Grip strength predicts death better than blood pressure. A gripper won't change your odds.
In half a million people, weaker grip tracked with higher mortality — a strong marker. But it reads out whole-body strength and health, not a number you move by training your hands.
| Group | Value (value) |
|---|---|
| All-cause mortality | 1.16 |
| Cardiovascular mortality | 1.17 |
| Stroke | 1.09 |
| Myocardial infarction | 1.07 |
Grip strength is one of the most reliable predictors of death that medicine has, and it is measured with a device you squeeze for a few seconds. That has made it a fixture of longevity content and a target people now try to train directly. The evidence for grip as a predictor is genuinely strong. The idea that squeezing a hand gripper will lower your mortality risk is a misreading of what that predictor is actually measuring.
How good a predictor it is
The landmark evidence comes from the PURE study, which measured grip strength with a dynamometer in a large international population and followed them for a median of four years [s1]. Each 5-kilogram reduction in grip strength was associated with a 16% higher risk of all-cause mortality (hazard ratio 1.16, 95% confidence interval 1.13 to 1.20), a 17% higher risk of cardiovascular death (1.17), and smaller but significant increases in the risk of stroke (1.09) and myocardial infarction (1.07) [s1]. Strikingly, grip strength was a stronger predictor of all-cause and cardiovascular mortality than systolic blood pressure [s1].
A later analysis in more than 502,000 UK Biobank participants, followed for around seven years, found the same thing: each 5-kilogram lower grip strength was associated with roughly 16–20% higher all-cause mortality, alongside higher risks of cardiovascular disease, respiratory disease and several cancers [s2]. Two very large, independent cohorts, the same conclusion: weaker grip, worse odds.
Why it works as a predictor — and why that is the catch
Here is the crucial point that the "train your grip" framing misses. Grip strength is powerful precisely because it is a cheap proxy for something much larger: overall muscle strength, physical function, nutritional status and the accumulated toll of chronic disease. A firm grip is a readout of a body that is, broadly, robust; a failing grip often signals frailty, undernutrition, or illness that has begun to erode strength throughout the body. The hand is just the convenient place to take the measurement.
That is why isolating and training the grip itself does not follow from the data. Nothing in these studies tested hand-strengthening, and there is no evidence that increasing grip in isolation — with a gripper, say — moves any of the outcomes the studies measured. You would be improving the instrument reading without necessarily changing the underlying thing it is reading. The PURE investigators were explicit about the open question, noting that further research was needed to identify what determines muscular strength and to test whether improving strength actually reduces risk [s1]. In other words, even the researchers who established grip as a predictor did not claim it was a proven lever.
What the studies did and did not find actionable
The observational design is the limit. These are associations, not experiments; grip strength was measured once and outcomes tracked forward, so the studies cannot establish that low grip causes death rather than marking bodies already on a worse trajectory. Reverse causation is plausible throughout — undiagnosed disease weakens people before it kills them. It is also worth noting what grip did not predict in PURE: it showed no significant association with incident diabetes, with injury from falls, or with fracture [s1], which cuts against treating it as a universal risk gauge.
The honest use of the number
Grip strength is best understood as an inexpensive, fast screening tool — a way for clinicians to flag people whose overall strength and resilience may be declining, and who might benefit from closer assessment [s1] [s2]. That is a real and useful role, and it is why grip measurement is built into frailty and sarcopenia assessments.
What it is not is a training target. If the underlying driver of the grip–mortality link is whole-body strength and general health, the intervention that plausibly matters is the one that builds those broadly — progressive resistance exercise and staying active — not hand-squeezing. Training your grip will make your grip stronger. Whether it changes anything the grip was predicting is exactly the question the evidence has not answered, and the mechanism suggests it would not. Grip is the smoke alarm, not the fire.
This article is informational and is not medical advice.
Sources
- Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study — The Lancet , July 1, 2015
- Associations of grip strength with cardiovascular, respiratory, and cancer outcomes and all cause mortality: prospective cohort study of half a million UK Biobank participants — BMJ , May 8, 2018
More on
Age-related muscle loss is now a diagnosable disease. One treatment has strong backing.
Sarcopenia got a formal definition and diagnostic cut-offs. The clinical guidelines that followed back resistance exercise strongly — and are far more cautious about protein powders, vitamin D and drugs.
Metformin's case as an anti-ageing drug now rests on three null results
Twenty-one years of cancer follow-up, multimorbidity data to 2021, and a randomised trial in frail older adults. In each, metformin did not beat placebo. Lifestyle change did.
Two October papers on creatine in older adults point at strength, not muscle mass
A meta-analysis of 20 trials found a two-kilogram gain in one-rep max and no effect on bone density. A small crossover trial found functional strength gains largely independent of muscle mass.
Which muscles hold their strength through weight loss? A gene hunt
A transcriptomic sub-study of the CALERIE calorie-restriction trial links 151 genes to preserved leg strength during weight loss, independent of how much body mass was lost.