Consensus conferences for the purposes of producing practice guidelines are occurring with increasing frequency both nationally and internationally. The international collaboration of national sciences in these efforts could have a dramatic impact on international standards of care. Too little emphasis is given to conference evaluations in terms of validity of methods, quality of recommendations, and influence on clinical practice and patient outcome. This article provides an overview of consensus methods used to produce guidelines in critical care. It also discusses the strengths and weaknesses of these methods, and how these may influence consensus guidelines. Finally, a brief overview of theoretically sound methods that can serve as benchmarks to evaluate current methods, and the bases for the development of improved methods is provided.