Muscle Strength as a Predictor of Gait Variability after Two Years in Community-Living Older Adults

Stride-to-stride fluctuations, or gait variability, can be captured easily using body worn inertial sensors. Previously, sensor-measured gait variability has been found to be associated with fall risk and central nervous changes. However, further research is needed to clarify the clinical relevance of this method. In this study, we look at how gait variability is associated with muscle strength, measured two years earlier. This is study of longitudinal associations. Participants were community-dwelling volunteers between 70-81 years. Participants were tested while walking with a single sensor at their lower back, and they walked back and forth over a distance of 6.5 meters under four conditions: at preferred speed, at fast speed, with an added cognitive task, and while walking across an uneven surface. Gait variability in the anteroposterior (AP), mediolateral (ML) and vertical (V) directions was identified. A muscle strength score was composed by transforming hand grip strength, isometric knee extension strength and the 30 second chair rise-test to z-scores and adding them. 56 individuals were analysed (mean age at baseline 75.8 (SD 3.43), 60 percent women). In a backwards regression method using age, gender and baseline walking speed as covariates, muscle strength predicted gait variability after two years for AP variability during preferred speed (Beta= .314, p=.025) and uneven surface walking (Beta=.326, p=.018). Further, muscle strength was associated with ML variability during preferred speed (Beta=.364, p=.048) and fast speed (Beta=.419, p=.042), and V variability during preferred speed (Beta=.402, p=.002), fast speed (Beta=.394, p=.004) and uneven surface walking (Beta=.369, p=.004). Sensor-measured gait variability tended to be associated with muscle strength measured two years earlier. This finding could emphasize the relevance of this relatively novel measure of gait in older adults for both research and clinical practice.