For adults under 60, mortality risk keeps dropping up to about 8,000–10,000 steps a day; for adults 60 and older, the benefit plateaus around 6,000–8,000. That is the core finding of a 2023 meta-analysis by Paluch and colleagues in The Lancet Public Health, which pooled data on roughly 47,000 adults. More steps beat fewer steps at every level — but the steep part of the curve happens early, and the famous 10,000 figure is a floor of benefit, not a finish line you must reach.
This site publishes information, not medical advice. If you have a heart condition, joint pain, or any medical concern, talk with a clinician before you change how much you walk.
Where did 10,000 steps come from?
Not from a clinical trial. The number traces to a 1960s Japanese pedometer campaign built around a device called the manpo-kei — literally "10,000-step meter." It was marketing that stuck, and for decades it drifted through fitness culture without an evidence base.
The actual research asks a different question: at what step count do health risks stop falling fast? The 2023 Lancet Public Health meta-analysis sorted the answer by age. Younger adults kept gaining benefit up to about 10,000 steps. Older adults hit their plateau earlier, around 6,000–8,000. The dose–response curve is real, but it bends.
How few steps still help?
Fewer than you might think. A 2019 study by I-Min Lee and colleagues in JAMA Internal Medicine followed about 17,000 older women and found that women averaging roughly 4,400 steps a day had significantly lower mortality than those averaging about 2,700 — a 41 percent difference. Benefit climbed steadily until about 7,500 steps, then leveled off.
Read that as good news, not a ceiling. The lesson from both studies is that the jump from 2,500 to 5,000 steps buys far more health than the jump from 9,000 to 12,000. If you currently move little, your first thousand extra steps are the most valuable ones you will take.
| Group | Steps where mortality benefit plateaus | Source |
|---|---|---|
| Adults 60+ | ~6,000–8,000 | Paluch et al., 2023 meta-analysis |
| Adults under 60 | ~8,000–10,000 | Paluch et al., 2023 meta-analysis |
| Older women | ~7,500 | Lee et al., 2019 |
Does walking speed matter?
Some, but far less than volume, and the evidence is thinner. The Paluch meta-analysis grouped findings by pace and found that higher stepping rates were associated with better outcomes — but the sources did not establish that intensity replaces volume, only that it may add to it. A practical read: walk at a pace that still lets you talk, and let total daily steps carry the main load of the benefit.
One honest limit: these are observational cohorts. They show strong associations between step counts and mortality, not laboratory proof that steps alone cause the effect. People who walk more differ from people who walk less in other ways too. Still, the association held across multiple countries and devices, which is about as solid as observational data gets.
Related stories: Exercise and Sleep Quality: What the Research Actually Shows · Why Group Exercise Keeps People Coming Back — What the Research Shows.
How do you actually add steps?
Attach walking to things you already do. Willpower is a bad strategy; friction removal is a good one.
- Measure first. Wear your phone or a tracker for three normal days and note your baseline.
- Add one fixed walk. Ten minutes after lunch is the easiest slot to defend — roughly 1,000–1,200 steps for most paces.
- Move the car once. Park at the far end of the lot, get off transit a stop early, take the stairs for one flight.
- Break up sitting. Stand and walk two minutes every hour of desk time.
- Recheck in two weeks. If the average moved up 1,000–2,000 steps and stayed there, add the next block.
The Centers for Disease Control and Prevention still frames the overall physical activity target as 150 minutes of moderate activity per week — walking counts fully toward it, per CDC guidance.
What if you can only manage 3,000 steps?
Start there and defend it. The dose–response curve means 3,000 beats 2,000, and 4,000 beats 3,000, with no cutoff where benefit begins. Missed days cost almost nothing in the long run — consistency across weeks is what the cohorts capture. If pain stops you from walking, that is a clinician conversation, not a discipline problem.
The bottom line
Treat 10,000 as a slogan, not a prescription. The evidence-backed picture: most adults capture the large majority of the mortality benefit somewhere between about 6,000 and 10,000 steps depending on age, and every step above your current baseline counts. Pick a number you can hit on your worst day, then build up from it.
FAQ
Is walking 10,000 steps a day bad for you?
No. For healthy adults it is simply more than most people need for the measured mortality benefit. The 2023 Lancet Public Health meta-analysis found benefit continuing up to the plateau for younger adults, with no signal of harm from higher step counts in the data. Joint pain or a heart condition changes the conversation — clear higher volumes with your clinician first.
Do steps from housework or shopping count?
Yes. The big cohort studies counted total daily steps from any source, not gym walking specifically. Carrying groceries, pacing during phone calls, and yard work all register the same way on a pedometer. What matters for the measured association is the daily total, so incidental movement genuinely counts toward it.
How many steps equal 30 minutes of walking?
Roughly 3,000–4,000 steps for most adults at a moderate pace, depending on stride length. That single brisk half-hour would move a sedentary person close to the 6,000–8,000 range where the benefit plateaus for older adults. The CDC's 150 weekly minutes can be built entirely from blocks like this.
What is a good step goal for a 70-year-old?
The Paluch meta-analysis puts the mortality-benefit plateau for adults 60 and older around 6,000–8,000 steps, and the 2019 JAMA Internal Medicine data leveled near 7,500 for older women. A practical target: whatever your current average is, plus 1,000–2,000 steps, then reassess. Any movement above a very low baseline still helps.