Background: Adapt gait is abilities to adjust the leading foot in response to the requirement of dynamic environments during walking. Accurate adjustments of the minimum toe clearance (MTC) height and step length can prevent older people from falling when walking and responding to hazards. Although older diabetic patients fall more frequently than healthy older adults, none of the previous studies have quantified the abilities of adaptive gait. This study aimed to investigate the effects of diabetes mellitus on step length and MTC adjustments using a non-immersive virtual-reality system.
Methods: Sixteen young adults (26 ± 5, 7 females), 16 healthy older adults (68 ± 5, 6 females), and 16 older diabetic patients (70 ± 5, 6 females) completed adaptability tests while walking on a treadmill. A computer system visualised a continuous real-time signal of absolute step length and MTC on a monitor. Each person responded to four discrete subject-specific step length and MTC visual targets that presented on the same signal. Tasks were to match the peaks of interest on each signal to presented targets. Targets were 10% longer or shorter than the mean baseline step length, 2.5 cm, and 3.5 cm higher than the mean baseline MTC. When a target was displayed, it remained unchanged for 10 consecutive foot displacement adaptation attempts. Then the target was removed and a new target or the same target was present after 10 consecutive steps and remained for 10 steps. Each target was randomly presented three times (3 × 10). Average absolute and constant errors of step length and MTC adaptation were compared among groups.
Results: In adaptability tests, the diabetic group had a mean positive error during step shortening; they struggled to walk with shorter steps. The diabetic group also walked with steps shorter than step length targets. Compared with other groups, mean constant and absolute errors for step length and MTC adaptation in older adults with diabetes were significantly greater.
Conclusions: Impaired adaptive gait can increase the risk of falling in pathological populations. The presented virtual-reality system with visual feedback has merits for training gait adaptability.