Table 4.

GRADE system for grading quality of evidence for an outcome

Step 1: starting grade for quality of evidence based on study design	Step 2: reduce grade	Step 3: raise grade	Final grade for quality of evidence for an outcomea
High for randomized controlled trials Moderate for quasi-randomized trial Low for observational study Very low for any other evidence	Study quality –1 level if serious limitations –2 levels in very serious limitations Consistency –1 level if important inconsistency Directness –1 level if some uncertainty –2 levels if major uncertainty Other –1 level if sparse or imprecise data –1 level if high probability of reporting bias	Strength of association +1 level is strong,b no plausible confounders, consistent and direct evidence +2 levels if very strong,c no major threats to validity and direct evidence Other +1 level if evidence of a dose-response gradient +1 level if all residual confounders would have reduced the observed effect	High Moderate Low Very low

GRADE, grading of recommendations assessment, development, and evaluation; RR, relative risk.

^aThe highest possible grade is “high” and the lowest possible grade is “very low.”

^bStrong evidence of association is defined as “significant RR of > 2 (< 0.5)” based on consistent evidence from two or more observational studies, with no plausible confounders.

^cVery strong evidence of association is defined as “significant RR of > 5 (< 0.2)” based on direct evidence with no major threats to validity.

Modified with permission from Uhlig K, Macleod A, Craig J, et al. Grading evidence and recommendations for clinical practice guidelines in nephrology. A position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int. 2006;70:2058–2065.¹⁷¹