Objective: The primary aim of the present study was to study the reliability of postural stability measures during whole body dynamic conditions, so they can be used in postural stability evaluations, investigations of injury mechanisms and assessments of treatment efficacy.

Materials & Methods: Twelve healthy volunteers and 12 low back pain patients performed repeated trunk bending motions in eight different conditions of high and low speed, in symmetric and asymmetric planes and with or without external load. All measurements were repeated on a second session, 7-10 days later. Center of pressure (COP) data were used to calculate the standard deviation of amplitude and that of velocity in anterior-posterior and medial-lateral directions, the mean total velocity and the path length. Relative reliability was assessed using the intra class correlation coefficient (ICC) and absolute reliability using the standard error measurement, minimum detectable change and coefficient of variation.

Results: Results in both low back pain and healthy subjects revealed that the mean total velocity showed good to excellent reliability, with ICC ranging from 0.6 to 0.91. In the healthy group other parameters such as standard deviation of amplitude and standard deviation of velocity in the anterior-posterior direction showed good to excellent reliability (range: 0.66-0.97). In the low back pain group standard deviation of amplitude in the anterior-posterior direction in symmetric test conditions (range: 0.61-0.75) and standard deviation of amplitude in the medio-lateral direction in asymmetric test conditions (range: 0.62-0.81) were good or excellent.

Conclusion: Mean total velocity is suggested as a good parameter for evaluation of postural stability in dynamic situations such as lifting and complex trunk movement and also for identifying low back pain patients from healthy subjects.