Biomechanical Mechanisms of Improved Balance Recovery to Repeated Backward Slips Simulated by Treadmill Belt Accelerations in Young and Older Adults

Aim: Exposure to repeated gait perturbations improves the balance of older adults (OAs) and decreases their risks of falling, but little is known about the underpinning mechanical adjustments. We aimed to quantify the changing temporo-spatial and kinetic characteristics of balance recovery following repeated backward slips to better understand the mechanical adjustments responsible for improved balance. Methods: We exposed 17 young adults (YAs) (25.2 ± 3.7 years) and 17 OAs (62.4 ± 6.6 years) to 10 backward slips simulated on an instrumented treadmill by unilateral backward belt accelerations. We measured the balance of the participants (margin of stability: MoS), balance recovery (n(steps): number of steps necessary to return to a steady gait for at least three consecutive steps), temporo-spatial (step length), and kinetics [ground reaction force (GRF) angle, lower limb joint moments] for 15 steps following each slip. The results were compared with baseline. Results: Participants in both groups improved their MoS and n(steps) with repeated exposure to the slips, but no significant effect of age was detected. During the perturbed step, the GRF vector was directed more posteriorly during mid-stance and more anteriorly during push-off than baseline, which resulted in a longer step. These adjustments were maintained from the first (Slip01) to the last (Slip10) slip, and by Slip10 were correlated with better balance (MoS) on the second recovery step. During the first recovery step following Slip01, participants developed lower plantarflexor and larger knee extensor moments whilst taking a shorter step, these adjustments were correlated with poorer balance and were not maintained with repeated slips. Joint moments and step length of the first recovery step returned to normal levels by Slip10. Conclusion: Young adults and OAs improved their balance with repeated slips. The adjustments that were positively correlated with balance (changes in step length, GRF angle) were maintained whilst those that were not (changes in joint moments) were discarded. All the responses observed in Slip10 were observed in Slip01. The observed balance improvements were achieved by refining the initial strategy rather than by developing a new one. The underlying mechanics were correlated with step length of the first recovery steps, which was associated with balance and should be monitored in fall prevention interventions.