Effect of age on the variability and stability of gait: A cross-sectional treadmill study in healthy individuals between 20 and 69 years of age

•We designed a cross-sectional study to analyze the effect of age on gait features.•Trunk accelerometry was used to characterize the gait of 100 healthy individuals.•The results show increased instability that begins as early as 40-50 years of age.•Lower balance capabilities seen in elderly seem already detectable earlier in life.

Falls during walking are a major health issue in the elderly population. Older individuals are usually more cautious, walk more slowly, take shorter steps, and exhibit increased step-to-step variability. They often have impaired dynamic balance, which explains their increased falling risk. Those locomotor characteristics might be the result of the neurological/musculoskeletal degenerative processes typical of advanced age or of a decline that began earlier in life. In order to help determine between the two possibilities, we analyzed the relationship between age and gait features among 100 individuals aged 20-69. Trunk acceleration was measured during a 5-min treadmill session using a 3D accelerometer. The following dependent variables were assessed: preferred walking speed, walk ratio (step length normalized by step frequency), gait instability (local dynamic stability, Lyapunov exponent method), and acceleration variability (root mean square [RMS]). Using age as a predictor, linear regressions were performed for each dependent variable. The results indicated that walking speed, walk ratio and trunk acceleration variability were not dependent on age (R2 < 2%). However, there was a significant quadratic association between age and gait instability in the mediolateral direction (R2 = 15%). We concluded that most of the typical gait features of older age do not result from a slow evolution over the life course. On the other hand, gait instability likely begins to increase at an accelerated rate as early as age 40-50. This finding supports the premise that local dynamic stability is likely a relevant early indicator of falling risk.