The estimation of uncertainty of measurement of glycated hemoglobin as an analytical performance specification and in the interpretation of its results

Abstract

Background: Glycated hemoglobin (HbA1c) plays a key role in monitoring the glycemic state of an individual. Uncertainty of measurement (U) indicates the magnitude of the doubt about a measurement result. To properly classify an individual as under either good or poor glycemic control, it has been suggested that U of an HbA1c result should not exceed ±0.5%.

Methods: The statistical method used to calculate uncertainty of measurement was the "top down" approach suggested by EURACHEM/CITAC. This approach allows the inclusion of imprecision, bias, uncertainty of bias and uncertainty of the calibration of the HbA1c method. The value of bias was obtained using data generated from the external quality assessment of the Randox International Quality Assessment Scheme and that of the Unity data management software system. Imprecision was calculated after the daily analysis of two levels of control sera.

Results: Calculation of uncertainty of measurement of HbA1c was a straightforward procedure used to calculate U. Due to the different bias results obtained using two different external quality programs, the results of U were significantly different (±0.19% vs ± 0.43%) from each other; however, in both cases, the U results were below the maximal suggested uncertainty of ±0.5%.

Conclusions: The calculation of U of HbA1c by the EURACHEM/CITAC method is a practical approach that can be used as an additional analytical goal in the measurement of HbA1c. In addition, this information can aid clinicians to determine the level of confidence that can be placed in the test results.