La vitesse à laquelle vous marchez à votre allure habituelle est l'un des indicateurs les plus simples et les plus puissants du bon fonctionnement de l'organisme. Elle prédit la survie, l'autonomie et les capacités cognitives futures assez bien pour que les cliniciens l'appellent le sixième signe vital. Une marche normale sollicite à la fois le cœur, les poumons, les muscles, les nerfs et l'esprit, si bien que l'allure reflète l'état de tout le système. Contrairement à beaucoup de marqueurs, elle répond à l'entraînement.
Trois réserves comptent. La vitesse de marche reflète votre santé mais ne la crée pas. Marcher plus vite n'est pas automatiquement plus sûr face aux chutes. Et une marche dont le schéma devient soudainement anormal constitue un avertissement différent qui nécessite un médecin.
Findings & Outcomes
What It Is
Walking speed is how fast you cover ground at your usual, unhurried pace, measured in meters per second. A normal walk depends on far more than the legs. The heart and lungs supply it, the muscles power it, the joints and nerves coordinate it, and balance and attention keep it steady. One easy walk draws on all of those together, so the pace mirrors the health of the whole body. Clinicians place it beside pulse, blood pressure, temperature and breathing rate. A single free measurement, taken with a hallway and a phone, predicts survival and independence about as well as a full workup of medical history and test results.
Walking speed belongs to the wider family of home self-checks, and the self-measures hub sets it beside grip strength, the chair stand and blood pressure.
What It Predicts
A slow usual pace predicts several outcomes at once: higher mortality, more disability and lost independence, and a greater chance of hospital or care admission. It also predicts a higher risk of cognitive decline and dementia. The pace tracks general health more than any single organ, so one number reaches many outcomes. In pooled data on 34,485 older adults, each 0.1 m/s faster came with about 12% lower mortality. Survival improved steadily across the whole range of speeds, with no single cutoff where risk suddenly jumped.
But that same broad reach is also the weakness. A slow walk is a common result of heart failure, lung disease, arthritis, neuropathy, low mood or plain deconditioning. The number alone cannot say which. Most of these findings are observational, and they share the same catch: the people who walk slowly are often already sick.
A slow reading marks the body's current state as much as it predicts the future.
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.
Progress Markers
Chaque augmentation de 0.1 m/s de la vitesse de marche s'accompagnait d'une baisse d'environ 12 % de la mortalité
Dans des données regroupées portant sur 34,485 personnes âgées, chaque augmentation de 0.1 m/s de la vitesse de marche habituelle s'accompagnait d'une baisse d'environ 12 % du taux de mortalité, et la relation se maintenait sur toute la plage, et non à un seul seuil.
Studenski et ses collègues ont regroupé des données individuelles issues de 9 études de cohorte portant sur des adultes vivant en communauté âgés de 65 ans et plus, 34,485 personnes avec une vitesse de marche habituelle moyenne de 0.92 m/s, suivies pendant 6 à 21 ans, période durant laquelle 17,528 sont décédées. La vitesse de marche était associée à la survie dans chaque cohorte (rapport de risque regroupé 0.88 par 0.1 m/s, IC à 95 % : 0.87 à 0.90). La survie augmentait sur toute la plage des vitesses, sans effet de seuil. Prédire la survie à partir de l'âge, du sexe et de la vitesse de marche était à peu près aussi précis qu'un modèle construit à partir de l'âge, du sexe, des maladies chroniques, des antécédents tabagiques, de la pression artérielle, de l'indice de masse corporelle et de l'hospitalisation combinés.
Who this may not transfer to:Survival curves are reported separately for men and women; the direction holds in both while the absolute speeds differ.
The study · 1
Studenski et al., gait speed and survival in older adults · JAMA 2011;305(1):50-58
Those whose speed rose over a year had 31.6% eight-year mortality against 49.3%
In 439 older adults, the ones whose walking speed improved over a year were much more likely to be alive eight years later, 31.6% had died against 49.3% of those whose speed never improved, and gait speed was the only one of six measures where improvement tracked survival.
Hardy and colleagues followed 439 adults aged 65 and older, assessing six measures of health and function at baseline and quarterly for a year, then linking a meaningful one-year change to 8-year mortality. Only improved usual gait speed (a rise of at least 0.1 m/s) predicted survival. Eight-year mortality was 31.6%, 41.2% and 49.3% across the improved, transiently improved and never-improved groups, and the benefit persisted after adjustment for demographics, comorbidity, cognition and hospitalization (hazard ratio 0.42, 95% CI 0.29 to 0.61), consistent across age, sex, ethnicity and initial speed. The authors call for research into whether interventions that raise gait speed change survival.
Who this may not transfer to:The survival benefit of improvement was consistent across sex subgroups in the cohort.
The study · 1
Hardy et al., improvement in usual gait speed predicts better survival in older adults · J Am Geriatr Soc 2007;55(11):1727-1734
Across 27 studies, a slow pace consistently forecast disability, care admission and death
Pooling the strongest studies, a slow usual walking pace consistently forecast losing independence, being admitted to care, falling and dying, and it did the job at least as well as longer batteries of tests. The 4-metre walk at your normal pace is the method behind most of that evidence.
An International Academy on Nutrition and Aging task force reviewed the literature and retained 27 studies meeting four criteria: usual-pace gait speed measured as a single item over a short distance, in autonomous community-dwelling older people, with adverse outcomes tracked over time. Across them, gait speed was a consistent risk factor for disability, cognitive impairment, institutionalisation, falls and mortality, and predicted these at least as well as composite instruments. The 4-metre usual-pace walk was the most frequently used protocol, which the group described as quick, safe, inexpensive and highly reliable.
Who this may not transfer to:The synthesised cohorts include men and women; norms and cut-points are applied separately by sex.
The study · 1
Abellan van Kan et al., gait speed at usual pace as a predictor of adverse outcomes (IANA Task Force) · J Nutr Health Aging 2009;13(10):881-889
A slow gait with a memory complaint roughly doubled dementia risk across 26,802 people
When a slow walk is paired with a memory complaint, the combination (present in about 1 in 10 older adults) roughly doubled the risk of later cognitive impairment and dementia, across 26,802 people in 17 countries.
The motoric cognitive risk syndrome pairs a measured slow gait with a subjective cognitive complaint, needing no laboratory test. Verghese and colleagues pooled 26,802 adults aged 60 and older without dementia from 22 cohorts across 17 countries, finding a prevalence of 9.7% with no sex difference. In prospective sub-cohorts, the syndrome predicted incident cognitive impairment (adjusted hazard ratio 2.0, 95% CI 1.7 to 2.4) and dementia (1.9, 1.5 to 2.3), persisting after excluding people with possible baseline impairment. The value is that two cheap observations together sharpen the signal that either gives alone.
Who this may not transfer to:Prevalence showed no sex difference; the syndrome and its dementia risk are reported across both sexes.
The study · 1
Verghese et al., motoric cognitive risk syndrome: multicountry prevalence and dementia risk · Neurology 2014;83(8):718-726
Slower walkers had about 60% higher dementia risk, up to 7 years before diagnosis
In a 9-year study of 3,663 older adults, slower walkers and those whose pace was dropping fastest were more likely to develop dementia, and the slowing showed up as much as 4 to 7 years before diagnosis.
The Three-City study in France followed 3,663 dementia-free adults (mean age 73.5) for 9 years, measuring gait speed over 20 feet (6 metres) four times, with 296 developing dementia. In age and sex adjusted models, 1-SD (0.204 m/s) lower baseline speed raised the hazard of dementia to 1.59 (95% CI 1.39 to 1.81), still evident when the walk was recorded 4 years (1.46) and 7 years (1.30) before onset. Independently, a steeper decline in speed carried a hazard of 3.39 per SD. The pattern places slowing gait among the early motor signs that precede a cognitive diagnosis.
Who this may not transfer to:Models were adjusted for sex; the association is reported across the mixed cohort.
The study · 1
Dumurgier et al., gait speed and decline in gait speed as predictors of incident dementia · J Gerontol A Biol Sci Med Sci 2017;72(5):655-661
The fastest walkers were not automatically the safest from falls
Among older people who had already fallen, walking faster did not mean fewer falls: the fastest walkers fell more than the moderate-paced, and in those low on balance confidence both the fastest and the slowest were at higher risk. Gait speed reads health, but for falls it is not a case of faster is always safer.
Tsang and colleagues followed 461 community-dwelling adults aged 65 and older (median age 69) who had fallen in the prior year for 12 months; 65 people (14%) reported 83 falls. Against a moderate-speed reference (0.81 to under 1.30 m/s), the high-speed group (1.30 m/s or more) carried an increased fall risk (adjusted odds ratio 1.84 to 2.37). Balance confidence modulated the pattern: it was linear in confident walkers and U-shaped in those with low confidence, where both the fastest and slowest walkers were at elevated risk. The finding cautions against reading a fast walk alone as protection from falls, a different question from what speed says about survival.
Who this may not transfer to:The cohort includes men and women; the finding concerns older adults already at risk of falls, not the general population.
The study · 1
Tsang et al., gait speed and falls in community-dwelling older adults · J Am Med Dir Assoc 2023;24(12):2002-2008
Muscle And Strength
Resistance training raised walking speed by about 0.08 m/s
Progressive resistance training raised older adults walking speed by a modest amount, about 0.08 m/s across 24 trials, while building substantial strength. The number responds to training, so a low reading is a signal to act on, not a fixed fact.
Liu et Latham ont regroupé 121 essais randomisés d'entraînement en résistance progressif chez des personnes âgées (6,700 participants). Sur les 24 essais qui l'ont rapportée (1,179 participants), la vitesse de marche s'est améliorée en moyenne de 0.08 m/s, un changement modeste mais réel, tandis que la force augmentait de façon substantielle (différence moyenne standardisée de 0.84) et que la capacité physique s'améliorait (0.14). La lecture pratique est que la vitesse de marche augmente lorsque la force globale du bas du corps augmente, de sorte que l'entraînement en résistance agit sur le chiffre, non sur la marche elle-même.
Who this may not transfer to:The pooled trials include men and women; the response to training is broadly similar while starting speeds differ by sex.
Treat gait speed as a readout of whole-body condition. Build it with progressive lower-body resistance work (squats, step-ups, leg press) two or three times a week, add aerobic activity for the heart and lungs the walk also reflects, and train balance separately, since that is a distinct capacity. See resistance training for how to load and progress it.
The study · 1
Liu and Latham, progressive resistance strength training for improving physical function in older adults · Cochrane Database Syst Rev 2009;(3):CD002759
L'entraînement en résistance à haute intensité a le plus augmenté la vitesse de marche, de 0.13 m/s
En comparant les types d'exercice tête à tête, l'entraînement en résistance mené à haute intensité a le plus augmenté la vitesse de marche, de 0.13 m/s, une différence suffisante pour compter, l'exercice rythmique et les programmes mixtes apportant des gains moindres. Une force des jambes suffisante semble être l'ingrédient clé.
Van Abbema et ses collègues ont mené une méta-analyse de 25 essais randomisés sur l'exercice pour la vitesse de marche préférée chez des adultes de 65 ans et plus. L'entraînement en résistance progressif, poussé à 70 à 80 % de la répétition maximale, a donné l'effet le plus important, 0.13 m/s (IC à 95 % : 0.09 à 0.16), ce qui dépasse le seuil d'environ 0.1 m/s considéré comme un changement significatif. L'exercice comportant une composante rythmique a ajouté 0.07 m/s, et l'entraînement combiné en résistance, équilibre et endurance, 0.05 m/s, les autres modes n'étant pas significatifs. Les auteurs concluent qu'une force musculaire suffisante est une condition importante pour une marche plus rapide, ce qui explique pourquoi le travail de force ouvre les conseils d'amélioration.
Who this may not transfer to:Trials include men and women; the ranking of modes is reported across the pooled sample.
Placez l'entraînement en résistance progressif au centre si l'objectif est une marche plus rapide, en le menant vers une charge exigeante au fil du temps. Ajoutez un travail aérobie et, pour la stabilité, un entraînement à l'équilibre, qui sert la prévention des chutes et l'autonomie même là où son effet direct sur la vitesse est plus faible.
The study · 1
Van Abbema et al., what type, or combination of exercise can improve preferred gait speed in older adults? A meta-analysis · BMC Geriatr 2015;15:72
How it works
Ce qui compte comme lent dépend de votre âge et de votre sexe
Sur plus de 51,000 adultes en bonne santé, une allure de marche confortable variait d'environ 1.40 m/s chez les hommes d'âge moyen à environ 0.97 m/s chez les femmes de plus de 80 ans, de sorte qu'une vitesse brute n'a de sens que rapportée à son propre âge et sexe.
Les travaux normatifs de type Bohannon ont été actualisés par une revue systématique et méta-analyse portant sur 51,248 adultes apparemment en bonne santé vivant en communauté, stratifiant la vitesse de marche confortable par décennie et par sexe. La moyenne pondérée allait d'environ 1.40 m/s (hommes dans la quarantaine) à environ 0.97 m/s (femmes de 80 ans et plus). Les hommes marchaient plus vite à tout âge et leur vitesse se maintenait jusqu'au-delà de 50 ans, tandis que les femmes commençaient à ralentir au-delà de 30 ans. La région géographique ne modifiait pas significativement les valeurs, si bien que l'âge et le sexe sont les comparaisons pertinentes pour situer une mesure.
Who this may not transfer to:Values are stratified separately for men and women at each age; men are consistently faster.
The study · 1
Andrews and Vallabhajosula, normal gait speed varies by age and sex but not by geographical region, a systematic review · J Physiother 2023;69(1):47-52
How to Measure It
The reference protocol is the 13-foot (4-meter) walk at your usual pace, the method behind most of the outcome research. Mark a 13-foot (4-meter) course on a flat floor. Add about six feet (2 meters) of run-off at each end so you are moving at a steady pace when timing starts. From a standing start, walk the way you would down the street: no target, no counting out loud. Wear your normal shoes, and use your walking aid if you rely on one. Time from your first movement to crossing the far mark, and take the better of two walks. Divide 4 by your time in seconds: 13 feet (4 meters) in 5 seconds is 0.8 m/s, and 4 seconds is 1.0 m/s.
If you would rather not time a walk, everyday life gives a rough proxy. Notice whether you have started falling behind the people you walk with. Watch whether keeping up on a gentle hill has become work it was not a year ago. Watch whether a pedestrian crossing signal runs out before you reach the far curb. These give you no number to set against the 4-meter test. A clear, recent change in how easily you keep up is still worth acting on, especially alongside feeling generally weaker or more tired.
How to Improve It
Walking speed rises as the body behind it gets stronger, so train overall capacity. Strength has the clearest trial evidence. High-intensity progressive resistance training raised preferred walking speed by 0.13 m/s in a meta-analysis of 25 trials. That gain is bigger than the roughly 0.1 m/s that counts as a meaningful change. Leg strength appears to be the key ingredient. A broader Cochrane pooling put the average gain across mixed programs at 0.08 m/s. Build with lower-body work (squats, step-ups, leg press and loaded carries) two or three times a week, and the strength minimum dose page covers where to start.
Two other capacities the walk depends on need their own attention. Aerobic fitness supplies the heart and lungs, so regular brisk walking and other cardio belong in the mix. Balance is a separate capacity that walking speed does not train, and it matters as much for staying steady and independent as raw pace does. The balance and fall prevention page covers what works there. Treating pain, correcting your vision and reviewing medications that make you unsteady each clear away obstacles to walking that no amount of exercise can fix.
What It Is Not
Walking faster on command raises the number without changing the health it reflects. Walking speed reports on the body's state; it is not itself a control you can push.
The number captures only speed. A walk's pattern carries its own warning.
Reading Walking Speed Well
Everything to be aware of is here, in one place. This practice suits most healthy people; a few situations call for real care.
Un schéma de démarche nouvellement anormal, pas seulement lent, a presque doublé le risque ultérieur de démence
Verghese et ses collègues ont examiné 422 adultes vivant en communauté, de plus de 75 ans, sans démence, et les ont suivis pendant une médiane de 6.6 ans, période durant laquelle 125 ont développé une démence. Une anomalie neurologique de la démarche à l'inclusion (présente chez 85 personnes, couvrant des schémas instable, frontal, hémiparétique, neuropathique, ataxique, parkinsonien et spastique) prédisait la démence globale (rapport de risque de 1.96, IC à 95 % : 1.30 à 2.96) et la démence non-Alzheimer en particulier (3.51, 1.98 à 6.24), la démence vasculaire en étant la composante la plus importante (3.46). La démence d'Alzheimer n'était pas prédite. Pour une auto-évaluation, l'enseignement à retenir est que la forme de la marche, pas seulement sa vitesse, porte une information, et qu'un schéma nouvellement anormal signale des causes neurologiques ou vasculaires souvent traitables qui justifient une évaluation.Verghese et al., abnormality of gait as a predictor of non-Alzheimer's dementia
A slow number is a signal, not a diagnosis
A slow reading is a signal to check your fitness, strength and activity, not a diagnosis. Act on it, and see a clinician where warranted.
A sudden change needs a clinical look
A gradual slowing over years fits normal aging. A gait that changes suddenly (becoming unsteady, shuffling, dragging on one side, or paired with numbness, weakness, difficulty speaking or a drooping face) needs prompt medical attention. Several causes (stroke, nerve compression, fluid on the brain, Parkinsonism) are identifiable and some are treatable. That is why a changed gait is worth assessing.
Pain and injury change the reading on their own
A painful hip, knee, foot or back, a recent injury, or a flare of arthritis will slow the walk for reasons unrelated to whole-body decline. A one-sided change, or one that tracks a specific pain, points to a local cause. The number is not comparable while the pain lasts.
Faster is not automatically safer for falls
Among older people already prone to falling, the fastest walkers were not the safest. Among those unsure of their own balance, both the fastest and the slowest fell more often. A faster pace is not a fall-prevention plan.
Keep the test consistent, and stop if it hurts
Keep the course length, pace instruction, footwear and any walking aid the same. Otherwise a change in method can look like a change in speed. Do the walk where you can steady yourself. Stop and seek advice if it brings chest pain, unusual breathlessness, dizziness or an irregular heartbeat.
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.
Go Deeper
- Self-measures hub: walking speed alongside grip strength, the chair stand, blood pressure and the other home self-checks, each with the protocol its numbers came from.
- Grip strength: the other simple readout that predicts survival and independence, and why it too is a whole-body signal.
- Resistance training: the strongest lever on walking speed, and how to load and progress lower-body strength safely.
- Strength minimum dose: how little strength work it takes to start moving the number, and how to build from there.
- Balance and fall prevention: the separate capacity that keeps you steady; walking speed does not train it.
- Walking: what the step-count research does and does not support, and how to fit more of it into a real day.
Common Questions
What is a good walking speed for my age?
Read your number against your own age and sex. Pooled data from more than 51,000 healthy adults put a comfortable pace at roughly 1.40 m/s for men in midlife, down to about 0.97 m/s for women in their eighties. Men are faster at every age. As a rough guide to the outcome research, a usual pace around 0.8 m/s is where adverse-outcome risk rises in older adults, and about 1.0 m/s or faster is reassuring. Your own trend over months matters more than any single cutoff.
Why does walking speed predict how long people live?
In one analysis, predicting survival from age, sex and walking speed proved about as accurate as a fuller model. That model added chronic conditions, smoking, blood pressure, body mass index and hospitalization. A single walk stands in for the body's overall condition.
Can walking speed predict dementia?
Walking speed is linked with dementia, and the slowing can start years before a diagnosis. In a 9-year study of 3,663 older adults, slower walkers and those whose pace fell fastest were more likely to develop dementia. The slowing was visible 4 to 7 years before onset. A slow walk paired with a memory complaint, a pattern called motoric cognitive risk, roughly doubled dementia risk across 26,802 people in 17 countries. The evidence is observational, and early brain changes can slow the walk before they are noticed. A slow gait may be an early sign of that process itself.
Does walking faster on purpose make me healthier?
A year of faster walking tracked better survival in older adults. Whether pushing the pace itself changes the outcome has not been tested.
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
All 10 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.