Nearly 6 in 10 adults with high blood pressure in a study of 1.4 million in England had never been diagnosed

More than a third of adults whose blood pressure was measured for a large English health study met the definition of hypertension, and most of them did not know it. Of those with high blood pressure, 58.8% had never been told so by a health professional and were not taking medication for it. Among people who were on medication, half still had readings above the usual treatment target.

The analysis comes from Oxford Population Health, a department of the University of Oxford, and was published in September in the journal BMJ Public Health. It draws on 1.4 million adults who joined Our Future Health, a research project recruiting volunteers across the United Kingdom, between 2022 and 2025. “This isn’t a problem confined to a particular group; it’s substantial across almost every demographic we examined,” said Wenyu Liu, the medical statistician who led the work.

What counts as high blood pressure

A blood pressure reading has two numbers. The first, systolic, is the pressure in the arteries when the heart beats, and the second, diastolic, is the pressure between beats. Both are given in millimeters of mercury, written mm Hg.

The Oxford team used the World Health Organization definition of hypertension, the medical term for high blood pressure. A participant qualified if their systolic pressure was 140 mm Hg or greater, their diastolic pressure was 90 mm Hg or greater, or they were taking blood pressure lowering medication.

Every reading came from a single visit to one of the study’s clinics. Participants sat with their feet flat on the ground and the cuff on an upper arm held level with the heart. An automated device took two readings, and a third if the first two were very high or very low.

Anyone whose average came out high was told. Those between 140/90 and 179/109 were asked to seek repeat measurements at their local pharmacy within the next few weeks, and those above that range within two days.

The researchers then sorted people with hypertension into four groups: undetected, detected but not treated, treated but not controlled, and controlled. “Detected” meant the person reported a previous diagnosis from a health professional or was taking medication. “Controlled” meant a treated person’s reading was below 140/90.

About 8% of volunteers had a serious condition such as kidney failure, heart failure, liver failure or cancer. They were left out of the main analysis because such illnesses alter blood pressure and how it is treated. That left 1,312,551 people.

How many adults had undiagnosed high blood pressure

Among those participants, 37.2% had hypertension and the prevalence was greater in men than women. Applied to the study population, that is roughly 488,000 people. Adjusted to match the age and sex makeup of England, the figures were 41.9% of men and 28.6% of women.

The headline result concerns what those people knew. About 58.8% of those with hypertension were undetected, the authors report. On the study’s numbers that is roughly 287,000 adults who walked into a research clinic with high blood pressure that had not been diagnosed.

Age made the largest difference. Between the youngest group, aged 18 to 39, and the oldest, 80 and over, the share with hypertension rose from 23% to 73% in men and from 10% to 69% in women.

The young adults drew the authors’ attention. Among all participants aged 18 to 39, 15.3% met the definition, which the authors say is a cause for concern since young adults are ineligible for an NHS Health Check. That free checkup, offered by the National Health Service in England, starts at age 40.

Other patterns were familiar. Hypertension was almost 50% more common among men than women. The odds of hypertension were about 50% higher in Black than in White participants, a gap that narrowed to about 30% once weight, smoking and other factors were taken into account. The likelihood rose step by step with overweight and obesity.

Detection followed the same map. People whom doctors expect to have high blood pressure, such as older men and those with diabetes or heart disease, were more likely to have had it found. Conversely, younger, female, White and lean individuals with no family history of cardiovascular disease were less likely to be detected as having hypertension.

No group was spared. The authors write that there were no demographic subgroups in which the prevalence of hypertension was not substantial.

Half of those treated still had high blood pressure

A diagnosis did not settle matters. Among those who reported taking one or more antihypertensive medications, meaning drugs that lower blood pressure, a group of 182,876 people, 50.2% still had a reading of 140/90 or higher. That is about 91,800 people on treatment whose pressure was above target on the day.

Doctors have an old name for this pattern. The authors write that the “rule of halves” still broadly applies with approximately half of hypertension being undetected and half of those treated being under-treated.

Poor control is not confined to England. In surveys weighted to be representative of US adults, the share of people with hypertension whose pressure was controlled rose from 31.8% in 1999-2000 to 53.8% in 2013-2014, then declined to 43.7% in 2017-2018. That American figure counts everyone with hypertension, treated or not, so it is not a like-for-like comparison with the Oxford number.

“Many patients with high blood pressure are undetected or are sub-optimally treated,” said Iain Turnbull, a general practitioner and a co-author of the paper, who is deputy chief medical officer at Our Future Health.

One worry about treating more people, particularly older ones, is that pressure lowered too far leads to dizziness and falls. In this dataset there was no apparent relationship between reporting a history of falls and being on antihypertensive medications.

How England compares on undiagnosed high blood pressure

High blood pressure is common everywhere. One analysis of 135 studies estimated that in 2010, 31.1% of the world’s adults had hypertension, about 1.39 billion people.

Awareness is patchy everywhere too. The largest global study, which pooled 1,201 surveys covering 104 million participants, found that globally 59% of women and 49% of men with hypertension reported a previous diagnosis of hypertension in 2019. In that analysis treatment and control rates were highest in South Korea, Canada, and Iceland, where more than 70% of people with hypertension were treated and over half were controlled.

An earlier project, the PURE study, measured 142,042 adults aged 35 to 70 in 17 countries with a standard device. It found that 40.8% had hypertension and 46.5% of those were aware of the diagnosis. It also found that only a minority of those receiving treatment were controlled, about a third.

Set against those numbers, the Oxford result is high for a wealthy country. The 1,201-survey analysis estimated that in high-income western and Asia-Pacific regions, 27% to 34% of people with hypertension were not aware of their condition. The Oxford paper itself notes that smaller household surveys in England in 2015 to 2019 found the percentage of undiagnosed hypertension among adults with hypertension was 29%.

A share of 58.8% is about double either figure, and the paper does not explain the gap. Three differences in method could contribute. The Oxford study included adults from 18, while the global analysis covered ages 30 to 79. The Oxford study also relied on one visit, and its participants were volunteers.

The direction of travel is less disputed. The authors cite evidence that in England, after gains in the 2000s, progress has plateaued since around 2011.

Why treating high blood pressure matters

The reason any of this matters is what high blood pressure does over decades. “High blood pressure remains the leading modifiable risk factor for cardiovascular disease, linked to around half of heart attacks and strokes in the UK,” said Bryan Williams, chief scientific and medical officer at the British Heart Foundation, who was not an author of the study.

Treatment works, and the size of the benefit is well measured. A collaboration based at Oxford pooled individual records from 48 randomized trials with 344,716 participants. It concluded that a 5 mm Hg reduction of systolic blood pressure reduced the risk of major cardiovascular events by about 10%.

In absolute terms the gain depends on a person’s starting risk. In participants without previous cardiovascular disease at baseline, there were 31.9 major cardiovascular events per 1,000 people per year in the comparison groups and 25.9 in the treated groups. That is six fewer heart attacks, strokes or cases of heart failure a year for every 1,000 people treated. Those are raw rates pooled across trials of different sizes, in which treatment lowered systolic pressure by 6.3 mm Hg on average, so the gap does not map exactly onto the 10% figure.

Pushing harder brings more benefit and more harm. The SPRINT trial assigned 9,361 people at increased cardiovascular risk to a systolic blood-pressure target of less than 120 mm Hg or to the standard target of less than 140. Serious cardiovascular events occurred at 1.65% a year in the intensive group and 2.19% a year in the standard group. Episodes of very low blood pressure, fainting and acute kidney injury were more common with intensive treatment.

What a single blood pressure reading cannot show

The study’s main limit is in its design. Each person was measured on one occasion, and the authors acknowledge that such readings are only approximate estimates of a person’s long-term average.

Blood pressure moves from minute to minute, and it tends to run higher in a clinic than at home, an effect known as white coat hypertension. For that reason the National Institute for Health and Care Excellence, known as NICE, which writes clinical guidelines for England, asks for a diagnosis to be confirmed with home readings or a monitor worn for 24 hours, called ambulatory monitoring.

The paper cites NICE’s own check on how well a clinic reading predicts the confirmed result. Compared with ambulatory monitoring, the median sensitivity of the clinic measurement was 81% and the median specificity was 76%. Sensitivity is the share of true cases a test picks up. Specificity is the share of people without the condition whom it correctly clears, so a specificity of 76% means roughly one in four people without sustained hypertension would still be flagged by a clinic reading.

Some of the 287,000 “undetected” cases would therefore not be confirmed on retesting. The paper does not estimate how many.

Who took part matters as well. The authors list the non-random sampling among the study’s limitations, since participants chose to respond to an invitation. Recruitment was initially focused on English regions with high socioeconomic deprivation and ethnic diversity, to draw in groups that research usually misses. Medication use and prior diagnoses were reported by participants and could not be checked against medical records. So far the data cover England only.

What the researchers want changed on blood pressure checks

The authors go beyond description to an argument about policy. Their target is the NICE requirement for several confirmatory steps, and for lifestyle advice in some cases, before medication begins. Liu said that “the current approach to detecting and treating high blood pressure is not fit for purpose”.

Turnbull put the alternative this way: “There is a need for a more proactive approach to detection of high blood pressure and to consider initiation of treatment based on predicted risk of cardiovascular disease in addition to levels of blood pressure.”

That is a proposal, and the study did not test it. The confirmation steps exist because of the white coat effect that the paper itself describes, and a study that counts readings on one day cannot show whether a different rule would lead to better outcomes or to more people treated unnecessarily. The authors do point to one existing route: pharmacies in the United Kingdom have offered checks since 2021, and they write that it is possible that extending this approach may improve detection and treatment of hypertension.

Guidelines in England have not changed. In the study, a high reading led to a request for a repeat measurement, and whether repeated high readings amount to hypertension is assessed by a person’s doctor.

The study found hypertension in more than a third of 1.3 million adults, counting those already on medication, and no prior diagnosis in most of them, on the strength of one clinic visit each that the authors agree needs confirming before a diagnosis is made.