Intra-rater and inter-rater reliabilities of real-time acceleration gait analysis system

Abstract

Purpose: The purposes of this study were to construct a real-time acceleration gait analysis system equipped with software to analyse real-time trunk acceleration during walking and to examine the intra-rater and inter-rater reliabilities of the this system. Methods: This system has been comprised of an accelerometer, an acceleration amplifier, a transmitter, two foot switches, a receiver and a personal computer installed with the real-time acceleration analysis software. The acceleration signals received were analysed using the real-time acceleration analysis software, and gait parameters were calculated. The subjects were 20 healthy individuals and two raters. The intra-rater and inter-rater reliabilities of the measurement results obtained from this system were examined by performing intraclass correlation coefficients (ICC) and Bland–Altman analysis. Results: The intra-rater and inter-rater ICCs ranged from 0.61 to 0.92 in any gait parameters. In the Bland–Altman analysis, neither fixed nor proportional bias was found in any of the gait parameters. Conclusions: From the ICC and Bland–Altman analysis results, the gait measurement using this system clearly demonstrates that the intra-rater and inter-rater measurements had good reproducibility. Owing to this system, we can improve the clinical efficiency of gait analysis and gait training for physiotherapy.

Implication for Rehabilitation

• This study focused on the advantage of a gait analysis method using an accelerometer and constructed a gait analysis system that calculates real-time gait parameters from trunk acceleration measurements during walking. The gait analysis using this system has good intra-rater and inter-rater reliabilities, and using this system can improve the clinical efficiency of gait analysis and gait training.

• Accelerometer
• Bland–Altman analysis
• gait analysis
• real-time analysis
• reliability

Acknowledgements

The authors wish to thank Naoto Ozawa and Chimao Ozaki for their technical assistance. The authors also thank Tomoko Tanabe for making this work possible.

Declaration of interest

This study was supported by a grant-in-aid from 2012 Kawasaki University of Medical Welfare Scientific Research Fund.