Your readings at home sit comfortably in range. Then a nurse wraps the cuff and the number is fifteen points higher. This is one of the best-documented phenomena in blood pressure measurement, and it isn't a sign that you're bad at being calm.

Key takeaways

• Among people whose office readings are high, 15% to 30% have white coat hypertension: elevated in clinic, normal outside it (Muntner et al. 2019, American Heart Association).

• It appears to be an automatic alerting reflex rather than conscious anxiety (Grassi et al. 1999, Circulation).

• Untreated white coat hypertension does carry some added cardiovascular risk. In people already treated, the same pattern showed none (Cohen et al. 2019, Annals of Internal Medicine).

• The reverse exists too, and matters more: about 15% to 30% of adults with normal office readings run high outside the clinic.

What is the white coat effect?

It's the tendency for blood pressure to read higher when measured by a clinician than it does in ordinary life.

The original demonstration is striking. Researchers monitored blood pressure continuously through an artery while an unfamiliar doctor entered the room. Pressure climbed within about four minutes, peaking on average around 27 points on the top number and 14 on the bottom (Mancia et al. 1983). If those two numbers need a refresher, blood pressure numbers, explained gently covers them.

That's the acute spike. In everyday practice the habitual gap is smaller: in one study of 153 untreated adults the average difference between office and home readings was about 7 points on the top number, though the spread around that average was wider than the average itself (Satoh et al. 2023).

Two related terms are worth separating. White coat hypertension means office readings in the high range while out-of-office readings are normal, in someone not on medication. The white coat effect describes the gap itself, including in people already being treated. The distinction turns out to matter for risk.

How common is it?

Common enough that clinicians expect it.

Among people whose office blood pressure falls in the hypertensive range, 15% to 30% have white coat hypertension. Among people taking multiple blood pressure medications who still read high in clinic, 30% to 40% show the white coat effect (Muntner et al. 2019).

There's also a straightforward mechanical contributor that has nothing to do with clinics: single readings are just unreliable. Among US adults whose first office reading was in the 140 to 159 over 90 to 99 range, 35% averaged below 140/90 once three readings were taken. Only 3% moved the other way (Muntner et al. 2019).

So part of what looks like a white coat effect is simply the difference between one measurement and an average of several. More on that in blood pressure numbers, explained gently.

Is it just anxiety?

Probably not, and the honest answer is more interesting.

Researchers measured nerve activity directly in ten people while an unfamiliar doctor visited. They found a split response: nerve traffic to the skin rose sharply, while nerve traffic to muscle blood vessels actually fell (Grassi et al. 1999). That split pattern is characteristic of an alerting or defense reaction, not the generic stress response you'd expect from feeling nervous.

The American Heart Association attributes the phenomenon to the presence of an observer rather than to patient anxiety, though the literature isn't unanimous (Muntner et al. 2019).

The fair summary: an automatic reflex to being measured by a clinician, which is not the same thing as being consciously nervous. Telling yourself to relax is unlikely to help, and it isn't a personal failing.

Supporting evidence: when blood pressure is taken by an automated machine with nobody in the room, the effect largely disappears. Pooled across studies, readings taken by an unattended machine came out essentially identical to readings taken during ordinary daily life, while routine office readings ran about 14.5 points higher (Roerecke et al. 2019, JAMA Internal Medicine).

The observer, not the patient, appears to be the active ingredient.

Does white coat hypertension actually matter?

This is where precision matters, because the answer differs by group.

A meta-analysis of 27 studies covering more than 64,000 people found:

Source: Cohen et al. 2019, Annals of Internal Medicine

An independent pooled analysis of five population cohorts found the same split: white coat hypertension carried elevated cardiovascular risk, while the white coat effect in treated people did not (Muntner et al. 2019). Two separate datasets agreeing is about as good as this evidence gets.

The AHA adds an important nuance: much of the excess risk in white coat hypertension may be explained by other cardiovascular risk factors travelling alongside it. But people with white coat hypertension do progress to sustained hypertension faster than people whose readings are normal in both places, which is why annual follow-up is advised rather than dismissal.

So it isn't nothing, and it also isn't the same as having hypertension.

What about the opposite problem?

Masked hypertension is the mirror image: normal in the clinic, high outside it. The American Heart Association puts it at roughly 15% to 30% of adults with normal office readings (Muntner et al. 2019). A US analysis using national survey data landed lower, at 12.3%, which still works out to about 17 million adults (Wang et al. 2017). Either way it is common and, by definition, invisible in clinic.

Arguably it matters more, because nothing flags it. In the pooled cohort analysis, masked hypertension carried a cardiovascular risk at least as high as white coat hypertension (Muntner et al. 2019). The two estimates overlap, so treat it as comparable rather than clearly worse.

Which is the strongest practical argument for measuring at home: not to overturn a clinic reading you didn't like, but to catch the pattern nobody would otherwise see. It is the same reason we compare you to you rather than to a chart.

How do office and home readings compare?

They aren't interchangeable, and the correspondence isn't a fixed offset:

Source: Muntner et al. 2019

The widely quoted "office 140/90 equals home 135/85" is correct, but only at that row. At 130/80 the two are the same number, which is easy to get wrong and matters now that current guidance centres on 130/80.

What should you actually do about it?

Measure outside the clinic, then bring it to your clinician rather than acting on it.

The US Preventive Services Task Force gives its strongest recommendation to obtaining out-of-office measurements for diagnostic confirmation before starting treatment (USPSTF, 2021). The 2025 AHA/ACC guideline similarly says it's reasonable to rule out white coat hypertension with out-of-office monitoring before diagnosing (American College of Cardiology summary).

Two cautions worth carrying:

One home reading doesn't rebut one office reading. The white coat effect itself is poorly reproducible from visit to visit, so a single comparison proves little (Satoh et al. 2023). A week of readings does.

A normal home average isn't an all-clear. White coat hypertension progresses to sustained hypertension faster than normal blood pressure does, so it's a reason for regular follow-up rather than reassurance.

An app is genuinely useful here for one thing: holding a week of properly taken readings in a form you can show someone. It isn't useful for deciding what they mean.

Frequently asked questions

What is white coat hypertension?

Blood pressure that reads in the high range at a clinic but is normal outside it, in someone not taking blood pressure medication. It affects 15% to 30% of people whose office readings are elevated (Muntner et al. 2019).

Is the white coat effect caused by anxiety?

Mostly not, on current evidence. Direct nerve recordings during a doctor's visit showed an alerting-reflex pattern rather than a general stress response (Grassi et al. 1999), and readings taken by an automated machine with nobody present come out close to everyday values (Roerecke et al. 2019).

Is white coat hypertension dangerous?

It carries some added risk when untreated: about 36% higher cardiovascular event risk and 33% higher all-cause mortality versus normotensive people. In people already on treatment, the same pattern showed no significant excess risk (Cohen et al. 2019). It also tends to progress to sustained hypertension, so it warrants follow-up.

Which reading should I trust, home or clinic?

Neither alone. Guidelines use out-of-office readings to confirm a diagnosis before treatment (USPSTF 2021), but they're an input to a clinician's judgment rather than a verdict. Note also that home and office thresholds differ: office 140/90 corresponds to home 135/85, while at 130/80 the two are identical.

Can I have the opposite problem?

Yes, and it's easier to miss. Masked hypertension, normal in clinic but high outside, affects roughly 15% to 30% of adults with normal office readings and an estimated 17 million US adults (Wang et al. 2017). It carried higher cardiovascular risk than white coat hypertension in pooled analysis.

Related reading

Two things worth reading next: what actually lowers blood pressure, and how much your readings drift with the seasons, which is larger than most people expect.

Knit shows patterns, not diagnoses. Talk to a clinician about readings that concern you.