Description of a workshop entitled "Sharing Guidelines for Low Back Pain Between Primary Health Care Providers: Toward a Common Message in Primary Care" that was held at the Fifth International Forum on Low Back Pain in Primary Care in Canada in May 2002. Despite a considerable degree of acceptance of current evidence-based guidelines, in practice, primary health care providers still do not share a common message. The objective of the workshop was to describe the outcomes of a workshop on the sharing of guidelines in primary care. The Fifth International Forum on Low Back Pain Research in Primary Care focused on relations between stakeholders in the primary care management of back pain. Participants in this workshop contributed to an open discussion on "how and why" evidence-based guidelines about back pain do or do not work in practice. Ways to minimise the factors that inhibit implementation were discussed in the light of whether guidelines are mono-disciplinary or multidisciplinary. Examples of potential issues for debate were contained in introductory presentations. The prospects for improving implementation and reducing barriers, and the priorities for future research, were then considered by an international group of researchers. This paper summarises the conclusions of three researcher subgroups that focused on the sharing of guidelines under the headings of: (1) the content, (2) the development process, and (3) implementation. How to share the evidence and make it meaningful to practice stakeholders is the main challenge of guideline implementation. There is a need to consider the balance between the strength of evidence in multidisciplinary guidelines and the utility/feasibility of mono-disciplinary guidelines. The usefulness of both mono-disciplinary and multidisciplinary guidelines was agreed on. However, in order to achieve consistent messages, mono-disciplinary guidelines should have a multidisciplinary parent. In other words, guidelines should be developed and monitored by a multidisciplinary team, but may be transferred to practice by mono-disciplinary messengers. Despite general agreement that multi-faceted interventions are most effective for implementing guidelines, the feasibility of doing this in busy clinical settings is questioned. Research is needed from local implementation pilots and quality monitoring studies to understand how to develop and deliver the contextual understanding required. This relates to processes of care as well as outcomes, and to social factors and policymaking as well as health care interventions. We commend these considerations to all who are interested in the challenges of achieving better-integrated, evidence-based care for people with back pain.