Evaluating the certainty of evidence in evidence-based medicine

Abstract

Certainty of evidence (formerly known as quality of evidence) is defined as the extent to which our confidence in an estimate of the effect is correct or our certainty that such estimate supports a particular recommendation for a clinical practice guideline. Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) is a structured and reproducible framework for assigning a level of certainty on a per-outcome basis for evidence derived from randomized and nonrandomized studies. The level of certainty starts as high or low and can be increased or decreased after considering several criteria (eg, risk of bias, inconsistency of results, publication bias, dose-response gradient, large magnitude of effect, among others). Here we describe in brief the GRADE process for summarizing and assigning a certainty rating for evidence.