BackgroundThe need to improve spinal motor behaviour in chronic low back pain (CLBP) rehabilitation remains unclear. The objective of this study was to test if changes in spinal motor behaviour were associated with changes in disability after an interdisciplinary rehabilitation program (IRP) in patients with CLBP.MethodsSeventy‐one patients with CLBP participating in an IRP were included. Spinal motor behaviour was assessed with biomechanical (lumbar angular amplitude and velocity, erector spinae muscle activity and duration of the task), cognitive‐emotional (task‐specific fear [PRF]) and pain‐related (movement‐evoked pain [MEP]) measures during a lifting task before and after the IRP. Disability was measured before and after the IRP, and at 3‐month and 1‐year follow‐ups.ResultsAfter adjusting for confounders, changes in disability were significantly associated with MEP changes (β adj. = 0.49, p < 0.001) and PRF changes (β adj. = 0.36, p = 0.008), but not with changes in any of the biomechanical measures. MEP at the end of IRP was also associated with disability at 3 months (β adj. = 0.37, p = 0.001) and 1 year (β adj. = 0.42, p = 0.01). Biomechanical measures at the end of the IRP were not associated with disability, except for the duration of the task that was significantly associated with reduction of disability at 3 months (β non‐adj = 0.5, p < 0.001).ConclusionsPain‐related and cognitive‐emotional measures of spinal motor behaviour were associated with reduction in disability following an IRP. Future research is needed to further investigate causal relationships between spinal motor behaviour and disability.Significance statementThis study supports a multidimensional understanding and analysis of spinal motor behaviour, integrating the cognitive‐emotional, pain‐related and biomechanical domains. It also supports the consideration of spinal motor behaviour as a potentially important treatment target in chronic low back pain management. Moreover, it suggests that reducing movement‐evoked pain and task‐specific fear may have more influence on disability than changing lumbar amplitude, lumbar angular velocity or erector muscle activity, which may have important implications for rehabilitation.