Search Results

Postural instability, falls, and fear of falling that accompany frailty with aging and disease form major impediments to physical activity. In this article we present a theoretical framework that may help researchers and practitioners in the development and delivery of intervention programs aimed at reducing falls and improving postural stability and locomotion in older individuals and in those with disability due to disease. Based on a review of the dynamical and complex systems perspectives of movement coordination and control, we show that 1) central to developing a movement-based intervention program aimed at fall reduction and prevention is the notion that variability can play a functional role and facilitate movement adaptability, 2) intervention programs aimed at fall reduction should focus more on coordination and stability boundary measures instead of traditional gait and posture outcome variables, and 3) noise-based intervention techniques using stochastic resonance may offer external aids to improve dynamic balance control.

The development of a methodology to assess movement coordination has provided gait researchers a tool to assess movement organization. A challenge in analyzing movement coordination using vector coding lies within the inherent circularity of data garnered from this technique. Therefore, the purpose of this investigation was to determine if accurate group comparisons can be made with varying techniques of vector coding analyses. Thigh–shank coordination was analyzed using a modified vector coding technique on data from 2 groups of runners. Movement coordination was compared between groups using 3 techniques: (1) linear average completed with compressed data (0°–180°) and noncompressed data (0°–360°), (2) coordination phase binning analysis; and (3) a circular statistics analysis. Circular statistics (inferential) analysis provided a rigorous comparison of average movement coordination between groups. In addition, the binning analysis provided a metric for detecting even small differences in the time spent with a particular coordination pattern between groups. However, the linear analysis provided erroneous group comparisons. Furthermore, with compressed data, linear analysis led to misclassification of coordination patterns. While data compression may be attractive as a means of simplifying statistical analysis of inherently circular data, recommendations are to use circular statistics and binning methods on noncompressed data.

Step-based metrics provide simple measures of ambulatory activity, yet device software either includes undisclosed proprietary step detection algorithms or simply does not compute step-based metrics. We aimed to develop and validate a simple algorithm to accurately detect steps across various ambulatory and nonambulatory activities. Seventy-five adults (21–39 years) completed seven simulated activities of daily living (e.g., sitting, vacuuming, folding laundry) and an incremental treadmill protocol from 0.22 to 2.2 m/s. Directly observed steps were hand-tallied. Participants wore GENEActiv and ActiGraph accelerometers, one of each on their waist and on their nondominant wrist. Raw acceleration (g) signals from the anterior–posterior, medial–lateral, vertical, and vector magnitude directions were assessed separately for each device. Signals were demeaned across all activities and band-pass filtered (0.25, 2.5 Hz). Steps were detected via peak picking, with optimal thresholds (i.e., minimized absolute error from accumulated hand counted) determined by iterating minimum acceleration values to detect steps. Step counts were converted into cadence (steps/minute), and k-fold cross-validation quantified error (root mean squared error [RMSE]). We report optimal thresholds for use of either device on the waist (threshold = 0.0267g) and wrist (threshold = 0.0359g) using the vector magnitude signal. These thresholds yielded low error for the waist (RMSE < 173 steps, ≤2.28 steps/min) and wrist (RMSE < 481 steps, ≤6.47 steps/min) across all activities, and outperformed ActiLife’s proprietary algorithm (RMSE = 1,312 and 2,913 steps, 17.29 and 38.06 steps/min for the waist and wrist, respectively). The thresholds reported herein provide a simple, transparent framework for step detection using accelerometers during treadmill ambulation and activities of daily living for waist- and wrist-worn locations.