7,000 Steps a Day Is the New 10,000: What a 57-Study Review Actually Found

For most of the fitness-tracker era, 10,000 steps has been the number on the screen, the number in the app notification, and the number people feel guilty about missing. It was never a scientific threshold. It traces back to a Japanese pedometer launched around the 1964 Tokyo Olympics and marketed as the manpo-kei, the “10,000-step meter”. The number was chosen because it sounded good.

In July 2025, a team led by researchers at the University of Sydney published the largest attempt yet to find out what the real dose-response curve looks like. Their systematic review and meta-analysis in The Lancet Public Health pooled 57 studies covering more than 160,000 adults, and looked not just at death rates but at heart disease, cancer, type 2 diabetes, dementia, depression and falls. The headline is simple: by 7,000 steps a day, you have captured most of what walking can do for you.

What the review measured

Earlier step studies mostly asked one question: how does daily step count relate to dying from any cause? The 2025 analysis widened that to nine outcomes and, crucially, used 2,000 steps a day as the reference point rather than zero. That makes the comparison realistic. Two thousand steps is roughly what someone gets moving around a house and an office without any deliberate walking.

The authors then modelled risk at each 1,000-step increment. Because the studies used wearable devices rather than questionnaires, the step counts are objective, which removes one of the biggest sources of error in older physical-activity research.

The numbers at 7,000 steps

Compared with people averaging 2,000 steps a day, people averaging 7,000 steps had:

Outcome	Difference at 7,000 vs 2,000 steps
All-cause mortality	47% lower
Cardiovascular disease (new cases)	25% lower
Cardiovascular mortality	47% lower
Cancer mortality	37% lower
Type 2 diabetes	14% lower
Dementia	38% lower
Depressive symptoms	22% lower
Falls	28% lower
Cancer (new cases)	6% lower

Two things stand out. The dementia figure is larger than most people would guess, and it fits with a growing body of work linking regular movement to brain blood flow and vascular health. The cancer-incidence figure is small, which is honest: walking is not a cancer-prevention tool in the way it is a heart-disease one.

Why the curve flattens

The relationship between steps and risk is not a straight line. It is steep at the bottom and shallow at the top. The review found that for most outcomes the benefit kept rising up to about 7,000 steps and then levelled off. For all-cause mortality specifically, risk continued to fall slightly up to 10,000 steps, but the extra gain between 7,000 and 10,000 was modest compared with the gain between 2,000 and 7,000.

This is consistent with the 2022 meta-analysis of 15 cohorts by Amanda Paluch and colleagues, also in The Lancet Public Health, which found mortality benefits plateaued at around 6,000 to 8,000 steps for adults over 60 and 8,000 to 10,000 for younger adults.

The low end is where the leverage is

If there is one practical message in the data, it is this: the first few thousand steps you add are worth far more than the last few. Moving from 2,000 to 4,000 steps a day was associated with a bigger drop in risk than moving from 8,000 to 10,000. For someone who is mostly sedentary, a 20-minute walk at lunch is not a token gesture. It is the most valuable part of the whole curve.

This also reframes the guilt problem. A target of 10,000 steps fails a lot of people, and when people miss a target they tend to abandon it. A target of 7,000 is reachable for most adults with a normal commute plus one deliberate walk, and the evidence says it is not a compromise.

Where this fits with official guidance

The World Health Organization and most national guidelines still express activity in minutes, not steps: 150 to 300 minutes of moderate-intensity activity a week, or 75 to 150 minutes of vigorous activity, plus muscle-strengthening work twice a week. Walking at a brisk pace counts as moderate. Roughly speaking, 7,000 steps a day with some of them brisk gets a typical adult close to the lower end of that minutes target, which is one reason the two frameworks agree so well.

Step counts are easier for most people to track, which is why the authors argue they should sit alongside minute-based targets in future guidance rather than replace them.

Does pace matter?

The 2025 review did not separate fast steps from slow ones in its main analysis, so it cannot answer this directly. Other cohort work, including UK Biobank analyses, suggests that cadence adds something on top of volume: the same number of steps taken at a brisker pace tends to track with slightly lower risk. The sensible reading is that total steps are the main driver and pace is a bonus, not a requirement. If you can only manage a slow walk, the slow walk still counts.

What the study cannot tell you

The limitations are worth stating plainly.

• Observational design. None of the 57 studies randomly assigned people to step counts. People who walk more are different in ways the statistics can only partly adjust for.
• Device differences. Studies used a mix of pedometers and accelerometers, which count steps slightly differently.
• Fewer studies for some outcomes. The mortality and cardiovascular estimates rest on many studies. The dementia, depression and falls estimates rest on fewer, so their confidence intervals are wider.
• Mostly older, mostly high-income populations. The findings may not transfer perfectly to younger adults or to other settings.

None of this undermines the core finding. It just means the exact percentages are estimates, not guarantees.

How to use this

If you currently log fewer than 4,000 steps a day, aim to add about 1,000 steps a week until you reach 7,000, then hold it. One thousand steps is roughly ten minutes of walking. Stacking it onto something you already do, such as a phone call, a school pick-up or the far end of a car park, makes it stick better than a standalone workout.

If you already walk 7,000 or more, the review suggests you have the bulk of the benefit. More is not harmful, and for longevity specifically there is a little extra to gain up to 10,000. But if the choice is between a few thousand extra steps and adding two sessions of strength training a week, the strength training likely fills a bigger gap.

The 10,000-step number will probably outlive this study, because it is memorable and it is already built into millions of devices. But the evidence now says something more useful than “walk more”: it says where the return on effort is highest, and that the bar is lower than most people assumed.