Introduction: The European Commission has approved a program for the development of European guidelines for the management of low back pain (LBP), called COST B13 . The main objective of this COST action is to produce European guidelines for the prevention, diagnosis and treatment of non-specific low back pain with an evidence-based and multidisciplinary approach. To ensure an evidence-based approach, it was stated from the beginning in the protocol that recommendations should be based on Cochrane reviews and on other existing evidence-based sources. Besides, many of the members of the multidisciplinary panel are Cochrane reviewers or have been involved with the Cochrane Collaboration at some point. These European guidelines should be the basis for future national guidelines or future updates of existing national guidelines: therefore they can be seen as an intermediate step between theory (Cochrane reviews) and practice (local guidelines). For these reasons, we believe that they can be used as a peculiar standpoint to observe the impact of Cochrane reviews on a clinical field.

Methods: Recommendations of the COST B13 guidelines will be scanned to investigate the presence of Cochrane reviews to support the statement. The absence of reviews for specific questions, the need for an update (determined by the presence of subsequent evidence) and the agreement between Cochrane reviews and recommendations will be recorded. On the other hand, the number of Cochrane reviews actually used for the guidelines out of the total number of existing Cochrane reviews in the field of LBP will be calculated to appraise their clinical relevance. Finally the number on non-Cochrane reviews will be calculated and the ratio between Cochrane and non-Cochrane reviews will be calculated.

Results: At the time of preparing this abstract (April 2004) final recommendations for all the subgroups are not readily available, but they will be by the time of the Colloquium, when final figures will be presented. From the analysis of the present data 50-60% of the recommendations are based on Cochrane reviews, in 20-30% of the cases additional non-Cochrane reviews were available. Recommendations were always in agreement with Cochrane reviews, although in 15-20% of the cases a need for an update or a different formulation of the clinical question was needed.

Conclusion: Low-back pain is a field in which a big effort has been made to produce a number of relevant Cochrane reviews, with a dedicated review group. This has been achieved sometimes by splitting the evidence in many smaller, feasible reviews. The high methodological quality of the reviews has been acknowledged, incorporating in most cases their conclusions into the recommendation. Still, even in such a favourable condition, a great proportion of the clinical questions are uncovered by the existing reviews. Possible suggestions for future research will be discussed.