A Simple Approach for the Evaluation of Individual Bioequivalence

Abstract

The refinement of a procedure for assessing individual bioequivalence is presented. It is suggested that two drug products are inequivalent in individuals if a variation contrasting them substantially exceeds that recorded within the formulations, following their repeated administration to the subjects. In the implementation of the approach, first, the variance for the interaction between formulations and subjects is compared with the intraindividual variance estimated within drug products. The results are compatible with individual bioequivalence of the formulations if the upper, one-sided confidence limit for the ratio of estimated variances is less than a critical value (F_cr). If, moreover, the ratio itself is close to 1.0 (in practice, say, less than 2.0), then terms can be pooled and the intraindividual variances estimated between and within formulations compared. Individual bioequivalence of the contrasted drug products is declared if the upper confidence limit for the ratio of these variances is less than F_cr. The magnitudes of the significance level (α) and of F_cr are subject to regulatory considerations. Initially, α = 0.10 and F_cr = 4.0 could be suggested. The various variances can be estimated either in an analysis of variance or by straightforward, direct calculations. Thus, the proposed procedure is not only effective but also simple.