Abstract
This paper discusses a procedure for independent evaluation of commercially available blood glucose meters. The evaluation exercise provides a basis for assessment of manufacturer's claims. It also provides solid experience in the reliability of blood glucose data obtained from a specific meter under field conditions.
We report measurements of the precision of three capillary blood glucose measurement devices, commercially available in the USA, used under field conditions. We have compared the measurements of blood glucose samples analysed by three different instruments. The method of analysis involves making two blood glucose measurements from a single finger puncture. This experimental design in available to any clinician who would wish to try the expeiriment. We report values for the precision (reproducibility) of a single device and comparisons for pairs of meters using standard statistical techniques. The coefficient of variation for the difference between the first and second measurements obtained for three instruments was: A, Β±28.7%; B, Β±36.5%; C, Β±10.2%. The data show that instrument C worked much better than either A or B for reproducing blood glucose values on the same sample under field conditions.