Measuring blood pressure properly
Last reviewed November 17, 2026. OwieClear editorial team. Waiting for clinician review. Not medical advice. Talk with your doctor about your own care.
Almost everything about hypertension depends on the number being right, and the number is easy to get wrong by 10, 20 or even 30 mmHg without doing anything that feels careless.
Measurement errors run mostly in one direction, upward, which means poor technique leads to treatment people do not need, and occasionally the reverse.
Related pages: High blood pressure (hypertension)
The bottom line
- An undersized cuff overestimates blood pressure substantially.
- An unsupported arm, or one below heart level, raises the reading.
- Validated upper-arm monitors are recommended over wrist and finger devices.
- A week of twice-daily home readings is more informative than a single clinic reading.
- How much of the gap between clinic and home readings is technique rather than white-coat effect.
- How accurate cuff devices are in atrial fibrillation, where they are known to be less reliable.
The story, step by step
What the evidence labels mean
Preparation that changes the number
What happens
The half hour beforehand.
What the research shows
- No caffeine, smoking or exercise for 30 minutes beforehand. Each can raise systolic pressure by around 10 mmHg.
- Empty your bladder. A full bladder adds roughly 10 mmHg.
- Sit quietly for 5 minutes first, not moving or talking, and not looking at your phone.
- Do not talk during the measurement, and do not be talked to. Conversation adds around 10 mmHg.
- Measure before taking medicines in the morning, not after, unless told otherwise, so readings are comparable.
- Sit on a chair with a back, not a stool or the edge of an examination table. Unsupported back adds several mmHg.
- Feet flat on the floor, legs uncrossed. Crossed legs add around 5 to 8 mmHg.
- Put the cuff on bare skin, not over a sleeve, and do not roll a sleeve up tightly above the cuff.
Why it matters
These add up. A reading taken immediately after rushing in, on a full bladder, talking, with crossed legs, can be 20 to 30 mmHg higher than the same person's true value.
What's still uncertain
- The exact contribution of each factor, which is measured in small studies and varies by person.
Read more
- CDC: Measure your blood pressureGovernment
- MedlinePlus: How to measure blood pressure at homeGovernment
- American Heart Association: hypertension guideline resourcesNonprofit
Next: The cuff and the arm
The two biggest errors
What happens
Where most of the inaccuracy sits.
What the research shows
- Cuff size matters more than anything else on this page. A cuff too small for the arm overestimates blood pressure, sometimes by a great deal, and undersized cuffs are extremely common in people with larger arms.
- Measure your upper arm circumference and buy the right cuff. Many monitors are sold with a standard cuff that does not fit a substantial proportion of adults, and larger cuffs can be bought separately.
- An oversized cuff underestimates, which matters less often but matters.
- The arm must be supported at heart level, resting on a table. Holding the arm up, letting it hang, or resting it on your lap all change the reading, and recent work found lap or unsupported positions overestimate substantially.
- The cuff's artery marker should sit over the brachial artery on the inside of the upper arm, snug enough that two fingers fit underneath.
- Measure both arms the first time. A consistent difference of more than 10 to 15 mmHg between arms is worth reporting, and thereafter use the higher arm.
- Wrist and finger monitors are less reliable because position relative to the heart changes the result so easily. Upper-arm devices are recommended.
- Use a device that has been independently validated. Validated-device lists are published and many monitors on sale have never been validated.
Why it matters
Cuff size and arm position are the two things most likely to be wrong and are the easiest to fix.
What's still uncertain
- How many people are being treated on the basis of undersized cuffs, which is likely substantial.
Read more
- CDC: Measure your blood pressureGovernment
- MedlinePlus: How to measure blood pressure at homeGovernment
- American Heart Association: hypertension guideline resourcesNonprofit
Next: A proper week
How home monitoring is actually done
What happens
The protocol that gives a usable answer.
What the research shows
- Measure twice daily: morning before medicines and before breakfast, and evening before dinner or bed.
- Take two readings each time, a minute apart, and record both. If they differ by more than about 5 mmHg, take a third.
- Do this for 7 consecutive days, ideally, and at minimum 3.
- Discard the first day entirely. First-day readings run high, and including them skews the average.
- Average everything that remains. That average is the number that matters, not the highest reading or the one that worried you.
- Record everything, including the readings you do not like. Selective recording is common and makes the log useless.
- Do not measure repeatedly after a high reading. Anxiety raises pressure, and chasing a number upward is a well-worn path to distress.
- Bring the monitor to appointments once a year so it can be checked against the clinic device.
Why it matters
A week of averaged readings is a better guide to risk than any single clinic measurement, which is why the protocol exists.
What's still uncertain
- The optimal number of days, where 7 is conventional and 3 performs nearly as well.
Read more
Ambulatory monitoring and special cases
What happens
What else is used, and when.
What the research shows
- Ambulatory monitoring uses a cuff worn for 24 hours that inflates automatically through the day and night. It is the reference standard for diagnosis.
- Its unique contribution is night-time pressure, which cannot be measured any other way and is the strongest single predictor of cardiovascular outcome.
- It is used to confirm a new diagnosis, to sort out white-coat and masked hypertension, to investigate apparently resistant hypertension, and when symptoms suggest pressure is dropping too low.
- It is uncomfortable and disrupts sleep, which is its main limitation.
- Automated cuffs are less accurate in atrial fibrillation, where the irregular rhythm confuses the algorithm. Several readings, or a manual measurement, may be needed.
- In pregnancy, measurement technique and thresholds differ, and blood pressure in pregnancy is monitored to a different standard entirely.
- Very large or conical arms, and arm differences from previous surgery or a dialysis fistula, all need specific arrangements rather than a best guess.
- Standing readings should be checked in older adults, anyone with diabetes, and anyone reporting dizziness, because a drop on standing changes the treatment plan.
Why it matters
If the clinic and the home monitor keep disagreeing, 24-hour monitoring is the way to settle it rather than arguing about which is right.
What's still uncertain
- How to measure reliably in atrial fibrillation, which remains an unsolved practical problem.