preventive services that are widely considered to be the “gold standard” for the process of guideline development (Guirguis-Blake et al., 2007). The task force maintains a rigorous process for contracting with evidence-based practice centers (EPCs) to produce systematic reviews and developing practice recommendations; it sets a high standard for other organizations.
The NIH also convenes expert panels to develop clinical recommendations. For example, the National Heart, Lung, and Blood Institute (NHLBI) launched the National Cholesterol Education Program in November 1985 and now sponsors a number of panels that produce guidelines in that area. The NIH Consensus Development Conferences also seek to inform clinical practice, and now contracts with EPCs for systematic reviews of the evidence, although they do not produce practice guidelines.
One of the challenges inherent in having such a decentralized, pluralistic process is that often multiple groups produce guidelines in the same clinical topic area. These guidelines may duplicate previous work or produce contradictory findings that may remain unresolved (Woolf et al., 1999). Box 5-1 illustrates a case in which two guideline development panels reviewed largely the same bodies of evidence and reached different conclusions about appropriate clinical practice.
The magnitude of this challenge is illustrated by the preponderance of guidelines related to hypertension and stroke. The NGC, for example,
Conflicting Guidelines for the Treatment of Epilepsy
Separate panels convened in the United States and the United Kingdom looked at the use of new antiepileptic drugs for the treatment of newly diagnosed epilepsy patients. Although both groups supported the efficacy and safety of the new drugs, they diverged on the appropriate management of these cases. The U.S. panel recommended that either the new drugs or the standard drugs be used (depending on the characteristics of the patient), whereas the U.K. panel was more restrictive, recommending that the new drugs be used only in more narrow circumstances (e.g., cases where the older drug is contraindicated). These discrepancies may be partially explained by the limited amount of information available on the new drugs and the different factors considered by the reviewers (e.g., the U.K. review considered cost and quality of life, but the U.S. review did not). It is also likely that more subjective judgments play a role in the recommendation process.
SOURCE: Beghi (2004).