A comparison of five glucometers in South Africa

Abstract

Objective: To assess the accuracy and precision of five currently available blood glucose meters in South Africa
Background: Since the introduction of glucometers, there has been an ongoing, competition-driven development in both meter and strip technology, which has allowed for greater accuracy and reliability of results. Despite the advances in technology, there is significant variation amongst these glucometers necessitating a proper evaluation before use.
Methods: Glucose levels in capillary blood samples from 115 patients attending the diabetic clinic at Tygerberg Hospital were measured with
each meter, and compared with the laboratory reference method.
Results: The coefficients of variation (CVs) (imprecision) of most meters were acceptable at less than 5%, with a bias ranging from 1.7 to 6.8%. None of the glucometers satisfied the American Diabetes Association (ADA) recommendation of less than 5% bias.
Conclusions: The study highlights the need for an objective and independent comparison of all glucometers in South Africa, as the variability observed can impact on patient care.

Keywords: glucometer; diabetes; self monitoring (SMBG); analytical error; Clarke error grid