sofidexFor patientsAre you a clinician? Clinical site

Health Library  ›  News  ›  Health Drive  ›  Is 10,000 steps a real number?

Is 10,000 steps a real number?

The most famous number in health came from a pedometer slogan, not a laboratory. The measured curve is kinder — and it starts far, far lower.

Health Drive

Is 10,000 steps a real number?

0:0027:23

Transcript

Welcome to the sofidex Health Drive podcast series. Today we are talking about 10,000 steps. The most famous number in health, the ring on a million wrists, and the question almost nobody asks about it: who actually measured that? Maybe there is a watch on your wrist right now, quietly disappointed in you. Maybe you bought one for somebody you worry about. Or maybe you gave up on the target months ago, and a small voice still files every day under ten thousand as a day you failed.

And the answers on offer come in the usual two unhelpful sizes. Either the number is gospel and anything less barely counts, or step counting is dismissed as a gimmick and you should ignore the watch entirely. Neither of those survives contact with the actual curve. So we want to ask the question properly, and not whether walking is good for you, because it is. The question is where the famous number came from, what the measured evidence really shows, and why the people furthest from the target have the most to gain.

We will follow one man and the watch his daughter gave him, from the evening it called him a failure. Along the way: a slogan from the nineteen-sixties, the studies that finally measured real steps against real lifespans, and the number where benefit actually begins, which is much, much lower than you think. And to say it before the worry grows: nothing tonight says your walking has been wasted, however little of it there is. The evidence runs the other way. The fewer steps you take today, the more each new one is worth.

One thing before we start. This is health information, not medical advice, and nothing in it is a substitute for talking to your own doctor about your own treatment. What we can tell you is where all of it came from — 7 published sources, all listed with the episode. I want to start with a birthday present. And a man we'll invent so nobody's privacy is involved. Call him Marcus. He's fifty-eight. His daughter gave him a fitness watch, because she worries about him, and the watch came out of the box with a goal already set. 10000 steps. And every evening it shows him a ring that didn't close. He's averaging about 4000. Roughly half. And last week he told his daughter the watch had proved he was too far gone for walking to matter, so he might as well stop.

And that sentence is the whole reason we're doing this episode. Because Marcus has it exactly backwards, and so does his watch. The man averaging 4000 steps isn't the person the evidence has given up on. He's the person the evidence is most excited about. But to see why, you have to find out where that 10000 came from. And it didn't come from a laboratory.

Okay, where did it come from? Because it feels like medicine. It has that round, official sound.

Well, it came from an advertising department. Truly. And I want to be precise about what's documented, because this episode polices other people's numbers, so it has to police its own. Researchers who reviewed the step-count literature put it plainly. The 10000 figure gained its popularity through the media and practice, and it can be traced to Japanese walking clubs and a business slogan more than thirty years back. That's the pedigree. A slogan. A pedometer sold in Japan in the nineteen-sixties, marketed around a nice round number, and walking clubs that adopted it. So the most famous health target on earth started life as a slogan. Nobody measured ten thousand against nine thousand. Or against four. Nobody measured. And the number was chosen because it's round, and it stuck because it's round.

But hold on. Marketing can be accidentally right. Lots of folk numbers turn out fine. Did this one?

Partly. And that's as far as it goes. And the same reviewers said as much. For healthy adults, 10000 steps a day looked like a reasonable estimate of an active day. So the slogan wasn't absurd. But reasonable-for-whom is the problem. Because those researchers also flagged, back in 2004, that the goal may not be sustainable for some groups. Older adults. The chronically ill. Which is to say, the exact people most likely to be handed a fitness watch by a worried daughter. So the slogan fit fit people. And then the fit people's number was installed, as a factory default, on the wrist of every Marcus on earth.

Before the proper studies. Did those same 2004 reviewers at least try to grade the number? Or was it slogan all the way down?

They graded it, actually. And their scale is worth hearing, because of what 10000 means inside it. They proposed cutting adults into bands. Fewer than 5000 steps a day, a sedentary lifestyle index. From 5000 to 7499, low active. That's ordinary daily life without much deliberate exercise. From 7500 to 9999, somewhat active. At 10000 or more, active. And past 12500, highly active. So notice what 10000 was doing even in the friendliest academic treatment it ever got. It was a label. A description of people near the top of the range. It was never a line between healthy and unhealthy, because in 2004 nobody had mortality data fine enough to draw that line. The watch turned a description of the active into an exam for the inactive. Same number. Entirely different sentence.

So for decades the number sat there untested. When did somebody finally measure it properly?

Properly? 2019. And the way they did it matters. Because a Harvard group took the Women's Health Study. Over 16000 older women, average age seventy-two. And instead of asking them how much they walked, which is where step folklore comes from, they put a real accelerometer on each of them for a week. Then they waited. Four years, nearly. Mean follow-up, 4.3 years. Then they counted who had died, and lined the deaths up against the measured steps. And the paper opens with the quiet demolition of our whole subject. It notes that the 10000 goal is commonly believed by the public to be necessary for health, but the number has limited scientific basis. That's the setup. The results are better than the setup.

Go on. Because this is the study I've heard quoted at dinner parties. The one about 4400 steps.

That's the one. So they split the women into quarters by step count. And the least active quarter averaged about 2700 steps a day. The next quarter up averaged about 4400. And just that gap came with mortality rates 41 per cent lower. Not a typo. Forty-one. The hazard ratio was 0.59. And then, as the steps kept climbing, the benefit kept climbing, but more gently, until it levelled off at around 7500 steps a day. Past that, in these older women, more steps didn't buy measurably more survival. So read the shape of that curve. Carefully. The steepest part, the biggest win per step, sits at the bottom. Between the sofa and the corner shop. Which is exactly where Marcus lives.

Which is the opposite of what his watch implies. The ring treats 9999 as failure and 10000 as health. Nothing in that curve looks like a cliff at 10000.

Well, nothing in any of these curves looks like a cliff anywhere. Because that's not how bodies work. It's a slope. Always. A steep slope at the start, flattening as you climb. And a watch that paints the day red until 10000 has drawn a cliff on top of a slope. And for a fit thirty-year-old, harmless. Because she'll close the ring anyway. But for Marcus, the default does something actively cruel. It takes the range where his gains are largest, everything from 4000 up, and labels all of it failure. So the watch is wrong about him. Not slightly wrong. Wrong in direction.

One study, though. Older women, in one country. Before we rebuild Marcus's watch around it, who else has measured this?

Nearly everyone, since. And the results rhyme. Loudly. So, the year after, 2020, an American team did the same trick on a representative sample of US adults from a national health survey. Average age about fifty-seven. Men and women, both. Accelerometers again, about 4840 people, followed for a mean of ten years. Compared with taking 4000 steps a day, taking 8000 was associated with half the all-cause mortality. Half. A hazard ratio of 0.49. And taking 12000 steps was associated with 65 per cent lower. But notice where the halving happened. Down low. Between 4000 and 8000. The second doubling, from 8000 to 12000, bought much less. Same shape again. Steep at the bottom, flat at the top.

And then presumably somebody pooled everything. Because this show has taught me that one good study is a hint and fifteen are an answer.

Well, fifteen is literally the number. A 2022 meta-analysis in the Lancet's public health journal gathered 15 cohorts. 47471 adults, more than 3000 deaths, a median of seven years of follow-up. And because they pooled individual-level data, they could draw the whole curve at once. And risk of dying fell progressively with more steps. But the levelling-off point depended on age. So, for adults sixty and older, the benefit flattened at around 6000 to 8000 steps a day. For adults under sixty, at around 8000 to 10000. So the famous target turns out to be roughly where the curve goes flat for young people. A ceiling. Almost. The slogan accidentally described the top of the curve, and the world installed it as the entry requirement.

Say the quiet part, then. For a sixty-five-year-old, the evidence-based version of the ring would close at six or seven thousand.

And for many of them even that's the wrong place to point the arrow. Because there's a newer meta-analysis, 2023, specifically hunting the bottom of the curve. Seventeen cohorts this time, 226889 people. And they asked the bluntest version of the question. Where does benefit actually start? And their answer: the risk of dying of anything starts falling measurably once you're past about 3867 steps a day. For dying of heart and vessel disease specifically, past about 2337. That second number deserves a moment of silence. 2337. Steps. Because that's twenty minutes of unhurried pottering. The received wisdom put the gates of health at 10000. The measured door is at 2337.

So where does that leave Marcus and his 4000? Run his numbers against everything you've just said.

Well. Above the cardiovascular threshold already. Right at the all-cause threshold. And standing at the bottom of the steepest stretch of the whole curve. The same 2023 analysis worked out what each increment buys. Every extra 1000 steps a day was associated with 15 per cent lower all-cause mortality. Every extra 500 steps, 7 per cent lower cardiovascular mortality. So Marcus, at 4000, adding a 15-minute walk after dinner — that's roughly 1500 steps — is buying more expected benefit per step than almost anyone else on the street. The man his watch calls a failure is, statistically, the best investment in the neighbourhood. So his daughter bought the right present. Because it just shipped with the wrong number in it.

Marcus has a daughter, and she's the one who bought the watch. Thirty-one, runs a bit, closes her ring most days. Is her 10000 wrong too?

No. And that's worth saying cleanly, because this episode isn't a debunking of walking a lot. For her age group the pooled curve kept falling until around 8000 to 10000 steps a day. So her default lands roughly where her curve flattens. For her, the ring is a perfectly sensible place to stop optimising. The number isn't wrong. It's wrongly universal. One factory setting, two different bodies, and the damage all lands on one of them. Her watch tells her the truth and his watch tells him a lie, and they're the same watch.

But I want to poke the evidence before we rebuild his mornings around it. All of this is observational. Cohorts. Accelerometers and death certificates. Nobody was randomised to walk. How do we know the walking is doing the saving? Sick people walk less. Dying people walk much less.

And that's the right objection, so let's give it its full weight. Because reverse causation is the ghost in every step study. Illness cuts steps. Death follows illness. The steps take the credit. And the researchers know this, and they fight it. They throw out deaths in the first year or two of follow-up, so people already dying don't drag the low-step groups down. They adjust for smoking, weight, diet, diagnosed disease, self-rated health. And the association survives all of it. It behaves like a real effect, too. It's graded, step by step, dose by dose, the same shape in older women in America, in national surveys, in fifteen cohorts across multiple countries. But you're right that graded and consistent is still not proof. No one has run the twenty-year randomised trial, and no one ever will, because you can't assign people a decade of walking. So the honest statement is this. The case for steps is circumstantial, enormous, and consistent in a way that would be very hard to fake. And the case for the specific number 10000 is a slogan. Those are different grades of evidence, and the second one is the one on his wrist.

Give me the scale of this, though. Because we're talking as if Marcus is unusual. How many people is the default actually mislabelling?

Most people, give or take. And the American guideline committee counted. Approximately 80 per cent of US adults and adolescents are insufficiently active, by their own standard. Four in five. So picture the watch's intended audience honestly. The majority of wrists it lands on belong to people below the official activity floor, let alone below 10000 steps. And for that entire population, the first rungs of the ladder are where the treasure is. A default that shames the bottom of the curve isn't mislabelling an unlucky few. But it's mislabelling the typical owner. The product was calibrated to its healthiest customers, and sold to everyone else.

Which raises a practical question. The official advice, the stuff governments publish — does any of it actually say 10000?

No. None of it. And people find this genuinely hard to believe. So check the two biggest. The American physical activity guidelines, 2018 edition, written off the back of a full evidence review. And their headline for adults is time. Not steps. At least 150 to 300 minutes a week of moderate activity, or 75 to 150 of vigorous, plus muscle work twice a week. The World Health Organization, 2020, lands in the same place. 150 to 300 moderate minutes a week. Neither sets a step target. Neither. And both say the thing the watch doesn't. The American document puts it almost tenderly. The people doing the least gain the most from even modest increases. And the WHO's version has become a slogan of its own, a better one. Some physical activity is better than none. That's the official line. Not a ring. A direction.

Translate 150 minutes into steps for me, though. Because Marcus owns a step counter, not a stopwatch.

Roughly. And only roughly. Because a brisk ten minutes is about a thousand steps. So 150 moderate minutes spread over a week is in the ballpark of 2000 extra purposeful steps a day on top of ordinary pottering. Which, for Marcus, means 4000 becoming about 6000. And look where 6000 sits in everything we've said. Past the thresholds. High on the steep part. Inside the flattening zone the Lancet analysis found for people his age. The official time-based advice and the step curves agree with each other almost embarrassingly well. The only document in his life that disagrees with all of them is the factory default on his watch.

And what about days the steps don't come from walking? And he swims on Thursdays. The watch gives him a few hundred steps for it and looks disappointed.

Which is the watch being a step counter, honestly. Not a health auditor. Because the official recommendations never mention steps at all, remember. They're written in minutes of moderate or vigorous activity, and a swim is moderate activity by any sane definition. So is cycling. So is digging the garden. The step curves are so useful because walking is how most people move, and steps are what we can measure cheaply on tens of thousands of people for years. But the underlying commodity is movement, not footfall. So a Thursday swim counts in the only ledger that matters, and scores zero in the one on his wrist. When the two disagree, believe the ledger. The watch is an instrument, and instruments have blind spots.

What about speed? He asked his daughter whether his slow walking even counts. The watch keeps celebrating something called brisk minutes.

It counts. All of it. And this result surprised the researchers themselves, so it's worth hearing. Both the Harvard study and the American survey study measured intensity. How fast people stepped at their peak, not just how much. And faster walkers did die at lower rates. But here's the thing. Once you accounted for the total number of steps, the speed effect mostly dissolved. In the older women, the intensity associations were attenuated and most were no longer significant after adjusting for steps per day. And in the national sample, same story. No significant association between intensity and mortality after total volume. The pooled analysis found a little residual benefit at the brisker end, but the overwhelming message is the same. The engine is volume. Full stop. Accumulate the steps, at whatever pace your knees permit, and you've captured most of what's on offer. So slow steps aren't discount steps.

One worry before the turn, because it's the one he'd actually raise. His knees. He's fifty-eight, they ache on stairs, and somewhere he picked up the idea that walking wears a joint out like tread on a tyre.

And the tyre model is the wrong model, though his knees deserve a proper answer. Briefly: cartilage is living tissue, and moderate loading is part of how it stays healthy. Movement is treatment for most ordinary joint ache, not a tax on it. But I'm deliberately not going further than that, because a specific aching knee belongs to a specific examination. So the honest division of labour is this. The step evidence tells him walking, at his pace and his volume, is associated with living longer. What his particular knee can carry this month is a question for his own doctor, and it's a good question to bring to the same appointment where the aspirin people are bringing theirs. What he shouldn't do is let the watch make the decision by demoralising him out of moving at all. That outcome is the worst one on the table, and it's the one he was heading for.

Alright. The turn, then. Because I think I can feel it coming. This was never really an episode about a number.

Well, it's an episode about what a default does to a person. Because the deep problem isn't that 10000 is wrong. As a description of a very active day, it's fine. The problem is that a target tuned to the fittest got installed as the definition of trying, and everyone below it was handed daily evidence of failure. And defaults have power. Real power. Marcus believed his watch over his own legs. He was ready to stop walking entirely, the one change that would genuinely hurt him, because a slogan from the nineteen-sixties told him 4000 was nothing. The evidence says 4000 is not nothing. The evidence says the next thousand steps he takes are worth more than the ten-thousandth step of the woman jogging past him. And that's not consolation. That's arithmetic.

While we're correcting slogans. Sitting is the new smoking. Marcus has a desk job. Does a day of sitting undo the walk?

No. But the sitting question is real, so let's be exact. The WHO looked at it in the same 2020 guidelines. And they do recommend reducing sedentary behaviour, across all age groups. Formally. On the official list. But read their fine print, because it's a model of honesty. The evidence was insufficient to quantify a threshold. Meaning: nobody can tell you how many sitting hours is too many, and anyone who does tell you is rounding a hunch. What the evidence does support is the trade. Some of the sitting, swapped for some moving, helps. And the people it helps most are, again, the people doing the least. So no, the desk doesn't cancel the walk. The walk is the thing that answers the desk. Smoking has no equivalent of a lunchtime lap of the block. Sitting does.

And the cramming question. Suppose his week collapses and it all lands on Saturday. A long walk with the dog, nothing else for six days. Wasted?

Not wasted. No. Because the weekly guidelines are deliberately written as weekly. 150 to 300 minutes over the week, arranged however your life permits. And the step cohorts measured averages over days, not perfect daily streaks. A man averaging 6000 because Saturday is enormous and Tuesday is nothing is still a man averaging 6000. Now, spreading it has advantages. Because habits survive better as daily grooves than weekly heroics. Joints prefer regular moderate loads to occasional big ones. So I'd gently push him toward the boring version. But the arithmetic doesn't punish the messy version. Because the curve doesn't check your calendar. It checks your total.

Last practical one. Can he trust the number at all? It's a watch on a wrist. He stirs a pan, it awards him steps. He pushes a trolley, it awards him nothing.

And both of those happen. Genuinely. Wrist counters misread stirring, clapping, trolleys, prams. So the count is an estimate. And on any single day it can be off by hundreds. But here's why that doesn't sink anything we've said. The errors repeat. The pan-stirring inflation happens every week the same way. And what the evidence rewards is change against your own baseline. Add a real after-dinner walk, and the number goes up by that walk, whatever the baseline error was. Because you're not competing with the Harvard women. You're competing with your own last month, measured on the same wrist by the same flaws. And for that race, the watch is a perfectly good referee. It's only as an absolute judge that it lies. Which, conveniently, is the one job we've already fired it from.

So. Monday morning. Marcus, the watch, the unopened trainers by the door. What are the take-homes?

Three things. And they're all small. First. Change the goal in the settings. Tonight. Take your current average, whatever it is, and add a thousand. That's the target with actual evidence behind it. Beat it for a few weeks. Then add another thousand. The curve rewards every one of those rungs, and for most people over sixty it quietly flattens around 6000 to 8000, so there's a finish line in sight that isn't 10000. Second. Stop auditing your pace. Volume first. The dog-walking amble and the station march land in the same column, and the column is the thing that counts. And third. The general rule hiding under tonight's specific one. When a health number feels like a verdict, ask where it came from. Because some numbers come from curves. And some come from slogans. The ones from curves bend kindly at the bottom. It's only the slogans that deal in pass and fail.

Does it matter when the steps happen? He's heard they have to come in proper walks. Twenty minutes minimum, or they don't count.

They count. And this is folklore with an expiry date on it, because older American guidance used to wrap activity into ten-minute parcels before anything counted. And the 2018 revision stopped counting parcels. Moving more and sitting less, their words, benefits nearly everyone, and the minutes add up however they arrive. Pacing on the phone. The far end of the car park. Stairs instead of the lift. The accelerometer studies couldn't care less whether steps came in ceremonial walks or in crumbs. A step is a step. Which is liberating for a man like Marcus, because 2000 extra steps sounds like a regime, but it's actually six small decisions scattered through an ordinary day.

Make the first month concrete for him, though. Because change the goal in the settings is one evening. What do the next four weeks actually look like?

Smaller than he fears. Week one, nothing changes except the target. His average is 4000, so the ring now closes at 5000, and the only project is finding his thousand. And it should be attached to something that already happens. The kettle, the school run, the evening news. A walk that needs a decision every day dies by November. A walk that's welded to the news bulletin survives. Week two, same target. Let it get boring. Boring is the goal state. Weeks three and four, if the ring's been closing without heroics, nudge to 6000. And if it hasn't, don't. Hold at 5000 until it's dull. Because the curve will still be there in the spring. What can't be recovered is the version of Marcus who quit in week two because the plan was built for somebody fitter. The evidence rewards the tortoise here. Properly, numerically rewards it.

And the bigger take-home, above the watch.

Don't let a default define failure for you. Because somewhere a device has shipped you a target that was never measured on anyone like you. The step evidence is unusually cheerful once you actually read it. Because it says start anywhere. It says the first improvements are the cheapest and the most valuable. And it says the gap between you and benefit is not 10000 of anything. It's a walk after dinner, started this week, continued next week. The ring was never the finish line. The kerb outside your door is the starting line, and it's a lot closer.

Three things to ask your doctor

  • Given my joints and my heart, is there any reason I should not walk more than I do now?
  • What would a realistic daily step target be for me, starting from my current average?
  • I cannot manage long walks. Do several short ones spread through the day count the same?

Sources

  1. 1Tudor-Locke C, Bassett DR Jr. How many steps/day are enough? Preliminary pedometer indices for public health. Sports Med. 2004;34(1):1-8. doi:10.2165/00007256-200434010-00001. PMID 14715035.The 10,000 steps-a-day figure can be traced to Japanese walking clubs and a business slogan rather than to evidence, is a reasonable description of an active healthy adult but may not be sustainable for older adults or the chronically ill
  2. 2Lee IM, Shiroma EJ, Kamada M, Bassett DR, Matthews CE, Buring JE. Association of step volume and intensity with all-cause mortality in older women. JAMA Intern Med. 2019 Aug 1;179(8):1105-1112. doi:10.1001/jamainternmed.2019.0899. PMID 31141585.In older women (mean age 72), roughly 4400 steps a day was associated with 41 percent lower mortality than roughly 2700, with benefit levelling at about 7500 steps a day, and stepping intensity mattered little after total steps
  3. 3Saint-Maurice PF, Troiano RP, Bassett DR Jr, et al. Association of daily step count and step intensity with mortality among US adults. JAMA. 2020 Mar 24;323(12):1151-1160. doi:10.1001/jama.2020.1382. PMID 32207799.In a representative sample of US adults, taking 8000 steps a day versus 4000 was associated with half the all-cause mortality (HR 0.49), 12000 with still lower (HR 0.35), and step intensity was not significant after total steps
  4. 4Paluch AE, Bajpai S, Bassett DR, et al. Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. Lancet Public Health. 2022 Mar;7(3):e219-e228. doi:10.1016/S2468-2667(21)00302-9. PMID 35247352.Across 15 cohorts and 47,471 adults, mortality fell progressively with more daily steps, levelling at about 6000-8000 steps for adults 60 and older and 8000-10,000 for younger adults
  5. 5Banach M, Lewek J, Surma S, et al. The association between daily step count and all-cause and cardiovascular mortality: a meta-analysis. Eur J Prev Cardiol. 2023 Dec 21;30(18):1975-1985. doi:10.1093/eurjpc/zwad229. PMID 37555441.Across 17 cohorts and 226,889 people, each 1000-step increment was associated with 15 percent lower all-cause mortality and each 500-step increment with 7 percent lower cardiovascular mortality, with benefit beginning above about 3867 steps (all-cause) and 2337 steps (cardiovascular)
  6. 6Piercy KL, Troiano RP, Ballard RM, et al. The Physical Activity Guidelines for Americans. JAMA. 2018 Nov 20;320(19):2020-2028. doi:10.1001/jama.2018.14854. PMID 30418471.The US Physical Activity Guidelines recommend 150-300 minutes a week of moderate activity (or 75-150 vigorous) plus muscle-strengthening, set no step target, and note that the least active benefit most from even modest increases
  7. 7Bull FC, Al-Ansari SS, Biddle S, et al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. 2020 Dec;54(24):1451-1462. doi:10.1136/bjsports-2020-102955. PMID 33239350.The WHO 2020 guidelines recommend 150-300 minutes a week of moderate activity for adults, set no step target, reaffirm that some physical activity is better than none, and recommend reducing sedentary behaviour while noting evidence was insufficient to quantify a threshold

How this was made

Information, not medical advice. Every claim in this episode has a published source, listed above. It is not a substitute for advice about your own treatment — for that, talk to your doctor. Published 2026-10-26.