Counts describe this library, not the strength of evidence.
How to read this evidence
The step-count literature moves from specific older populations to broader international cohorts and multiple outcomes. Associations support attention to sustained activity; they do not establish a universal daily threshold or an individual survival forecast.
Compared with 2000 steps a day, 7000 steps was associated with a hazard ratio of 0.53 (95% CI 0.46 to 0.60) for all-cause mortality across 14 studies, 0.75 (0.67 to 0.85) for cardiovascular disease and 0.62 (0.53 to 0.73) for dementia; cancer incidence was not significantly different (0.94, 0.87 to 1.01). Several outcomes rest on only two to four studies, and the certainty of evidence ranged from moderate to very low.
Among 25,241 UK Biobank adults who reported no leisure-time exercise, a median of 3 vigorous bouts of 1 or 2 minutes a day, picked up by a wrist accelerometer, was associated with a 38% to 40% lower risk of all-cause and cancer death and a 48% to 49% lower risk of cardiovascular death than no such bouts, over an average of 6.9 years and 852 deaths. This is an observational study using one week of wrist data and an algorithm to define vigorous effort, so it cannot show that adding such bursts causes the lower risk.
Pooling 15 cohorts (47,471 adults, 3,013 deaths, median follow-up 7.1 years), the second, third and fourth quartiles of daily steps (medians 5,801, 7,842 and 10,901) had hazard ratios for death of 0.60, 0.55 and 0.47 against the lowest quartile (median 3,553). The curve flattened at about 6,000 to 8,000 steps a day for adults aged 60 and over and 8,000 to 10,000 for younger adults. The data are observational, from a single week of step counting, mostly with waist-worn research devices.
Among 316 cancer patients in Germany, more positive feelings about physical activity correlated with higher life satisfaction (r = 0.212, p < 0.001) and more strongly with health-related expectations of exercise (r = 0.580). Self-reported sport participation before diagnosis was not linked to life satisfaction, and a one-time questionnaire cannot show cause and effect.
In 327 newly diagnosed Hong Kong patients with non-small cell lung cancer, objective cognitive impairment affected 17.5% to 29.1% over 6 months, and consistently high or declining-from-high activity was linked to 93.8% and 85.9% lower risk than consistently low activity. Activity was grouped from patterns over time in an observational study, so it cannot show that exercise protects thinking.
Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer
Among 1732 people in a 12-month digital lifestyle program, physical activity rose from 2.16 to 3.06 MET-h/day, and higher activity was associated with better quality of life, mostly through a direct path rather than weight loss (84.9% of the total link with physical functioning). Follow-up was incomplete and individual associations were modest, so this is observational and not proof that activity caused the gains.
Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation
In a 2024 US national survey (N = 3,486), 33.9% of cancer survivors were inactive and 38.0% met the aerobic guideline of at least 150 minutes a week. Activity was self-reported in a cross-sectional survey, so the links with walking environment and health status show association only.
In 435 Swiss adults aged 19-84, age had the strongest positive association with fat mass index and visceral fat, and physical activity was negatively associated with them, especially in men. Diet quality related to body composition in women but not men, and the cross-sectional design cannot show that diet or activity caused the differences.
Among 206 university students at Shaqra University, higher self-reported physical activity went with higher eating-disorder symptom scores (B = 0.64, p < 0.001), and a small indirect path via smartphone addiction was borderline significant (B = 0.02, p = 0.049). The data are cross-sectional, so they cannot show that exercise causes eating-disorder symptoms or that smartphone addiction mediates anything.
In a survey of 1,083 people with multiple sclerosis, physical activity was linked to higher quality of life in both those diagnosed within 5 years and those diagnosed longer ago, and very active people reported lower walking disability. These are cross-sectional self-reports, so they cannot show that activity or diet caused the differences.
Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology
In 180 women followed after a hypertensive pregnancy, reported vegetable intake rose from 2.7 servings/day at 12 months to 4.2 at 3-4 years, and median moderate-to-vigorous activity rose to 515 minutes/week (all p < 0.001), while motivators fell and barriers rose. Different questionnaires were used at the two time points, so part of the rise may be a difference in measurement.
Among 2072 informal caregivers in the Spanish National Health Survey 2023, 34.5% screened positive for depressive symptoms, and the odds were lower with more perceived social support (OR 0.79 per point) and higher with never doing physical activity (OR 1.82). The data are a single cross-sectional snapshot, so they cannot show whether support and activity protect against low mood or whether low mood reduces them.
In 10 randomized trials with 446 healthy women over 60, physical activity was associated with higher global cognitive scores (Hedges' g 0.63, 95% CI 0.12-1.13) and higher BDNF, with no significant effect on BMI. The certainty of evidence was very low and heterogeneity was substantial (I2 77.82%), so the true effect could be small or absent.
In interviews with 24 adults aged 34-66, knowledge, attitudes and daily routines around exercise varied widely, and the authors built a framework of 20 dimensions to guide personalizing exercise apps and wearables. It is a small qualitative study, so it describes what these people said, not whether personalized apps actually raise activity.
Compared with senior high school students, students in grades 1-9 reported 43.1 more minutes of moderate-to-vigorous physical activity per week (95% CI 33.8-52.3; p < 0.001) and a 3.9 percentage-point lower overweight/obesity rate (95% CI -7.2 to -0.6; p = 0.022) after the policy. Activity was self-reported and only two pre-policy waves were available to check trends, so a causal reading needs care.
In national survey data on 7,982 Chinese children aged 10-18, self-reported exercise time rose by 10.2 min/day (95% CI 7.5-12.9) between 2020 and 2022, while boys were more active than girls in every wave. A questionnaire change in 2020 limits comparisons across that point, and activity was self-reported.
In 1,086 female Chinese university students followed over 12 months, higher physical activity at the start was associated with lower body dissatisfaction at the end (β = -0.127, p < .001), partly through psychological flexibility and resilience (total indirect β = -0.243). Self-reported questionnaires and a statistical mediation model show association, not that activity changed body image.
In a national sample of 8,653 Chilean adults, nearly the same share with and without disabilities said activity spaces were available nearby (92.1% vs. 91.4%), yet only 29.9% of those with disabilities used them versus 42.7% of those without. This is a self-reported cross-sectional survey, so it cannot show why the gap exists or whether changing spaces would close it.
In a pilot of 35 Swedish adults aged 60 or older, overall attrition over 26 weeks was 37% (13/35), with no significant difference between online and on-site Otago exercise (P=.29). The trial was small, cluster-randomized by site without blinding and not powered to judge effectiveness.
In 1,016 Chinese college students, ego depletion was associated with more cyberaggression (total effect b = 0.522), and physical activity and anger rumination together accounted for 54.02% of that link. The data are one-time self-reports, so the pathways describe statistical associations and not proven causes.
In 33 people with early Parkinson's disease, 12 weeks of twice-weekly dance classes produced no significant between-group differences in accelerometer-measured activity versus usual care. Stepping time and step count rose with moderate-to-large effect sizes (d = 0.72 and 0.91), but the study was too small to confirm them.