Clinical guidelines are viewed as an important tool in the quest to promote evidence-based practice. Consequently there has been a proliferation in guideline development at both national and local levels. However, there are relatively few health care issues that have a well developed evidence base from which to derive recommendations. Therefore decisions in some areas have to be made even though no evidence has been determined, as limiting recommendations to those based on specific evidence could reduce the scope of guidelines, thus limiting their usefulness to clinicians. Guideline developers must rely on various sources of evidence and adapt their methods accordingly. This paper outlines a method for guideline development which incorporates a formal consensus process devised to tackle the challenges of a weak evidence base.