Statistically Significant versus Practically Relevant Trend in Stability Data

Abstract

How to decide whether a statistically significant trend is of practical relevance? In the context of stability data of pharmaceuticals, this publication provides a way forward to use different measures of (method) variability to compare to the observed changes over time. A panel of analytical experts assessed whether statistically significant trends were of practical relevance or not. For different types of assessing variability, recommendations for decision criteria were derived that best matched these assessments, i.e., finding a suited balance between not detecting a relevant trend and between flagging a trend wrongly as relevant. For this purpose, more than 60 data sets from Biosimilar projects of Sandoz/Novartis were leveraged. Hence, this article provides a scientific way to assess whether a statistically significant trend is of practical relevance or not, and a case study is presented and discussed.

Keywords: Stability; TOST; equivalence test; relevant change over time; significant change over time.