The 10,000-Step Target Came From a Name

A single worn walking shoe on a dark surface, lit in olive-chartreuse

In a meta-analysis pooling fourteen studies and 161,176 adults, walking 7,000 steps a day instead of 2,000 was associated with a 47% lower rate of death from any cause — hazard ratio 0.53, 95% CI 0.46 to 0.60.^[1] Going from 7,000 to 10,000 moved that to 0.52.

Almost everyone has been told to hit 10,000. Almost nobody is told that the great majority of the benefit has already arrived by 7,000 — or that the number 10,000 was, in all likelihood, a product name.

The 60-second answer

The mortality benefit of daily walking rises steeply at low step counts and flattens well before 10,000. Across the main dose-response analyses, the inflection sits somewhere around 5,000 to 8,000 steps a day, varying by age and by which outcome you look at.^[1]^[5] The 2026 American College of Cardiology scientific statement on exercise prescriptions says the same thing in clinical language — that meaningful benefit occurs well below 10,000 steps per day, with roughly 5,000 to 7,000 steps associated with lower cardiovascular and all-cause mortality for many adults — and adds two things worth knowing: resistance training should complement aerobic exercise for most adults, and routine exercise testing is not required before starting exercise for most asymptomatic or clinically stable people.^[3] The practical upshot: if you are currently doing 3,000 steps a day, getting to 6,000 buys you most of what is available. The last 3,000 to reach 10,000 is worth roughly another percentage point in relative risk. Do it if you enjoy it. Don't treat failing to reach it as failure.

Where 10,000 came from

The honest answer is that nobody can point to a trial.

The best source for the origin is a 2019 analysis in JAMA Internal Medicine, which states that the origin of the 10,000-step goal "is unclear" and that it "likely derives from the trade name of a pedometer sold in 1965 by Yamasa Clock and Instrument Company in Japan" — the Manpo-kei, which the authors translate as "10,000 steps meter."^[4] A 2004 review in Sports Medicine reaches the same place from a different direction, saying the 10,000 figure "can be traced to Japanese walking clubs and a business slogan 30+ years ago."^[2]

Note the hedge, because it belongs there: likely derives. The researchers who report this do not claim to have documented it. The sources behind the claim are a Japanese-language walking journal from 1997 and that 2004 narrative review. There is a good deal of colourful detail circulating about the engineer, the marketing campaign and a Tokyo Olympics connection, and I would leave all of it alone.

What is solid is the negative: the figure did not come from a dose-response study, because the dose-response studies came decades later. And when they arrived, they did not land on 10,000.

What the dose-response data actually show

Four large analyses, and they agree more than they disagree.

Paluch and colleagues, 2022 pooled 15 international cohorts — 47,471 adults, 3,013 deaths, median follow-up 7.1 years. Compared with the lowest quartile (median 3,553 steps/day), hazard ratios for all-cause mortality were 0.60 (95% CI 0.51–0.71) at a median of 5,801 steps, 0.55 (0.49–0.62) at 7,842, and 0.47 (0.39–0.57) at 10,901. Crucially, they looked at where the curve flattens by age: progressively lower risk until approximately 6,000–8,000 steps per day in adults aged 60 and older, and approximately 8,000–10,000 steps per day in adults younger than 60.^[5]

Figure 1

More steps, lower mortality — and where the curve flattens

Reference categoryHigher step quartiles

The steepest drop is between the first and second quartile — from about 3,500 to about 5,800 steps a day. After that the curve flattens. Paluch and colleagues report progressively lower risk until roughly 6,000–8,000 steps a day in adults aged 60 and over, and roughly 8,000–10,000 in adults under 60. Bars show adjusted hazard ratios with 95% confidence intervals; the first bar is the reference category.

Source: Paluch et al., Lancet Public Health 2022;7:e219–e228 (15 cohorts, 47,471 adults, 3,013 deaths).

Ding and colleagues, 2025 reviewed 57 studies from 35 cohorts and meta-analysed 31. Comparing 7,000 steps with 2,000, they found the 47% mortality reduction quoted above, plus cardiovascular disease incidence HR 0.75 (0.67–0.85), cardiovascular mortality 0.53 (0.37–0.77), cancer mortality 0.63 (0.55–0.72), dementia 0.62 (0.53–0.73), depressive symptoms 0.78 (0.73–0.83) and falls 0.72 (0.65–0.81). Cancer incidence did not reach significance at 0.94 (0.87–1.01), and type 2 diabetes came in at 0.86 (0.74–0.99). Their inflection points differ by outcome: about 5,391 steps for all-cause mortality, 5,422 for cardiovascular mortality and 4,794 for cancer mortality, but 7,802 for cardiovascular events and roughly 8,800 for dementia and falls. Their own conclusion is worth quoting: 10,000 steps a day "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]

Stens and colleagues, 2023 pooled 12 studies and 111,309 participants. A detectable benefit began at just 2,517 steps/day (adjusted HR 0.92, 0.84–0.999) — which is to say, barely moving at all beats not moving. Risk then fell progressively to an optimal dose at 8,763 steps/day (aHR 0.40, 0.38–0.43).^[6]

Banach and colleagues, 2023 pooled 17 cohorts and 226,889 participants and is the dissenting voice: a 1,000-step increment was associated with a 15% lower risk of all-cause mortality (HR 0.85, 0.81–0.91), and the authors describe the relationship as "more the better" above about 3,867 steps a day, with their highest quartile — a median of 11,529 steps — carrying 67% lower all-cause mortality than the reference.^[7]

So: four analyses, one of which finds benefit continuing well past 10,000 and three of which find it flattening between roughly 7,500 and 10,000. The claim "most of the benefit arrives well before 10,000" is safe across all four. The claim "there is nothing above 7,000" is not — Ding found further, smaller gains above 7,000 for several outcomes, and Paluch found no plateau below 8,000 in adults under 60.

Does walking faster help? Unclear, honestly. Paluch found that peak 30-minute stepping rate was still associated with lower mortality after adjusting for total steps (HR 0.67, 0.56–0.83), but time spent walking at 40 steps per minute or faster was not (1.12, 0.96–1.32), nor at 100 steps per minute or faster (0.86, 0.58–1.28).^[5] A 2019 analysis of 16,741 older women found that once total steps were accounted for, the intensity associations were attenuated and most were no longer significant; the authors concluded that stepping intensity "was not clearly related to lower mortality rates after accounting for total steps per day."^[4] Stens found that moving from a low to a moderate or high cadence was independently associated with lower mortality.^[6] Total volume does most of the work; whether cadence adds anything independent is genuinely unsettled.

The two other things cardiology just said

The ACC statement is not a trial, and I want to be clear about that — it is a consensus document summarising evidence, not new data.^[3] But two of its positions are worth knowing because they contradict common practice.

Resistance training is not optional. The statement says resistance work should complement aerobic exercise for most adults. The supporting evidence is better than most people realise: a meta-analysis of 16 cohort studies found that any muscle-strengthening activity was associated with 15% lower all-cause mortality (RR 0.85, 0.79–0.93), 17% lower cardiovascular disease (0.83, 0.73–0.93), 12% lower total cancer (0.88, 0.80–0.97) and 17% lower diabetes risk (0.83, 0.77–0.89).^[8]

The dose-response is the surprising part. Maximum benefit arrived at roughly 30 to 60 minutes a week — all-cause mortality bottomed out at RR 0.83 (0.79–0.86) at about 40 minutes per week, cardiovascular disease at 0.82 (0.76–0.90) at 60 minutes, and total cancer at 0.91 (0.85–0.97) at 30 minutes — with risk remaining below baseline out to roughly 130–140 minutes weekly, beyond which the data thin out.^[8] Forty minutes a week. Two short sessions. For diabetes risk the curve kept falling with more, so the "less is enough" finding is outcome-specific rather than a general rule against training hard. The honest caveat the authors attach to all of this is that they graded the certainty of the evidence as very low for mortality, cardiovascular disease and cancer, and low for diabetes — these are observational cohorts, not trials.

Figure 2

Resistance training: the curve bottoms out at about 40 minutes a week

Any muscle-strengthening activity vs none

Lowest risk, and the weekly dose it occurred at

Left: the simple any-versus-none comparison. Right: the lowest point of each non-linear dose-response curve, and the weekly volume at which it occurred. Benefit is maximal at roughly 30–60 minutes a week and the estimate remains below baseline out to approximately 130–140 minutes weekly, beyond which the data thin out — the authors do not report that benefit reverses above that, only that the evidence becomes sparse and the influence of higher volumes is unclear. They grade the overall certainty of this evidence as very low for mortality, cardiovascular disease and cancer, and low for diabetes.

Source: Momma et al., Br J Sports Med 2022;56:755–763 (16 prospective cohort studies).

You probably don't need a stress test first. The statement's own abstract says it discusses when exercise testing is useful "while recognizing that routine preparticipation testing is not required for most asymptomatic individuals," and ACC's summary adds that testing can still inform an individualised prescription when clinically appropriate.^[3] This matters because the practical effect of requiring a test is delay, and a false positive actively discourages someone from starting.

I would add the obvious clinical caveat the statement itself makes room for: if you have symptoms — chest discomfort, unusual breathlessness, exertional dizziness or palpitations — or known cardiac disease, that is a conversation with your physician before you start, not after. If you want somewhere sensible to begin, our performance guidance covers how to structure that.

The thing that outranks step counts

If you want the single number with the largest association with how long you live, it is not steps. It is cardiorespiratory fitness.

In 122,007 patients undergoing treadmill testing at the Cleveland Clinic, followed for a median of 8.4 years, being in the least-fit group carried an adjusted mortality hazard ratio of 5.04 (4.10–6.20) compared with elite fitness. In the same statistical model, smoking carried 1.41 (1.36–1.46), diabetes 1.40 (1.34–1.46) and coronary artery disease 1.29 (1.24–1.35).^[9] And there was no upper limit: elite fitness beat merely high fitness, HR 0.77 (0.63–0.95).

Figure 3

Low fitness sits beside the risk factors we screen for

Cardiorespiratory fitnessClinical risk factors

These are separate coefficients within one adjusted model, not a head-to-head comparison between risk factors, and this was a cohort referred for treadmill testing rather than a general population. What the figure supports is narrower than the headline version of it: low cardiorespiratory fitness is associated with mortality on the same order of magnitude as the risk factors clinicians routinely screen for and treat.

Source: Mandsager et al., JAMA Netw Open 2018;1:e183605 (122,007 patients, median follow-up 8.4 years).

A second study in 750,302 US veterans, followed a median of 10.2 years with 174,807 deaths, found the same shape — lowest mortality risk around 14 METs, and the least fit carrying roughly four times the risk of the extremely fit (HR 4.09, 3.90–4.20) — and concluded that being unfit carried greater risk than the cardiac risk factors they examined.^[10]

I want to be careful how that is read. Those risk-factor hazard ratios are separate coefficients in a model, not a head-to-head contest, so the sharper version of that sentence is a stronger claim than either paper makes. What both papers do support is this: low fitness is a risk factor of the same order of magnitude as the ones we screen for and treat aggressively, and it is the one nobody measures. It is also one of the more modifiable inputs to long-term healthspan.

Steps are a reasonable proxy and a very good place to start. Fitness is the thing steps are a proxy for.

What I would actually do

If you're under about 4,000 steps a day, the most valuable thing available to you is getting to 6,000. The curve is steepest exactly where you are.

If you're already around 7,000 to 8,000, you have most of the mortality benefit that walking offers. More is mildly better. The bigger returns now come from adding two short resistance sessions — 40 minutes a week total is where the curve bottoms out — and from doing something hard enough, often enough, to move your cardiorespiratory fitness.

And if you've been treating 10,000 as a pass mark you keep failing, stop. It was a pedometer's name.

The practical bottom line

Under 4,000 steps a day? Getting to 6,000 captures most of the available mortality benefit. That is the steepest part of the curve.
Around 7,000–8,000? You already have most of it. The last 3,000 steps to 10,000 are worth roughly one percentage point of relative risk.
Pace is secondary to volume. Get the steps; don't agonise over cadence.
Add resistance training. Thirty to sixty minutes a week — two short sessions — is where the mortality curve bottoms out.
Cardiorespiratory fitness is the stronger signal. Steps are a proxy for it, not a substitute.
Most asymptomatic, clinically stable adults do not need an exercise test before starting. Cardiac symptoms or known heart disease are the exception — speak to your physician first.

Frequently asked questions

How many steps a day do I actually need?

Most of the mortality benefit arrives between about 5,000 and 8,000 steps a day. A 2025 meta-analysis found 7,000 steps associated with a 47% lower rate of death from any cause compared with 2,000, and moving from 7,000 to 10,000 changed the hazard ratio only from 0.53 to 0.52. Inflection points differ by age — roughly 6,000–8,000 for adults over 60 and 8,000–10,000 for those under 60.

Where did the 10,000-step rule come from?

Not from research. The most-cited account, in JAMA Internal Medicine, says the origin is "unclear" and "likely derives" from the trade name of a pedometer sold in Japan from 1965 — the Manpo-kei, or "10,000 steps meter." A 2004 review traces it to Japanese walking clubs and a business slogan. The dose-response studies came decades later and did not land on 10,000.

Is walking faster better than walking more?

Total volume does most of the work. Some analyses find an independent benefit from faster cadence after adjusting for total steps; others find that intensity associations largely disappear once total steps are accounted for. The evidence is genuinely unsettled, so the practical advice is to get the steps and not worry much about pace.

Do I need to lift weights too?

The 2026 ACC scientific statement says resistance training should complement aerobic exercise for most adults. A meta-analysis of 16 cohort studies found any muscle-strengthening activity associated with 15% lower all-cause mortality, with maximum benefit at roughly 30–60 minutes a week — around 40 minutes for mortality specifically. The authors rate the certainty of that evidence as very low, so treat it as a reason to do some, not a precise dose.

Do I need a stress test before I start exercising?

The 2026 ACC statement's position is that routine preparticipation testing is not required for most asymptomatic individuals, though testing can help individualise a prescription where clinically appropriate. If you have cardiac symptoms or known heart disease, speak to your physician first.

References

Disclaimer: This article is for education and is not medical advice. If you have cardiac symptoms, known heart disease, or have been advised to limit activity, speak to your own physician before changing your exercise routine.

Back to Blog

The 10,000-Step Target Came From a Pedometer's Name