A comparison of two motion analysis devices used in the measurement of lumbar spinal mobility

Objective. The aim of the study was to compare lumbar range of motion deter mined using two computerised dynamic motion analysis devices. Background. Measures of spinal motion are currently used in biomechanical, epidemiological and clinical studies of the low back. It is essential that the various devices used to measure mobility yield similar results, particu larly when the absolute values are to be used to assess job suitability, th e extent of injury or the need for rehabilitation. Methods . Eleven volunteers took part. The ranges of lumbar flexion, extens ion, lateral bending and axial rotation were measured using the CA6000 Spin e Motion Analyser and the Polhemus Fastrak system, using standardised proto cols. Results. Each device showed good test-retest reliability in itself (R > 0.8 2). The absolute values for range of flexion in a standing posture were sig nificantly higher with the CA6000 than with the Fastrak (though well correl ated); those recorded in sitting were comparable for the two devices. Value s for lateral bending using the two devices were well correlated, although small (but significant) differences in the absolute values were found. For extension and axial rotation, the devices gave significantly different valu es that were also poorly correlated. The 'lmits of agreement' for the two d evices (calculated to examine whether they could be used interchangeably) w ere rather wide, especially for extension and axial rotation. Conclusion. The two devices do not always yield comparable measures for spi nal mobility. The accuracy of each, in relation to true angular movements o f the vertebrae, remains unknown.