Evaluation of the Linearity of Quantitative Measurements Procedures with Respect to In Vitro Diagnostic Devices

Abstract

Linearity is one of the key characteristics for evaluation of the accuracy in assay validation. The accuracy in terms of linearity in vitro validation of quantitative analytical diagnostic procedures is defined as its ability (within a given range) to obtain the test results, which are directly proportional to the concentration (amount) of the analyte in the test sample. According to the approved Clinical Laboratory Standard Institute (CLSI) EP6-A guideline, a statistically significant nonlinearity does not show where that nonlinearity is, nor do they show the magnitude of the bias. Therefore, even if the best-fitted model is a nonlinear model, it does not necessarily imply that the assay can not be concluded linear. With respect to the approved CLSI EP6-A guideline, when the best-fitted model is non-linear, the linearity of an analytical method can still be concluded if the deviation from the linearity as measured by the difference in the predicted means between the best-fit nonlinear model and the linear model at each concentration is within some pre-specified goal. The current estimation method for evaluation of the linearity recommended by the Clinical Laboratory Standard Institute (CLSI) guideline EP6-A directly compares the point estimates with the pre-specified allowable limit and completely ignores the sampling error of the point estimates. An alternative method for evaluation of linearity proposed by Kroll, et al. considers the statistical test procedure based on the average deviation from linearity (ADL). However this procedure is based on the inappropriate formulation of hypothesis for evaluation of the linearity. Consequently, the type I error rates of both current methods may be inflated. To overcome these deficiencies, we propose some appropriate procedures with new criteria for assessment of linearity in assay validation. The simulation results and numerical examples will be also provided. Keywords: Linearity, Accuracy, Assay validation, Generalized Pivotal Quantities