Walking is free, needs no equipment, and is one of the most-measured health habits there is: around half a million adults wearing step counters sit behind the numbers here. People who walk more die at a lower rate and get less heart disease, diabetes, dementia and depression, and most of that benefit has already arrived by about 7,000 steps a day. The famous 10,000 came from a 1960s pedometer advertisement, not a study, and the measured curve is the better guide: risk falls steeply from very low counts and flattens between 6,000 and 8,000 steps a day in adults over 60 and 8,000 to 10,000 under 60, so older adults reach the flat part at a lower count than younger people.
Walking does not do everything: it does almost nothing for spine bone density and it maintains muscle without building it, so a couple of sessions of something heavier each week cover what it leaves out. Most people can start today by adding one walk where there was none.
Findings & Outcomes
What It Is
Walking is ordinary movement done on purpose. The step-count research rests on adults wearing accelerometers, none of whom were assigned a number to hit, and that single fact limits everything the data can show: it records what people already did and follows what happened to them next.
The 10,000-step target came from a marketing slogan. Researchers who traced it in 2004 found Japanese walking clubs and a 1960s pedometer called the Manpo-kei, a name that translates as ten-thousand-step meter. The figure is a round number from an advertisement. Someone walking 4,000 steps who reads 10,000 as the pass mark counts a failure, when 4,000 already sits on the steep part of the curve where each extra step returns the most.
What It Does
More walking goes with a longer life, and the size of that link is large. Pooling 15 international cohorts of about 47,000 adults, people in the second-lowest step group died at a lower rate than the least active, with an all-cause mortality hazard ratio of 0.60 (95% confidence interval 0.51 to 0.71), and the rate kept falling as steps rose. Most of the drop had arrived by 6,000 to 8,000 steps a day in adults over 60 and 8,000 to 10,000 in adults under 60. Older adults reach the flat part at a lower count than younger people, the opposite of how step advice is usually pitched. Each extra 1,000 steps a day tracks with roughly a 15% lower death rate across the range, and the first steps out of near-total inactivity count for the most.
For someone who is sedentary, the first walk matters more than any target. Going from about 3,500 to 5,800 steps a day cut the death rate by 40% in the pooled cohorts, a larger drop than the whole climb from 5,800 to 10,900.
The same pattern runs through the individual diseases. Against 2,000 steps a day, 7,000 tracks with about 25% less heart disease, 38% less dementia, 22% fewer depressive symptoms, and lower rates of type 2 diabetes and death from cancer. Walking tracks with surviving a cancer more than with avoiding one: the association with getting a diagnosis in the first place does not reach significance. In randomized trials, where people were assigned to walk, a walking program lowered blood pressure and raised aerobic fitness. That is the one part of this evidence that does not depend on watching who happens to walk.
Walking holds up against harder exercise. Matched for the same energy burned, walking and running produced similar drops in blood pressure, cholesterol, diabetes and heart disease, so the choice between them comes down to time and joints. Each finding below is graded at the strength of its own evidence.
The Research & Studies
Everything here is based on the research we have collected and checked, sorted into groups and ordered with the strongest evidence first. Click any claim to open the studies behind it.
Longevity And Mortality
Walking from about 3,500 to 5,800 steps a day tracked with a 40% lower death rate, flattening by 6,000 to 8,000 steps
In a pooling of 15 studies covering about 47,000 adults, people who walked more died at a lower rate over the next seven years, and most of the drop had arrived by 6,000 to 8,000 steps a day. Going from about 3,500 steps up to about 5,800 cut the death rate by 40%, the single biggest step of the whole curve. Older adults reached the flat part sooner, at 6,000 to 8,000 steps, while people under 60 kept gaining out to 8,000 to 10,000.
Against the lowest quartile (median 3,553 steps a day), all-cause mortality hazard ratios were 0.60 (95% CI 0.51 to 0.71) at a median 5,801 steps, 0.55 (0.49 to 0.62) at 7,842, and 0.47 (0.39 to 0.57) at 10,901. Restricted cubic splines placed the flattening at 6,000 to 8,000 steps a day in adults 60 and over and 8,000 to 10,000 under 60. Nobody was assigned a step count, so this is an association, and undiagnosed illness that lowers walking before it is recorded as disease is the main thing pulling the low end down.
Who this may not transfer to:68% of participants were women, which is unusual for physical-activity research and runs the other way from the usual male skew. The age-stratified splines are reported for the whole sample rather than separately by sex, so whether the flattening point differs between men and women is not established here.
The study · 1
Paluch et al., daily steps and all-cause mortality, a meta-analysis of 15 international cohorts (Steps for Health Collaborative) · Lancet Public Health 2022;7(3):e219-e228
Counts once: this finding and 1 other here come from the same source, so they are one body of evidence, not separate confirmations.
7,000 steps a day against 2,000 tracked with 47% lower death from any cause and lower rates across eight outcomes
A 2025 review pulling together 57 studies compared 7,000 steps a day against 2,000 and found lower rates across the board: death from any cause down 47%, heart disease down 25%, dementia down 38%, depression down 22%, falls down 28%, and smaller drops for diabetes and cancer death. For death, heart disease, dementia and falls the gains mostly leveled off around 5,000 to 7,000 steps, so 7,000 is a sensible target rather than 10,000.
Against 2,000 steps a day, 7,000 was associated with all-cause mortality HR 0.53 (95% CI 0.46 to 0.60), cardiovascular disease incidence 0.75 (0.67 to 0.85), cardiovascular mortality 0.53 (0.37 to 0.77), cancer mortality 0.63 (0.55 to 0.72), type 2 diabetes 0.86 (0.74 to 0.99), dementia 0.62 (0.53 to 0.73), depressive symptoms 0.78 (0.73 to 0.83) and falls 0.72 (0.65 to 0.81). Certainty ranged from moderate down to very low by outcome, and several rows rest on two or three studies. Every contributing cohort carries reverse causation, since people already ill walk less.
Who this may not transfer to:Sex composition is not reported at the pooled level across the 24 contributing cohorts, and no outcome is broken down by sex, so whether the 7,000-step figure sits in the same place for men and women is not answered here.
The study · 1
Ding et al., daily steps and health outcomes in adults, a systematic review and dose-response meta-analysis · Lancet Public Health 2025;10(8):e668-e681
Counts once: this finding and 5 others here come from the same source, so they are one body of evidence, not separate confirmations.
The most active quarter for light activity, mostly walking, died at less than half the rate of the least active
Across eight studies using motion sensors, even light activity, which for most people is mostly walking, tracked with a substantially lower death rate, while more sitting tracked with a higher one. People in the most active quarter for light activity died at less than half the rate of the least active quarter.
Against the least active quarter, mortality hazard ratios across quartiles of light-intensity activity were 0.60 (95% CI 0.54 to 0.68), 0.44 (0.38 to 0.51) and 0.38 (0.28 to 0.51). Moderate-to-vigorous activity gave 0.64, 0.55 and 0.52; total activity of any intensity 0.48, 0.34 and 0.27; and sedentary time ran the other way at 1.28, 1.71 and 2.63. Light and moderate activity are separated by a sensor cut point that shifts the comparison, quartiles describe rank rather than an absolute dose, and follow-up of 5.8 years is short. Frailty lowers light activity and raises sitting at once, which is the steep bottom of the range.
Who this may not transfer to:72.8% of participants were women. The pooled quartile estimates are not reported separately by sex.
The study · 1
Ekelund et al., dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality, systematic review and harmonised meta-analysis · BMJ 2019;366:l4570
Each extra 1,000 steps a day tracked with a 15% lower death rate, with benefit first detectable near 3,900 steps
Looking at the bottom of the curve across 17 studies and 227,000 people, each extra 1,000 steps a day was linked to a 15% lower death rate, and a benefit first became measurable at roughly 3,900 steps for death from any cause and even lower, around 2,300 steps, for heart-related death. The message is that the first steps out of near-total inactivity count for the most.
Each additional 1,000 steps a day was associated with a 15% lower rate of all-cause death (HR 0.85, 95% CI 0.81 to 0.91) and each additional 500 steps with a 7% lower rate of cardiovascular death (0.93, 0.91 to 0.95). Detectable thresholds were about 3,867 steps for all-cause and about 2,337 for cardiovascular mortality. A per-1,000-step figure is the average slope of a curve that is not straight, so it overstates the return at the top and understates it at the bottom. At these very low counts much of the risk reflects existing illness and frailty, not the walking itself.
Who this may not transfer to:49% female, the most balanced sample on this page. The pooled slope is not reported separately by sex.
The study · 1
Banach et al., the association between daily step count and all-cause and cardiovascular mortality, a meta-analysis · Eur J Prev Cardiol 2023;30(18):1975-1985
In women averaging 72, death risk fell until about 7,500 steps a day; 4,400 steps tracked with 41% lower mortality
In about 16,700 women averaging 72 years old, those who walked more lived longer, and risk kept dropping until roughly 7,500 steps a day before it leveled off. Even a modest lift, from about 2,700 to about 4,400 steps, went with a 41% lower death rate. The count that mattered was lower than the usual 10,000, which fits the pattern that older adults reach their own flat point sooner.
Against a median 2,718 steps a day, mortality hazard ratios were 0.59 (95% CI 0.47 to 0.75) at 4,363 steps, 0.54 (0.41 to 0.72) at 5,905, and 0.42 (0.30 to 0.60) at 8,442, with risk levelling around 7,500 steps. Follow-up was a mean 4.3 years and there were 504 deaths, a modest basis for a spline, and participants were health-study volunteers who are healthier than average. Reverse causation is at its strongest at this age, since a woman walking 2,700 steps a day is often doing so because of arthritis, heart failure or frailty not yet recorded as a cause of death.
Who this may not transfer to:Men were not studied here. The same shape, with a lower flattening point in older adults, appears in the mixed-sex pooled analysis of 15 cohorts and in a mixed-sex US national sample, so the pattern transfers even though this particular set of numbers was measured in women only.
The study · 1
Lee et al., association of step volume and intensity with all-cause mortality in older women · JAMA Intern Med 2019;179(8):1105-1112
Counts once: this finding and 1 other here come from the same source, so they are one body of evidence, not separate confirmations.
Once daily steps are counted, whether a faster cadence adds anything is unresolved; the three best datasets disagree
Whether walking faster helps beyond just walking more is unsettled: the three best datasets disagree once total daily steps are accounted for. For reference, about 100 steps per minute is a moderate pace and about 130 is brisk. Total steps is the part the evidence supports most firmly; a faster pace may add a little, and it finishes the same distance sooner either way.
In 16,741 older women, every intensity measure lost significance after adjusting for steps per day (peak 1-minute cadence HR 0.87, 95% CI 0.68 to 1.11; peak 30-minute 0.86, 0.65 to 1.13). In 4,840 US adults, peak 30-minute cadence gave 0.90 (0.65 to 1.27) after the same adjustment. In the pooled 15-cohort analysis, peak-30 and peak-60 cadence both survived adjustment (0.67, 0.56 to 0.83 and 0.50 to 0.90). Cadence and total steps are strongly correlated, so adjusting one for the other asks a question the data can barely answer. A person's habitual speed is itself a marker of fitness, joints and cognition.
Who this may not transfer to:Two of the three mortality datasets are mixed-sex and one is women only. The CADENCE-adults calibration used 80 people, 10 men and 10 women in each five-year age band, which is balanced but small, and the 100 and 130 steps per minute thresholds have not been established in adults over 60.
Build total daily steps first, since that is the part the evidence backs most consistently. Treat a brisker pace as a bonus rather than the goal: it may add a little on top of volume and it gets the same distance done in less time, which is an advantage on a full day. There is no case here for pushing pace at the expense of how far or how often you walk.
The studies · 5
Lee et al., association of step volume and intensity with all-cause mortality in older women · JAMA Intern Med 2019;179(8):1105-1112
Saint-Maurice et al., association of daily step count and step intensity with mortality among US adults · JAMA 2020;323(12):1151-1160
Paluch et al., daily steps and all-cause mortality, a meta-analysis of 15 international cohorts · Lancet Public Health 2022;7(3):e219-e228
Tudor-Locke et al., walking cadence (steps/min) and intensity in 41 to 60-year-old adults, the CADENCE-adults study · Int J Behav Nutr Phys Act 2020;17(1):137
Tudor-Locke et al., how fast is fast enough? Walking cadence as a practical estimate of intensity in adults, a narrative review · Br J Sports Med 2018;52(12):776-788
Counts once: this finding and 1 other here come from the same source, so they are one body of evidence, not separate confirmations.
Self-described average or brisk walkers died at 20 to 24% lower rates than self-described slow walkers
Among 50,000 walkers, those who described their usual pace as average or brisk died at a lower rate over the following years than self-described slow walkers, about 20 to 24% lower for death from any cause, with no difference between men and women. Pace was self-reported, and a slow pace is one of the most sensitive early signs of undiagnosed illness.
Against a slow pace, an average pace carried 20% lower all-cause mortality (95% CI 12% to 28%) and 24% lower cardiovascular mortality (9% to 36%); a brisk pace carried 24% lower all-cause (13% to 33%) and 21% lower cardiovascular mortality (1% to 38%). There was no association with cancer death, and no interaction by sex or body mass index. Pace was self-rated as slow, average or brisk rather than measured. Gait speed is a strong marker of undiagnosed disease, and dropping the first two years of follow-up does not cover slowly progressive conditions such as heart failure, Parkinson's and osteoarthritis that slow a person years before diagnosis.
Who this may not transfer to:The analysis tested for interaction by sex and found none, which is a stronger statement than most rows on this page can make: the pace association was of similar size in men and women.
The study · 1
Stamatakis et al., self-rated walking pace and all-cause, cardiovascular disease and cancer mortality, individual participant pooled analysis of 50,225 walkers from 11 population British cohorts · Br J Sports Med 2018;52(12):761-768
Heart And Vascular
7,000 steps a day against 2,000 tracked with 25% less heart disease and 47% lower cardiovascular death
More walking tracks with less heart disease. Compared with 2,000 steps a day, 7,000 went with 25% fewer heart-disease diagnoses and 47% fewer heart-related deaths, and in a large UK study the benefit kept building out to about 10,000 steps.
7,000 steps a day against 2,000 was associated with 25% lower cardiovascular disease incidence (HR 0.75, 95% CI 0.67 to 0.85) and 47% lower cardiovascular mortality (0.53, 0.37 to 0.77). In 78,500 UK Biobank adults each additional 2,000 steps carried a mean rate of change of -0.10 (95% CI -0.15 to -0.06) for cardiovascular death, continuing to about 10,000 steps. The mortality estimate rests on three studies graded low certainty, and the UK cohort is 97% White and healthier than the general population. Established heart disease itself limits walking through angina and breathlessness, so low counts are partly a symptom of the outcome being counted.
Who this may not transfer to:The UK Biobank accelerometer sample was 55% female and neither source reports the cardiovascular results separately by sex, so whether the step-count association differs in size between men and women is not established here.
The studies · 2
Ding et al., daily steps and health outcomes in adults, a systematic review and dose-response meta-analysis · Lancet Public Health 2025;10(8):e668-e681
Del Pozo Cruz et al., prospective associations of daily step counts and intensity with cancer and cardiovascular disease incidence and mortality and all-cause mortality · JAMA Intern Med 2022;182(11):1139-1148
Counts once: this finding and 5 others here come from the same source, so they are one body of evidence, not separate confirmations.
Matched for energy burned, walking and running produced similar drops in blood pressure, cholesterol, diabetes and heart disease
Matched for the same energy burned, walking and running produced similar drops in high blood pressure, high cholesterol, diabetes and heart disease. So the choice is mostly about time and joints, not about one being fundamentally better, though the two groups differed in ways a head-to-head trial would not.
Per metabolic equivalent hour per day, walking was associated with 7.2% lower incident hypertension, 7.0% lower high cholesterol, 12.3% lower diabetes and 9.3% lower coronary heart disease; running gave 4.2%, 4.3%, 12.1% and 4.5%. These are two separate self-selected cohorts compared against each other, not one randomized comparison, and the groups differ: male walkers averaged 61.8 years against 48.3 for male runners. Outcomes were self-reported over 6.2 years. Energy-matching also hides the time cost, since the same energy takes far longer to accumulate walking than running.
Who this may not transfer to:The two cohorts have opposite sex balances: the walkers were 79% women and the runners roughly half men. That makes the walking estimates better grounded in women and the running estimates better grounded in men, which is a limit on the head-to-head comparison rather than on either figure alone.
For the same health return you can walk or run; pick the one you will actually keep doing and that your joints tolerate. Running gets the energy done faster, so it suits a tight schedule, while walking spreads the same work over more time but is gentler on knees and hips and is the activity most people with joint or heart concerns are cleared to do. If time is the limit, run or walk briskly; if joints or comfort are the limit, walk more minutes.
The study · 1
Williams and Thompson, walking versus running for hypertension, cholesterol, and diabetes mellitus risk reduction · Arterioscler Thromb Vasc Biol 2013;33(5):1085-1091
In randomized trials, walking lowered systolic blood pressure by 3.6 mmHg and raised aerobic capacity by 3 mL/kg/min
This is the one part of the walking evidence that comes from trials rather than from watching who walks. Pooling 32 randomized trials of inactive adults, a walking program lowered systolic blood pressure by about 3.6 mmHg and diastolic by about 1.5, raised aerobic fitness, and trimmed weight, waist and body fat, while leaving blood cholesterol unchanged. Because people were assigned to walk, this is the finding least troubled by the worry that runs through the rest of the page, that people who are already ill simply walk less.
Across 32 randomized controlled trials, walking reduced systolic blood pressure by 3.58 mmHg (95% CI -5.19 to -1.97) and diastolic by 1.54 mmHg (-2.83 to -0.26), raised aerobic capacity by 3.04 mL/kg/min (2.48 to 3.60), and lowered weight by 3.0 lb (1.37 kg), waist by 0.6 inches (1.51 cm), body fat by 1.22% and body mass index by 0.53 kg/m2, with no change in blood lipids. A second review of 37 trials in 2,001 adults found the same pattern, including for fasting glucose. The trials are small and short and the outcomes are risk-factor surrogates rather than events, but they are randomized, so the effects are caused by the walking.
Who this may not transfer to:The larger of the two reviews was 81% women, so the blood-pressure and fitness effects are better grounded in women than in men, and neither review reports the pooled effects separately by sex.
The studies · 2
Murtagh et al., the effect of walking on risk factors for cardiovascular disease, an updated systematic review and meta-analysis of randomised controlled trials · Prev Med 2015;72:34-43
Oja et al., effects of frequency, intensity, duration and volume of walking interventions on cardiovascular disease risk factors, a systematic review and meta-regression analysis of randomised controlled trials among inactive healthy adults · Br J Sports Med 2018;52(12):769-775
Blood Sugar
7,000 steps a day against 2,000 tracked with 14% less type 2 diabetes, the smallest of the step-count links
Walking more goes with a somewhat lower chance of type 2 diabetes, though this is the weakest of the step-count links. Compared with 2,000 steps, 7,000 a day was associated with 14% less diabetes, and a separate study found no extra benefit above roughly 8,000 to 9,000 steps.
7,000 steps against 2,000 was associated with a 14% lower rate of type 2 diabetes (HR 0.86, 95% CI 0.74 to 0.99), the smallest pooled effect and the one whose interval comes closest to 1. In 6,042 US adults wearing their own Fitbits, the association was non-linear with no further reduction above about 8,000 to 9,000 steps. That subgroup rests on 156 cases and is heavily self-selected. Body weight sits on both sides of this link, lowering step counts and raising diabetes risk, so adjusting for it can remove part of the effect, and early undiagnosed diabetes reduces activity through fatigue.
Who this may not transfer to:The All of Us Fitbit sample was 73% female. Neither source reports the diabetes association separately by sex.
The studies · 2
Ding et al., daily steps and health outcomes in adults, a systematic review and dose-response meta-analysis · Lancet Public Health 2025;10(8):e668-e681
Master et al., association of step counts over time with the risk of chronic disease in the All of Us Research Program · Nat Med 2022;28(11):2301-2308
Counts once: this finding and 5 others here come from the same source, so they are one body of evidence, not separate confirmations.
Cancer Risk And Outcome
7,000 steps a day against 2,000 tracked with 37% lower death from cancer
More walking tracks with fewer cancer deaths: 7,000 steps a day against 2,000 went with 37% lower cancer mortality. This sits next to a finding of no clear effect on getting cancer in the first place, which points at surviving a cancer rather than avoiding one, and undiagnosed cancer and smoking both muddy the picture.
7,000 steps against 2,000 was associated with 37% lower cancer mortality (HR 0.63, 95% CI 0.55 to 0.72). In the UK Biobank cohort, each additional 2,000 steps carried a mean rate of change of -0.11 (95% CI -0.15 to -0.06). Cancer death pools every type into one outcome with no shared mechanism, and it sits alongside a null for cancer incidence in the same review. Undiagnosed cancer lowers activity for months to years through weight loss, fatigue and pain, and smoking is the largest shared cause of both low activity and cancer death, surviving statistical adjustment.
Who this may not transfer to:55% female in the accelerometer cohort. Neither source separates cancer mortality by sex, which matters more than usual here because the cancer mix differs between men and women.
The studies · 2
Ding et al., daily steps and health outcomes in adults, a systematic review and dose-response meta-analysis · Lancet Public Health 2025;10(8):e668-e681
Del Pozo Cruz et al., prospective associations of daily step counts and intensity with cancer and cardiovascular disease incidence and mortality and all-cause mortality · JAMA Intern Med 2022;182(11):1139-1148
Counts once: this finding and 5 others here come from the same source, so they are one body of evidence, not separate confirmations.
Walking was not clearly linked to getting a cancer diagnosis: 7,000 steps against 2,000 gave a 6% lower rate that missed significance
Walking more was not clearly linked to getting a cancer diagnosis in the first place. Comparing 7,000 steps a day with 2,000 showed a 6% lower rate that did not reach significance, so on current evidence walking looks better at outliving a cancer than at preventing one. This rests on just two studies, the thinnest evidence on the page.
7,000 steps against 2,000 was associated with a 6% lower cancer incidence that missed significance (HR 0.94, 95% CI 0.87 to 1.01, 2 studies), graded low or very low certainty. The same review found the association with cancer death clearly significant, so the gap between the two is the point. An interval running to 1.01 is a failure to detect an effect, not proof none exists, and a pooled all-cancers outcome would dilute an effect on any single site. Detection bias runs the other way here, since more active people have more screening contact, which can raise recorded incidence without raising the true rate.
Who this may not transfer to:Sex composition is not reported for the two contributing studies at the pooled level.
The study · 1
Ding et al., daily steps and health outcomes in adults, a systematic review and dose-response meta-analysis · Lancet Public Health 2025;10(8):e668-e681
Counts once: this finding and 5 others here come from the same source, so they are one body of evidence, not separate confirmations.
Mood & stress
Walking 7,000 or more steps a day tracked with a 31% lower rate of new depression
Across 33 studies and about 96,000 adults, people who walked more had fewer depressive symptoms, and in the studies that tracked people over time, 7,000 or more steps a day went with a 31% lower rate of new depression. Most of the data measured mood and steps on the same day, so it cannot show which came first.
Against fewer than 5,000 steps a day, cross-sectional pooling gave standardized mean differences in depressive symptoms of -0.26 (95% CI -0.38 to -0.14) at 10,000 or more, -0.27 (-0.43 to -0.11) at 7,500 to 9,999 and -0.17 (-0.30 to -0.04) at 5,000 to 7,499. In three prospective cohorts, 7,000 or more steps carried a relative risk of new depression of 0.69 (0.62 to 0.77), and each extra 1,000 steps 0.91 (0.87 to 0.94). Twenty-seven of the 33 studies were same-day, and reduced activity is itself a feature of depression, so much of the cross-sectional link measures the same thing twice.
Who this may not transfer to:The review does not report a pooled sex breakdown across the 33 studies, and it states that the optimal step count may vary by sex and age without being able to estimate that variation.
The study · 1
Bizzozero-Peroni et al., daily step count and depression in adults, a systematic review and meta-analysis · JAMA Netw Open 2024;7(12):e2451208
Bone Density
Walking on its own did not raise bone density at the spine, and only borderline at the hip
Walking on its own did not meaningfully build bone in postmenopausal women. Across eight trials it produced no clear gain at the spine and only a borderline one at the hip, because walking is a load the skeleton is already used to. None of these trials measured actual fractures.
Prescribed walking produced no significant change in lumbar spine bone mineral density (weighted mean difference 0.007 g/cm², 95% CI -0.001 to 0.016, P = 0.09). At the femoral neck the estimate was 0.014 g/cm² (0.000 to 0.028, P = 0.05), on the significance boundary with the interval touching zero. Eight small, short trials, and bone density is a surrogate for the fractures nobody measured. Bone adapts to strain magnitude and rate rather than to repetition of a familiar load, which is why walking moves it so little.
Who this may not transfer to:Men were not studied. The mechanism this rests on, that bone adapts to strain magnitude and rate rather than to repetition of a load it is already used to, is not sex-specific, and there is no reason to expect walking to build a man's spine when it does not build a woman's. Men also lose bone with age and are underserved by a literature that studied this almost entirely in postmenopausal women.
Keep walking for the heart, metabolic and longevity benefits, but do not count on it for bone. To load the skeleton you need something it is not already used to: resistance training two or three times a week, or impact such as jumping or stair work where it is safe. Men were not studied here, and the mechanism is not sex-specific, so the same applies to them.
The study · 1
Martyn-St James and Carroll, meta-analysis of walking for preservation of bone mineral density in postmenopausal women · Bone 2008;43(3):521-531
Muscle And Strength
For building muscle, walking lost to resistance training; it maintains the muscle you have without adding to it
Walking maintains the muscle you have without adding to it. In a head-to-head of 38 trials, aerobic training beat resistance training for fitness and fat loss but lost on building muscle, and the authors' conclusion was to do both.
Aerobic training beat resistance training on VO2max (mean difference 1.80 mL/kg/min, 95% CI 0.96 to 2.64) and six-minute walk distance (61 feet (18.58 m), 95% CI 10.38 to 26.78 m) and lowered body mass more (-2.7 lb (-1.23 kg), 95% CI -1.98 to -0.47 kg), while change in lean body mass favored resistance training. The abstract gives the direction of the lean-mass result without a pooled estimate or interval, so its size cannot be read from it, and the aerobic arm covers cycling and swimming as well as walking.
Who this may not transfer to:The abstract does not give the sex composition of the 1,682 participants across the 38 trials, so the balance is unknown from the published record.
Use walking for cardiovascular fitness and fat loss, and add resistance training two or three times a week for the muscle it does not build. The two do different jobs, so combining them is the better plan rather than choosing one, which is what the trial authors themselves concluded.
The study · 1
An, Su and Meng, effect of aerobic training versus resistance training for improving cardiorespiratory fitness and body composition in middle-aged to older adults, a systematic review and meta-analysis of randomized controlled trials · Arch Gerontol Geriatr 2024;126:105530
Cognition
The lowest dementia rate sat near 9,800 steps a day, 51% below the reference, with clear benefit already by about 3,800
In about 78,000 UK adults, those who walked more were later diagnosed with dementia less often, with the lowest rate, 51% below the least-active group (hazard ratio 0.49), near 9,800 steps a day and a clear benefit already at about 3,800. This is early evidence, and it is the finding most vulnerable to the disease itself quietly lowering activity years before diagnosis.
The lowest dementia rate sat at 9,826 steps a day (HR 0.49, 95% CI 0.39 to 0.62), with 3,826 steps already at 0.75 (0.67 to 0.83); a peak 30-minute cadence of 112 steps per minute carried 0.38 (0.24 to 0.60). Pooled across two studies, 7,000 steps against 2,000 gave 0.62 (0.53 to 0.73). There were 866 cases at a mean baseline age of 61, young for a disease of the eighties. Dementia has a preclinical phase of a decade or more in which reduced activity is an early sign, so a low step count at 61 may be the first symptom of the disease it appears to predict, and 6.9 years of follow-up cannot separate those.
Who this may not transfer to:55.3% of the cohort were women. The dose-response curve is not published separately by sex, and women carry roughly two thirds of dementia cases worldwide, so a sex-specific curve would be worth having and does not exist here.
The studies · 2
Del Pozo Cruz et al., association of daily step count and intensity with incident dementia in 78,430 adults living in the UK · JAMA Neurol 2022;79(10):1059-1063
Ding et al., daily steps and health outcomes in adults, a systematic review and dose-response meta-analysis · Lancet Public Health 2025;10(8):e668-e681
Counts once: this finding and 5 others here come from the same source, so they are one body of evidence, not separate confirmations.
Joint And Arthritis Pain
In knee osteoarthritis, walkers were less likely to develop new frequent knee pain over four years (odds ratio 0.6)
In adults over 50 who already had knee osteoarthritis, those who walked for exercise were less likely to develop new frequent knee pain and less likely to show the joint narrowing further over four years. This is early evidence, and much of it may reflect that people whose knees hurt less walk more to begin with.
Among 1,212 adults with knee osteoarthritis, walkers were less likely to develop new frequent knee pain (odds ratio 0.6, 95% CI 0.4 to 0.8) and less likely to show progression of medial joint space narrowing (0.8, 0.6 to 1.0) over 48 months. Walking history was collected retrospectively at the 96-month visit, after the outcomes it is used to predict, which is the wrong order for causal reading, and the joint-space interval reaches 1.0. The authors present it as proof of concept warranting a trial. Reverse causation is plain here, since people whose knees hurt more walk less.
Who this may not transfer to:45% of participants were men. Results are not reported separately by sex, which matters because knee osteoarthritis is more common and more severe in women.
If you have knee osteoarthritis, this is reassurance that walking for exercise is not damaging the joint and may help, so it is reasonable to keep walking within comfort rather than avoiding it out of fear of wear. It is not proof that walking reverses the disease. Build distance gradually, and treat pain that is sharp, swelling, or lasting well beyond the walk as a reason to ease off and get assessed.
The study · 1
Lo et al., association between walking for exercise and symptomatic and structural progression in individuals with knee osteoarthritis, data from the Osteoarthritis Initiative cohort · Arthritis Rheumatol 2022;74(10):1660-1667
How It Works
Anatomy of the Practice
1The walk itself
Contracting muscle pulls glucose out of the blood through a route that barely needs insulin, so a walk after a meal blunts the sugar rise that follows it. Heart rate and breathing settle into a moderate load, and blood moves faster through the body.
2Over weeks
Walked most days, resting blood pressure eases, the arteries respond better to demand, and the body gets more efficient at burning fat and sugar for fuel. Mood tends to lift and sleep often steadies.
3Over years
People who walk more have lower rates of death, heart disease, diabetes and dementia. Most of that signal comes from watching large groups over time, not from trials that assign people to walk, so it is a strong and consistent association; the blood-pressure and fitness gains have been confirmed in trials.
Walking is moderate aerobic movement, and most of what it does follows from working the muscles at a steady load. Contracting muscle draws glucose from the blood through a pathway that hardly needs insulin, which is why a walk after the largest meal of the day flattens the blood-sugar rise more than the same walk at any other hour; that timing has its own protocol at the post-meal walk. The steady demand on the heart and vessels lowers resting blood pressure over weeks, and randomized trials have confirmed that part directly.
Almost all of the step-count evidence is observational: researchers measured whatever people happened to walk and followed them for years. The main problem is reverse causation. People who are already ill walk less, and illness begins before it is diagnosed, so a low step count can be an early symptom of the disease it appears to predict. Healthy-user patterns pull the same way, since people who walk more also tend to smoke less, eat better, and have more money and safer streets.
The better studies drop everyone who died in the first two years of follow-up, on the reasoning that undetected illness is concentrated there. That shrinks the association and leaves it standing, which narrows the problem without removing it. Dementia is the outcome most exposed to it, since its silent phase runs a decade or more and reduced activity is one of its earliest signs.
For someone currently sedentary, the bottom of the range is where the data is most useful: the gap between doing nothing and doing something is far larger than the gap between walking and harder exercise.
What Walking Does Not Do
Walking leaves two gaps:
- Bone. Across eight controlled trials in postmenopausal women, prescribed walking produced no significant change in spine bone mineral density. Walking is a load the skeleton is already used to, and growing bone takes an unfamiliar strain.
- Muscle. Walking maintains the muscle you have and does not add to it, losing to lifting on lean mass in head-to-head trials.
Both gaps are filled by resistance training two or three times a week, which can be done at home; the mechanism for bone is on the mechanotransduction page. Balance is a third thing walking does not train, because balance improves only when it is challenged, and balance and falls covers what works there. None of this is a reason to walk less. It is why walking is a foundation and not a whole program.
Ways to Do It
Walking needs no equipment and no membership, so cost barely enters into it; the work is fitting the steps into a day that is already full. The reliable way to add them is to hang the walking on something you already do. Find your current average over a normal week first, then build from there.
The single walk that was not happening yesterday. The pooled data starts detecting a lower death rate somewhere around 2,300 to 4,000 steps a day, so the first walk a mostly-still person adds lands on the steepest part of the curve. Ten easy minutes is roughly 1,000 steps.
Take the phone call walking, park at the far end, get off a stop early, or walk a short commute you would drive. Ten minutes each way is about 2,000 steps a day, five days a week, and the time was already spoken for, so the habit holds.
Ten minutes on foot after your biggest meal lowers the blood-sugar rise that follows it by more than the same ten minutes at another hour. It is the highest return per minute here and it has its own protocol at the [post-meal walk](/go/integrative/practice/post-meal-walk).
Once a walk is a habit, add about 1,000 steps a day to your average and hold it until it stops feeling like a change, then add again. Aim at around 7,000 if you are over 60 and 8,000 to 9,000 if you are younger, and treat that as the place the curve levels off; it is not a pass mark you have failed until you hit it.
The one paid option, and it earns its place only when weather, safety or a desk keep you indoors for months. An under-desk walking pad turns work time into steps; a treadmill does the same with a larger footprint and more cost. Neither does anything the pavement does not; you are buying convenience.
Go Deeper
- Insulin and glucose: how a working muscle clears sugar from the blood, the mechanism behind the post-meal walk.
- Post-meal walk: the ten minutes after your biggest meal that return the most per minute.
- Resistance training: what covers the muscle and spine bone that walking leaves untouched.
- Balance and falls: why walking does not train balance, and what does.
- Type 2 diabetes: where walking sits in managing blood sugar day to day.
- Arthritis and joint pain: the knee question in full, for anyone walking with joint pain.
- Nature exposure: the greenspace research, since a walk in a park is a walk either way.
- Morning light: walk early and the same trip sets your body clock.
- Walking or running: the energy-matched head to head, including the knee question.
- Gait speed: how walking speed is measured on its own terms.
The Chinese Medicine View
The tradition's name for deliberate therapeutic movement is 導引 (daoyin), guiding and pulling, the ancestor of tai chi and qi gong, and walking is the plainest member of that family. The classics endorse it directly. The Su Wen's instruction for spring is 廣步於庭, to walk with long strides in the courtyard, hair loose and body unhurried. Movement here is held to move Qi and Blood, and because the Spleen governs the four limbs, using them is held to support digestion.
The dose is set by what a person has to spend, not by a target. The governing phrase is 形勞而不倦, the body should labor without becoming exhausted. And the tradition names a specific limit: among the five injuries of overuse in the Su Wen is 久行傷筋, prolonged walking injures the sinews. The sinews belong to the Liver, and it is Liver Blood that keeps them supple, so the person that caution is written for is someone with Liver Blood deficiency, whose picture is calf cramps at night, dry eyes and floaters, brittle nails and tight tendons. For that person the classical advice is short, flat, frequent walks and less total distance. Someone who is Qi deficient, tired out of proportion to the effort and breathless on speaking, does better with several short walks than one long one, and someone who runs cold is walked warm, out of the wind, and well short of heavy sweating.
That is a description of walking in the tradition's own vocabulary. The cohorts measured deaths, diagnoses and accelerometer counts. The two accounts describe the same act, and neither one certifies the other.
Cautions For This Practice
Everything to be aware of is here, in one place. This practice suits most healthy people; a few situations call for real care.
Chest symptoms that come on with walking
Chest pain, pressure or tightness that comes on while walking and eases within a few minutes of stopping, or the same feeling in the jaw, neck or arm, is the pattern of exertional angina until a doctor says otherwise. Breathlessness out of proportion to the effort, and fainting, near-fainting or palpitations while walking, belong in the same group. Stop for these and get assessed; do not walk through them or test again to see if they repeat.
Calf pain that a short rest relieves
Cramping pain in the calf, thigh or buttock that starts after a fairly predictable distance and settles within a few minutes of standing still is the classic pattern of intermittent claudication, the main symptom of peripheral arterial disease. It calls for assessment, for two reasons: it signals narrowed arteries elsewhere in the body, and the usual treatment is supervised walking, so the diagnosis tends to come first and the walking continues after it. Pain in the foot at rest, or a wound on the foot or leg that is not healing, is more pressing and should be seen quickly.
Get cleared first if your heart is already a question
If you have unstable angina, had a heart attack recently, have an uncontrolled arrhythmia, decompensated heart failure, severe aortic stenosis, or blood pressure that is uncontrolled at rest, talk to a doctor before starting. For almost everyone else, walking is the activity people are cleared to do, and it needs no permission.
Diabetes, reduced feeling in the feet, and footwear
If you have diabetes with reduced sensation in the feet, walking is appropriate and footwear matters more. Well-fitted shoes, and a look at your feet each day, because a blister you cannot feel is how a foot ulcer begins. New unsteadiness, a foot that catches or drops, or numbness spreading in the feet is worth having looked at before adding distance.
Heat is the underrated risk on a walk
Heat catches more walkers out than the walking itself does. Older adults, anyone on diuretics, beta blockers, anticholinergics or antipsychotics, and anyone not used to the temperature do better moving the walk to the cool end of the day and carrying water. [Hydration and electrolytes](/go/integrative/practice/hydration-and-electrolytes) covers the rest.
Start slow, be smart, read the research, and consult a professional if you have any concerns. This is here to inform your choice, not make it for you.
Common Questions
How many steps a day do I actually need?
Around 7,000 for most adults, and 6,000 to 8,000 if you are over 60. Pooled cohorts show the death rate dropping steeply from very low counts, and the curve flattens in that range, with the flattening at 8,000 to 10,000 in people under 60. The steepest returns sit at the bottom of the curve: going from about 3,500 to about 5,800 steps a day carried a larger drop than going from 5,800 all the way to 10,900 did. The first steps out of near-total inactivity are the ones that count for the most.
Is 10,000 steps the right target?
It is a reasonable goal for someone already active, and it sits above the point where most of the measured benefit has already arrived. Researchers who traced the figure in 2004 found a Japanese walking-club slogan and a pedometer brand behind it, and even then flagged that 10,000 steps might not be sustainable for older adults or people living with chronic illness. A sedentary reader is better served by adding 1,000 steps a day to their own average than by chasing a number they are 6,000 short of.
Does walking fast matter more than walking far?
Total volume is what the evidence best supports, and whether cadence adds anything on top of it is unresolved, with the studies disagreeing. Roughly 100 steps per minute is a moderate pace and about 130 is brisk. In one large cohort of older women every intensity measure lost significance once daily steps were in the model, while a fifteen-cohort pooled analysis found peak cadence surviving that same adjustment. Walking faster also finishes the same distance sooner, which is a reason on its own.
Is walking bad for my knees?
The evidence points the other way. Among adults over 50 who already had knee osteoarthritis, those who walked for exercise were less likely to develop new frequent knee pain over four years, and less likely to show the joint narrowing further, than non-walkers. This is early evidence, and part of it reflects that people whose knees hurt less walk more to begin with, so it is not proof that walking reverses the disease. It is reassurance that walking within comfort is not wearing the joint out. Build distance gradually, and treat pain that is sharp, swelling, or lasting well beyond the walk as a reason to ease off and get assessed.
Does walking build bone or muscle?
Not much of either. On its own, walking produced no clear gain in spine bone mineral density across eight trials, because it is a load the skeleton is already used to, and it maintains muscle without adding to it. That is the case for pairing it with resistance training two or three times a week, which loads both the bone and the muscle that walking leaves alone.
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
All 27 sources on this page independently checked and cross-referenced.
Thomas Dehli, Founder & Editor, Sacred Lotus
Sacred Lotus has published Chinese medicine reference material since 2001. Integrative pages are held to the same standard as the herb and formula library: cite the source, grade the claim at its real strength, and say where the research has not looked. This page is educational and it is not medical advice. Last reviewed and updated August 10, 2026.
Evidence strength
How confidently the research supports a claim. Strength describes the evidence, not our endorsement.