Do you need to walk before you start running? For many people moving away from inactivity, walking is a useful starting point. But there is no universal number of weeks that prepares everyone for running, nor a mandatory sequence that works for all. The transition depends on current fitness and health, goals, known medical conditions, symptoms and how exertion is introduced.[1,2,3,4]
Walking deserves to be valued as a complete activity in its own context, not as a waiting room for running. It offers benefits of its own and can remain part of a routine even after the first bouts of jogging. For those who want to run, the issue is not simply improving stamina: it is gradually bringing the body closer to the demands of a different task.[1,4,5,6]
To understand this transition, it is essential to distinguish three goals: improving overall health, adapting to the specific demands of running and reducing the occurrence of injuries. These outcomes are related, but they are not interchangeable. A person may improve their blood pressure without developing all the strength needed to run; a measurable change in a tendon, in turn, does not prove that the likelihood of injury has decreased.
Walking already provides clinically relevant benefits
Fitness and cardiometabolic risk factors
Oja and colleagues brought together 37 randomized trials involving 2,001 previously inactive adults and interventions lasting at least eight weeks. Walking had favorable average effects on aerobic fitness, blood pressure, fasting glucose, body mass, body mass index and body fat. Not all outcomes improved: there was no significant effect on some measures of fat distribution and lipid variables. The optimal dose for each person remained undefined.[1]
A Cochrane review by Lee and colleagues, published in 2021, included 73 trials and 5,763 participants from 22 countries. The synthesis for systolic blood pressure, involving 5,060 participants, estimated an average reduction of 4.11 mmHg compared with the control, with a 95% confidence interval of 3.01 to 5.22 mmHg of reduction and moderate certainty. Diastolic blood pressure showed an average reduction of 1.79 mmHg, with low certainty. These results show that the benefits of walking do not depend on progressing to running; they do not guarantee the same individual response or replace hypertension treatment.[7]
A targeted update in 2026, involving 17 trials and 1,493 people with hypertension, also found favorable effects of brisk walking and explored dose–response relationships. However, estimating a dose range in a meta-analysis is not the same as directly testing all doses against one another or defining the best training for a healthy beginner. The findings should not be turned into a mandatory number of minutes before starting to run.[8]
Acute effects are not the same as chronic adaptations
In a 2026 review, Gale and colleagues assessed interruptions to sitting time in laboratory studies lasting less than 24 hours. There were 53 eligible studies, with 39 included in the meta-analysis. Short active breaks, especially walking breaks, improved post-meal glucose and insulin responses compared with remaining seated. This describes metabolic responses after meals; it does not demonstrate that a single walk prevents diabetes, lowers glycated hemoglobin or prepares a tendon for running.[9]
This distinction between acute and chronic effects helps in interpreting health messages. It is possible to obtain a metabolic response on the same day and still need sustained practice to develop functional capacity. There is no contradiction: the studies are measuring different processes and timescales.
Step counts: an activity indicator, not a readiness test
In 2025, Ding and colleagues brought together 57 studies from 35 cohorts; the meta-analyses used 31 studies from 24 cohorts. Higher daily step counts were associated with better health outcomes. In the all-cause mortality analysis, 7,000 steps per day, compared with 2,000, corresponded to an HR of 0.53, with a 95% CI of 0.46 to 0.60. This is an observational association: prior health, function, frailty and other factors can influence both step counts and outcomes.[10]
This result cannot be interpreted as proof that adding steps will cause the same risk reduction in everyone. Nor do 7,000 or 10,000 steps constitute a test of fitness to run. Counting steps helps describe activity; deciding on the transition to running requires a different question.
Comparisons between activities depend on what is matched
Williams and Thompson compared two large cohorts, involving 33,060 runners and 15,945 walkers, analyzing activity in terms of energy expenditure. They found similar associations for some cardiometabolic outcomes when energy expenditure was comparable. This does not mean that walking and running for the same number of minutes are equivalent interventions: time, distance and energy represent different doses. Exposure and diagnoses were self-reported, and the observational design does not establish causal equivalence or identical prevention for each person.[11]
Walking and running are not the same task at different speeds
During typical walking, at least one foot remains in contact with the ground; typical running includes a flight phase. The organization of ground support, force production and use of elastic energy all change. This is why an improvement in walking performance does not transfer fully to running, just as good cycling fitness does not, by itself, demonstrate tolerance for thousands of running foot contacts.[5,6]
Kharazi and colleagues studied 11 adults walking and running on a treadmill, combining ultrasound, movement analysis and an individualized estimate of the Achilles tendon's force–elongation relationship. The study showed greater tendon demands and differences in energy storage and return during running. It is an acute mechanics experiment, not a trial of longitudinal adaptation or injury prevention.[6]
Force estimates vary with speed, method and model. It is not appropriate to turn a laboratory value into a universal multiplier of impact or risk. Greater loading does not automatically mean damage; lower loading does not guarantee protection either. The clinical meaning depends on exposure, the person's capacity and what happens over time.
A study by Meardon and colleagues involving 40 recreational runners makes this distinction more concrete. The researchers estimated tibial stress using musculoskeletal models and individual imaging and movement data. Peak stresses were higher during running; however, the compressive and tensile stress integrals over each stance phase were lower than during walking. Peak stress and exposure over time are therefore not interchangeable. This experiment did not measure injury incidence or demonstrate acquired bone adaptation.[12]
Muscles, tendons and bones respond to different stimuli
Muscle strength and aerobic capacity are complementary
Walking requires muscular action and can improve the capacity to perform activities, but this does not mean developing all the strength needed for every task. Walking does not automatically replace a strengthening program. World Health Organization guidelines recommend combining aerobic activity and muscle strengthening; for older adults, they add multicomponent activity focused on balance and functional capacity.[4]
As population-level guidance, the WHO recommends that adults undertake 150 to 300 minutes of moderate-intensity aerobic activity per week, or 75 to 150 minutes of vigorous-intensity activity, or an equivalent combination, as well as muscle strengthening on at least two days of the week. These are general health targets, not a mandatory starting dose. Inactive people can start below these levels and progress according to their condition; reaching the full weekly target is not necessary before gaining benefits or discussing short bouts of running.[4]
It would also be incorrect to say that walking never improves strength. In a five-month randomized study involving 246 middle-aged and older adults, Nemoto and colleagues observed gains in isometric knee strength and aerobic capacity with higher-intensity interval walking. Adherence to the targets was incomplete, requiring caution when generalizing the results. This finding should not be extrapolated to gentle strolling, sufficient overall strength or injury prevention. The more precise conclusion is that effects depend on intensity and the task, and walking does not automatically replace progressive strengthening.[13]
Tendon adaptation: mechanical properties are not injury outcomes
The review by Lazarczuk and colleagues included 61 articles and 763 participants, with resistance interventions predominating. Adaptations occurred in the mechanical and material properties of tendons, particularly with higher-intensity resistance stimuli. In this context, the term tendon stiffness describes the relationship between force and elongation; it is not synonymous with the clinical sensation of a locked or painful joint.[5]
This analysis also distinguishes structural stiffness, material properties and geometry. The elastic modulus describes the behavior of the material; cross-sectional area describes the size of the tendon's cross-section. Changes in stiffness were more closely associated with material properties than with increased area. However, the analysis of these possible mechanisms was based on a subset of studies and does not demonstrate individual causality. A tendon can change its properties without a large increase in thickness; none of these measures serves as a certificate of readiness.[5]
The results were more consistent with resistance training than with aerobic activities alone. This does not establish tendon adaptation to running in any given person, nor does it establish how many weeks make each person ready to run. It also does not justify claiming that increasing a particular mechanical property necessarily reduces all injuries. The link between mechanism and prevention requires clinical studies of its own.[5]
Bone health: activity type, population and the region assessed matter
A network meta-analysis published in the BMJ in 2026 brought together 124 trials and 18,429 participants aged 40 and older. It found benefits of certain types of activity for bone mineral density, with generally low to moderate certainty. One crucial caveat is that brisk walking and jogging were grouped in the same category: the result for this group cannot be attributed to walking alone.[14]
Furthermore, improving mineral density in one region is not the same as demonstrating individual tolerance for repeated foot contacts or prevention of stress injury. Bone health, task adaptation and injury risk are related but distinct dimensions. The data do not support a universal clock by which the cardiovascular system becomes ready and bone or tendon becomes safe after a fixed period.
What trials show about a preparatory phase
Bredeweg and colleagues randomized 432 beginners to a four-week phase of walking and jumping or to a group without this preconditioning. After that interval, both groups began the same nine-week running program. The study analyzed 362 participants and reported injury incidences of 15.2% versus 16.8%, with no statistically significant difference. The package tested therefore did not demonstrate the expected preventive benefit.[2]
This design does not directly compare walking for a few weeks with immediately starting to alternate walking and running, because both groups began running on the same date. Nor does it prove that every possible form of preparation is useless. The conclusion must respect the intervention actually tested, losses between randomization and the start, and the injury definition used.
Harrison and colleagues compared preparatory walking with strength training and plyometrics in women who were beginners, followed by a running program. Of the 57 participants initially included, 21 completed the analysis. The main outcome was kinematics, not injury incidence. The study may contribute to understanding movement, but it does not determine the ideal duration of walking or demonstrate clinical protection.[15]
Why injury figures require context
In a 2015 review, Videbæk and colleagues estimated 17.8 injuries per thousand hours of running in beginners and 7.7 in recreational runners. The 95% confidence intervals were 16.7–19.1 and 6.9–8.7, respectively. These values are incidence rates in the groups studied, not individual probabilities, and should not be read as 17.8% or 7.7% of participants.[16]
Definitions of a beginner and an injury, follow-up duration and the way exposure was recorded varied. Runners who remain injury-free and continue running may also become a selected group over time. This comparison should therefore not be used to promise someone that their risk will automatically fall after a given number of weeks.
In a 2024 review of reviews, Correia and colleagues analyzed 13 evidence syntheses on risk factors, all rated as low or critically low quality. This calls for caution when turning an isolated finding on strength, alignment or movement into a sufficient cause of injury. The literature should be read as a set of factors and contexts, not as a list of bodily defects that need correcting before any running.[17]
A scoping review by Linton and colleagues, published in 2025, mapped 106 studies on support for runner-centered prevention. It broadens the discussion to education, behavior and implementation of strategies. As a mapping exercise, it does not, by itself, demonstrate that a multifactorial package or supervision is superior in a causal comparison. It supports the importance of also studying how a plan is applied and sustained.[18]
Load progression: why the 10% rule is not a law
Buist and colleagues randomized 532 beginners to a graded 13-week program based on the 10% rule or to an eight-week program. Among the 486 participants analyzed, injury incidences were 20.8% and 20.3%, with no significant difference. Both programs included walking and running: the trial does not compare alternating the two with continuous running.[19]
In a review of 36 articles involving 23,047 runners, Frédette and colleagues found conflicting evidence on distance, duration, frequency, intensity and recent changes in training. This does not mean that any increase is safe; it means the literature does not support a simple, universal formula for all runners.[20]
A 2025 cohort study added an important distinction. Frandsen and colleagues followed 5,205 runners and analyzed 588,071 sessions. The distance of each session was compared with the longest distance run in the preceding 30 days. Increases above 10% in a single session were associated with a higher rate of overuse injuries; in the category of increases above 10% and up to 30%, the HRR was 1.64, with a 95% CI of 1.31 to 2.05. The same association was not found for the week-to-week ratio.[21]
This observation does not revive the 10%-per-week rule. The study assessed a different exposure, was observational and predominantly included experienced runners. It does not identify the dose for the first jog of someone with no recent running history. Nor does it allow increases below the cutoff to be treated as a guarantee of safety. The lesson is to pay attention to individual sessions as well as the weekly total, without turning an association into a causal prescription.
In practice, external load can be described by running time or distance, frequency, speed and the difficulty of the route. A person's response is reflected in perceived exertion, recovery, symptoms and functional capacity. Recording these aspects helps organize decisions; it does not create a validated injury-prediction algorithm.
Alternating walking and running: usefulness and limitations
Alternating makes it easier to divide running into blocks and control how much running actually takes place. The NHS Couch to 5K program offers an operational example: in the first week, one minute of running and one and a half minutes of walking, with non-running days between sessions. The progression may take nine weeks or longer. This does not mean nine weeks of walking before the first jog, nor is it a suitable dose for every beginner.[22]
In a study involving 42 marathon runners, Hottenrott and colleagues observed less muscle pain and fatigue when running and walking were alternated, without a demonstrated advantage in the selected cardiac markers. Comfort at the end of a marathon does not answer whether beginners will have fewer injuries over months. The evidence examined does not establish universal preventive superiority of alternating walking and running over an appropriate dose of continuous running.[23]
There are clinical trials of alternating walking and running in other populations. ASTEROID randomized 40 adults with chronic low back pain to an individualized walk–run program with remote support or a waiting list for 12 weeks. Benefits were observed in pain and disability outcomes, but there were also non-serious adverse events. This study does not compare alternating walking and running with walking alone or continuous running and does not test primary prevention in healthy beginners. It is relevant clinical evidence, addressing a different question with a different comparator.[24]
The breaks need to be part of the total dose. They should not be used as permission to run the running segments much faster or to extend training excessively. One minute of running may represent vigorous exertion for a deconditioned person: walking between blocks does not automatically make the whole session moderate in intensity.
Strengthening: there is preventive evidence, but not for every program
Wu and colleagues brought together nine trials involving 1,904 participants in a review published in 2024. Overall, the exercise programs showed no statistically significant difference in injury risk or rate. The favorable result for supervised studies came from a post-hoc analysis; supervision was not randomized on its own. Seven studies were considered low quality. In addition, the search ended in January 2023: the review does not include the later trials by Leppänen and Naderi and cannot be used to negate them.[25]
The Run RCT trial, published by Leppänen and colleagues in 2024, randomized 370 participants; 325 began the interventions and were analyzed. A hip- and trunk-focused program, performed alongside the running program and supervised by a physiotherapist, showed a lower occurrence of lower-limb injuries than the stretching group, with an HR of 0.66 and a 95% CI of 0.45 to 0.97.[26]
This result does not mean 34 percentage points fewer injuries or an identical reduction for every person. The women and men participating could already run three kilometers or 20 minutes and do not represent everyone moving away from inactivity. In addition, the ankle-and-foot program tested in the same trial did not show the same benefit. It cannot be concluded that any strengthening routine or any single session produces the effect of the package studied.[26]
Another trial, first published in 2025 and appearing in a 2026 journal issue, assessed 264 recreational runners and an online multicomponent program. The abstract reports overall injury rates of 4.62 versus 8.71 per thousand hours, with a rate ratio of 0.53 and a 95% CI of 0.37 to 0.77. These figures refer to a multicomponent program and do not demonstrate a specific effect of walking breaks. Interpretation of this study remains limited to the data in its abstract, as indicated in the references.[27]
The contrast also appears in a pilot trial by Nguyen and colleagues involving 74 university students who were beginners. Replacing some running with bodyweight strengthening did not demonstrate a significant difference in injury incidence. Attrition was high: 52.6% in the strengthening-and-running group and 41.7% in the running group. A small pilot with attrition does not prove general ineffectiveness; moreover, both training content and running volume changed. The studies do not test identical interventions.[28]
A trial by Abran and colleagues, published in 2026 and involving 140 runners, offers another lesson: neither adopting minimalist footwear nor a program to change to softer running demonstrated a significant reduction in the overall injury rate compared with the control. Some differences in injury location emerged in secondary analyses. This result does not invalidate every individual intervention, but it shows why changing mechanics or redistributing load does not automatically equate to preventing injuries.[29]
Weight, age and medical conditions change how evidence applies
Obesity: interpreting the dose without creating a barrier to movement
Bertelsen and colleagues studied 56 beginners with a BMI between 30 and 35, comparing three and six kilometers of running in the first week; both groups alternated running and walking. The cumulative risk difference in the primary analysis was −16.3 percentage points, with a 95% CI of −43.8 to +11.3 and p=0.25. The wide interval includes benefit and no benefit; the trial fell short of its initially planned size.[30]
A per-protocol analysis favored the lower dose, but it depends on selection after randomization and should not replace the primary analysis. The study does not define three kilometers as a safe ceiling, does not prove a protective effect of walking and does not require losing weight to reach an ideal weight before starting activity. Strength, symptoms, medical conditions, experience and recovery need to be considered alongside weight.
If walking causes substantial pain or limitation, tolerated alternatives, such as cycling or water-based exercise, can form part of an individual plan while the cause is assessed. This is a practical option for adaptation, not a proven equivalent form of preparation for running.
Aging: independence may be the priority outcome
The LIFE study randomized 1,635 sedentary people aged 70 to 89 with functional limitations. The physical activity program combined walking, strength training and other components. Major mobility disability occurred in 30.1% of the exercise group and 35.5% of the education group, with an HR of 0.82 and a 95% CI of 0.69 to 0.98. The benefit belongs to the multicomponent program; it cannot be attributed to walking alone or translated into a need to run.[31]
A review of walking alone in older adults included nine studies and 1,309 participants, finding favorable results for walking endurance and quality of life, but very low certainty for the outcomes. Strength and balance also matter for preserving independence. Age alone does not prohibit running, but falls, frailty, bone health, medications and medical conditions can change goals and follow-up.[4,32]
Proportionate cardiovascular screening, without automatic testing for everyone
The American College of Sports Medicine screening model considers current activity, symptoms or known cardiovascular, metabolic and kidney diseases, as well as the intended intensity. Age and obesity alone do not create a universal requirement for tests before light-to-moderate walking in someone who is asymptomatic and has no known disease. For someone who is inactive and has known cardiovascular disease, diabetes or kidney disease, assessment before starting is recommended. The intention to undertake vigorous exertion also enters the screening process, but it does not automatically make an exercise stress test mandatory for everyone.[3,4]
When to consider the transition to the first bouts of jogging
There is no single test that certifies readiness for every beginner. Comfortable walking, controlled exertion, adequate recovery and the absence of progressively worsening localized pain or swelling are useful information. Being able to walk for around 30 minutes can serve as an operational reference, not a biological boundary: completing that time does not automatically provide clearance, and falling short of it does not, by itself, rule out a short, guided attempt.
The Running Readiness Scale illustrates the difference between measuring an attribute and predicting an outcome. The study involving 56 women who were beginners assessed reliability and relationships with running kinematics in a cross-sectional design. Its results do not validate prediction of future injuries or a universal safety cutoff. Tests used after surgery, fracture or tendon rupture should not be imported as requirements for every healthy adult either.[33]
As practical guidance, running can be introduced in small doses, leaving room for recovery and repeating stages. Avoiding simultaneous increases in time, speed and route difficulty makes the response easier to interpret. These decisions are individualized: they are not a single prescription or a proven method of eliminating injuries.
How to organize this transition in practice
The evidence does not provide a single training plan for everyone, but it helps organize decisions. For those who want to start running, the plan can be considered in five points, adjusted to each person's condition and goals:[3,4,22,26]
- Recognize the starting point. What is already comfortable: walking in daily life, walking at a faster pace or running short segments? Medical conditions, symptoms and limitations need to be part of this assessment.
- Control how much running takes place. Alternating can help distribute exposure; walking time should not be used to justify running too fast or extending the session excessively.
- Repeat a stage when necessary. The calendar is a guide, not an obligation. The response to exertion and recovery help determine when to increase, maintain or reduce the dose.
- Keep strengthening as a component of the plan. Content, intensity, frequency and the need for supervision depend on the person; a generic routine does not automatically reproduce the results of a trial.
- Observe progress, rather than simply completing the workout. Progressively worsening localized pain, swelling, a change in stride or declining function require a review of the plan and, as appropriate, clinical assessment.
These points are guidance for organization, not a validated prevention protocol. They do not replace assessment of symptoms or guarantee an injury-free transition.
Running is not synonymous with automatic knee wear
Greater mechanical demands do not justify claiming that running inevitably destroys cartilage either. A 2023 review brought together 17 studies involving 7,194 runners and 6,947 non-runners and found no consistent association with worsening radiographic signs or osteoarthritis outcomes over the period assessed. Another meta-analysis of nine observational studies estimated an OR of 0.97, with a 95% CI of 0.56 to 1.68, for osteoarthritis prevalence in runners versus controls.[34,35]
These syntheses have important limitations: selected populations, differences in age and follow-up, and risk of bias in several studies. They do not demonstrate that any volume is safe, that running treats all osteoarthritis or that the result applies equally to previously injured knees. The useful conclusion is to abandon the notion of inevitable, automatic wear, without replacing that myth with a promise of protection.
Warning signs should not be normalized as adaptation
Localized pain that increases, changes the stride, occurs with swelling or worsens the ability to perform usual activities warrants reviewing the load and assessing the cause. The absence of a fixed progression timeline does not mean persisting despite deterioration. Substantial difficulty bearing weight on the leg requires assessment, not a self-administered running test.
Chest pain, fainting or severe and disproportionate shortness of breath require stopping exercise and seeking assessment, urgently when acute or persistent. These symptoms should not be interpreted merely as deconditioning or compensated for with walking breaks.[3]
A summary for practice and learning
Walking before running is a useful option, especially for moving away from inactivity, but not a universal obligation. Walking offers benefits of its own; running adds specific demands. Metabolic benefit, tissue adaptation and injury prevention need to be assessed separately.
The evidence favors an approach combining consistency, a dose compatible with current condition, appropriate strengthening and attention to recovery. The calendar should serve the person, not the other way around. There is room for goals and planning, but not for selling a number of weeks, steps or percentages as a guarantee of safety.
From a scientific perspective, the best answer is not to choose between walking and running as though one activity were superior in every situation. It is to define which question is being answered, which outcome matters and how far the studies allow conclusions to go. This care makes guidance more useful, more individual and more honest.
How to interpret the cited studies
This article brings together evidence in a critical narrative review, with a targeted literature update through 12 November 2026. It is not a registered systematic review or a new meta-analysis. Clinical trials, reviews, cohorts, laboratory experiments and official guidance answer different questions: a biomechanical study explains the task; a cohort identifies associations; a trial tests a strategy in the population, at the dose and for the outcomes studied.
It is also necessary to distinguish denominators. An average reduction in blood pressure is not a reduction in the likelihood of injury. A rate per thousand hours of running is not the percentage of people who will be injured. And a result without a statistical difference does not prove that two strategies are equivalent. The figures presented should be interpreted with their confidence intervals, precision and the context of each study.
