Development of injury mitigation techniques for dynamic compression of the spine relies on local dynamic response. Attention needs to be paid to the ability of the material model to capture rate-dependent responses of the spine, and to the attainment of whole spine models based on FSU data. This study assesses the influence of dynamic compression rates on the behaviour of the thoracolumbar spine and spinal segments as well as the relationship of spinal segments to the thoracolumbar spine response. Post-mortem human subject thoracolumbar spines and FSUs from the same donors were subjected to three dynamic compression rates of 200, 400, and 600 N/ms resulting in an average thoracolumbar stiffness of 360 ± 93, 424 ± 99, 505 ± 106 N/mm. No statistically significant differences on stiffness across the proposed loading rates was observed for thoracolumbar spines or FSUs. A nonlinear FSU response (fourth order polynomial) improved the full spine reconstruction by 28.6% over a linear model. Though the nonlinear model underpredicted the thoracolumbar response, the deviation from the mean of the composite spine overlapped with the standard deviation of the thoracolumbar experimental response. The normalised deviation from the mean between the nonlinear reconstruction and the thoracolumbar spine averaged 8.4%.