How Many Daily Steps Reduce Mortality: Evidence from Six Meta-Analyses

Where did the "10,000 steps" figure come from?

In 1965, a Japanese company released a pedometer called "manpo-kei" - literally "ten-thousand-step meter." The device was timed to coincide with the 1964 Tokyo Olympics. No epidemiological or physiological research stood behind this figure - it was a marketing slogan.

Five to six decades later, "10,000 steps" became firmly embedded in fitness tracker recommendations and popular guides. What does the data actually say?

Meta-analyses: a dose-dependent relationship

Banach et al. (European Journal of Preventive Cardiology, 2023; 17 prospective cohorts, 226,889 participants, mean age 64, median follow-up 7.1 years) - the most comprehensive quantitative meta-analysis as of 2023. Key results:

• Every +1,000 steps per day: all-cause mortality HR=0.85 (95% CI: 0.81-0.91) - 15% risk reduction.
• Every +500 steps: cardiovascular mortality HR=0.93 - 7% reduction.
• Fourth quartile (~11,529 steps/day) vs. first (~3,867 steps/day): all-cause mortality 67% lower.
• The relationship is nonlinear: the steepest risk reduction occurs between 5,000 and 10,000 steps/day; the curve flattens with further increases.

Ding et al. (The Lancet Public Health, 2025; systematic review of 57 studies; meta-analysis of 31 studies; n>160,000; search through February 2025) - the most current data with a wide spectrum of outcomes. At 7,000 steps/day compared to a sedentary level (~2,000 steps/day):

• All-cause mortality: reduction of 47% (14 studies).
• Cardiovascular morbidity: reduction of 25% (6 studies, n=111,349).
• Cardiovascular mortality: reduction of 47% (3 studies, n=120,758).
• Dementia: reduction of 38%.
• Depression: reduction of 22%.
• Falls: reduction of 28%.

At 10,000 steps/day, the mortality risk reduction was approximately 48% - a minimal additional benefit compared to 7,000 steps.

Minimum threshold: at what step count does benefit begin?

Ahmad et al. (Journal of the American College of Cardiology, 2023; 12 studies, 111,309 participants): significant reduction in all-cause mortality begins at just 2,517 steps/day vs. 2,000 (HR=0.92). The optimal point on the dose-response curve is 8,763 steps/day (HR=0.40 - 60% lower risk compared to ~2,000 steps). Above 10,000 steps, no statistically significant additional reduction was found.

Saint-Maurice et al. (JAMA, 2020; 4,840 US adults aged 40+, mean follow-up 10.1 years): 8,000 steps/day vs. 4,000 - HR=0.49 (95% CI: 0.44-0.55), mortality 51% lower; 12,000 steps vs. 4,000 - HR=0.35, 65% lower. Key finding: walking pace (step intensity per minute) after adjustment for total step count was not an independent predictor of mortality. Volume is the primary factor, not speed.

Age differences: the plateau shifts

Paluch et al. (The Lancet Public Health, 2022; harmonized meta-analysis of 15 international cohorts with individual participant data) identified age-specific features of the dose-response curve:

• Participants under 60: risk plateau at around 8,000-10,000 steps/day.
• Participants over 60: plateau at around 6,000-8,000 steps/day.

This means that for older adults, reaching 6,000-7,000 steps per day can yield risk reductions comparable to 10,000 steps in younger people. Setting realistic goals is more effective than maximizing a number.

Caveats: all of this is observational data

All six meta-analyses are based on prospective cohort studies. In none of them were participants randomly assigned to physical activity levels. This means that the observed associations, despite their consistency and effect size, do not prove causation in the strict sense.

The main threat is reverse causality: people with chronic diseases walk less because of their illness, not the other way around. Most studies address this through excluding deaths in the first 1-2 years of follow-up, adjusting for baseline health status, and sensitivity analyses - but fully neutralizing this without a randomized design is impossible.

Nevertheless, the consistency of data across six independent meta-analyses covering different countries, age groups, and activity measurement methods makes a biologically plausible causal role of physical activity highly likely.