Balance evaluation using accelerometers has been compared with comprehensive clinical balance assessments in healthy older subjects and idiopathic fallers [17]. For comparison a variety of clinical balance tests (variants of Romberg’s tests, heel-toe straight line walking and a functional reach test) and quantitative motor co-ordination tests (rapid stepping tests and heel-toe transitions) were performed. Acceleration amplitude data was obtained while standing on the floor or compliant foam with eyes open and eyes closed. A significant difference was found between the faller and non-faller groups for the acceleration values. However, the only clinical balance test score to distinguish between the two groups was Romberg’s test using the right leg alone with eyes open. The motor co-ordination tests showed a significant difference between the groups for rapid stepping but not for heel-toe transition. Therefore, clinical evaluation methods alone show deficiencies in identifying patients at high risk of falling.
In a separate study, the gait of older adults with and without stability problems and young subjects was compared using trunk accelerometers [18]. Subjects were classified as having stability problems if they had a history of falling or reported that they felt unsteady when standing or moving. An index of smoothness was calculated in this study and peak acceleration amplitude was measured. The older individuals with stability problems were shown to be significantly different from the younger controls and the older individuals without stability problems (i.e. less smooth trunk movement and lower peak acceleration amplitude).
These studies suggest that accelerometers are useful for assessing balance and that they detect definite abnormalities in the gait of fallers. However, as of yet they have not been prospectively investigated to determine their effectiveness in predicting future falls risk.