How to interpret the certainty of evidence based on GRADE (Grading of Recommendations, Assessment, Development and Evaluation)

Abstract

GRADE (Grading of Recommendations, Assessment, Development and Evaluation) is awidely used approach in the fields of medicine and public health to assess the outcome-specific certainty of the evidence in systematic reviews. To make the GRADE approach comprehensible in order to facilitate the reading, understanding and interpretation of GRADE assessments in systematic reviews. Presentation of the procedure of the GRADE approach using the example of aCochrane review on selenium supplements in the prevention of prostate cancer. GRADE provides criteria for rating the certainty of evidence. GRADE's approach to rating the certainty of the evidence is based on afour-level system (high, moderate, low, very low). The GRADE approach classifies bodies of randomized controlled trials as initially starting at high certainty and bodies of observational studies at initially starting at low certainty. By assessing the five domains (risk for bias, inconsistency, indirectness, insufficient precision and publication bias), certainty can be rated down or, in the case of large effects, existing dose-response relationships or plausible confounders, rated up. GRADE is aconsistent and transparent approach for rating the certainty of abody of evidence by offering explicit key questions.