Data Analysis to Evaluate Which Specific Gait Measures Are Associated with Risk of Injurious Falls Evaluating Gait Measures Associated with the Risk of Injurious Falls Through Data Analysis

Condition(s) studied

Investigational drug(s) / intervention(s)

Daily Activity Patterns Using Wearable Tri-Axial Sensors: This intervention uniquely focuses on the prediction of injurious falls by combining daily life gait (DLG) measures (e.g., gait speed, cadence, variability) with daily life physical activity (DLPA) measures (e.g., activity levels, activity fragmentation). Unlike other studies, this analysis leverages data from a large cohort of older women (n=17,466) enrolled in the Women's Health Study (WHS), where participants wore a tri-axial accelerometer for 1 week. Additionally, the study links accelerometer data to long-term health outcomes, specifically fall-related injuries from Centers for Medicare \& Medicaid Services (CMS) records. This is the first study to explore whether combining DLG and DLPA measures, derived from wearable technology, can predict fall-related injuries in an aging population, applying advanced machine learning techniques to this large, anonymized dataset.

Study summary

The goal of this study is to understand if specific gait and activity measures can help predict injurious falls in older women. The main questions it aims to answer are:

Can combining daily gait (DLG) and daily physical activity (DLPA) measures more accurately predict the risk of injurious falls? How effective is wearable technology and machine learning in analyzing these activity measures for fall prediction? Researchers will analyze data from the Women's Health Study (WHS), using wearable technology to track daily walking patterns and physical activity, and apply machine learning to assess the likelihood of harmful falls.

Eligibility

Primary outcome measure(s)

• Association of Gait Speed with Risk of Injurious Falls (AIM1) — njurious falls within 1 year after baseline assessment using time-to-event analyses.
  The study will evaluate the association between gait speed (measured in meters per second) and the risk of injurious falls within one year following the accelerometer assessment.
• Association of Cadence with Risk of Injurious Falls (AIM1) — Injurious falls within 1 year after baseline assessment using time-to-event analyses.
  The study will assess the association between cadence (measured in steps per minute) and the risk of injurious falls within one year following the accelerometer assessment.
• Association of Gait Variability with Risk of Injurious Falls (AIM1) — Time Frame: Injurious falls within 1 year after baseline assessment using time-to-event analyses.
  The study will assess the association between gait variability (measured as the standard deviation of step times) and the risk of injurious falls within one year following the accelerometer assessment.
• Association of Overall Activity Levels with Risk of Injurious Falls (AIM2) — Injurious falls within 1 year after baseline assessment using time-to-event analyses
  The study will evaluate the association between overall activity levels (measured in average accelerometer counts per minute) and the risk of injurious falls within one year following the baseline assessment.
• Association of Activity Fragmentation with Risk of Injurious Falls (AIM2) — Injurious falls within 1 year after baseline assessment using time-to-event analyses.
  The study will assess the association between activity fragmentation (measured by the fragmentation index) and the risk of injurious falls within one year following the baseline assessment.
• Combined DLG and DLPA Measure for Predicting Risk of Injurious Falls (AIM3) — Time Frame: Injurious falls within 1 year after baseline assessment, using combined predictive models.
  his outcome will evaluate a single combined score derived from both daily life gait (DLG) and daily life physical activity (DLPA) measures to assess the association with the risk of injurious falls. The combined score will be created incorporating DLG measures (e.g., gait speed, variability) and DLPA measures (e.g., overall activity levels, fragmentation) into a unified predictor.

Trial sites (1)

More Tel-Aviv Sourasky Medical Center trials in Israel