Different statistical approaches are proposed to facilitate such evaluations. There is required a regulatory consensus on a quantitative and scientifically justified definition and an underlying hypothesis of statistically equivalent quality. It is essential to calculate the operating characteristics of different approaches. This article was written to propose a hypothesis for establishing statistically equivalent quality which is concordant with current regulations. Along with that this article also describes a tool that allows comparisons of different statistical approaches or tests by calculating the operating characteristics for false acceptance and false rejection rates of a claim for statistically equivalent quality. Error rates should be low to allow a meaningful application of a statistical approach in regulatory decision making. The described tool can be used to compare different statistical approaches for their suitability and may also facilitate the discussion and development of statistical approaches for comparing quality attributes in similarity assessments in general. Statistical approaches may facilitate the comparison of quality attributes by identification of statistical differences, which require further scientific evaluation before drawing a conclusion on whether a claim of high similarity is fulfilled or not.
Reference: https://aapsopen.springeropen.com/articles/10.1186/s41120-019-0033-9
