Accuracy of glucose meters is a common topic of clinical concern. Blood glucose meters must meet accuracy standards set by the
International Organization for Standardization (ISO). According to ISO 15197 Blood glucose meters must provide results that are within 20% of a laboratory standard 95% of the time (for concentrations about 75 mg/dL, absolute levels are used for lower concentrations). However, a variety of factors can affect the accuracy of a test. Factors affecting accuracy of various meters include calibration of meter,
ambient temperature, pressure use to wipe off strip (if applicable), size and quality of blood sample, high levels of certain substances (such as
ascorbic acid) in blood,
hematocrit, dirt on meter,
humidity, and aging of test strips. Models vary in their susceptibility to these factors and in their ability to prevent or warn of inaccurate results with error messages. The
Clarke Error Grid has been a common way of analyzing and displaying accuracy of readings related to management consequences. More recently an improved version of the Clarke Error Grid has come into use: It is known as the
Consensus Error Grid. Older blood glucose meters often need to be "coded" with the lot of test strips used, otherwise, the accuracy of the blood glucose meter may be compromised due to lack of calibration.