The 10,000 steps myth, and what the data actually shows.

There is a very specific cultural authority granted to the number 10,000. It blinks on watches, it organises office step challenges, it appears in the small talk of people who have never read a paper on physical activity. It is also, in its origins, a marketing slogan. In 1965 the Japanese company Yamasa launched a pedometer it called the manpo-kei — literally, the '10,000 steps meter'. The number was chosen because the kanji for 10,000 looks a little like a person walking, and because it was a round, memorable target. It was not derived from clinical evidence. There was no clinical evidence to derive it from.

Sixty years later, that marketing target has become the default cultural unit of daily exercise. The good news is that the science has, belatedly, caught up — and what it shows is more useful, more forgiving, and a great deal more interesting than the wrist tells you.

§ What the modern step-count studies actually find

The most important recent paper is the 2022 Lancet Public Health meta-analysis by Amanda Paluch and colleagues, which pooled fifteen prospective cohorts covering nearly 50,000 adults wearing accelerometers, with mortality outcomes tracked for an average of seven years. The findings are quietly devastating for the 10,000-step orthodoxy.

For adults aged 60 and older, all-cause mortality risk dropped sharply between roughly 2,500 and 6,000 steps a day, then flattened. There was no statistically meaningful additional benefit from 7,000 to 10,000 steps. For adults under 60, the curve continued slightly further — benefit accrued up to around 8,000 to 10,000 steps — but the additional gain from 8,000 to 10,000 was small compared with the gain from 4,000 to 8,000.

Earlier work by I-Min Lee at Harvard, on older women, had pointed in the same direction: meaningful mortality reduction by 4,400 steps, plateauing around 7,500. A separate 2020 JAMA paper by Saint-Maurice on US adults found a 51 percent lower mortality rate at 8,000 steps versus 4,000, with diminishing returns after 12,000.

The pattern across studies is consistent. The biggest yield is in the lower part of the curve — moving sedentary adults to moderately active ones. Once a person is moderately active, the marginal cost of every additional thousand steps is high and the marginal mortality benefit is low.

"The most important step is the one that takes you from sedentary to lightly active. Everything after that is icing."

§ Cadence: the variable nobody puts on their watch

Total step count is the headline number. Cadence — how fast you walk — is the variable that better tracks intensity, and intensity is what produces most of the cardiovascular adaptation that matters.

Saint-Maurice's analysis found that, after adjusting for total step volume, faster cadence (in particular, the steps accumulated in periods of 100+ steps per minute, which corresponds roughly to a brisk walk) was independently associated with lower mortality. A more recent Nature Medicine paper by Stamatakis introduced the concept of 'vigorous intermittent lifestyle physical activity' — short bursts of one to two minutes of stair-climbing, fast walking, or carrying — and found that as little as three or four such bouts a day was associated with substantial reductions in cardiovascular mortality.

The practical translation: if you cannot increase total step count, you can still benefit by increasing the intensity of the steps you already take. Walk faster. Take stairs at pace. Pick up the bags and walk briskly from the car.

§ Why the 10,000 number persists

Three reasons, none of them scientific. First, round numbers stick. 10,000 is memorable; 7,432 is not. Second, the wearable industry has a commercial interest in a target most users will not hit, which keeps engagement up. Third, public-health communicators are understandably nervous about lowering a target that, even if arbitrary, was getting people to walk more.

The last point has weight. The behavioural argument for keeping the 10,000-step myth alive is that it is a higher target, and so produces more activity than a 'true' 7,000-step target would. This may be empirically right. But it is paternalistic, and it has a cost: it tells the older adult who reliably walks 5,500 steps a day that they are failing, when on the actual evidence they have captured almost all of the longevity benefit available from walking.

§ What this means for an actual person

If you are sedentary today: the highest-value change you will ever make to your daily activity is the first 3,000 to 4,000 step jump. This is not metaphor. The mortality curves are nearly L-shaped. Getting from 2,000 to 6,000 steps a day buys you more years of expected life than getting from 8,000 to 12,000.

If you are already at 7,000 to 8,000 steps a day: the marginal returns to more steps are small. The marginal returns to faster steps are not. Consider:

• →One or two ten-minute brisk walks a day at a cadence that makes conversation slightly effortful.
• →Taking stairs whenever they exist, deliberately at pace.
• →A weekly 'long walk' of 60 to 90 minutes, which adds an aerobic stimulus that incidental walking does not.
• →Two strength sessions per week, which step counts cannot substitute for and which independently predict mortality.

If you are tracking nothing at all: a cheap pedometer or any phone in a pocket gives you the only metric that matters, which is the trend line of your own behaviour against last month. The absolute number is much less important than the direction.

§ A closing point on metrics

The relationship between modern adults and the metric on their wrist is one of the more peculiar developments in personal health. People will hit a target and stop, or fall short by a few hundred and feel they have failed a moral test. The number is a tool, and a noisy one. The underlying behaviour — movement, ideally a mix of slow and brisk, accumulated across the day, every day — is what the trials are actually measuring.

Walk more than you did yesterday. Walk faster when you can. Do not let a slogan from a 1965 pedometer ad tell you whether your day counted.