There are thousands of clinical practice guidelines authored by hundreds of medical groups. Clearly criteria for culling are needed. Consider, for example, the Institute of Medicine’s standards for trustworthiness:
1. Transparent process: The processes by which a clinical practice guideline is developed and funded should be described transparently.
2. Conflicts of interest: Potential guideline development group members should declare conflicts. None, or at most a small minority, should have conflicts, including services from which a clinician derives a substantial proportion of income. The chair and co-chair should not have conflicts. Eliminate financial ties that create conflicts.
3. Guideline development group composition: The group should be composed of methods experts, clinicians, representatives of stakeholders, and affected populations.
4. Systematic reviews: Essential to the process, systematic reviews must meet the IOM’s methodological standards.
5. Evidence quality and recommendation strength: Explain the reasoning behind each recommendation, summarize evidence for benefits and harms, characterize the quality and quantity of relevant evidence and the role of subjective judgments. Rate the level of evidence and the strength of the recommendation. Describe differences of opinion about recommendations.
6. Articulating recommendations: Describe the action recommended by the guideline and when it should be used; wording should facilitate measurement of adherence.
7. External review: Essential to the process, external review should include a full spectrum of stakeholders, reviewers not identified by name, explain all changes done in response to reviewers, and post for public comment.
8. Updating: Document the dates of the guideline, systematic review, and planned update; monitor the literature and update the guideline when new evidence suggests the need for change.
No, not every medical society producing adheres to these standards. Related, here’s a critique of the Cochrane Systematic Reviews in mental health, which references others (shared without endorsement of the conclusions).
