Steps per day: why "10,000 steps" does not come from the evidence
What the data say about steps and mortality, where the curve appears, why it saturates depending on age, and why 10,000 is not evidence-based.
“Walk 10,000 steps a day” is probably the most repeated health recommendation and the least derived from a study. This article does not reconstruct its origin, which is commercial, but what the data say.
Where the number comes from
A meta-analysis from the Steps for Health group pooled 15 international cohorts (7 published and 8 unpublished), with 47,471 adults aged 18 or over, 3,013 deaths and a median follow-up of 7.1 years [1]. The sample has a mean age of 65.4 years and more than two thirds women [1].
| Steps/day quartile | Median | Mortality vs. the lowest quartile |
|---|---|---|
| Q1 (lowest) | 3,553 | Reference |
| Q2 | 5,801 | 40 % less |
| Q3 | 7,842 | 45 % less |
| Q4 (highest) | 10,901 | 53 % less |
The last row is the one that matters: the median of the highest quartile is 10,901 steps, and the number repeated everywhere is 10,000. That number did not come out of a study: the meta-analysis itself opens by saying there is little evidence to support that recommendation [1], and a later review places the start of the benefit at around 3,000 steps/day [4].
Where the benefit saturates
Here is the figure that changes the advice. When adjusted with splines, the risk of mortality falls progressively up to around 6,000-8,000 steps/day in people over 60, and up to 8,000-10,000 steps/day in those under 60 [1].
In other words: the curve flattens out, and the flattening point depends on age. For a 65-year-old person, the difference between 7,000 and 12,000 steps is much smaller than the difference between 2,000 and 7,000.
A 2024 umbrella review, more recent and broader (11 reviews with meta-analyses and 14 cohorts), reaches a compatible conclusion: the association is not linear, with a protective threshold around 3,000 steps/day and a 9 % reduction in risk for every additional 1,000 steps/day [4]. The optimal dose still depends on age and sex [4].
The detail almost nobody mentions
The study also looked at whether time spent walking at a specific cadence mattered, adjusting for the total number of steps [1]:
- 30 minutes of peak movement a day: around 33 % less mortality.
- 60 minutes of peak movement: around 33 % less mortality.
- Walking at 40 or at 100 steps per minute: no association with the outcome.
Translation: the specific cadence was not an independent factor once the total steps were taken into account [1]. These are observational associations, not proof of causality.
Why the WHO does not talk about steps
The WHO fact sheet states the target in minutes, not in steps: at least 150 min/week of moderate activity [2]. The range of 150-300 min/week (or 75-150 vigorous) is the one from the 2020 guideline [3].
And the WHO is explicit about the framework: some activity is better than none. That is why the correct translation of this study into public health language is not “walk 10,000 steps” but “more steps than fewer, and the benefit saturates depending on age” [1][3].
A nuance about “peak movement”
What the study did not find was that walking at a specific cadence (40 or 100 steps per minute) changed the outcome once adjusted for the total [1]. That nuance is small but it matters: the recommendation to “walk fast” is not the same as “walk more”.
And what the study did find, adjusting for the total number of steps, opens a door: not all activity minutes have to be equivalent. There can be days with a lot of movement spread thin and days with little, and it is the whole that is associated with the outcome [1].
In summary
- The 10,000 does not come from any study: the meta-analysis of 47,471 adults opens by saying there is little evidence for that recommendation, and a 2024 review places the protective threshold at around 3,000 steps/day [1][4].
- The benefit drops progressively up to 6,000-8,000 steps in people over 60 and 8,000-10,000 in those under 60 [1].
- Cadence (40 or 100 steps/min) had no association of its own once adjusted for total steps [1].
- The WHO talks about minutes, not steps, and its message is about relative progress [2][3].
Counting steps is fine as a tool for knowing whether you move. As a health target, the figure the evidence supports is “more than before”. And be careful about reading the most active quartile as a promise: it is an observational cohort, not causality.
More in: the running page, with the figures and the sources of the Running section.
Related: how much weight you can lose in a week, which separates the figures in the guidelines from internet promises, and cardio or weights for weight loss.
Related articles
Sources for this article (5) 5 references
- Paluch AE, Bajpai S, Bassett DR, et al. (2022) Daily steps and all-cause mortality: a meta-analysis of 15 international cohorts. The Lancet Public Health 7(3):e219-e228. See source (opens in a new tab) DOI: 10.1016/S2468-2667(21)00302-9 Metaanálisis de 15 cohortes (47.471 adultos): el riesgo baja hasta 6.000-8.000 pasos/día en mayores de 60 y 8.000-10.000 en menores de 60; el beneficio se satura según la edad Licence: Artículo de acceso abierto en PMC (Elsevier, The Lancet Public Health). Se cita con enlace. Consulted:
- Organización Mundial de la Salud (OMS) (2024) Physical activity — hoja de datos. Organización Mundial de la Salud (OMS). See source (opens in a new tab) Recomendación oficial de la OMS (actividad física en adultos) Licence: OMS: el contenido de esta web está disponible bajo licencia CC BY-NC-SA 3.0 IGO. Se cita con enlace. Consulted:
- Bull FC, Al-Ansari SS, Biddle S, et al. (2020) World Health Organization 2020 guidelines on physical activity and sedentary behaviour. British Journal of Sports Medicine 54(24):1451-1462. See source (opens in a new tab) DOI: 10.1136/bjsports-2020-102955 Guía de la OMS 2020 (recomendaciones con GRADE; texto libre en PMC) Licence: Artículo de acceso abierto (BMJ, CC BY 4.0). Guía oficial en PDF: who.int/publications/i/item/9789240015128. Se cita con enlace. Consulted:
- Rodríguez-Gutiérrez E, Torres-Costoso A, Del Pozo Cruz B, et al. (2024) Daily steps and all-cause mortality: An umbrella review and meta-analysis. Preventive Medicine 189:108047. See source (opens in a new tab) DOI: 10.1016/j.ypmed.2024.108047 Revisión paraguas y metaanálisis de 2024 (11 revisiones y 14 cohortes): asociación no lineal, con un umbral protector en torno a 3.000 pasos/día Licence: Artículo de revista con derechos de autor. Se cita con enlace al editor y al registro PubMed 38901742; no se reproduce texto salvo cita breve verificable. Consulted:
- Pedisic Z, Shrestha N, Kovalchik S, et al. (2020) Is running associated with a lower risk of all-cause, cardiovascular and cancer mortality, and is the more the better? A systematic review and meta-analysis. British Journal of Sports Medicine 54(15):898-905. See source (opens in a new tab) DOI: 10.1136/bjsports-2018-100493 Revisión sistemática y metaanálisis de 14 estudios de 6 cohortes prospectivas (232.149 participantes; 25.951 muertes): correr se asocia a menos mortalidad por todas las causas, cardiovascular y por cáncer; la metaregresión NO halló tendencia dosis-respuesta Licence: Artículo de revista con derechos de autor («© Author(s) 2020. No commercial re-use», BMJ). La editorial devuelve 403 a bots y no hay PMC. Se cita con enlace al DOI; no se reproduce texto salvo cita breve verificable. Consulted:
How to cite
Pan y Hierro (2026). Steps per day: why "10,000 steps" does not come from the evidence. https://panyhierro.com/en/blog/steps-per-day/. Updated October 11, 2026.
Citation copied