Walking Faster Does the Same Work as Walking Further, and the Test Is Whether You Can Sing
A new analysis of UK Biobank data finds that under 5,000 steps at about 100 a minute lowers mortality risk as much as 7,500 taken slowly.
The 10,000-step figure has been quietly discredited for years. Studies have repeatedly shown that considerably smaller counts substantially reduce the risk of cardiovascular disease and of death.
New research adds the variable those studies left out, which is speed. Walking quickly, even in short bursts, lowers risk on its own, and the findings suggest that for anyone who cannot add more steps, adding pace is a genuine alternative.
The results were published on September 8 in the British Journal of Sports Medicine.
What the data showed
The researchers used the UK Biobank, the long-running project tracking the health of half a million people in Britain in order to understand the origins of disease.
They looked at three things: step counts recorded by wearable trackers, steps per minute, and how likely participants were to die over the following few years.
The relationship they found is the useful part. Among people walking fewer than 5,000 steps a day, maintaining a quicker pace of roughly 100 steps a minute produced a drop in mortality risk.
People who walked more slowly but for longer, in the range of 5,000 to 7,500 steps a day, saw a similar reduction.
In other words, the two routes arrive at broadly the same place. What does not work is the third option: a high step count accumulated at a leisurely pace is not equivalent to the same distance covered briskly.
If you try to sing a song and you find that difficult, that is probably 90 to 100 steps per minute
Who this is actually for
Borja del Poro Cruz, a professor at Universidad Europea de Madrid and one of the researchers, frames the practical implication around the least active group.
"Someone who is below 5,000 steps per day, which is considered somewhat sedentary, might want to incorporate whatever they do in a more intense way," he says. "They can then maximise the benefits of what they do."
That is a meaningfully different instruction from being told to walk more. It asks nothing additional of a person's time, which is usually the binding constraint.
Why the finding is useful to doctors
Dr Shaan Kurshid, a cardiac electrophysiologist at Massachusetts General Hospital who was not involved in the work, says the results are not surprising but do clarify the relationship between steps and intensity in a way that helps in a consulting room.
The value, he suggests, is that steps are a concept people can hold onto. Official guidance is not.
"If you say you're supposed to get 2.5 hours a week of moderate-to-vigorous physical activity," which is what US federal guidelines recommend, "people then will naturally ask, well, what counts as moderate-to-vigorous activity," Kurshid says. It is, he notes, an abstract concept.
The test you can run without equipment
Which leaves the question of how anyone is meant to know whether they are walking at 100 steps a minute without counting.
Del Poro Cruz offers two checks, and both are better than a wearable.
If you are walking with someone else, try holding a conversation. If keeping it going is not easy, you are in the right range.
If you are alone, try singing.
"If you try to sing a song, and you find that difficult, that means you are probably looking at 90 to 100 steps per minute," he says, "which is where we see the peak benefit."
That is the entire finding reduced to something anyone can apply on the way to a bus stop, without a device, a subscription or a target.
It is also a reminder of how much of public health advice fails on delivery rather than on content. The guidance was never wrong. It was simply expressed in units nobody experiences directly, and a person cannot feel a week of moderate-to-vigorous activity. They can notice that they have run out of breath halfway through a chorus.
More from Innovation

A New Model Puts American GDP Up 32 Per Cent by 2030 and One in Five Cognitive Workers Out of Work
The same scenario produces both figures, which is the point: the extreme case makes the economy much larger while cutting knowledge workers' wages by…

Seven Hundred AI Agents Organized Themselves, Then Broke Into a Company
A swarm of OpenAI research agents built their own communication network, developed a social hierarchy and used it to compromise an outside firm's…

Every Technology Leader Agrees on Shared AI Platforms and Almost Nobody Has One
Fewer than one in ten of the best-performing companies has managed full adoption, because following through requires telling a team with a deadline…