Daily step count and mortality

Two large meta-analyses agree that more daily steps track with lower mortality, that the famous round number is arbitrary, and that the curve flattens earlier than most people think.

goodWell supported, with the usual caveats about effect size.Revised 28 Sept 2026

Definition

Daily step count: the number of steps recorded per day, usually by a waist- or wrist-worn device. It is a crude measure of ambulatory activity — it says nothing about intensity, terrain, or what the rest of the day looked like — and that crudeness is part of why the findings below are interesting. Even this blunt instrument predicts mortality well.

The association

The best summary comes from Paluch and colleagues, who pooled participant-level data from 15 cohorts: 47,471 adults, median follow-up 7.1 years, 3,013 deaths. Split into quartiles by daily steps, with medians of roughly 3,500, 5,800, 7,800 and 10,900 steps, the adjusted hazard ratios for all-cause mortality against the least-active quartile were 0.60, 0.55 and 0.47. Read that again: the second quartile — people averaging under 6,000 steps — already carried about 40% lower mortality risk than the least active group.

A second meta-analysis by Banach and colleagues (17 studies, nearly 227,000 participants) found the same shape: each 1,000-step increment associated with roughly 15% lower all-cause mortality, each 500-step increment with roughly 7% lower cardiovascular mortality, with benefits appearing from as few as 2,500–4,000 steps a day and continuing up to the 20,000 steps a day their data covered.

The shape of the curve

The relationship is nonlinear, and the nonlinearity is the practical point. Risk falls steeply at the bottom and then flattens: in Paluch’s analysis the curve levelled off around 6,000–8,000 steps a day for adults over 60 and 8,000–10,000 for younger adults. The biggest return on walking sits between “almost nothing” and “a moderate amount”. Past the plateau, extra steps are fine — Banach found no upper limit where the association reversed — but they buy little additional mortality benefit.

Two corollaries worth stating plainly. First, 10,000 steps has no special status. Paluch’s authors note explicitly that little evidence supports that threshold; it survives because it is round, not because the data put it there. Second, stepping rate (cadence) added little once total volume was accounted for. Brisk is nice; more is what matters.

What the data do not show

Everything above is observational. Step counts were measured, not assigned, and people who walk more differ from people who walk less in ways no adjustment fully captures — health, wealth, neighbourhood, habits. No randomised trial has shown that prescribing steps extends life, and this literature cannot say whether adding 3,000 steps to a sedentary 65-year-old’s day produces the risk differences the cohorts display. Devices, wear locations and wear-time rules also varied between studies, so the exact step numbers are softer than they look. Treat the thresholds as zones, not lines.

Practical reading

The WHO’s floor — 150–300 minutes of moderate activity a week — remains the reference target. Steps are best read as a compliance instrument for that target, not a rival to it: a tracker turns an abstract weekly minute goal into a number you can check before dinner. For a mostly sedentary older adult, the evidence suggests the first few thousand steps above baseline are the ones that count most, which is a more encouraging message than any round number.