Grip Strength Predicts Death Better Than Blood Pressure. The Fix Is Unproven.

How hard you can squeeze predicts whether you are going to die better than your blood pressure does.

That is not a headline we are dressing up. It is what the paper says, in plainer words, after measuring 139,691 people in 17 countries. [1]

The finding

The PURE study handed a dynamometer — a handheld gauge you squeeze — to nearly 140,000 adults across seventeen countries rich and poor, then followed them. [1]

For every 5 kg of grip you were missing — that is about eleven pounds of squeeze, a difference you would not particularly notice in daily life — the risk of dying in the follow-up period was 16% higher. Death from cardiovascular disease, the heart and blood-vessel diseases that include stroke, 17% higher. Having a stroke, 9% higher.

And the line everyone quotes, correctly: “Grip strength was a stronger predictor of all-cause and cardiovascular mortality than systolic blood pressure.” [1]

The effect held across rich and poor countries alike, which is unusual and makes it harder to explain away as an artefact of one health system.

It has held up elsewhere, too. UK Biobank, which followed half a million adults in the UK aged 40 to 69, put grip beside systolic blood pressure (the top number) in a comparison it added after its main analysis, and got the same answer for deaths. For whether people went on to develop cardiovascular disease in the first place, the two predicted about equally well [3].

Now the part that usually gets skipped

Every few months this study resurfaces, and underneath it somebody is selling a hand gripper.

Here is the sentence from the same paper, from the interpretation section, whose point six of the eight first-page web results we read for “grip strength longevity” on 7 October 2026 leave out:

“Further research is needed to identify determinants of muscular strength and to test whether improvement in strength reduces mortality and cardiovascular disease.” [1]

The researchers are saying, plainly, that nobody knows whether getting stronger lowers your risk of dying or of cardiovascular disease. A later PURE paper, in 2020, says the same: it is not known whether changing strength itself would directly affect deaths [4].

Our reading, which is reasoning rather than a trial result: grip strength is most likely not doing much by itself. It is a cheap readout of something much larger — total muscle, nerve function, nutrition, whether you have an illness nobody has found yet. Squeezing a spring will move the readout. We found no trial showing that it moves the thing being read: no trial of grip training that we could find set out to measure deaths or heart attacks (PubMed, OpenAlex and two trial registries, searched 30 September 2026). A 2026 trial of grip exercise in 54 people with advanced kidney disease did record one death, in its control group, among the reasons people left the trial; deaths were not what it measured. Some researchers leave room for strength itself to protect, as reserve the body can draw on when illness arrives [5]. That idea has been put to trials, of strength training in general, and they have not settled it; they are set out below.

It is a thermometer, not a thermostat.

One kind of squeezing does have trial evidence behind it, though for blood pressure, not for survival. Isometric handgrip training means squeezing and holding, rather than pumping a gripper through reps. In seven small randomised trials of adults with high blood pressure, fewer than 200 people in all, it lowered resting systolic pressure, the top number, by about 6.7 points (mmHg) more than in the control groups, the people in each trial who were not given the training; the range the true figure probably sits in runs from 3.4 to 10.0 points [6]. That is a risk factor moving, not a death rate, and a set of reps on a spring gripper is not what those trials tested. Our blood-pressure verdict reports the 2023 ranking of exercise types, in which isometric training came first, and the later rankings, most of which put it lower.

So has anyone tested the fix?

Yes, in a way, and the answer so far is no clear difference. No trial we found set out to test whether getting stronger lowers deaths in healthy adults (searched 30 September 2026). But trials of strength training count deaths along the way, and those counts have been added up. A Cochrane review of strength training in older people pooled the 13 trials that reported deaths: 20 in the training groups and 21 in the comparison groups, among 1,125 people, nearly all of them frail, in care or living with a diagnosed condition [7]. That is far too few deaths to show an effect in either direction, and most of the programmes in that review lasted only two or three months. A second review looked at trials of exercise lasting a year or more in people over 60, most of them mixing strength work with aerobic and balance training. Pooling the 29 with usable counts of deaths, 11,441 people, it found no fewer in the exercise groups; in the ten of those trials that enrolled people with a particular disease, deaths leaned lower with exercise, right on the line researchers use for “probably not chance” [8]. A larger review the next year, of 93 trials and 28,523 older people, found the same: no clear difference in deaths overall, and fewer deaths in the trials of patients [16].

The nearest trial we found that was built to count deaths in mostly healthy older people tested endurance exercise, not strength. In Norway, 1,567 people aged 70 to 77 were assigned to five years of twice-weekly interval training, steady training or the national advice on activity. Deaths came out almost level: 35 among the 787 who trained, 37 among the 780 given the advice, who, left to themselves, exercised at an intensity between the two training groups [15].

Where single trials have counted fewer deaths, they were small, the people were already ill, and the strength training came bundled with other care. After hip-fracture surgery, 4 of 62 patients died during a year of supervised weight training plus treatment for a dozen other problems, from nutrition to eyesight, against 8 of 62 given usual care [9]. Among 39 people with cirrhosis, one of those who had done 12 weeks of resistance training was dead three years later, against six of those who had not; that count was made afterwards from medical records and was not something the trial was set up to measure [10]. And in 60 people having radiotherapy for cancer in the spine, resistance training made no difference to survival [11]. Those are counts from small trials of sick people, not odds for anyone else.

Genetic studies come at cause another way, by asking whether people born with gene variants for a stronger grip die later or get less heart disease. They have split. The first large one found the variants in 195,180 people, then tested them against deaths and heart disease in other large sets of data and found no sign of either [12]. Later ones using UK Biobank found more deaths among people whose genes predict a weaker grip [13] and less coronary heart disease among those whose genes predict a stronger one [14]. The authors of the deaths study still wrote that it is not yet clear whether improving strength would lower the risk. Being born stronger is not the same as getting stronger.

Four years is not very long

Here is the objection that should worry you most. The researchers worried about it too.

The median follow-up was four years. [1]

Serious illness weakens your grip long before it kills you. Over a four-year window, some unknown share of “weak grip predicts death” is simply undiagnosed disease showing up in the hand first. That is not a flaw in the study; it is a limit on what a four-year study can tell you, and it means the arrow may point backwards from the direction people assume.

Other studies went looking for it. UK Biobank threw out every death in the first two years and still found weaker grip going with higher death rates: for every 5 kg weaker, 20% higher in women and 16% higher in men [3]. A study in Hawaii followed 6,040 men of Japanese ancestry, aged 45 to 68, for 30 years. It kept only men who reported no heart disease, stroke, cancer or diabetes at the start, and left out anyone who died in the first three years. After allowing for age, smoking, physical activity and height, the weakest third still died at a rate 24% higher than the strongest third [5]. And PURE itself kept going: at a median of 9.5 years, in 155,722 people with no history of cardiovascular disease, low grip was still as strong a marker of death as several standard risk factors, or stronger [4].

Hidden illness is part of the story, then, but not the whole of it. The Hawaii team saw it from both sides: the link held in men who were healthy at the start, and with the men who were already ill left in, grip looked an even stronger predictor, because illness had weakened them first [5]. What none of these studies can show is cause and effect, and UK Biobank’s authors say so themselves: they cannot rule out hidden illness entirely, and the link may not be causal [3]. Grip may simply be standing in for something else — overall muscle, fitness, nutrition. That is the thermometer point, and it still stands.

The result that cuts the other way

In high-income countries, the study found something awkward: stronger grip went with more cancer, not less. [1] The association did not appear in middle- and low-income countries.

In UK Biobank, a far larger cohort from a high-income country, cancer went the other way: weaker grip went with more cancer, in new diagnoses and in deaths, and the deaths ran 17% higher per 5 kg in women and 10% in men [3]. Its authors flagged the clash with PURE and called the reason “not clear”. Where they did see stronger grip going with more cancer was in prostate cancer diagnoses, and they suggest, without having tested it, that stronger, more health-conscious men may simply see their doctor more [3]. Whether anything like that lies behind PURE’s result, we cannot tell, and we have no explanation of our own. We are reporting it because it is in the paper, and because a page that prints the tidy rows and quietly drops the awkward one is doing the thing we spend most of our time here catching other people doing.

So what actually has evidence behind it

If grip is the readout, the sensible question is what moves the underlying thing. And there the answer is better.

A meta-analysis of 16 cohort studies looked at muscle-strengthening activity — actual resistance training, not hand squeezing — and found it associated with a 10 to 17% lower risk of dying, of cardiovascular disease, of cancer and of diabetes. [2]

Then the interesting bit. The curve was J-shaped: benefit climbed steeply, peaked, and then flattened out or bent slightly back. And it peaked at roughly 30 to 60 minutes a week. [2]

Not a day. A week.

Beyond about an hour a week, the authors say the effect on mortality, cardiovascular disease and cancer is “unclear”. [2] More is not obviously better, and might not be better at all.

This is cohort data, not trials, so it carries the same caution as everything else on this page — people who lift weights differ from people who do not in a hundred ways. But the seven cohorts in its estimate for deaths all pointed the same way, at a dose most people would consider almost embarrassingly small. Its authors still graded that evidence “very low” certainty, the bottom of the scale used to say how far a finding can be trusted, mainly because most of the cohorts were American. And every one of the 16 relied on people’s own account of how much they lifted [2].

The verdict

Grip strength predicts death, and it does so better than blood pressure. That is Supported — 139,691 people across seventeen countries, consistent across income levels.

Training your grip is not the same as changing your risk, and the researchers who produced the finding list that as an open question rather than a conclusion. Trials of strength training and of longer exercise programmes in older people have counted deaths and found no clear difference [7] [8] [16], and we found none built to answer the question in healthy adults.

If the number moved you to do something, the evidence that goes with lower risk is for strength work in general. A 2022 review of 16 cohort studies found the most benefit at roughly 30 to 60 minutes of it a week, and its authors said that makes the usual advice of at least two days a week look reasonable [2]; a 2026 analysis of three large American cohorts found the lowest risk at 90 to 119 minutes, with no further gain beyond two hours [17]. An hour or two a week, then, not an hour a day. That is cohort evidence graded very low certainty, but the cohorts on deaths all point the same way. Your hands will get stronger as a side effect — which is rather the point.

And if you are worried about your own reading, that is a conversation with a clinician. We read the research; we are not your doctor.

Related: does lifting weights burn fat · creatine benefits · how many steps a day · does metabolism slow with age · all the habits we have checked