matters in both directions

#

High blood pressure is the largest single contributor to strokes and heart attacks worldwide, and it produces no symptoms until it produces a catastrophe. That makes measurement genuinely important — and it makes measurement error important too, in both directions.

A falsely high reading can put someone on lifelong medication they don’t need, drop their pressure too low, and produce dizziness, a fall, a head injury. A falsely reassuring reading leaves real disease untreated for years. Neither error is benign, and both are common.

Blood pressure also isn’t a fixed property. It’s a continuously moving variable, and a single number is a snapshot of a moving target under whatever conditions happened to apply.

The seven errors

#

1. Rushing. Sit quietly for 5–10 minutes before measuring. Leave about 30 minutes after caffeine, nicotine, or exercise. And empty your bladder first — a full bladder activates the sympathetic nervous system and can add 10–15 points.

2. Talking. Conversation during inflation raises sympathetic activity and can lift the reading by around 10 mmHg — enough to move 130 into 140 and change a clinical category. Treat the measurement as a silent two minutes: phone down, nobody talking to you.

3. Arm position. The midpoint of the cuff should sit at heart level with the forearm supported on a surface. The ARMS crossover randomised trial (Liu et al., 2024, JAMA Internal Medicine) tested this directly against a supported-on-desk reference: resting the arm in the lap overestimated systolic by roughly 4 mmHg, and letting it hang unsupported at the side overestimated by nearly 7 mmHg.

4. Cuff size. This is the largest single error available. The Cuff(SZ) randomised crossover trial (Ishigami et al., 2023, JAMA Internal Medicine) found that a cuff one size too small overestimated systolic pressure by nearly 5 mmHg, and a cuff two sizes too small by close to 20 mmHg. An oversized cuff underestimates. Measure your mid-upper-arm circumference once and buy the matching cuff.

5. Posture. Back supported, feet flat on the floor, legs uncrossed — crossing adds roughly 2–8 points systolic. Bare the arm rather than shoving a sleeve up, since a rolled sleeve can act as a tourniquet.

6. Trusting one number. Take two or three readings a minute apart, discard the first — it’s reliably the highest — and average the rest. Better still, measure morning and evening across several days and read the pattern rather than any single value. No clinic should diagnose hypertension from one reading unless it’s extreme.

7. Relying on the clinic alone. Roughly one in five people have white coat hypertension — elevated at the surgery, normal in life. The more dangerous mirror image is masked hypertension: normal in clinic, quietly high the rest of the time, and therefore untreated. Out-of-office readings predict strokes and heart attacks better than clinic readings, which is why home monitoring or a 24-hour ambulatory monitor is the resolution when numbers are borderline or inconsistent.

The compounding problem

#

These errors don’t occur one at a time. A realistic bad measurement — slightly small cuff, arm in the lap, legs crossed, a bit of conversation, no rest beforehand, full bladder — stacks numbers that individually look like rounding error into something that changes clinical decisions.

That’s the actual argument for caring about technique. Not perfectionism, but the recognition that a handful of small procedural details determine whether you get an accurate picture of the single largest modifiable risk factor for stroke.

Doing it properly at home

#

Buy a validated upper-arm monitor — wrist devices are less reliable, and a pharmacy is a reasonable place to get a properly validated brand. Match the cuff to your measured arm circumference.

Then run the same protocol every time: bladder empty, five minutes seated and quiet, back supported, feet flat, legs uncrossed, bare arm supported on a table with the cuff at heart level, no talking. Three readings a minute apart, discard the first, average the other two. Morning and evening, several days running, then look at the trend rather than any one value.

Checking weekly from your thirties is reasonable — cardiovascular disease frequently begins accumulating in that decade, and this is one of the very few major risk factors you can watch directly, at home, for the cost of a single device.

Summary

#

Blood pressure readings are highly sensitive to technique, and the errors are large enough to change diagnoses. A cuff two sizes too small adds nearly 20 mmHg; an unsupported arm nearly 7; a lap-rested arm about 4; a full bladder 10–15; talking around 10; crossed legs 2–8. Roughly one in five people have white coat hypertension, and the reverse — masked hypertension, normal in clinic and high in daily life — is the more dangerous version because it goes untreated. Out-of-office readings predict cardiovascular events better than clinic readings.

What to actually do:

1. Measure your mid-upper-arm circumference and buy the matching cuff size. This is the biggest error and the easiest to eliminate permanently.
2. Use a validated upper-arm monitor, not a wrist device.
3. Run the protocol every time: empty bladder, 5 minutes quiet, back supported, feet flat, legs uncrossed, bare arm at heart level, no talking.
4. Take three readings, discard the first, average the rest.
5. Measure morning and evening over several days and judge the pattern, not a single number.
6. Don’t accept a diagnosis from one clinic reading. Ask for home monitoring or a 24-hour ambulatory monitor if the number is borderline.

Sources & further reading

#

• Liu H., Zhao D., Sabit A. et al., Arm position and blood pressure readings: the ARMS crossover randomized clinical trial, JAMA Internal Medicine 184(12):1436–1442 (2024) — JAMA Internal Medicine / PubMed
• Ishigami J. et al., Effects of cuff size on the accuracy of blood pressure readings: the Cuff(SZ) randomized crossover trial, JAMA Internal Medicine (2023) — PubMed / full PDF
• Johns Hopkins Medicine, Commonly used arm positions can substantially overestimate blood pressure readings — Johns Hopkins
• There is no 1-size-fits-all to blood pressure measurement — cuff size matters, JAMA Internal Medicine editorial (2023) — PubMed
• Hypertension in obesity: impact of miscuffing in blood pressure measurement, American Journal of Medicine — AJM
• American College of Cardiology, Effects of cuff size on accuracy of blood pressure readings — ACC