The influence of dual-task conditions on postural control and instrumented timed up and go performance in fallers and non-fallers

Date of Award

2016

Degree Name

Ph.D. in Mechanical Engineering

Department

Department of Mechanical and Aerospace Engineering

Advisor/Chair

Advisor: Kimberly E. Bigelow

Abstract

One in three older adults fall each year; many falls resulting in moderate to severe injuries. Falls are a multi-faceted problem, with risk factors that include balance and gait impairments. Balance and movement assessments are often used to identify individuals at risk of falls by identifying a change in the center of pressure excursions or movements. This study examined two fall risk assessments, posturography analyzed through traditional time-domain measures and newer non-linear measures and the instrumented Timed Up and Go (iTUG), under standard and dual-task conditions, to determine better ways to distinguish individuals with subtle deficits contributing to fall risk. One hundred fifty older adult fallers and non-fallers performed quiet-standing posturography and iTUG methodologies. Test conditions included standard testing conditions, cognitive dual-task, manual dual-task, and cognitive+manual dual task. Five traditional postural sway parameters, four non-linear postural sway parameters were calculated, and eight iTUG parameters were calculated. One-way multivariate analysis of variance (p<0.05) was used to compare fallers versus non-fallers and to compare each type of dual task. Effect sizes were calculated using the Cohen's d method. Stepwise logistic regression was performed to identify the postural sway and iTUG parameters that best differentiated fallers from non-fallers for the traditional Timed Up and Go Test, iTUG test, posturography test and a combined model including the iTUG and posturography tests. Results demonstrated that not just one dual-task prevailed over the overs, rather when analyzing posturography data through traditional measures the manual dual-task provided greater differentiation between fall risk groups, when analyzing posturography data through non-linear measures the cognitive dual-task provided greater differentiation between fall risk groups, and when utilizing the iTUG test the cognitive+manual dual-task affected the iTUG parameters the most, with fall risk differentiation seen in the sit-to-stand measures. A stepwise logistic regression model was created, with all of the posturography, traditional and nonlinear, parameters and all iTUG output parameters input into the model. The resulting fall risk model has a max re-scaled R2 value of 0.3244, sensitivity of 54.3% and specificity of 82.7%. The parameters included in the model are height, sit-to-stand duration, stand-to-sit duration, turn peak velocity, and A/P sway range. Dual-tasks and non-linear analysis measures were valuable additions to posturography and iTUG fall risk assessments. Future work is necessary to extend exploration of dual-tasks and how they affect fallers and non-fallers differently.

Keywords

Falls (Accidents) Risk assessment, Falls (Accidents) in old age Risk assessment, Human beings Attitude and movement, Biomechanics, Mechanical Engineering, Posturography, Instrumented Timed Up and Go, iTUG, Older Adults, Dual-task

Rights Statement

Copyright © 2016, author

Link to Full Text

Share

COinS