Does grip strength predict how long you live?

The PURE study measured grip in 139,691 adults aged 35–70 in 17 countries and followed them for a median of four years, during which 3,379 died (Leong 2015). Each 5 kg of lower grip went with a hazard ratio for death of 1.16. Grip predicted all-cause and cardiovascular death more strongly than systolic blood pressure did.

UK Biobank measured 502,293 adults aged 40–69 and followed them for a mean of 7.1 years, with 13,322 deaths (Celis-Morales 2018). Per 5 kg of lower grip, the hazard ratio was 1.20 in women and 1.16 in men.

A meta-analysis pooled 42 cohorts and 3,002,203 people (Wu 2017). It found 1.16 per 5 kg again, with no difference between men and women. Three independent sources landing on the same figure is unusual, and it is the main reason grip is the best-evidenced strength test there is.

A hazard ratio compares the rate of an event between groups at any point during follow-up. A ratio of 1.16 per 5 kg means that, after the study’s adjustments, someone with 5 kg less grip died at a rate about 16% higher than otherwise similar people. It is relative. If your risk of dying in a given year is small, 16% more of a small number is still small.

The bracket after each figure is the 95% confidence interval: the range the true value plausibly sits in. For PURE it was 1.13–1.20. Because the whole range sits above 1, the association is unlikely to be chance.

Comparing extremes gives bigger numbers. In the meta-analysis, the weakest category against the strongest had a hazard ratio of 1.41 for all-cause mortality. A separate pooled analysis put the weakest quarter against the strongest at 1.67 (Cooper 2010).

The link is broad. In PURE, 5 kg less grip went with a hazard ratio of 1.17 for cardiovascular death and 1.17 for non-cardiovascular death, and smaller ratios for having a heart attack (1.07) or a stroke (1.09). In UK Biobank the strongest link was with respiratory death: 1.31 in women and 1.24 in men.

Cancer is the inconsistent one. UK Biobank found 1.17 (women) and 1.10 (men) for all cancer, while the meta-analysis found no significant link with cancer (1.01). PURE also found no association between grip and diabetes, admission for pneumonia or COPD, falls, or fracture.

A dose-response meta-analysis of 48 cohorts and 3,135,473 adults aged 35–85 found that higher grip went with lower all-cause mortality across 26–50 kg, in a close-to-linear way (López-Bueno 2022). Cardiovascular risk fell between 24 and 40 kg. Cancer mortality trended U-shaped, with lower risk between 16 and 33 kg. The earlier meta-analysis found a linear relation up to 56 kg.

These ranges are pooled across both sexes and many devices. They are not personal targets. They say the association does not stop at "not weak"; they do not say what number you should aim for.

All of these studies are observational. Nobody was assigned to get stronger. Illness can lower both grip and survival, which is called reverse causation, and PURE’s four-year follow-up leaves room for it. The UK Biobank authors say reverse causation and residual confounding cannot be excluded, and their volunteers were healthier than the general population.

Prediction for one person is also modest. Adding grip to an office-based risk score in UK Biobank raised the C index, a measure of how well a model ranks who dies first, by 0.013. That helps a population model a little and says much less about you.

Two findings make strength itself more plausible as something that matters. In Health ABC, adults in their 70s with weaker legs died sooner, while muscle size was not strongly related to death; grip gave similar estimates (Newman 2006). And in the Honolulu Heart Program, men in the lowest grip third in midlife had 2.87 times the odds of walking very slowly 25 years later (Rantanen 1999). Both are still associations.

The Honolulu result is the one closest to everyday life. The middle grip third also had higher odds of very slow walking (1.79), and the risk of needing help with self-care more than doubled in the lowest third. These were men of one ethnic group, measured once in midlife, so it shows a pattern rather than a personal forecast.

Grip is worth measuring because the evidence for the link is strong and consistent. It is not worth converting into a life expectancy. Longstanding reads your grip against age and sex norms and against the line you need for the goal you chose at 85. It does not turn grip into a risk score, and it does not tell you that a stronger grip will add years, because no trial has shown that.

It also does not use 26–50 kg as a target, or the weakness cut-offs of 26 kg (men) and 16 kg (women) as a pass mark. Those mark weakness in older adults; for most age-85 goals the line sits well above them. Your line comes from your goal and the norms at each age. The calculator shows where a result sits and which table scored it. This is not medical advice.