Human Movement & Mobility Research
Sony RX0-II

Age-Related Differences in Joint Kinematics and Spatiotemporal Parameters During Ramp Walking and Level Walking: Cross-Sectional Study

JMIR Biomedical Engineering
•
2026
•
Original research
Theia3D version:
v2021.2.0.1675
More Articles

Study at a glance

In 20 young and 20 older healthy adults assessed in a simulated living laboratory, Theia3D-derived joint kinematics revealed substantially more pronounced age-related differences during ramp walking than during level walking, even after statistically accounting for differences in walking speed between the two age groups.

Summary

Disclaimer: The following summary details independent academic research. Theia3D is intended for research, sports performance and educational use only. It is not cleared as a medical device and is not intended to be used for the diagnosis, treatment, mitigation or prevention of any disease, injury or medical condition.

Editor's note: Theia3D version v2021.2.0.1675, as stated in the paper. This study uses Theia3D as a measurement tool to derive lower-limb joint kinematics and spatiotemporal parameters for an age-related gait research question, rather than evaluating Theia3D's own kinematic accuracy.

Study Overview

Design

A cross-sectional study comparing lower-limb sagittal-plane joint kinematics and spatiotemporal gait parameters between young and older healthy adults during level walking, ramp ascent, and ramp descent in a simulated living laboratory environment, using markerless motion capture.

Participants

Forty healthy adults: 20 young adults (15 men, 5 women; mean age 31.3 years; mean height 1.68 m; mean weight 63.5 kg), who were staff members at the National Center for Geriatrics and Gerontology, and 20 older adults (11 men, 9 women; mean age 64.2 years; mean height 1.65 m; mean weight 63.7 kg), recruited through a work-support organization for older adults in Japan. Participants with any neurological or musculoskeletal condition affecting gait were excluded.

Data Collection

Testing took place in a living laboratory combining indoor and outdoor spaces, with a 7-meter level walking path and a 3.25-meter ramp at a 7-degree incline fitted with handrails that participants were asked not to use. Each participant completed five trials each of level walking, ramp ascent, and ramp descent, in that fixed order, at a self-selected pace, starting with the right foot. Twenty-seven synchronized Sony DSC-RX0M2 cameras recorded each trial at 60 Hz and were processed through Theia3D (inverse kinematics with a two-degree-of-freedom knee model) and exported to Visual3D for spatiotemporal and kinematic analysis. Gait events were identified using an established foot-kinematics-based algorithm with manual visual confirmation, and sagittal-plane hip, knee, and ankle joint angles were calculated and time-normalized to the gait cycle; frontal- and transverse-plane angles were not analyzed.

Analysis

A linear mixed model was used to compare representative joint angle values between age groups within each walking condition, with walking velocity included as a covariate to adjust for the confounding effect of the age groups' differing walking speeds, age group as a between-participant factor, and walking condition as a within-participant factor. The false discovery rate method corrected for multiple comparisons, and effect sizes were calculated as Cohen d to evaluate whether age-related differences were more pronounced during ramp walking than level walking. Spatiotemporal parameters (cadence, velocity, and step length) were compared between groups using independent two-tailed t tests with the same correction and effect-size approach.

Key Findings

  • Older adults showed significantly shorter step length than younger adults across all three walking conditions (level walking: 0.67 m vs 0.73 m, P=.01; ramp ascent: 0.64 m vs 0.72 m, P=.003; ramp descent: 0.59 m vs 0.67 m, P=.003), each with a large effect size, while cadence did not differ significantly between groups in any condition.
  • Walking velocity was significantly lower in older adults during ramp ascent and ramp descent, but the same age-related slowing during level walking did not reach statistical significance (ramp ascent: 1.08 m/s vs 1.24 m/s, P<.001; ramp descent: 1.11 m/s vs 1.27 m/s, P=.004; level walking: 1.28 m/s vs 1.39 m/s, P=.06), underscoring that ramp walking placed a more detectable demand on older adults' gait speed specifically.
  • After statistically adjusting for walking velocity, level walking showed significant age-group differences in only 2 of 12 joint angle parameters (hip flexion angle at mid-stance and knee extension angle at stance), despite several other level-walking parameters showing large effect sizes (|d| = 0.80-1.18) that did not reach statistical significance, suggesting level walking may not fully reveal the underlying age-related motor adaptations present in this sample.
  • Ramp ascent and ramp descent showed substantially more extensive and larger age-related kinematic differences than level walking, even after adjusting for walking velocity. During ramp ascent, 4 of 12 parameters reached significance with large effect sizes (|d| = 1.36 to 2.37), including markedly increased knee flexion during early stance and reduced ankle plantarflexion range of motion in older adults; during ramp descent, 5 of 12 parameters reached significance with large effect sizes (|d| = 1.46 to 1.87), including increased hip and knee flexion alongside reduced knee extension and ankle plantarflexion in older adults.
  • The pattern of differences observed across ramp ascent and descent reflected coordinated adaptations across multiple joints rather than isolated single-joint changes, with older adults showing increased hip extension at terminal stance during ascent (consistent with a shift toward more hip-dominant, proximal power generation) alongside reduced ankle plantarflexion and range of motion (consistent with reduced ankle push-off capacity).

What This Means for Movement Science Researchers and Mobility Assessment Teams

For researchers and teams studying age-related mobility decline, this study offers a specific, practically useful finding: ramp walking revealed substantially more age-related kinematic differences than level walking, even after accounting for the fact that older adults walked more slowly, suggesting that standard level-walking assessments alone may understate the gait adaptations occurring with age. The specific, coordinated multi-joint patterns identified, reduced ankle plantarflexion and range of motion during ascent, and reduced knee extension and ankle plantarflexion during descent, offer concrete targets for future work examining whether these patterns predict functional decline or fall risk over time. More broadly, this study continues to build out the case for living-laboratory testing environments combined with markerless motion capture as a practical way to study age-related mobility under conditions that more closely resemble the varied demands of everyday walking, extending prior reliability work establishing Theia3D's measurement consistency in this same living laboratory setting.

The Role of Theia3D in Movement Analysis

Theia3D was used as the markerless motion capture processing system throughout this study, generating the lower-limb joint kinematics and spatiotemporal parameters used to compare age-related gait adaptations across level walking, ramp ascent, and ramp descent in a 27-camera living laboratory setup. Because this study used Theia3D as an established measurement tool to investigate an age-related mobility research question rather than to test the software's own measurement accuracy, it builds directly on companion research from the same research group establishing Theia3D's day-to-day measurement reliability for level walking, ramp walking, and stair walking in this same living laboratory environment, offering a practical demonstration of that established reliability being applied to a substantive aging-and-mobility research question.

Study Limitations and Future Directions

The authors note that this study analyzed only kinematic data; kinetic measures such as joint moments and forces were not assessed, which the authors identify as a necessary next step for more fully understanding the age-related gait strategies observed, potentially in combination with musculoskeletal modeling. The study's cross-sectional design means that while it demonstrates age-group differences at a single point in time, it cannot establish whether these kinematic patterns predict future functional decline or fall risk, a question the authors identify as requiring longitudinal research with follow-up periods of five to ten years. The older-adult sample was relatively young and narrowly distributed in age (mean 64.2 years, SD 0.8 years), limiting how well these findings generalize to frailer or substantially older populations, and sex distribution was unbalanced between groups, which the authors note may have influenced results given well-documented sex-related differences in gait biomechanics. The authors also note that several potential confounding factors, including habitual physical activity, strength, other health conditions, medications, footwear, and fear of falling, were not measured or controlled for, and that cognitive and perceptual factors known to influence gait, such as fear of falling and attentional load during dual-task walking, were not examined in this study and are identified as important directions for future research.

Citation: Shimotori, D., Fujisawa, S., Nishimura, M., Yoshimi, T., & Kato, K. (2026). Age-related differences in joint kinematics and spatiotemporal parameters during ramp walking and level walking: Cross-sectional study. JMIR Biomedical Engineering, 11, e86036. https://doi.org/10.2196/86036

Link: Read the full study

Does Ramp Walking Reveal Age-Related Gait Changes That Level Walking Misses?

This study uses Theia3D markerless motion capture in a living laboratory setting to show that ramp walking reveals substantially more pronounced age-related kinematic differences than level walking, even after accounting for walking speed. If your work involves studying age-related mobility decline, fall risk, or real-world gait adaptation, talk to our team about applying Theia3D to your own aging and mobility research.

Recent Posts