Non-Linear Center-of-Pressure Features Associated with Fall History in Older Adults: An Exploratory Analysis

Area of Science:

• Biomechanics and Motor Control
• Gerontology
• Data Science in Health

Background:

• Postural sway, measured by center-of-pressure (CoP) trajectories, is a key indicator for assessing balance and fall risk in older adults.
• Traditional linear metrics of CoP primarily focus on sway magnitude, potentially neglecting the complex temporal dynamics of balance control.
• The loss-of-complexity hypothesis suggests that reduced complexity in physiological signals may be associated with aging and disease.

Purpose of the Study:

• To investigate the association between fall history and both linear and non-linear CoP metrics in older adults.
• To explore the utility of advanced non-linear analytical techniques in characterizing balance control beyond conventional measures.

Main Methods:

• Analysis of quiet-standing CoP data from an open-access dataset of adults aged 60 years and older, categorized by fall history (fallers vs. non-fallers).
• Application of propensity score matching to control for potential confounding factors including age, sex, BMI, and health status.
• Examination of a comprehensive suite of linear and non-linear CoP metrics, including Recurrence Quantification Analysis, Detrended Fluctuation Analysis, and Multiscale Entropy.

Main Results:

• No significant differences were found between fallers and non-fallers across all examined CoP metrics after propensity score matching.
• Exploratory analysis using SHAP values indicated that non-linear features, particularly those reflecting temporal structure and multiscale organization, were more influential in predictive models than magnitude-based linear metrics.
• These findings suggest a potential, albeit not statistically significant in this cohort, for non-linear dynamics to differentiate balance control characteristics.

Conclusions:

• While no single CoP metric differentiated fallers from non-fallers in this matched cohort, non-linear measures showed potential in capturing complex balance dynamics.
• The exploratory nature of the findings highlights the need for hypothesis generation regarding the role of temporal complexity in postural control and fall risk.
• Further research with larger sample sizes is warranted to validate the contribution of non-linear CoP dynamics to understanding balance deficits and predicting falls in older adults.