THE VERTICAL DISPLACEMENT OF THE CENTER-OF-MASS DURING WALKING - A COMPARISON OF 4 MEASUREMENT METHODS

Measuring the vertical displacement of the center of mass (COM) of the the body during walking may provide useful information about the ener gy required to walk. Four methods of varying complexity to estimate th e vertical displacement of the COM were compared in 25 able-bodied, fe male subjects. The first method, the sacral marker method, utilized an external marker oil the sacrum as representative of the COM of the bo dy. The second method, the reconstructed pelvis method which also util ized a marker over the sacrum, theoretically controlled for pelvic til t motion. The third method, the segmental analysis method, involved me asuring motion of the trunk and limb segments. The fourth method, the forceplate method, involved estimating the COM displacement from groun d reaction force measurements. A two-tailed paired t-test within an AN OVA showed no statistically significant difference between the sacral marker and the reconstructed pelvis methods (p = 0.839). There was als o no statistically significant difference between the sacral marker an d the segmental analysis method (p = 0.119) or between the reconstruct ed pelvis and the segmental analysis method (p = 0.174). It follows th at the first method which is the most simple, cart provide essentially the same estimate of the vertical displacement of the COM as the more complicated second and third measures. The forceplate method produced data with a lower range and a different distribution than the other t hree methods. There was a statistically significant difference between the forceplate method and the other methods (p < 0.01 for each of the three comparisons). The forceplate method provides information that i s statistically significantly different from the results of the kinema tic methods. The magnitude of the difference is large enough to be phy siologically significant and further studies to define the sources of the differences and the relative validity of the two approaches are wa rranted.