Blind grading is a quality method in which two or more raters grade the same adverse event independently, without seeing each other's grades, so their agreement can be measured and their disagreements resolved on the evidence. Sites use it to check grading consistency, calibrate new graders and build a reference set to compare an AI suggestion against. In Burna AI, an administrator builds a case set from sample notes, imported cases or a case entered by hand, assigns raters and an adjudicator, and each rater sees only their own work until the case is settled. Where raters agree, the case closes; where they disagree, it goes to an adjudicator. Each submission is signed under 21 CFR Part 11.
Interactive demo
Try Burna right now, no sign up required.