The number on most step counters has an odd history. The 10,000-step target is usually traced to a Japanese pedometer marketed in the 1960s, not to any study of health outcomes. It stuck because it was round and memorable, and for decades it has been the default goal on watches and phones.
Researchers have spent the past ten years testing what step counts actually predict. The largest summary of that work so far, published in The Lancet Public Health1 in 2025, suggests the useful target for most adults is lower.
What the review measured
A team led by Melody Ding at the University of Sydney gathered prospective studies published between January 2014 and February 2025 in which adults wore a device to count their steps and were then followed for health outcomes. They found 57 studies drawn from 35 cohorts, and 31 of those studies had data suitable for pooling.
Instead of looking only at deaths or heart disease, as earlier reviews had, they examined a wider set of outcomes. These included cardiovascular disease, cancer, type 2 diabetes, dementia, depressive symptoms, physical function and falls.
The numbers at 7,000 steps
The comparison point was 2,000 steps a day, roughly the level of someone who moves very little. Compared with that baseline, 7,000 steps a day was associated with2:
- a 47 percent lower risk of death from any cause
- a 25 percent lower risk of developing cardiovascular disease
- a 37 percent lower risk of dying from cancer
- a 14 percent lower risk of type 2 diabetes
- a 38 percent lower risk of dementia
- a 22 percent lower risk of depressive symptoms
- a 28 percent lower risk of falls
For deaths from any cause, cardiovascular disease, dementia and falls, the curve was not a straight line. Benefit rose steeply at first and then began to flatten at around 5,000 to 7,000 steps4. For outcomes such as cancer and diabetes, the relationship was closer to linear, so more steps kept helping at about the same rate.
So is 10,000 pointless?
No. The authors were clear that 10,000 steps was linked with somewhat lower risk than 7,000 for several outcomes. The extra gain was simply smaller. Dementia risk, for example, fell 38 percent at 7,000 steps and only a few points more at 10,000, according to the university's summary3.
Melody Ding, the lead author, put it plainly. For people who already reach 10,000, there is no reason to cut back. For the many people who are far from it, 7,000 delivers most of the measured benefit and is a goal they are more likely to reach.
The review also found that moving from 2,000 to 4,000 steps a day was associated with real improvements. That matters for people who are recovering from illness, managing arthritis or simply starting from a low baseline. A goal that feels out of reach often leads to no change at all.
Making it work in a New England winter
Step counts in Rhode Island tend to fall from November through March, when sidewalks ice over and daylight ends before most workdays do. A few strategies hold up in practice.
Short walks add up. A ten-minute walk after each meal comes to roughly 3,000 steps for most people and has the added benefit of blunting the rise in blood sugar after eating. Indoor options count as much as outdoor ones, whether that is a mall loop, a gym track or laps of a large store. On clear days, paved paths such as the East Bay Bike Path offer long, flat stretches away from traffic.
It also helps to know your real starting point. Look at the past two weeks on your phone or watch rather than a single day, then aim to add 1,000 steps to that average and hold it for a few weeks before adding more.
What the numbers cannot tell you
All of these findings come from observational studies. People who walk more may also sleep better, eat differently or have fewer illnesses that limit movement in the first place. The authors adjusted for many of these factors, but no adjustment is perfect.
Even so, the direction of the evidence is consistent, the risks of walking are low, and the target is concrete. For most adults, 7,000 steps is a better daily goal than a number chosen for a pedometer box.
