Learn physiotherapy · BPT-207 Biomechanics and Kinesiology
The Gait Cycle
You will be asked to memorise the phases of gait and the percentage of the cycle each one occupies. The phases are real and worth knowing. The percentages are not constants — they move with walking speed, and speed changes almost every gait variable you will measure.
In one line. The gait cycle is one complete sequence from initial contact of one foot to the next initial contact of the same foot, divided into a stance phase, when that foot is on the ground, and a swing phase, when it is in the air.
Stance is conventionally subdivided further: a loading period at the start, a foot-flat period in the middle, and a pushing period at the end leading to toe-off. [2] Those divisions are what a gait analysis system actually measures.
Speed is the variable that moves everything else. A systematic review of 20 studies and 587 healthy people found that walking speed affected spatiotemporal parameters, joint kinematics, joint kinetics and ground reaction forces, with values decreasing at slow speeds and increasing at fast speeds relative to comfortable speed. [1] Effect sizes reached beyond 4 standard deviations in young adults. Any gait figure quoted without its speed is incomplete.
What the evidence shows
| Question | Finding | Source and quality |
|---|---|---|
| How much does speed change gait? | Substantially, across every category measured. Effect sizes for slow and fast walking against comfortable speed reached -3.61 to 0.59 in children, -3.56 to 4.06 in young adults and -1.76 to 1.43 in older adults | 20 studies, 587 healthy people (227 children, 310 young adults, 59 older adults). The authors conclude speed must be accounted for when comparing patients with controls [1] |
| What does a gait system actually measure? | Speed, cycle duration and its variability, asymmetry, stride length, cadence, double support, stance, swing, and the loading, foot-flat and pushing subdivisions of stance | Definitions as used in a 545-participant normative study: stance is the time the foot is on the ground during a cycle, swing the time it is in the air [2] |
| How does gait change with age? | Speed and stride length decline; cycle duration variability and asymmetry increase. The difference reached significance in those aged 80 to 91 compared with younger groups | 629 healthy normal-weight adults aged 20 to over 90, 545 analysed, evenly distributed across seven decades with about 40 men and 40 women per decade [2] |
| Is gait one thing or several? | Several. Factor analysis of nearly two dozen parameters identified five independent domains: rhythm, phase, variability, pace, and base of support | 294 participants over 70 from the Mayo Clinic Study of Aging. Reference values for 23 parameters are published. Domains differed by sex and across age groups [3] |
| Do muscles act as clean agonist/antagonist pairs? | Rarely. Tibialis anterior and gastrocnemius lateralis acted as pure agonist/antagonists, with no co-contraction, in only 19.2% of strides (SD 10.4). Co-contraction appeared in early stance (46.5% of strides), mid-stance (28.8%), pre-swing (15.2%) and swing (73.2%) | 100 healthy children aged 6 to 11, about 16,500 strides in total, roughly 165 strides analysed per child [5] |
| Do normative values agree between laboratories? | No. The authors report disparities in reference values in recent literature even between similar research groups, attributed to differences in method, walking speed and laboratory environment | Their explicit recommendation: check more than one source of normative values when looking for reference data [4] |
Where students get this wrong
1. Treating the phase percentages as constants
The proportion of the cycle spent in stance is not a property of human beings; it is a property of human beings walking at a particular speed. Because speed changes spatiotemporal parameters systematically, [1] a percentage learned for comfortable walking does not describe the same person walking slowly after a stroke. Learn the phases and what defines them; treat the proportions as speed-dependent.
2. Comparing a patient with a textbook rather than with a matched control
The systematic review's own conclusion is that speed effects "should be considered when comparing the gait analysis of pathological individuals with normal or control ones". [1] A slow-walking patient will differ from normative data on nearly every variable, and most of that difference may be speed rather than pathology.
3. Quoting one set of normative values as though it were definitive
Reference values disagree between laboratories studying similar groups, because of method, speed and environment. [4] The 2011 Mayo dataset covers people over 70; [3] the 2024 Swiss cohort covers 20 to over 90. [2] They are not interchangeable. Name the source of any normative figure you quote, and check a second.
4. Picturing muscles switching cleanly on and off
The diagram in which each muscle fires during its phase and is silent otherwise is a simplification. In healthy children, ankle dorsiflexors and plantarflexors behaved as pure agonist/antagonist pairs in under a fifth of strides; co-contraction was the norm, and was present in nearly three quarters of strides during swing. [5] Co-contraction in a patient is therefore not automatically abnormal — it is part of how healthy gait is controlled.
5. Reducing gait to walking speed
Speed is one of five statistically independent domains — alongside rhythm, phase, variability and base of support. [3] Two patients can walk at the same speed with completely different rhythm and variability profiles. Speed is the most responsive single measure and it is not a summary of gait.
6. Assuming age changes gait uniformly
The pattern is not a steady decline from midlife. In a cohort spread evenly across seven decades, the significant difference in speed and stride length appeared in the 80 to 91 group. [2] Variability and asymmetry rose with age while other parameters held up longer. "Gait declines with age" is true and too coarse to use clinically.
What the evidence supports — and what it does not
Supported
- Dividing the cycle into stance and swing, with loading, foot-flat and pushing subdivisions. [2]
- Recording walking speed alongside every other gait variable. [1]
- Using more than one normative source. [4]
- Reporting several domains, not speed alone. [3]
- Expecting increased variability and asymmetry with age. [2]
- Treating co-contraction as part of normal control, not automatically as pathology. [5]
Not supported
- Fixed phase percentages independent of speed. [1]
- Comparing a patient to normative data without matching speed. [1]
- Treating any single laboratory's reference values as definitive. [4]
- Using gait speed as a summary of gait. [3]
How certain is this?
Evidence grade: Moderate.
The speed finding is the most secure: a systematic review of 20 studies and 587 people across three age groups, reporting consistent direction with large effect sizes. [1] The normative datasets are substantial for their populations — 545 adults across seven decades [2] and 294 adults over 70 [3] — and the domain structure comes from a formal factor analysis rather than from convention. [3]
What limits the grade is that normative gait data are laboratory-specific by the field's own admission, [4] and the largest datasets here are of older adults, so reference values for younger patients rest on smaller samples. The 2018 study contributing the disagreement finding used 28 women of one age. [4]
Sourcing note. This is a curriculum resource. It is written against sources that can be opened and checked, and it does not reproduce figures, tables or diagrams from any textbook.
Common questions
What percentage of the cycle is stance?
This page will not give you a single number, because the evidence says the honest answer is "it depends on speed". [1] What is stable is the definition: stance is the portion of the cycle in which that foot is in contact with the ground, subdivided into loading, foot-flat and pushing. [2] If your examination expects a specific figure, give it and add that it applies at comfortable walking speed.
How do I compare my patient to normal?
Match the speed, or accept that most of the difference you see may be speed. [1] Then name the normative source you are comparing against and be aware it may disagree with another laboratory's. [4] Where possible, use the patient as their own control across time rather than against a published mean.
Which gait measures should I record?
At minimum speed, stride length, cadence and a measure of variability, because these sit in different independent domains. [3] Recording three parameters from the same domain tells you less than three from different ones. Double support time and step width add the phase and base-of-support domains.
How does this connect to observational gait analysis?
Directly. Observational scoring is how most physiotherapists assess gait, and its reliability is better for distal segments than proximal, and better from video than in real time — see observational gait analysis, which reports the reliability figures and the minimal detectable change.
References
- Fukuchi CA, Fukuchi RK, Duarte M. Effects of walking speed on gait biomechanics in healthy participants: a systematic review and meta-analysis. Systematic Reviews. 2019 Jun 27;8(1):153. doi:10.1186/s13643-019-1063-z PMID 31248456 Systematic review and meta-analysis
- Rössler R, Wagner J, Knaier R, et al. Spatiotemporal gait characteristics across the adult lifespan: Reference values from a healthy population - Analysis of the COmPLETE cohort study. Gait & Posture. 2024 Mar;109:101–108. doi:10.1016/j.gaitpost.2024.01.005 PMID 38290395 Cross-sectional normative study
- Hollman JH, McDade EM, Petersen RC. Normative spatiotemporal gait parameters in older adults. Gait & Posture. 2011 May;34(1):111–8. doi:10.1016/j.gaitpost.2011.03.024 PMID 21531139 Factor analysis and normative reference values
- Fryzowicz A, Murawa M, Kabaciński J, et al. Reference values of spatiotemporal parameters, joint angles, ground reaction forces, and plantar pressure distribution during normal gait in young women. Acta of Bioengineering and Biomechanics. 2018;20(1):49–57. PMID 29658524 Normative reference study
- Di Nardo F, Mengarelli A, Burattini L, et al. Normative EMG patterns of ankle muscle co-contractions in school-age children during gait. Gait & Posture. 2016 May;46:161–6. doi:10.1016/j.gaitpost.2016.03.002 PMID 27131195 Normative dataset, statistical gait analysis
About this resource
How to use this
Written to be learned from, not memorised.
This page teaches the phases as definitions and the percentages as speed-dependent, because that is what the evidence supports. Faculty may use this page in teaching with attribution. It carries its review date and its next review date, so you can see at a glance whether it is current before you put it in front of a cohort.