Error grid analysis
The error grid analysis ( EGA for short ; also error grid analysis ) is used to evaluate the clinical accuracy of the blood sugar measured values in comparison to a reference value .
It was developed by W. Clarke (Department of Pediatrics, University of Virginia Medical School, Charlottesville) in the 1970s. Originally, it was used to determine the clinical accuracy of the patient's opinion about the current blood glucose value compared to the value that the patient had received with his device for self-monitoring of blood glucose (SMBG). Later this analysis was increasingly used to compare the clinical accuracy of the values determined with the measuring devices for blood sugar self-monitoring with a reference value. A description of the EGA appeared in the journal “ Diabetes Care ” in 1987. Finally, the Error Grid Analysis became one of the “gold standards” for determining the accuracy of blood glucose meters.
The grid (grid, raster) arranges the determined blood sugar values of a measuring device to be assessed and the reference device in five regions.
- Region A contains all values that show a maximum deviation of 20% from the reference value.
- Region B has all values that deviate by more than 20% from the reference value, but do not lead to a decision that has an incorrect effect on the patient's therapy.
- Region C, here are the results that lead to unnecessary decisions.
- Region D shows all values that indicate a potentially dangerous error in the detection of hypo- or hyperglycaemia and in the
- Region E are all results which would lead to confusion in the treatment of hypoglycemia or hyperglycemia and vice versa.
Individual evidence
- ↑ " Evaluating clinical accuracy of systems for self-monitoring of blood glucose " at diabetesjournals.org
- ^ Graphic of the Food and Drug Administration ( Memento from September 24, 2005 in the Internet Archive )
