A brief overview of the GRADE approach for rating strength of evidence and practice recommendations

Abstract

The medical fraternity is faced with an avalanche of medical literature. While it is desirable that the current best evidence inform practice guidelines and recommendations, it is true that much of the evidence is not of optimal quality, and recommendations based on poor quality evidence may result in undesirable outcomes. Moreover, many context specific considerations, such as costs and patient preferences may be key concerns in formulating recommendations. These issues call for a systematic framework to rate the quality of evidence and strength of recommendations. The GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) offers such a framework for any given clinical question, and offers advantages over traditional approaches to guideline development. Crucial elements for ensuring the trustworthiness of the guideline development process, such as recruiting an appropriate panel, excluding conflicts of interest, systematically reviewing the best evidence, rating certainty in evidence, and incorporating patient values and preferences, are rendered due importance in the process.