How Many Steps a Day? About 7,000 — and It Buys More Than Your Heart.
The 10,000 came from a pedometer. A Japanese company launched the first commercial step counter around the Tokyo 1964 Olympics and called it the Manpo-Kei — “10,000 steps metre”. [5] We told that story in full on our walking and weight verdict, so here we will leave it at one line and answer the other question instead.
Which is: how many steps actually buy you something, and what do they buy?
There is now a real answer, and it arrived in 2025.
| Outcome | Change | Confidence |
| Dying of anything | −47% | moderate · 14 studies |
| Dementia | −38% | moderate · 2 studies |
| Depressive symptoms | −22% | moderate · 3 studies |
| Heart disease | −25% | moderate · 6 studies |
| Type 2 diabetes | −14% | moderate · 4 studies |
| Falls | −28% | VERY LOW · 4 studies |
| Getting cancer | no change | low · 2 studies |
About 7,000
A dose-response meta-analysis published in The Lancet Public Health pooled 31 studies from 24 separate groups of people, and looked at nine different health outcomes rather than just death. [1]
Compared with a fairly sedentary 2,000 steps a day, 7,000 steps was associated with 47% lower risk of dying from any cause — and the trials agreed with each other reasonably well while saying it.
The shape matters more than the single number. For death, heart disease, dementia and falls, the curve is not a straight line. It falls steeply and then bends, and the bend sits somewhere around 5,000 to 7,000 steps. Past that, you keep gaining, but you gain far less per step than you did climbing out of the bottom.
The authors’ own conclusion is worth quoting because it is unusually un-preachy: 10,000 “can still be a viable target for those who are more active”, but 7,000 “might be a more realistic and achievable target for some.” [1]
The part nobody puts on the box
Steps get sold as a weight and heart thing. Look at what else moved.
Dementia risk was 38% lower. Depressive symptoms, 22% lower. Type 2 diabetes, 14% lower. [1] The dementia estimate came from only two studies — but those two studies agreed with each other completely, which is not nothing.
And one that did not move: getting cancer. The estimate was a 6% reduction with a range running from 13% lower to 1% higher, which means no effect is still comfortably on the table. [1] Cancer mortality was lower, which is a different claim about a different group of people.
We are reporting that null in the same breath as the wins because leaving it out would make the list look tidier than the evidence is.
Your age changes the number
Here is the finding that quietly rescues the 10,000.
An earlier meta-analysis pooled 15 cohorts — 47,471 adults, 3,013 deaths, a median of just over seven years each. [2] When they split by age, the curve bent in different places:
• 60 and over: risk kept falling until about 6,000 to 8,000 steps a day.
• Under 60: it kept falling until about 8,000 to 10,000.
So for a younger adult, 10,000 lands roughly where the evidence says the benefit stops accruing quickly. The number is approximately right for them — by coincidence, not by derivation. A product name got lucky.
For a 72-year-old it is well past the bend, which is exactly the population most likely to be told they are failing at it.
Does walking faster help?
Less than you would think, once you account for the fact that faster walkers also walk more.
In the study of 16,741 older women, higher stepping intensity looked strongly protective — until the researchers adjusted for total steps per day, at which point almost all of it evaporated. [3] The 15-cohort analysis found a bit more survived that adjustment, but not for every measure of pace. [2]
The honest summary: total steps is doing most of the work. If you have to choose between more steps and faster steps, the evidence points at more.
What would change our mind
All of this is observational. Nobody has randomly assigned people a step count and followed them for a decade, and realistically nobody will.
That leaves two objections open, and they are not small ones.
Reverse causation. Being ill makes you walk less, often long before it kills you. Some of the gap between the 2,000-step group and the 7,000-step group is people who were already sick, and no statistical adjustment fully removes that.
Residual confounding. People who walk more differ in other ways — income, smoking, whether they have a safe pavement. The 2025 authors list it as a limitation themselves, along with the small number of studies behind most of the individual outcomes. [1]
That is why this is rated Supported and not Established. The evidence is consistent, graded moderate by its own authors, and points the same way across three independent analyses. It is still not a trial.
The verdict
More daily steps track with living longer, and with less dementia, less depression and less diabetes. That is Supported. Most of the gain has landed by around 7,000 a day, and the steepest part of the curve is well below that — the biggest single win is getting off 2,000.
The 10,000 was a trademark. It is also, for the under-60s, not far off. Both of those are true, and the second one is an accident.
If you are currently taking 3,000 steps a day, the useful number is not 10,000 or 7,000. It is 4,000.
Related: does walking help you lose weight · does lifting weights burn fat · does metabolism slow with age · does lack of sleep cause weight gain · magnesium benefits
