To synthesise the evidence relating to the effectiveness of advice, the relevance of its content and frequency, and to compare the advice being offered to acute, subacute and chronic low back pain (LBP) patients. A systematic review of Randomised Controlled Trials (RCTs) using advice, either alone or with another intervention. The QUOROM guidelines and the Cochrane Collaboration Back Review Group Guidelines for Systematic Reviews were followed throughout: methodological assessment identified RCTs of 'high' or 'medium' methodological quality, based on their inclusion of at least 50% of the specified internal validity criteria. Outcome measures were analysed based on five recommended core outcome domains; pain, work disability, back-specific function, generic health status and satisfaction with care. Relevant RCTs (n=56) were scored for methodological quality; 39 RCTs involving 7347 patients qualified for inclusion, based upon their methodological quality. Advice as an adjunct to exercise was most effective for improving pain, back-specific function and work disability in chronic LBP but, for acute LBP, was no more effective for improving these outcomes than simple advice to stay active. Advice as part of a back school was most effective for improving back-specific function in subacute LBP; these trials generally demonstrated long-term positive results. Advice as an adjunct to exercise was the most common form of treatment for acute and chronic LBP; advice as part of a back school was most commonly used for subacute LBP. Fifteen percent of acute LBP trials had a positive outcome, compared to 86% and 74% of subacute and chronic LBP trials respectively. A wide variety of outcome measures were used, making valid comparisons between treatment outcomes difficult. The advice provided to patients with LBP within RCTs varied considerably depending on symptom duration. The findings of this review have important implications for clinical practice, and for the design of further clinical trials in this area. Advice to stay active is sufficient for acute LBP; however, it appears that RCTs do not commonly reflect these recommendations. No conclusions could be drawn as to the content and frequency of advice that is most effective for subacute LBP, due to the limited number and poor quality of RCTs in this area: this review provides preliminary support for advice as part of a back school approach. Given that the effectiveness of treatment for subacute symptoms will directly influence the development of chronicity, these results would suggest that education and awareness of the causes and consequences of back pain may be a valuable treatment component for this patient subgroup. For chronic LBP there is strong evidence to support the use of advice to remain active in addition to specific advice relating to the most appropriate exercise, and/or functional activities to promote active self-management. More investigation is needed into the role of follow-up advice for chronic LBP patients.