Cycling produced lower fatigue-weighted cumulative knee joint loading than running and was not greater than walking, with dose-equivalence analyses indicating that cycling would need to be performed approximately 1–2 hours to match 1 hour of walking and approximately 3–5 hours to match 1 hour of running.
Key Findings
Results
Cycling produced lower fatigue-weighted cumulative knee joint loading (wCKL) than running and was not greater than walking across all planes.
Fifty-six recreational cyclists participated: 19 with knee osteoarthritis (KOA), 19 age-matched controls, and 18 young controls.
Cycling was tested at nine cadence-power combinations (60/80/100 rpm × 157/210/261 W).
Walking was performed at 1.4 m/s and running at 3.0 m/s.
wCKL was calculated from sagittal-plane and frontal-plane external knee moments derived via OpenSim and scaled to 1 hour of activity.
The finding held 'across planes' for all group comparisons tested.
Results
In the sagittal plane, cycling and walking showed comparable wCKL across all groups.
No significant difference in sagittal-plane wCKL was found between cycling and walking across the KOA and control groups.
Mixed-effects linear models were used to test task- and group-dependent effects.
This pattern was consistent across all three participant groups (KOA, age-matched controls, young controls).
Results
Individuals with KOA exhibited reduced sagittal-plane wCKL relative to controls during running.
KOA participants showed significantly lower sagittal-plane wCKL compared to control groups specifically during running.
This group difference was not observed during cycling or walking in the sagittal plane.
The KOA group consisted of 19 individuals matched in age to 19 control participants.
Results
In the frontal plane, individuals with KOA demonstrated significantly elevated wCKL during walking and running but not during cycling.
KOA participants had higher frontal-plane wCKL than controls during both walking and running.
During cycling, the frontal-plane wCKL of KOA participants was not significantly elevated relative to controls.
This finding suggests cycling may attenuate the elevated frontal-plane loading burden associated with KOA during weight-bearing locomotion.
Results
Dose-equivalence analyses indicated that approximately 1–2 hours of cycling is needed to match 1 hour of walking, and approximately 3–5 hours of cycling to match 1 hour of running.
Dose-equivalence estimates were derived to relate cycling duration to equivalent walking or running knee joint load exposure.
Larger ratios were found in the frontal plane specifically for the KOA group.
These ratios imply that cycling accumulates substantially less cumulative knee joint loading per unit time compared to both walking and running.
The ratios provide a practical framework for prescribing cycling as a low-load alternative to weight-bearing exercise.
Methods
Three-dimensional kinematics and reaction forces were used to compute external knee moments via OpenSim for all three activities.
Ground reaction forces were used for walking and running; pedal reaction forces were used for cycling.
External knee moments in the sagittal and frontal planes were the primary biomechanical outcomes.
wCKL integrated moment magnitude and repetition frequency in a fatigue-weighted manner, scaled to 1 hour of activity.
Mixed-effects linear models were used for statistical comparisons of task and group effects.
What This Means
This research suggests that cycling places substantially less cumulative load on the knee joint compared to running, and roughly comparable load compared to walking, when measured over the same period of time. The study tested 56 recreational cyclists — including people with knee osteoarthritis (KOA) and healthy controls of different ages — performing cycling at various speeds and resistance levels, as well as walking and running. By calculating a 'fatigue-weighted' measure of knee loading that accounts for both the size of forces and how many times they are applied, the researchers found that an hour of cycling produces far less total knee loading than an hour of running, and similar loading to an hour of walking.
A particularly notable finding involves the frontal plane (side-to-side forces across the knee), where people with KOA showed significantly higher loading during walking and running compared to healthy individuals, but not during cycling. This suggests that cycling may specifically reduce one of the types of knee stress that is elevated in people with KOA. The dose-equivalence calculations further quantified these differences: a person with KOA would need to cycle roughly 1–2 hours to accumulate the same knee load as 1 hour of walking, and 3–5 hours to match 1 hour of running, with even greater differences seen in the frontal plane.
This research suggests that cycling may be a particularly suitable form of exercise for individuals with knee osteoarthritis because it allows for physical activity with substantially less cumulative knee joint stress than running and without exceeding the loading seen during walking. The dose-equivalence framework developed in this study could be useful for healthcare providers and researchers designing exercise prescriptions or comparing exercise interventions for people managing knee osteoarthritis.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Ebbecke J, Hansen L, Viellehner J, Edwards W, Naeckel N, Potthast W. (2026). Comparing Knee Joint Load Accumulation During Cycling, Walking, and Running in Individuals With and Without Knee Osteoarthritis.. Scandinavian journal of medicine & science in sports. https://doi.org/10.1111/sms.70369