Slip-induced fall-risk assessment based on regular gait pattern in older adults

Area of Science:

• Gerontology
• Biomechanics
• Kinesiology

Background:

• Aging increases fall risk, necessitating effective fall prevention strategies.
• Assessing fall risk in older adults, particularly after unexpected perturbations like slips, is crucial.
• Normal gait patterns offer potential biomarkers for predicting dynamic balance deficits.

Purpose of the Study:

• To develop and validate a prognostic model for predicting fall risk in healthy older adults following an over-ground slip perturbation.
• To identify key gait parameters from normal walking that best predict fall outcomes during a slip.
• To compare the predictive accuracy of a gait-based model versus a clinical/demographic model.

Main Methods:

• 112 healthy older adults participated in a controlled laboratory setting, experiencing a novel slip perturbation.
• Gait variables (step length, segment angles, center of mass, ground reaction force) were analyzed from slip and pre-slip natural walking trials.
• Stepwise logistic regression was employed to identify the optimal predictive model for fall risk.

Main Results:

• The optimal fall prediction model utilized right thigh angle at touchdown, maximum ground reaction force, and momentum change, achieving 75.9% overall accuracy.
• This gait-based model demonstrated high sensitivity (74.5% for falls) and specificity (77.2% for recoveries).
• A clinical/demographic model showed lower overall accuracy (62.5%) with poor specificity (47.4% for recoveries).

Conclusions:

• A prognostic model based on normal gait patterns can accurately predict fall risk in older adults during slip perturbations.
• Specific gait parameters during normal walking can serve as indicators for dynamic balance and fall susceptibility.
• These findings support the development of targeted interventions to enhance dynamic balance and prevent falls in at-risk populations.