Five More Minutes of Sleep and Two More Minutes of Exercise a Day Were Worth an Extra Year of Life
A study of 59,078 people found the minimum combined improvement associated with one more year of life was startlingly small. The same study found the maximum was nine years, which is the number nobody is quoting.
Tuesday, August 25, 2026/4 min read
Almost every piece of advice about living longer asks for a reorganisation of your life. Get eight hours. Do a hundred and fifty minutes a week. Change how you eat.
A cohort study in eClinicalMedicine asked a narrower and more useful question: what is the smallest combined change across sleep, movement and diet that shows up in the data at all? The answer was five minutes, two minutes, and a slightly better shopping basket.
The numbers, exactly
Precision matters here, because this is a finding that gets rounded into nonsense within about two shares.
The study followed 59,078 people in the UK Biobank, recruited between 2006 and 2010, median age 64, of whom 45.4 per cent were men. Sleep and physical activity were measured with wearable devices rather than by asking people to estimate, which is the study's principal strength.
A combined improvement of five minutes a day more sleep, 1.9 minutes a day more moderate to vigorous physical activity, and a five point rise in a diet quality score was associated with one additional year of lifespan. The confidence interval ran from 0.69 to 1.15 years, which is tight.
A larger combination, 24 minutes more sleep, 3.7 minutes more activity and a 23 point diet improvement, was associated with four additional years of healthspan, meaning years lived without major illness or disability limiting daily function. That confidence interval ran from 0.50 to 8.61 years, which is not tight at all, and anyone quoting the four should quote the range with it.
The number nobody is repeating
Nine and a bit years.
Compared with the least favourable third of the cohort, people in the most favourable third on all three measures had 9.35 more years of lifespan and 9.45 more years of healthspan. The thresholds for that group were 7.2 to 8.0 hours of sleep, more than 42 minutes a day of moderate to vigorous activity, and a diet quality score between 57.5 and 72.5.
That is the actual headline, and it is being ignored in favour of the small number because the small number is more comforting. Both are in the same paper. The five minute finding tells you the gradient starts immediately. The nine year finding tells you where the gradient goes.
What two minutes of exercise means
Not two minutes of exercise. Two minutes more than you currently do, of activity hard enough to raise your breathing.
This is the most misread part of studies like this one. The dose is a marginal change from wherever you are, and the people in this cohort were mostly not doing very much. If you already train, another 1.9 minutes will not buy you a year, and nothing in the paper claims it will. The gradient is steepest at the bottom, which is a general feature of exercise epidemiology and the reason the first small increment always looks so cheap.
The same applies to sleep. Five extra minutes for someone averaging six hours is not the same intervention as five extra minutes for someone averaging eight.
What the study cannot tell you
Whether any of it is causal, and the authors are clear about this.
It is observational. Behaviour was measured once, and the device data covered only three to seven days, which is a thin sample of a life. Diet was self reported, and collected three to nine years before the activity measurements, so the two are not describing the same period.
The UK Biobank cohort is also healthier and more affluent than the British population it was drawn from, which biases this kind of estimate in a predictable direction. And the harder problem, which no statistical adjustment fully solves, is that sleeping well, moving often and eating properly are all downstream of having the time, money and security to do them. Some of what looks like the effect of behaviour is the effect of circumstance.
So what is it good for?
As a correction to the way this advice is usually delivered.
The standard framing is a threshold: hit 150 minutes a week, get eight hours, follow the diet, and you qualify. Thresholds are useful for public health messaging and terrible for individual motivation, because they tell the person doing 40 minutes a week that they have failed rather than that they are two thirds of the way to something.
A dose response curve says something different. There is no cliff. Every increment counts for something, the early ones count for more, and the return does not require you to become a different person first. That is a more accurate description of the evidence and a more useful one, and it is the same argument we made about the backlash against optimisation culture in the wellness reckoning.
It does not replace the things with the strongest individual evidence behind them. Resistance training still earns its place for reasons we set out in what strength training does to the body over decades, and aerobic base work still matters for the reasons in the zone two evidence. What this study adds is the floor, not the ceiling: the smallest change that is not nothing.
Published in The Outspoken Digest
Newsletter
The Digest, in your inbox
One edition, sent when it is ready. No noise, and your address is never passed on.
Read Next
More Lifestyle →
We Wrote That Third Places Were Coming Back. The Census Data Says They Were Disappearing.
Aug 24, 2026/4 min read
More Companies Ordered People Back to the Office This Year, and the Best Studies Still Do Not Support It
Aug 22, 2026/4 min read
Every Sleep Tip You Have Read Is the One Treatment Guidelines Do Not Recommend on Its Own
Aug 21, 2026/3 min read
Houseplants Do Not Purify Your Air, and the Real Reason to Keep Them Is Better Than the Myth
Aug 21, 2026/3 min read