Purpose <p><?tk 2?>High body mass leads to changes in gait to accommodate the increased mechanical demands. Thus, previous research has examined how body mass affects gait variability (i.e., stride-to-stride fluctuations). However, these studies have been focused on the magnitude (extent of these fluctuations) of gait variability, while the effects of body mass on the temporal structure (how fluctuations are organized over time) remain unexplored. This study examined how body mass, quantified using body mass index (BMI), affects both magnitude and temporal structure of gait variability in young (24.66 ± 2.79 years), middle-aged (45.77 ± 5.85 years), and older adults (64.70 ± 7.65 years).</p> Method <p><?tk 2?>We analyzed stride interval and stride length of 117 participants in NONAN GaitPrint dataset. Magnitude of gait variability was quantified using standard deviation (SD) and coefficient of variation (CoV). Temporal structure was assessed by Hurst exponent (H), which characterizes gait variability correlation over time. Linear mixed models examined the effects of BMI, age, and their interaction.</p> Results <p><?tk 2?>Higher BMI significantly predicted greater magnitude of variability for both stride interval and stride length, independent of age. For the temporal structure, a significant interaction of BMI and age was observed for stride length <i>H</i>; higher BMI significantly predicted lower <i>H</i> in older adults (left: β = −0.293, <i>p</i> = 0.002; right: β = −0.231, <i>p</i> = 0.013) but not in young and middle-aged groups.</p> Conclusion <p>These findings suggest BMI affects the temporal structure of gait variability is age-dependent, which may have implications for gait assessment in ageing populations.</p>