Implementation in France,
the Netherlands, and Great Britain
The relationship between outcomes data and what physicians do in everyday practice is far from straightforward. David Eddy has pointed to one of the sources of this problem: outcomes data are rarely generated within the patient population that physicians treat. They originate mostly from a highly artificial situation: controlled clinical trials or similar activities. Thus, family physicians treating eczema may not be helped by outcomes data generated within a tertiary-care hospital where only the more serious cases are seen.
A "device" is needed to connect outcomes data to the practical situation, and practice guidelines are such a device (there are many other names, such as "consensus recommendations," "protocols," or "standards," which I will not use here). So, one of the functions of practice guidelines is translating outcomes data into medical decisions. Just a few words on the term "translation" may be in order here. 1 Because of its double meaning, "translation" is a good term for describing what happens when scientific information or a technical innovation is transported to users. It involves not only rendering the same thing in a different language but also shifting and transforming and adapting it to the circumstances and desires of users. The acceptance of scientific insights or technical innovations thus involves their modification. This, I think, is what happens to outcomes data when they are translated into practice guidelines.
Guidelines have been developed, published, and disseminated in most Western countries during the last two decades, not only with the purpose of connecting data to practice but also for other goals: reduction of unjustified practice variations, improving the quality of care, allowing for medical audit, helping physicians negotiate with patients, and, finally (the most tricky one), cost reduction. The problem with the last goal is that practice guidelines sometimes do not make health care cheaper but more expensive (and, hopefully, better). 2