To measure blood pressure at home correctly: use a validated upper-arm cuff sized to your arm, empty your bladder, then sit quietly for five minutes — back supported, feet flat, legs uncrossed, bare upper arm resting at heart level. No caffeine, exercise or smoking for 30 minutes beforehand, and no talking during the reading. Take two readings one minute apart and average them. Repeat morning and evening for about seven days when your clinician wants a real picture. A single reading — high or low — is a snapshot, not a diagnosis.
This matters more than it sounds. In a randomized crossover trial published in JAMA Internal Medicine in 2023, using a standard-size cuff on people who actually needed an extra-large cuff pushed systolic readings up by 19.5 mm Hg on average — enough to move someone from normal to stage 2 hypertension on paper alone. A 2024 trial found that simply resting your arm in your lap instead of on a desk added about 4 mm Hg, and letting it hang at your side added 6.5 mm Hg. These are not rounding errors. They are the difference between a conversation and a prescription.
Why home readings are worth doing at all
The US Preventive Services Task Force gives blood pressure screening in adults an A recommendation and — importantly — advises confirming an elevated office reading with measurements taken outside the clinic before hypertension is diagnosed and treatment started. Ambulatory monitoring (a device worn for 24 hours) is the reference standard; home self-measurement is the practical alternative most people can actually access.
The reason is two mismatches that only out-of-office readings can catch. White-coat hypertension is a clinic reading at or above 130/80 mm Hg with home readings below it — the clinic environment itself is raising the number, and treating it would mean treating an artifact. Masked hypertension is the opposite and more dangerous: a reassuring clinic reading with genuinely elevated pressure the rest of the time. Neither is visible from a single hurried check in a waiting room where you sat down 40 seconds ago with your coat still on.
The correct method, step by step
| Step | Why it matters |
|---|---|
| Use a validated upper-arm cuff monitor | Upper-arm devices are what guidelines are built on. Wrist devices are far more sensitive to position and wrist angle; they're a fallback for specific situations, not the default. Cuffless and smartwatch BP features are not validated for diagnosis. |
| Match the cuff to your arm circumference | The single most common fixable error. A cuff that is too small falsely raises readings — by nearly 20 mm Hg systolic in people who need an extra-large cuff. A cuff that is too large falsely lowers them. Measure your mid-upper arm with a tape and check the range printed on the cuff. |
| No caffeine, exercise or smoking for 30 minutes | All three transiently raise blood pressure. A reading taken 10 minutes after coffee tells you about the coffee. |
| Empty your bladder first | A full bladder raises blood pressure measurably through sympathetic activation. |
| Sit quietly for 5 minutes before the first reading | Blood pressure falls during quiet rest. Skipping this is why clinic readings so often run high. |
| Back supported, feet flat, legs uncrossed | An unsupported back and crossed legs both push readings upward. |
| Arm supported at heart level, on a table | Arm in the lap adds roughly 4 mm Hg; arm hanging at your side adds about 6.5 mm Hg. Prop the arm on a cushion if the table is low. |
| Cuff on a bare arm, lower edge just above the elbow crease | Measuring over a sleeve — or over a sleeve rolled into a tight band above the cuff — distorts the reading. |
| Don't talk, scroll or watch anything | Talking during a measurement raises blood pressure immediately. |
| Two readings, one minute apart — then average | The second reading is usually lower than the first. Averaging is the number your clinician wants; the lone first reading is the least reliable one you'll take. |
| Same arm, same time of day | Arms can differ by several mm Hg. Check both once, then use the higher-reading arm consistently. |
How many days, and at what times?
For a one-off curiosity check, a single well-taken pair of readings is fine. For anything your clinician will act on — confirming a diagnosis, judging whether a treatment is working — the standard is a seven-day series, often called the "7-2-2" protocol: 7 consecutive days, 2 readings per session, 2 sessions per day (morning before medication and food, evening before bed). Day one is typically discarded, because first-day readings run high while people are unfamiliar and slightly anxious with the device. That leaves 24 readings; the average is what counts. Shorter runs of three to four usable days give a reasonable estimate, just a noisier one.
Write every reading down, or use a monitor that stores and exports them. Don't edit out the ones you don't like — the outliers are information too, and your clinician can see whether the pattern is stable or erratic.
What do the numbers mean?
These are the American Heart Association / American College of Cardiology categories from the 2025 hypertension guideline. Treat them as reference ranges, not a diagnosis — diagnosis belongs to a clinician who sees your full history, and treatment decisions in the 2025 guideline also factor in overall cardiovascular risk, not just the number.
| Category | Systolic (top) | Diastolic (bottom) |
|---|---|---|
| Normal | Under 120 | and under 80 |
| Elevated | 120–129 | and under 80 |
| Stage 1 hypertension | 130–139 | or 80–89 |
| Stage 2 hypertension | 140 or higher | or 90 or higher |
| Hypertensive crisis | Over 180 | and/or over 120 |
An honest note on home targets
You will see two different home thresholds quoted, and both are defensible. Older guidance and many research protocols use 135/85 mm Hg as the home equivalent of a 140/90 clinic reading, on the logic that home readings run a few points lower than office readings. The 2025 AHA/ACC guideline simplifies this and uses 130/80 mm Hg for both office and out-of-office measurement, defining white-coat and masked hypertension against that single line. If your average lands between those two figures, that is exactly the grey zone where a clinician's judgement — not a chart — decides what happens next. Anyone telling you the threshold is settled to the millimetre is overstating the evidence.
Choosing a device without guessing
Skip the brand rankings — including ours. Two independent, non-commercial registries publish lists of monitors that have passed formal accuracy validation against published protocols: STRIDE BP (international) and ValidateBP.org (US, backed by the AMA). Check your model against those lists before buying, and buy the cuff size that fits your arm rather than assuming the included cuff is right. Bring your monitor to your next appointment so a reading on it can be compared side by side with the clinic device — that five-minute check is worth more than any review.
Cuffless wearables and smartwatch "blood pressure" features are an active research area, but STRIDE BP does not list them and no cuffless device currently substitutes for a validated cuff in diagnosis or management. Don't make decisions on one.
Why this matters more in midlife
Blood pressure in women tends to be lower than in men through the reproductive years, then climbs — the rise in systolic pressure is steepest in the decade following menopause, and prevalence of hypertension in women catches up with and eventually overtakes men's. The mechanisms are debated (loss of estrogen's vasodilatory effect, weight and body-composition change, sleep disruption, arterial stiffening) and no single one explains it. What is not debated is that high blood pressure in women is under-recognized and under-treated relative to men, with control rates that remain poor.
That combination — a real physiological rise, plus symptoms like palpitations and poor sleep that get filed under "it's just menopause" — is precisely the setting where masked hypertension hides. Home monitoring is the cheapest way to close that gap. Related reading: high blood pressure in women, menopause and heart health, and can water lower blood pressure (short answer: not meaningfully, but hydration status does affect individual readings). You can also work through our heart risk check to see which numbers are worth bringing to an appointment.
When to see a doctor
Call 911 or go to emergency care now if a reading is above 180/120 mm Hg and you have any of: chest pain or pressure, severe headache, sudden vision change, shortness of breath, weakness or numbness on one side, difficulty speaking, or back pain. That combination is a hypertensive emergency, not something to recheck in an hour. Note that heart attack symptoms in women can be subtler than crushing chest pain — jaw or back pain, nausea, unusual fatigue — so don't wait for a textbook presentation.
If a reading is above 180/120 with no symptoms, rest five minutes and repeat once. If it stays that high, contact your clinician the same day.
Make a routine appointment if your seven-day average sits at or above 130/80; if readings swing widely between sessions; if you feel dizzy, faint, or unusually fatigued on standing (which can signal pressure that is too low, especially on treatment); if you are pregnant or postpartum and see any elevated reading, since thresholds and urgency differ in pregnancy; or if you are already on treatment and your averages have drifted. Bring the full log.
Nothing here is a reason to start, stop, switch or adjust any medication or supplement on your own — those are prescriber decisions, and the 2025 guideline's treatment pathways depend on overall cardiovascular risk, not a home number in isolation. Your job is to produce accurate data. What happens with it is a shared decision. More on the wider picture in our women's heart health guide and across heart health.
