This thesis investigates routine physiotherapy management of patients with subacute and chronic non specific low back pain. In a pragmatic multi-centre trial patients were randomised to receive a course of physiotherapy treatment or advice following a bio-psychosocial model. Disease specific, patient specific and generic measures were used to assess outcome. The 286 patients recruited in the trial had, on average, minimal to moderate low back pain disability. Patients reported enhanced perceptions of benefit in the physiotherapy group but there was no evidence of a long term effect in any other outcomes. There were no differences between the groups in NHS costs although patients in the physiotherapy group incurred significantly higher out of pocket expenses. Further analysis of the outcome data confirmed that the primary outcome measure (Oswestry Disability Index) was the most responsive instrument because it was able to detect deterioration as well as improvement. As the trial demonstrated no additional benefit of physiotherapy over brief advice, it was important to investigate the effectiveness of the latter. A systematic review found limited evidence that brief bio-psychosocial advice was more effective in reducing fear avoidance and improving back beliefs in patients with acute and subacute low back pain compared with traditional medical advice. There was no direct evidence to support the use of brief bio-psychosocial advice (2 sessions or less) for reducing pain or disability. This thesis describes research that has contributed to European guidelines for the management of chronic low back pain and reviews extensively the literature that seeks to evaluate physiotherapy practice. The clinical implication of this research is that for patients with non specific low back pain of mild severity, brief advice is likely to be as effective as prolonged physiotherapy intervention. The extent to which a single session of advice is more effective than no intervention needs further assessment.