The desire for quality assurance by medical professional organisations has led to the development of guidelines by national and international societies, formulated by expert panels and based on current evidence in the literature. Governments and third party payers have increased their scrutiny of what physicians do and how they do it. As the costs of medical care grow, the call for assessment of appropriateness and cost effectiveness of medical procedures increases. In evaluating clinical practice, a number of studies have pointed to the large variations in daily practice between countries, centres, and individual doctors.

Governments, insurers, and doctors would like to see this variance reduced. This has led to the development of guidelines for clinical practice which can serve the purposes of both physicians and administrators; if used wisely, antagonism between the two can be avoided.

Guidelines formulated by expert panels have several limitations. First, discordance is observed between panels addressing the same issue.1 When more evidence is available, less discordance is expected, but it is unknown if different expert panels show better concordance with increasing evidence. Paradoxically, variability in opinions (and practice patterns) may in fact increase with increasing evidence. Similarly, the need for guidelines may increase with increasing evidence—for example, cholesterol lowering treatment.

Second, the evidence from the literature is usually derived from highly selected cohorts of patients, making extrapolations to the general population difficult or impossible.

Third, the advances in some clinical fields are so rapid that guidelines may be outdated by the time they are published; therefore, they need to …