Background: Measurement of temporal-spatial gait variables is common in aging research with several methods available. This study investigated the differences in temporal-spatial gait outcomes derived from two different programs for processing instrumented walkway data.
Method: Data were collected with GAITRite® hardware from 86 healthy older people and 44 older people four months following surgical repair of hip fracture. Temporal-spatial variables were derived using both GAITRite® and PKmas® processing programs from the same raw footfall data.
Results: The mean differences between the two programs for most variables were negligible, including for Speed (mean difference 0.3 ± 0.6 cm/sec, or 0.3% of the mean GAITRite® Speed). The mean absolute percentage difference for all 18 gait variables examined ranged from 0.04% for Stride Duration to 66% for Foot Angle. The ICCs were almost perfect (≥0.99) for all variables apart from Base Width, Foot Angle, Stride Length Variability, Step Length Variability, Step Duration Variability and Step Width Variability, which were all never-the-less above 0.84. There were systematic differences for Base Width (PKmas® values 1.6 cm lower than GAITRite®) and Foot Angle (PKMAS® values 0.7° higher than GAITRite®). The differences can be explained by the differences in definitions and calculations between the programs.
Conclusions: The study demonstrated that for most variables the outcomes from both programs can be used interchangeably for evaluation of gait among older people collected with GAITRite® hardware. However, validity and reliability for Base Width and Foot Angle derived by PKMAS® would benefit from further investigation.