Assessing parallelism prior to Determining Relative Potency
With nearly 50 citations, this paper summarizes the unanimous agreement of a USP committee, that the method for assessing parallelism that is currently presented in 〈111〉 and in the European Pharmacopeia’s Chapter 5.3 is flawed and should be replaced. The procedure adopted is the one that was originally published by me and my co-author Dr. Jan Callahan in 2003.
This article explains why a difference testing approach is not appropriate and presents an argument for testing the null hypothesis for a specified difference (ie using an equivalence approach) to statistically ascertain parallelism.