Physical activity

How Many Steps a Day Do You Really Need? What the Large Studies Show About 7,000 and 10,000

Physical activity · · Published 2026-11-17

A person in a grey hooded top, shorts and trainers walking away from the camera along a narrow cobbled street between pale buildings
Everyday walking counts: most of the steps in the studies were ordinary steps, not exercise sessions. Photo by Nilo Velez, CC0.

Ten thousand steps a day is probably the best-known health target in the world, built into phones, watches and workplace challenges. Yet no public health body set it, and for years there was little evidence behind it. That has changed. Since about 2019, large studies using step counters worn for a week have followed tens of thousands of adults for years. They give a clearer answer to a simple question: how many steps does it take to make a real difference to health, and is there a point beyond which more steps add little?

Where 10,000 came from

The figure is usually traced to Japan in the 1960s, when a pedometer was sold under a name that translates as "10,000-step meter". It was a round, memorable number, and it stuck. When the epidemiologist I-Min Lee and colleagues published a study of step counts in older women in 2019, they opened by noting that the 10,000 goal is commonly believed to be necessary for health but has limited scientific basis. As our guide to walking for health points out, the official guidelines do not mention steps at all: they ask adults for at least 150 minutes of moderate activity a week, plus muscle-strengthening on two days, as set out in the physical activity guidelines for adults.

That leaves step counters, which most people now carry, without an agreed target. The newer studies try to fill the gap.

What the large studies found

Older women: benefits start well below 10,000

Lee's study, published in JAMA Internal Medicine, followed 16,741 women with an average age of 72 who wore a step counter for a week. Over about four years, women averaging around 4,400 steps a day had a significantly lower death rate than those averaging about 2,700. The risk kept falling with more steps until roughly 7,500 a day, then levelled off.

Fifteen cohorts pooled: the plateau depends on age

In 2022, the Steps for Health Collaborative combined individual data from 15 studies covering 47,471 adults, published in The Lancet Public Health. People in the most active quarter, with a median of about 10,900 steps a day, had roughly half the risk of death of those in the least active quarter, who took about 3,550. The shape of the curve differed by age: the risk kept falling up to about 6,000–8,000 steps a day in adults aged 60 and over, and up to about 8,000–10,000 in younger adults, before flattening.

The broadest review: 7,000 as a realistic target

The most complete summary so far is a systematic review and dose–response meta-analysis led by Melody Ding at the University of Sydney, published in The Lancet Public Health in 2025. It gathered 57 studies from 35 cohorts and looked beyond death rates to heart disease, cancer, diabetes, dementia, depression and falls. Compared with 2,000 steps a day, 7,000 steps was associated with:

OutcomeLower risk at 7,000 vs 2,000 stepsCertainty of evidence
Death from any cause47%Moderate
Dementia38%Moderate
Death from cancer37%Moderate
Falls28%Very low
Cardiovascular disease25%Moderate
Depressive symptoms22%Moderate
Type 2 diabetes14%Moderate

For death from any cause, heart disease, dementia and falls, the curve bent at around 5,000 to 7,000 steps a day: below that, each extra thousand steps mattered a great deal; above it, the gains became smaller. The authors concluded that 10,000 steps remains a fine goal for people who are already active, but that 7,000 is associated with clinically meaningful improvements and may be a more realistic target for many.

What these numbers can and cannot tell you

All of these are observational studies: they compare people who walk more with people who walk less. Researchers adjust for age, smoking, existing illness and other factors, but they cannot rule out that some people walk less because they are already unwell. The Ding review lists residual confounding and the small number of studies for some outcomes among its limitations. The percentages describe associations across large groups, not a guaranteed effect for any one person.

Step counters also differ. A phone in a pocket misses the steps taken when it is on the desk; wrist devices can count arm movements; different brands give different totals for the same day. Compare your own counts with each other over weeks rather than with someone else's device.

Does pace matter?

Intuitively, brisk steps should be worth more than slow ones. The evidence is less clear than you might expect. In Lee's study of older women, measures of walking intensity were linked to lower mortality, but once the total number of steps was taken into account, most of those links were no longer significant. In the 2022 pooled analysis, a faster peak pace stayed modestly associated with lower risk after adjusting for total steps, while time spent walking at faster cadences did not. The practical reading: the total matters most, and some brisk walking is a sensible way to also meet the minutes-based guidelines. Research on walking cadence often uses about 100 steps a minute as a rough marker of moderate intensity.

How to raise your count

Start from where you are. Wear your counter for a normal week, note your typical daily total, and aim to add around 1,000 steps a day for a couple of weeks before adding more. If you take 3,000 steps now, reaching 5,000 is a bigger health change than going from 9,000 to 11,000.

Steps also play a part in weight management, but walking alone rarely produces large weight loss; its value lies in health benefits that hold at any weight. If you have a heart or lung condition, joint problems or have been inactive for a long time, check with a GP before a big increase.

This page is general information from published research and public health guidance, not personal medical advice.

Sources