Real-world gait speed distributions were typically not bimodal, and gait bout length could not explain bimodality, so the findings do not support the idea that distinct components of the gait speed distribution reflect short versus long walking bouts.
Key Findings
Results
Well-separated bimodal gait speed distributions were uncommon, found in only 12 of 73 participants (16%).
Remote walking data were analyzed from four cohorts: healthy adults (n = 16), chronic stroke (n = 20), end-stage knee osteoarthritis (n = 21), and incomplete spinal cord injury (n = 16).
Participants wore a sensor on each foot and one on the lower back for three days.
Gaussian mixture modeling characterized each individual's gait speed distribution from all recorded strides, and bimodality was quantified using Ashman's D, with D > 2 indicating well-separated components.
Results
The occurrence of well-separated bimodal gait speed distributions did not evidently differ between people with and without gait impairment.
The abstract reports 'no evident differences between people with and without gait impairment' among the 12 of 73 participants with D > 2.
Cohorts compared were healthy adults and people with chronic stroke, end-stage knee osteoarthritis, and incomplete spinal cord injury.
Results
Across participants, mean gait speed increased and gait speed variability decreased with longer bout length.
This was an overall pattern across the cohorts, as reported in the abstract.
Bout length was derived from the same remote three-day sensor recordings.
Results
Among participants with bimodal distributions (D > 2), bout lengths did not differ significantly between the slow and fast speed components, although the difference approached significance.
Slow speed component: median 11 strides [IQR: 7.5, 31.8].
Fast speed component: median 34.5 strides [IQR: 20.8, 122].
The comparison yielded p = 0.056, in a sample of only 12 participants.
Conclusions
The authors conclude that the findings do not support the idea that distinct components of the real-world gait speed distribution reflect short versus long walking bouts.
Real-world gait speed distributions were typically not bimodal.
Gait bout length could not explain the bimodality that was present.
What This Means
Some researchers have suggested that everyday walking consists of two distinct types: slower walking in short bursts (for example, moving around the house) and faster walking in longer stretches. If true, a person's walking speeds across a day would form two separate peaks when plotted. This study tested that idea using sensors worn for three days by 73 people: healthy adults and people with stroke, severe knee osteoarthritis, or incomplete spinal cord injury.
Only 12 of the 73 participants (16%) showed two clearly separated speed groups, and this was not clearly different between people with and without walking problems. Overall, people walked faster and with less variation in speed during longer walking bouts. But even among the 12 people with two speed groups, the slow group did not have significantly shorter bouts than the fast group (median 11 vs 34.5 strides, p = 0.056), though this was borderline and the sample was small.
This research suggests that real-world walking speed usually does not split into two neat categories tied to short versus long bouts. For researchers using wearable sensors to measure walking in daily life, this implies that simple two-type models of walking may not capture how people actually walk, and other factors may matter.
Check Your Own Numbers
Upload your bloodwork. We'll cross-reference your results against this study and 4,700 others.
Boekesteijn R, van Mierlo M, Hannink G, Keijsers N, Smulders K. (2026). The distribution of real-world gait speed and its relation to gait bout length in people with and without gait impairment.. Journal of biomechanics. https://doi.org/10.1016/j.jbiomech.2026.113581