Anyone with a blood pressure cuff at home can run this experiment on themselves: take a reading, drink a strong cup of coffee, and take another an hour later. Many people will see the numbers climb. That observation is real, and it is the reason doctors have long told people with hypertension to go easy on coffee. It is also only half of the story, because the question that matters for long-term health is not what one cup does this morning, but what years of daily cups do to your baseline. Those turn out to be different questions with different answers, and this article takes them one at a time.
First, what blood pressure numbers mean
Blood pressure is written as two numbers in millimetres of mercury, or mmHg. The first, systolic pressure, is the peak pressure in your arteries while the heart contracts. The second, diastolic, is the pressure while the heart relaxes between beats. Both move around constantly through the day with posture, stress, exercise, sleep and food. Hypertension is a diagnosis of pressure that stays high over time, and it matters because sustained high pressure damages blood vessels and raises the risk of stroke, heart disease and kidney disease. A brief rise that settles back down is a different thing from a sustained one, and keeping that distinction in mind is what makes the coffee research make sense.
The short-term spike is real, especially if you already have hypertension
The clearest measurement of the acute effect comes from a systematic review by Mesas and colleagues, published in the American Journal of Clinical Nutrition in 2011[1], which looked specifically at people who already had hypertension. Pooling 5 controlled trials, they found that a dose of 200 to 300 mg of caffeine, roughly what you would get from one or two strong brewed coffees, raised systolic blood pressure by a mean of 8.1 mmHg and diastolic pressure by 5.7 mmHg. The rise appeared within the first hour after caffeine and lasted for at least three hours.
That is not a trivial change for someone whose pressure is already high. It is the reason a clinic reading taken shortly after a coffee can look worse than your usual numbers, and why some clinicians ask patients to skip caffeine before a measurement. The usual explanation offered is that caffeine blocks adenosine receptors, and that adenosine normally helps blood vessels relax, although the trials summarised here measured the rise in pressure rather than testing that mechanism.
But the effect shrinks when the coffee becomes a habit
The same review contained a second, quieter finding. In 3 studies that looked at the longer-term effect of drinking coffee for two weeks, Mesas and colleagues[1] found no increase in blood pressure when coffee was compared with a caffeine-free diet or with decaffeinated coffee. Across 7 cohort studies, habitual coffee drinking was also not associated with a higher risk of cardiovascular disease in people with hypertension. Their conclusion drew exactly the line this article is about: caffeine produces an acute rise for at least three hours, but the evidence did not support a link between longer-term coffee drinking and higher blood pressure.
The most likely explanation is tolerance. Regular caffeine users adapt to many of its effects, and the blood pressure response appears to be one of them. Someone who rarely drinks coffee and has a double espresso will usually see a bigger rise than a person who drinks the same amount every morning. Tolerance is not total, and it varies from person to person, which is part of why individual readings after coffee are so inconsistent.
Coffee is not the same as caffeine pills
A meta-analysis of randomised controlled trials by Noordzij and colleagues in the Journal of Hypertension[2] is useful because it separated two kinds of trial. It pooled 16 randomised studies lasting at least 7 days, covering 1,010 participants. Overall, coffee and caffeine together raised systolic pressure by 2.04 mmHg and diastolic by 0.73 mmHg. But when the authors split the trials, the difference was striking. Trials giving caffeine on its own, at a median dose of 410 mg a day, raised systolic pressure by 4.16 mmHg and diastolic by 2.41 mmHg. Trials giving coffee, at a median of 725 ml a day, raised systolic pressure by only 1.22 mmHg, and the diastolic change of 0.49 mmHg was not statistically clear.
The authors concluded that regular caffeine intake does increase blood pressure, but that when caffeine arrives through coffee, the effect is small. Why coffee should behave more gently than the same caffeine in a capsule is not settled. Coffee contains hundreds of other compounds, including chlorogenic acids and minerals such as potassium and magnesium, and one hypothesis is that some of these partly offset caffeine's effect on blood vessels. That remains a hypothesis rather than a demonstrated mechanism.
Does a daily coffee habit lead to hypertension?
This is the question most people actually mean when they search. The answer depends on long observational studies, because nobody can randomise thousands of people to drink coffee or not for a decade. One of the largest is the analysis of the Nurses' Health Studies by Winkelmayer and colleagues, published in JAMA in 2005[3]. They followed 155,594 US women without diagnosed hypertension for 12 years, during which 19,541 cases were reported in one cohort and 13,536 in the other.
They found no straight-line relationship between caffeine intake and new hypertension. Looking by category, the pattern was an inverse U shape: women in the middle fifth of caffeine intake had a 13% and 12% higher risk than those in the lowest fifth, but the risk did not keep climbing at higher intakes. When the researchers looked at individual drinks, habitual coffee consumption was not associated with an increased risk of hypertension at all. Cola was, and interestingly that held for both sugared and diet cola, which suggests the cola finding is not simply about caffeine either.
Pooling the long studies together
Several meta-analyses have since combined the cohort studies. Steffen and colleagues, also in the Journal of Hypertension[4], pooled 10 randomised trials and 5 cohort studies. In the trials, the pooled change in systolic pressure was -0.55 mmHg and in diastolic -0.45 mmHg, neither significant. In the cohort studies, the pooled risk ratio for developing hypertension was 1.03, again not significant. They described the evidence as low quality and said no recommendation could be made for or against coffee on blood pressure grounds, which is an unusually candid way to summarise a field.
A later dose-response meta-analysis by D'Elia and colleagues in the European Journal of Nutrition[5] pooled 4 prospective studies covering 196,256 participants and 41,184 diagnoses of hypertension. The relationship was not a straight line. Drinking one or two cups a day was not associated with any change in risk compared with not drinking coffee, while a small protective association appeared from three cups a day, with a relative risk of 0.97, and held at higher intakes. A 2023 review by Haghighatdoost and colleagues[6] reached a similar direction: combining 13 effect sizes from cohort studies, higher coffee consumption was associated with a 7% lower risk of hypertension, though results varied with study region, sex, quality and size.
So which answer is right?
Both, because they answer different questions. In the short term, caffeine raises blood pressure, and the rise is larger in people who already have hypertension and in people who do not normally drink caffeine. In the long term, people who drink coffee every day do not appear to develop hypertension more often than people who do not, and some large analyses even find a slightly lower risk at three or more cups. The tolerance that habitual drinkers build up is the most likely bridge between those two findings.
It is worth being careful with the reassuring half. A relative risk of 0.97 is a very small difference, and none of the long-term findings are strong enough to suggest drinking coffee to prevent hypertension. They are strong enough to say that for most people, a moderate coffee habit does not look like a meaningful cause of high blood pressure.
What the genetic evidence adds
Because the long-term studies are observational, researchers have turned to Mendelian randomisation to test whether coffee itself affects blood pressure. The method uses gene variants that make people drink slightly more or less coffee. Since genes are assigned at conception, they are far less entangled with lifestyle than a questionnaire answer, so they can act as a rough natural experiment.
A 2026 analysis by Zhong and colleagues in Nutrition, Metabolism and Cardiovascular Diseases[7] took gene variants linked to coffee drinking from UK Biobank data on 428,860 people and tested them against blood pressure data from large international consortia. Genetically predicted coffee consumption was associated with lower systolic pressure, an estimate of about 1.9 mmHg, and lower diastolic pressure, about 1.3 mmHg, and the results held in several sensitivity analyses. That is consistent with the cohort studies finding no harm and a small possible benefit. Mendelian randomisation depends on assumptions that are hard to fully verify, including that the genes affect blood pressure only through coffee, and this is a single analysis, so it strengthens the reassuring reading without making it certain.
A note on home blood pressure readings
The acute finding has one very practical consequence. If a single dose of caffeine can raise pressure for three hours or more in people with hypertension, as the review by Mesas and colleagues[1] found, then a reading taken shortly after your morning coffee is not measuring your usual resting pressure. It is measuring your pressure plus a temporary caffeine effect. That can make home readings swing from day to day depending on when coffee happened, and it can make a clinic visit after a café stop look worse than your true baseline.
The fix is simply consistency. Readings are most comparable when they are taken in the same conditions each time, including the same timing relative to coffee. Whatever measuring routine your doctor recommends, the coffee research is a good reason to follow it closely rather than to treat one high post-coffee number as a new diagnosis, or one low reading on a coffee-free day as proof everything is fine.
The limits of this evidence
Nearly all the long-term evidence is observational. It shows association, not cause, and people who drink coffee differ from people who do not in many ways that researchers can only partly adjust for. Intake was usually self-reported on food questionnaires, and a cup was rarely defined. There is also a quiet selection problem: people whose blood pressure rises uncomfortably with coffee, or who are told by a doctor to cut down, may drink less, which can make coffee drinkers look healthier than coffee itself makes them.
The trials have the opposite weakness. They are properly randomised but short, usually a week to a couple of months, and small. They can show what coffee does to blood pressure over weeks. They cannot show what it does to the risk of a stroke over twenty years. The review by Steffen and colleagues[4] rated the overall body of evidence as low quality for precisely these reasons.
Finally, what is in the cup matters. The studies above mostly describe coffee as drunk in the United States and Europe, typically black or with a little milk. A large sweetened café drink or a 3-in-1 sachet delivers sugar and sometimes sodium alongside the caffeine, and those carry their own links to weight and blood pressure that the coffee research was not designed to measure.
Practical reading, not medical advice
If you have normal blood pressure and drink coffee regularly, this body of research gives no strong reason to stop. If you have hypertension, the acute finding from the review of hypertensive patients[1] is the one to take seriously: a strong coffee can raise your pressure noticeably for a few hours, which matters for timing home readings and clinic visits, and your own doctor's advice outranks any population average. Decaf is a reasonable option for anyone who wants the drink without the acute caffeine effect, since the decaf comparison trials showed no long-term rise. For how much caffeine counts as moderate, see our article on the 400 mg daily figure, and for coffee's effect on heart rhythm rather than pressure, the article on the CRAVE trial covers palpitations and extra heartbeats. Nothing here replaces individual medical advice.