Gradering av kvaliteten på dokumentasjonen

Abstract

What is meant by the claim that something is "well documented"? How much confidence can we have in the results? The most commonly used method to grade the quality of the evidence is GRADE (Grading of Recommendations Assessment, Development and Evaluation). GRADE is used in systematic reviews, where the effects of different treatments or options are being compared, and in guidelines with recommendations about the competing options. GRADE makes a clear distinction between the quality of the evidence and the strength of recommendation. When grading the quality of the evidence with GRADE, the starting point is a well conducted systematic review of research on the question that is being assessed. Both randomized controlled trials and observational studies can be included in the evaluation. GRADE uses five criteria to judge for possible downgrading of the quality of the evidence: risk of bias, consistency, directness, precision and reporting bias. Observational studies may in addition be assessed for possible upgrading of the quality of evidence using the following three criteria: strong association, dose response relationship, and plausible confounding. All the judgments are noted and presented together with the results in tables.