Intra-GUM comparisons for the estimation of measurement uncertainty: Application to a chromatographic assay of four statins in a dry form

Abstract

The chromatographic method (HPLC) is well-known for its dependability and precision in pharmaceutical analyses for assessing the quality and efficacy of drug products. However, the characteristics of the HPLC method make sense only after a careful evaluation of the measurement uncertainty. Nevertheless, it has been observed that precise identification of the sources of uncertainty resulting from the operational mode using the HPLC technique to assess their standard uncertainty based on the ISO-GUM approach is not obvious. Because of our extensive bibliographic search, we noticed that many studies published on the estimate of the uncertainty in chromatographic methods (HPLC) using the Eurachem/Citac guide are somewhat diverse in their use of this approach. The documentary research revealed four mathematical models namely, GUM1, GUM2, GUM3, and GUM4 that are most used to estimate the measurement uncertainty of the HPLC technique. Hence, a comparison between the four GUM mathematical models was carried out graphically and statistically using one-way ANOVA followed by Tuckey and Scheffe tests. The comparison results show a significant difference, which may create confusion in the reporting of measurement results as well as problems in making sound decisions. This leads to the suggestion that alternative approaches, such as the approach of a total error or ISO 11325, be used to check the ISO-GUM method's calculations of measurement uncertainty.

Keywords: HPLC method; ISO-GUM; Measurement Uncertainty; Statins; Total Error.