Measures of postural stability are not predictors of recovery from large postural disturbances in healthy older adults

OBJECTIVES: To determine, in healthy older adults, the relationship between postural steadiness, stability limits, and the ability to recover balance from three postural disturbances requiring anteriorly directed stepping res ponses. DESIGN: Analysis of multiple motor tasks in a cross-sectional sample of hea lthy older adults. SETTING: A biomechanics research laboratory. PARTICIPANTS: Fifty women and 29 men aged 65 or older, all healthy, living in the community, participated in this study. Subjects were examined by a g eriatrician to identify the presence of exclusionary factors. MEASUREMENTS: Anterior-posterior and medial-lateral excursion distances of the center of pressure during quiet standing (postural steadiness), static leaning (static stability limits), and dynamic swaying (dynamic stability l imits) were determined from ground reaction forces measured by a strain gau ge forceplate. Within the same group of subjects, the maximum angle of forw ard lean from which a subject could recover with a single step, the ability to recover balance in response to an accelerated support surface, and the ability to recover balance after being tripped were determined. RESULTS: Re covery from the three types of postural disturbances were found to be stati stically independent. The pos rural steadiness and the stability limit vari ables were only weakly correlated. Postural steadiness and stability limits were not related to the maximum recoverable angle of lean. The average med ial-lateral center of pressure speed during the postural steadiness test wa s significantly slower for those who failed to recover after tripping than for the subjects who recovered successfully. However, a logistic regression model failed to achieve statistical significance, suggesting that the diff erence may not be functionally important. The anterior-posterior static sta bility limits were significantly larger for subjects who recovered successf ully than for those who failed to recover during the accelerated support su rface test. Although logistic regression suggested that a reduced anterior- posterior stability limit: represents a risk factor for failure to recover during this task, only nine of 28 failures could be properly classified, th us diminishing the functional importance of this finding. CONCLUSIONS: Because recovery following postural disturbances could not gen erally be predicted from measures of postural stability, these findings sug gest that these measures of postural stability are of limited utility in id entifying potential anteriorly directed fallers in healthy older adults.