How Many Steps You Really Need to Take Each Day, According to Science

The more you walk, the lower your risk of all-cause and cancer mortality, according to a new study, with the benefits leveling out once you reach 10,000 steps per day. So clearly, that is the number of steps to aim for—or is it?

Studies that compare health outcomes to step counts sound pretty compelling, because these days we all have step counters on our wrists or in our pockets. A step count number also sounds very concrete and precise: 10,000 steps equals health and happiness, and it gets measured for us automatically. Cool.

But already, I bet you’ve noticed some major caveats. Our bodies are messy meat machines, not neat step counters. If exercise is what matters, wouldn’t a cyclist have a lower step count than a runner, yet be just as healthy? For that matter, couldn’t a walker and a runner end up with similar step counts despite doing very different intensities of exercise that likely have different effects on the body?

On the other hand, there are some ways that step counts are a good way of tracking activity, so I don’t want to dismiss the idea entirely, even though I’m skeptical of how sharp a picture it provides. Step counts are higher for people who move around more in daily life (“incidental” activity, it’s sometimes called) even if they don’t do a lot of structured exercise. The steps are also counted automatically: You might not remember whether you were doing yard work for 20 minutes or 45, but your tracker probably has a good idea of how many steps you took.

There’s a further set of caveats: These studies are usually observational. They tell us that people who take more steps per day tend to be healthier. But is it cause or effect? People who are in poor health may have less energy to run errands and go for daily walks. And people who use wheelchairs or other mobility aids probably aren’t clocking step counts even when they do.

With that in mind, here are some step counts published in recent research, alongside some of their caveats.

For all-cause mortality and cancer mortality

This study found that the people who took 10,000 steps had lower risk than those who took 8,000, who in turn had lower risk than those who took 6,000, and so on. Step counts above 10,000 seemed to have a similar risk as 10,000. In other words, if this represents a true and causal relationship—which we can’t be sure of—increasing from 10,000 to 12,000 wouldn’t change your risk of cancer or death.

The 78,500 people tracked were from the UK, aged 40- to 79, and 97% white.

For dementia

This study found that participants’ risk of dementia decreased the more steps they got in, up to 9,800 per day, similar to the above study. (It was also conducted by the same team and drew from the same pool of subjects.) They also note that people who took 3,800 steps had about half the reduced risk of people who took 9,800, so perhaps that lower number would be a good target if you’re currently more sedentary. That said, this was also an observational study, and most of the participants were a bit young to start developing dementia.

For all-cause mortality in elderly women

This study found a reduced risk of death from any cause in women who took 4,400 steps compared to those who took 2,700 steps per day. More was better, up to about 7,500 steps, after which the chance of dying seemed to level off. Step count numbers come from quartiles: the 25% of people with the lowest step counts averaged around 2,700.

The participants were 16,741 women with an average age of 72. They come from the Women’s Health Study, which began as a 1990’s trial of aspirin and vitamins for prevention of heart disease and cancer. The participants are 95% white and most are nurses.

For mortality in middle-aged people

This study compared steps per day to the risk of death in middle age (41 to 65). It found that people who took more than 7,000 steps had a 50% to 70% reduced risk of mortality compared to people who took fewer than 7,000 steps per day. This number was chosen as a cutoff because it is the number that the American College of Sports Medicine estimates as a 30-minute walk each day plus a small amount of non-exercise activity.

The 2,110 participants were 57% women, 42% Black, and were followed for an average of about 11 years after the study.

For arterial stiffness

Stiffening arteries are a component of cardiovascular disease. This systematic review found that increasing steps by 2,000 per day seems to reduce arterial stiffness by about the same amount as starting a structured exercise program. The categories compared in the analysis ranged from those who took less than 5,000 steps to those who took more than 10,000. The authors write: “In layman’s terms, these findings suggest that some physical activity is better than none, but also that more is better than less.”

The results come from 20 previous studies. Most were cross-sectional (comparing groups of people based on how many steps they take) but a few were randomized controlled trials or prospective studies.

For diabetes risk in Latinx adults

This study found that each 1,000 steps more per day was associated with a 2% reduced risk of diabetes. People who took 10,000 to 12,000 steps per day had an 18% lower risk compared to those who took fewer than 5,000 steps per day.

The study participants were 6,634 Hispanic and Latino adults, half of them female, with an average age of 39.

For all-cause mortality, but at different ages

This study is interesting because it breaks down the results by age group. Data from 15 studies suggests that mortality decreases with more steps up to 6,000-8,000 steps for people aged 60 and up, but that the equivalent in younger adults is 8,000-10,000.

What do we make of all this?

I think it would be a mistake to take these top-line results completely at face value. Can you reduce your risk of death by a certain percentage just by deliberately walking a few thousand more steps per day? Nearly all of these studies compared people who already walked different amounts, rather than tasking groups of people with increasing their step counts and seeing how their health changed.

But the results do suggest that healthier people tend to have step counts that are toward the higher end of the typical range. In pretty much all of these studies (and others in this research area), people who take, say, 8,000 steps tend to be in a lower risk category than those who take, say, 2,000. So if you’re currently pretty sedentary, getting your step count up might be worth a try, even if there is no specific study saying you have to meet such-and-such number.

It’s also interesting, I think, to see that there’s no specific optimal number that these studies have identified, even though we like to talk about these studies in specifics. It’s not like you need to get to 10,000 because something different will happen than if you got 9,500.

The curves on the graphs in these papers tend to level off somewhere in the high four digits, but estimates also get less certain there because there just aren’t very many people who get more steps than this. A person who routinely gets 25,000 steps per day, for example, is just off the charts. They might be super fit, or they might have an active job that works them harder than they can easily recover from; these studies aren’t designed to tease out the difference.

The bottom line, then, is probably what you would have assumed even before checking the numbers: if you sit around a lot, moving more will probably be good for you. And if you want specific guidance, you can go with the good old 150+ minutes of exercise per week guideline, or follow the various guidelines from government projects that recommend 8,500 steps per day (U.S. presidential challenge), 7,000 to 10,000 (U.K.’s national obesity forum), or 8,000 to 10,000 (Japan).