A review of grading systems for evidence-based guidelines produced by medical specialties.

The development of evidence-based guidelines requires scrupulous attention to the method of critical appraisal. Many critical appraisal systems give 'gold standard' status to randomised controlled trials (RCTs) due to their ability to limit bias. While guidelines with a prominent research base consisting of RCTs have been well served by such systems, specialist societies with research bases consisting of a wide range of study designs have been at a disadvantage, potentially leading to inappropriately low grades being given for recommendations. A review of the Scottish Intercollegiate Guidelines Network, the Grading of Recommendations Assessment, Development and Evaluation, the Graphic Appraisal Tool for Epidemiology and the National Service Framework for Long Term Conditions grading systems was therefore undertaken. A matrix was developed suggesting the optimum grading system for the type of guideline being developed or question being addressed by a specialist society.