How to Build Grip Strength, and What It Really Buys You

Grip strength earned its reputation honestly. Across 139,691 people in seventeen countries, every five kilograms weaker went with 16 percent higher all-cause mortality, and per standard deviation it predicted death better than systolic blood pressure did.

Then a second sentence got attached that the first one does not support: therefore, train your grip.

A sign, or something you can change#

The investigators who ran that study wrote plainly that its observational design does not allow strong conclusions about whether muscular strength is causal. They describe grip as a risk marker.

That is not a technicality. Grip strength works as a predictor because it reports on things that are hard to measure directly: total muscle mass, nutritional state, nerve function, cumulative illness. A hand dynamometer is a cheap window onto a large amount of biology.

Squeeze a spring for six weeks and the number goes up. None of the biology behind it has changed. You have improved the reading on the instrument.

And the half of the finding that never travels: in the same study, systolic blood pressure predicted who would go on to develop cardiovascular disease better than grip did. Grip won on mortality. Blood pressure won on disease. Both halves are real.

What the training evidence actually shows#

A 2025 meta-analysis pooled 22 randomised trials and 1,017 participants aged 19 to 78, covering strength training, proprioceptive training and combinations, over two to thirteen weeks.

Overall effect on grip strength: small to moderate. Then the age split, which is the finding worth knowing:

Manual dexterity improved more than strength did, at 1.11. Pinch strength did not improve significantly at all. Sixty-four percent of the trials were at high risk of bias.

So the honest reading is that grip training works well in older adults, which is where the clinical interest is, and barely at all in younger healthy people, which is where most of the marketing is aimed.

There is also a body of work showing grip is oddly refractory: it frequently fails to move in programs where leg strength, chair-stand time and walking speed all improve substantially. One paper concluded that handgrip is not an appropriate measure for tracking change during an exercise program in frail older people, precisely because it lags what it is standing in for.

What to actually do#

Load your whole body through your hands.

Loaded carries. Pick up something heavy in each hand, walk thirty metres, put it down. Shopping bags, dumbbells, water containers. Three or four trips. Be honest about the evidence here: there are no randomised trials of loaded carries for any health outcome. What exists is a measurement study showing that carrying activates the trunk muscles more than standing and holding the same load, plus a great deal of coaching experience. It is a sensible way to load grip, trunk and walking at once, and it is not an evidence-based intervention.

Rows and deadlifts. Anything where the limiting factor at the end of a set is your hands. Do not use straps if grip is the thing you are training.

Hangs. Hang from a bar with your feet supported enough that you can hold thirty seconds. Three of those. The one randomised trial of hangboard training, in 27 intermediate climbers over eight weeks, found a small improvement in grip strength from weighted hangs and no significant effect from endurance-style hangs. Note what that means: the evidence for hangs building grip is one small trial in climbers.

While we are here, the widespread claim that hanging fixes shoulder impingement traces back to a single self-published, uncontrolled case series by one author, and the mechanism it proposes is itself contested in the shoulder literature. Hanging is low-risk and pleasant. It is not shoulder therapy.

Enough protein. Grip is a report on muscle, and muscle needs the raw material. Adults over 65 do better at around 1.0 to 1.2 grams per kilogram of body weight per day, which is above the population minimum and above what many older people actually eat.

Where the cutoffs sit, and who set them#

If you are testing yourself, the number only means something against a chart, and the charts disagree.

Which one you apply changes how many people get labelled by several times over. All three were derived in older adults, so a forty year old below them is weak for their age rather than meeting a definition that was written for them.

And under the European criteria, low grip strength alone establishes probable sarcopenia only. Confirming it needs a scan showing low muscle mass. It is a reason to raise the subject at an appointment, not a diagnosis to give yourself.

One last measurement warning#

British and most clinical figures are single hand, best of three. American CDC tables add both hands together, so they read roughly double. Halving before comparing is the single most common correction needed on this subject.

Test yourself properly against the chart for your age, and see what else reads the same reserve.