Data management in gait analysis for clinical applications

Objective. To study the reliability of gait analysis data obtained using the Calibrated Anatomical System Technique (CAST) protocol and to verify the suitability and repeatability of the extraction of a number of parameters from the waveforms obtained. Design. The experimental protocol and the parametric analysis technique were applied on a population of able-bodied subjects. Background. The clinical interpretation process of gait data still needs a more accurate analysis of the reliability and repeatability of the measurements and a suitable procedure for data reduction useful for data comparison. Methods. Gait analysis was performed in 20 able-bodied subjects using a stereophotogrammetric system and a forceplate. 124 parameters relative to time-distance, kinematic and kinetic variables were calculated by means of an automatic procedure and statistically analysed. Results. Most of the parameters were found to be normally distributed with relatively small range of variation. Few of them showed poor repeatability, mostly due to the experimental inaccuracies introduced. Correlation of several gait parameters with age, sex, and speed of progression was also identified. Relevance Gait data obtained by means of an innovative experimental protocol, CAST, are presented and evaluated through parametric analysis of curves. In the presence of a not dramatically altered morphology of the curves, parametric analysis is demonstrated to be a reliable method for analysing gait data, especially in the study of large populations. The least reliable gait parameters, much affected by experimental errors, are identified. These latter parameters should be handled with caution in the process of clinical decision-making using gait analysis. (C) 1998 Elsevier Science Ltd. All rights reserved.