Foot displacement but not velocity predicts the outcome of a slip induced in young subjects while walking

The purpose of the present study was to induce slips in healthy subjects as a means to determine if recovery from an induced slip is possible under co nditions in which the displacements and velocities of the slipping foot exc eed the generally accepted limits of 10 cm and 50 cm/s, respectively, and t o determine if there are gait-related variables that predispose an individu al to falling after a slip. Thirty-three young and barefoot adults, protect ed by an instrumented safety harness, were subjected to a single slipping t rial following a series of unperturbed walking trials. The slip was induced when the bare foot contacted a vinyl sheet coated with mineral oil. Lower extremity kinematics were acquired using a video-based motion capture syste m. Fourteen and 12 subjects could be unambiguously categorized as having fa llen or recovered, respectively. Four variables demonstrated significant be tween-group differences and two were used to compute the probability of the slip outcome using logistic regression. The variables were the displacemen t of the foot during the slip and the angle of the shank relative to the gr ound at the instant of ground contact just prior to the slip. Separate univ ariate logistic regressions using each variable were significant and correc tly classified about 70% of the slip outcomes. The results demonstrated tha t previously published values for the displacement and velocity of the slip ping foot, 10 cm and 50 cm/s, respectively, may not accurately represent th e upper limits beyond which recovery is not possible. The results also demo nstrated that heel-strike angle, reflective of stride length, exerts a sign ificant influence on the outcome of a slip. (C) 2000 Elsevier Science Ltd. All rights reserved.