Liquid chromatography is one of the main techniques used in pharmaceutical quality control analytical procedures. However, there will always be a measurement uncertainty (MU) associated with them, that can lead to the approval of an out of specification lot (consumer risk) or rejection of a lot within specification (producer risk). Thus, the aim of this study was to evaluate the performance of liquid chromatography analytical procedures based on their measurement uncertainty and to estimate the risk of false conformity decisions. The uncertainties of the analytical procedures were estimated based on the results of validation (trueness and precision). Then, the ratio between overall uncertainty and specification range (U/T%) was calculated. It was noted that in most cases (73%), random errors (precision) contributes more significantly to the overall uncertainty when compared to systematic errors (trueness). Monte Carlo method was used, generating different manufacturing processes scenarios, and analytical results based on the MU of each analytical procedure. Then, consumer's and producer's risks were estimated from the simulated values. Pharmaceutical dosage forms that require more steps in sample preparation had higher measurement uncertainties, often above the recommended target uncertainty. As most of the analytical procedures showed U/T% values above recommended, the majority presented high estimated risk values and did not fit for purpose. Therefore, it is important to considerate the measurement uncertainty as part of analytical procedures validation, since trueness and precision values affect directly the measurement uncertainty and the risk of false conformity decisions.
Keywords: Analytical procedure validation; Conformity assessment; Liquid chromatography; Measurement uncertainty; Monte Carlo method.
Copyright © 2019 Elsevier B.V. All rights reserved.