Early Warning Model of Fall Risk for the Elderly Based on Gait Characteristics / 医用生物力学

ABSTRACT

Objective To construct an early warning model of fall risk for the elderly based on six kinds gait parameters. Methods A digital field was used to collect parameters from six kinds of gait for the elderly with or without the history of falls, and the binomial logistic regression analysis was used to establish a regression equation for predicting the fall risks in the elderly, and an early warning model was constructed. Results The regression equations constructed according to the parameters from six kinds of gait were statistically significant. The overall correct rate was predicted from high to low walking forward with closed eyes (97.1%), walking backward with open eyes (92.9%), walking backward with closed eyes (88.6%), walking forward with open eyes (87.1%), turning head up and down with open eyes (85.7%), turning head left and right with open eyes(82.9%). The constructed early warning model for fall risk of the elderly mainly included five steps, namely, judgment, test, extraction, calculation and early warning, which was suitable for gait testing and evaluation of the elderly in the laboratory. Conclusions Parameters from six kinds of gait could predict the fall risk of the elderly. Among them, walking forward with closed eyes was best to predict the fall risk in the elderly. The established early warning model of fall risk for the elderly could be used to predict the fall risk of 65-75 year old people within one year, which could provide early warning based on the probability of falling, playing a positive effect on preventing falls in the elderly.