The objective of the proposed work is to develop a safe balance """"""""stress test"""""""" that can be used as a predictor of relative falling liability in elderly individuals. Such a test could then be used to identify """"""""high-risk"""""""" individuals and to study possible causal and preventative factors. The balance stress test is based on an hypothesis that the posture control system acts to maximize the """"""""stability margin"""""""", the distance between the center-of-pressure on the feet (COP) and the limits of the base-of-support (BOS). If this hypothesis is true, then it may be possible to quantify the relative falling liability of an indivudal in terms of the stability margin of the response to a transient perturbation (simulating a trip, slip or misstep). For safety reasons, the transient response cannot be measured directly. Instead, a safe small-amplitude random or pseudo random perturbation is used to identify an input-output model of the posture control system, which is then used to predict the transient response and stability margin. The proposed work will test the ability of the balance stress test to predict relative falling liability in a population of elderly subjects. The predicted falling liability will be compared with actual falling behaviour moitored prospectively over a period of one year through self-reporting, observation by clinical staff and interviews. To allow observation of falls, the subjects will be seleced from a population of institutionalized, yet ambulatory elderly. The test predictions and actual oberved incidence of falling will be compared, both with and without allowance for errors in the fall-reporting (i.e. non-falling fallers and falling non-fallers). In addition, the result will be compared to those obtained using conventional spontaneous sway mesures as the predictor falling liability, and to results obtained when consideration is given to other risk factors (physiological, activity/lifestyle, environmental and gait-related).
