The Influence of Stride Selection on Gait Parameters Collected with Inertial Sensors.

Different methods exist to select strides that represent preferred, steady-state gait. The aim of this study was to identify the effect of different stride-selection methods on spatiotemporal gait parameters to analyze steady-state gait. A total of 191 patients with hip or knee osteoarthritis (aged 38-85) wearing inertial sensors walked back and forth over 10 m for two minutes. After the removal of strides in turns, five stride-selection methods were compared: () include all strides, others removed () two strides around turns, () one stride around turns, () strides <63% of median stride length, and () strides that fall outside the 95% confidence interval of gait speed over the strides included in . Means and SDs of gait parameters were compared for each trial against the most conservative definition (). and definitions resulted in similar means and SDs compared to , while and definitions resulted in substantially higher SDs of all gait parameters. An in-depth analysis of individual strides showed that the first two strides after and last two strides before a turn were significantly different from steady-state walking. Therefore, it is suggested to exclude the first two strides around turns to assess steady-state gait.