"It's good for you" is a phrase that changes nothing for anyone. The useful question is another: how much? And on this, for once, there are big numbers.
654,827 people, and an answer in years
An analysis published in PLoS Medicine in 2012 pooled six large cohort studies from the National Cancer Institute: 654,827 people, followed for a median of ten years, with 82,465 deaths. The question was exactly this: how many years of life, after the age of 40, are gained at each level of physical activity?
| Activity in free time | Years of life gained after 40 |
|---|---|
| none | reference |
| up to 75 minutes a week of brisk walking | +1.8 years |
| the highest level measured: 450 minutes and over | +4.5 years |
The sharpest comparison is a different one: being active and of normal weight, compared with being inactive and obese, was associated with 7.2 years more life.
These are population averages, not a personal contract. But the order of magnitude is clear: the first step, 75 minutes a week, is worth almost two years.
The shape of the curve: the biggest jump is the first one
Three years later, the same network of cohorts, 661,137 people and 116,686 deaths, followed for 14.2 years on average, measured the dose-response in JAMA Internal Medicine. Compared with those who did no activity in their free time:
- less than the recommended minimum: mortality 20% lower;
- 1 to 2 times the minimum (that is 150 to 300 minutes a week): −31%;
- 2 to 3 times the minimum: −37%;
- 3 to 5 times the minimum: −39%, and here the curve flattens;
- 10 times the minimum or more: no excess risk, −31%.
Look at where the difference lies. Between doing nothing and doing less than the minimum there are twenty points. Between the minimum and the theoretical maximum there are eight.
Those who gain the most are not those who train a lot: they are those who stop sitting still. This is why, when someone tells us they do not have time for an hour at the gym, the answer is not to convince them to find it. It is to get them to do the first fifteen minutes.
And what if people lie on questionnaires?
It is the serious objection, because all the studies above rely on self-reported activity, and people overestimate how much they move.
In 2019 the BMJ published a systematic review with a harmonised meta-analysis of individual data, using only studies that measured movement with an accelerometer: eight cohorts, 36,383 people, average age 62.6 years, 2,149 deaths.
The numbers, divided into quarters from least to most active:
- total activity, second quarter: −52% mortality compared with the first;
- third quarter: −66%; fourth quarter: −73%;
- light activity only, most active quarter: −62%;
- time spent sitting, most sedentary quarter: a risk 2.63 times higher compared with the least sedentary quarter.
Two things here. The first is that when you remove the questionnaires the association does not deflate: it strengthens. The second is that light activity counts too: standing up, walking around the house, taking the stairs, standing. It is not a consolation prize for people who do not go to the gym: it is the part of the day on which almost everyone can still do something.
What these studies do not show
It has to be said, otherwise we are selling something.
They are observational studies. People were not randomly assigned to a "moves" group and a "stays still" group: they were followed while doing what they already did. So two problems always remain open:
- reverse causation: those brewing an illness move less for that reason, and die sooner for that reason. The authors exclude the first years of follow-up precisely for this, but it is not a perfect solution;
- residual confounding: those who move more also tend to smoke less, eat better, have more money and more education. The analyses correct for all of this, and correcting is not the same as eliminating.
A randomised study that assigned thirty thousand people to move or not for twenty years does not exist and never will. What can honestly be said is that the association is large, consistent across different cohorts in different countries, dose-dependent and biologically sensible. On falls and pain, where randomised trials do exist, the evidence is more solid and points in the same direction.
What we do with these numbers
Three practical things.
The first step is the most valuable. If you are starting from a standstill, the first 75 minutes a week, twenty-five minutes three times, or ten minutes every day, are the ones with the best return on the whole curve. You do not need a gym membership: you need them to exist.
Light movement is not wasted time. In the accelerometer data it is associated with an almost halved risk. If a pain stops you running, everything else remains, and our work often starts exactly there, getting back in motion whatever can move while treating what hurts.
Sitting time is a separate variable. The hour of activity is not enough if the other fifteen are spent on a chair: the most sedentary quarter had a risk 2.63 times higher. Interrupt, do not compensate.
On how this translates into a real pathway, and on where the line lies between a physiotherapist and a personal trainer, we have written movement partner, not personal trainer.
In summary
Seventy-five minutes of walking a week is worth, in the estimates across 654,827 people, almost two years of life. The maximum benefit is reached much earlier than people think, and there is no point at which moving becomes dangerous. These are observational studies, with the limits they have: but of all the things we are recommended in order to live longer, this is one of the few you can start today, for free, from any point.
On the back, the same thing measured on a precise outcome, relapse: what reduces back pain relapses.
And the flip side, measured over fourteen days: two weeks standing still, how much you lose.
And an indicator that, on its own, predicts mortality better than several tests: grip strength and health.
And after a cancer diagnosis, where the historical advice was rest: moving during and after cancer, the data.
And on a precise number, blood pressure, where the comparison with drugs has been made: high blood pressure and movement, 391 studies.
And on the brain, where a randomised trial exists and not just cohorts: preventing cognitive decline, the FINGER study.
Sources
- Leisure time physical activity of moderate to vigorous intensity and mortality: a large pooled cohort analysis, PLoS Medicine, 2012
- Leisure time physical activity and mortality: a detailed pooled analysis of the dose-response relationship, JAMA Internal Medicine, 2015
- Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis, BMJ, 2019
Last source check: September 2026.
Need a consultation?
Start with the Ri-Hub questionnaire: a few questions about your case, you sign up, and you see who will follow you: a physiotherapist on the Italian professional register, to whom you can write in chat whenever you like. If you prefer to talk it over by voice, the first consultation is free, lasts twenty minutes and commits you to nothing: it arrives with your written message already read, and together you decide whether and how to continue.
