Both critics and supporters of evidence-based medicine view clinical practice guidelines as an important component of this self-defined "new paradigm" whose goal is to rationalize medicine by grounding clinical decision-making in a careful assessment of the medical literature. We present an analysis of the debates within a guideline development group (GDG) that led to the drafting, revision and publication of a French cancer guideline. Our ethnographic approach focuses on the various aspects of the dispositif (or apparatus) that defines the nature and roles of participants, procedures, topics and resources within the GDG. Debates between GDG members are framed (but not dictated) by procedural and methodological rules as well as by the reflexive critical contributions of the GDG members themselves, who justify their (tentative) recommendations by relating to its (possible or intended) audiences. Guideline production work cannot be reduced to an exchange of arguments and to consensus-seeking between pre-defined professional interests. It is about the production of a text in the material sense of the term, i.e. as a set of sentences, paragraphs, statements and formulations that GDG members constantly readjust and rearrange until closure is achieved. As such, guidelines partake in the emergence and stabilization of a new configuration of biomedical knowledge and practices grounded in the establishment of mutually constitutive links between two processes: on the one hand, the re-formatting of clinical trials into a device for producing carefully monitored evidence statements targeting specific populations and clinical indications and, on the other hand, the increasingly pervasive role of regulatory processes.
(c) 2010 Elsevier Ltd. All rights reserved.