Five Minutes a Day: What the headlines said and what the papers found
The Headline vs The Paper
Earlier this year, the news told you that five minutes a day could add a year to your life. It was an extremely catchy headline that promised a simple action would provide a massive return. Five minutes of walking, an extra year - who wouldn't take that trade? But it sent my spidey senses off, as it's the kind of message that can spread without a clear audit trail. So I want to show you where it came from, because the real story is more interesting than the headline.
There were two studies, not one
Two papers were published on the same day, by researchers at the University of Sydney. The coverage merged them into a single claim. Almost nothing you read about "five minutes" was describing one study. The first, Koemel and colleagues, in eClinicalMedicine, looked at sleep, physical activity and diet together. Nearly 60,000 people from UK Biobank, wearing wrist accelerometers for a week, with diet assessed by questionnaire, followed for an average of eight years. The second, by Ekelund and colleagues, was published in The Lancet. It pooled individual data from more than 135,000 adults across seven cohorts in the UK, US, Norway and Sweden, looking specifically at physical activity and sedentary time.
Two different questions, two different designs but one attractive headline.
Where "five minutes" actually came from
In the Koemel paper, the five minutes was five minutes more sleep, not exercise. The finding was about a combination: five minutes more sleep, two minutes more moderate-to-vigorous activity, and half an extra serving of vegetables a day. Three changes, made together, not one.
In the Lancet paper, five minutes did refer to walking. But the outcome attached to it wasn't a year of life. So the famous number belongs to the sleep study, and the famous activity belongs to the mortality study, and the famous outcome belongs to neither of them in the form it was hitting the press.
What the "10% lower risk" really means
The Lancet paper found that an extra five minutes a day of moderate-to-vigorous activity might prevent up to 10% of all deaths. Is that really what the headlines were saying?
That figure is also a population attributable fraction. It answers the question: if everybody in the population made this change, what share of deaths might be avoided. It does not say your personal risk of dying early falls by 10% if you walk for five more minutes.
Important note - those are different claims. The first is a public health projection across millions of people, while the second is a personal promise. The headlines seemed to focus on the second. Interestingly, the paper also gives a lower figure, around 6%, for a "high-risk" approach targeting only the least active. That number appeared almost nowhere, possibly because it is less impressive and more realistic? Just my thoughts.
What about the extra year?
The Koemel paper did find something close to that, but with a condition attached that changes the positioning. The extra year applied to people starting from the poorest category for all three behaviours, worst sleep, lowest activity, poorest diet. If you are already sleeping reasonably, moving a little and eating some vegetables, the paper does not promise you a year for five more minutes of anything.
The same paper contains a figure I find far more interesting, and which barely made the news. People with seven to eight hours of sleep, more than forty minutes of moderate-to-vigorous activity a day, and a good diet lived around nine years longer in good health than those with the worst of all three. Nine years! That's a real finding, but more than 40 minutes seems a harder sell than just five.
What these studies can and can't establish
Both are observational. That is not a criticism as you cannot randomise people to a decade of poor sleep, but it does limit what can be concluded.
They can tell us that these behaviours travel together with longer, healthier lives across very large populations, with activity and sleep measured by device rather than self-report, which is a genuine strength. They cannot tell us that making the change causes the outcome. Nobody in either study was asked to add five minutes of anything, no one was followed to see what happened when they did. People who sleep well and walk more differ from those who don't in many ways such as income, health status, work patterns, existing illness. Statistical adjustment helps but it does not eliminate the fundamental problem of unmeasured confounding. UK Biobank participants are, on average, healthier and more affluent than the general population, which is a known limitation of the cohort. There is also reverse causation to consider. Someone in the earliest stages of an undiagnosed illness moves less and sleeps worse because they are becoming unwell. Both studies take steps to reduce this but neither can remove it entirely.
One more thing worth knowing - disclosure. A senior author on the eClinicalMedicine paper declares paid consultancy and equity in a company whose products relate to physical activity promotion. That is properly disclosed, and it does not invalidate the work, it is simply part of reading a research paper completely.
So what should you actually do with this?
Definitely incorporate more movement, better sleep and more vegetables, advice that many have heard before. The evidence that those three matter is overwhelming and long-standing, and these two papers add real weight to something already well established.
What I object to is not the advice, it's the promise it leads people to believe. When someone is told five minutes buys them a year and it doesn't visibly buy them anything, the conclusion they draw is not "that headline oversold a population statistic." It's "this doesn't work for me." I see the consequences of that in clinic constantly, people who have tried things that were sold to them as certainties, and who now believe their own body is the problem.
The honest version is less sexy but more encouraging. Small changes are worth making, they accumulate, and starting from a low base is where they matter most. Bigger changes, sustained, do considerably more. Nine years' worth, on this evidence. That is not as neat as five minutes for a year, but it is a good deal more honest.
References:
Koemel, N., Stamatakis, E., et al. (2026). Minimum combined sleep, physical activity, and nutrition variations associated with lifeSPAN and healthSPAN improvements: a population cohort study. eClinicalMedicine.
Ekelund, U., Tarp, J., Ding, D., et al. (2026). Deaths potentially averted by small changes in physical activity and sedentary time: an individual participant data meta-analysis of prospective cohort studies. The Lancet.
Comments