Validation and Estimation of Uncertainty for a Glucose Determination Method GOD-PAP Using a Multi-calibrator as Reference

• Published in: MĀPAN : journal of Metrology Society of India 2021-06, Vol.36 (2), p.269-278
• Main Authors: Rascón-Careaga, A., Corella-Madueño, M. A. G., Pérez-Martínez, C. J., García-Rojas, A. M., Souflé-Vásquez, S. Z., García-Moroyoqui, M. T., Córdoba-Beltrán, L. J., Cáñez-Carrasco, M. G., García-Alegría, A. M.
• Format: Article
• Language: English
• Subjects: Bias; Coefficient of variation; Correlation coefficients; Diabetes mellitus; Dilution; Glucose; Linearity; Mathematical analysis; Mathematical and Computational Physics; Mathematical Methods in Physics; Measurement Science and Instrumentation; Numerical and Computational Physics; Original Paper; Physics; Physics and Astronomy; Recovery; Reference materials; Reliability analysis; Reproducibility; Simulation; Theoretical; Uncertainty
• ISSN: 0970-3950; 0974-9853
• DOI: 10.1007/s12647-021-00441-5

When measuring glucose concentration, assuring the reliability of the results is critical, because the clinical diagnosis of pathologies related to diabetes mellitus, as well as their medical treatment, depends on them. One of the most widely used glucose determination methods in clinical chemistry is the GOD-PAP enzymatic method. Hence, a validation process based on the theory of errors, as well as the estimation of uncertainty based on the law of propagation of uncertainty, must be performed to obtain the required level of analytical reliability. For these purposes, the method was validated by assessing linearity, the limits of detection and quantification, precision under repeatability conditions, and intermediate precision, as well as assessing trueness based on bias and the recovery percentage. The level of uncertainty was estimated using the following sources of uncertainty: glucose concentration based on the calibration curve, volumetric material, dilution factor, analytical balance, the repeatability of the measurement, and reference material. According to the results obtained, the Pearson correlation coefficient for linearity was 0.9988, and the limits of detection and quantification were 0.48 and 1.60 mg/dL, respectively. Precision under repeatability conditions and the intermediate precision denoted a coefficient of variation of 2.1% and 1.9%, respectively, while bias was −0.85 mg/dL, and the recovery percentage was 99.15%. Finally, the estimated expanded uncertainty was 5.871 mg/dL.