JMIR Publications

ABSTRACT

Background:

Frailty affects over 50% of older adults in long-term care (LTC), and early detection is critical due to its potential reversibility. Wearable sensors enable continuous monitoring of gait and physical activity, and machine learning has shown promise in detecting frailty among community-living older adults. However, its applicability in LTC remains underexplored. Furthermore, dynamic gait outcomes (e.g., gait stability, symmetry) may offer more sensitive frailty indicators than traditional measures like gait speed, yet their potential remains largely untapped.

Objective:

This study aimed to evaluate whether frailty in long-term care (LTC) facilities could be effectively identified using machine learning models trained on gait and daily physical activity data derived from a single accelerometer.

Methods:

This study is a cross-sectional secondary analysis of baseline data from a two-arm cluster randomized controlled trial. Of the 164 individuals initially enrolled, 51 participants (mean age 83.2 ± 10.6, 47.0% females) met the inclusion criteria of completing all assessments required for this study, and were included in the final analysis. Frailty status was assessed using the FRAIL-NH scale. Participants completed a 5-meter walking task while wearing a 3D accelerometer. Following this task, the accelerometer was used to record daily physical activity over approximately one week. A total of 34 dynamic and spatiotemporal gait outcomes, three physical activity variables, and six demographic characteristics were extracted. Five machine learning models were trained to classify frailty status using a leave-one-out cross-validation approach. Model performance was evaluated based on accuracy and the area under the receiver operating characteristic curve (AUC). To enhance model interpretability, explainable AI (XAI) techniques were employed to identify the most influential predictive outcomes.

Results:

The eXtreme Gradient Boosting model demonstrated the optimal performance with an accuracy of 86.3% and an AUC of 0.92. XAI analysis revealed that frail older adults exhibited more variable, complex, and asymmetric gait patterns, which were characterized by higher stride length variability, increased sample entropy, and a lower gait symmetry index.

Conclusions:

Our findings suggest that dynamic gait outcomes may serve as more sensitive indicators of frailty than spatiotemporal gait outcomes (e.g., gait speed) in LTC settings, offering valuable insights for enhancing frailty detection and management.