Measurements of trunk flexibility have frequently been used to aid in diagnostic, prognostic and therapeutic processes. However, the speed at which these tests should be performed has not been established. Some authors have performed range of motion studies with subjects moving at their maximum speed, whilst others have asked subjects to move at slow speeds. McIntyre et al. (1991) suggested that tests of human performance should be performed at the subject’s preferred speed, since this is a more reliable measure of function. However, the influence of speed on movement characteristics has received limited attention. In addition the impact of speed on repeatability of measures of motion is unclear. The impact of test speed was investigated in a sample of normal control subjects and low back pain subjects with known spinal stenosis. Measurements of lumbar spine motion were obtained using a triaxial potentiometric analysis system (CA ∼ 6000 Spinal Motion Analyser, OSI, USA) during a simple antero-posterior flexion-extension test. Each subject was asked to perform a series of three flexion-extension tests at three different speeds; as fast as they could, slowly and at their preferred/ normal rate of movement. The results demonstrated that in both study populations the speed at which the test was performed had no significant effect on the resultant ROM of flexion. However, in the normal population it did have an influence on ROM extension, with the tests performed at a fast speed producing a significantly greater ROM extension ( p< 0.05) than tests performed at the subject’s preferred speed or at a slow speed. Not surprisingly the mean velocity characteristics of each test speed showed significant differences in both populations. In terms of test repeatability however, the results indicated that testing at a subject’s preferred speed produced more consistent readings of motion characteristics (in terms of both ROM and velocity). In addition all parameters of motion were significantly impaired in the spinal stenosis group with p< 0.01.
Read full abstract