Library / Testing Statistical Hypotheses of Equivalence and Noninferiority


Stefan Wellek “Testing Statistical Hypotheses of Equivalence and Noninferiority” (2010) // Publisher: Chapman and Hall/CRC. ISBN: 9780429092671. DOI: 10.1201/ebk1439808184


  title = {Testing Statistical Hypotheses of Equivalence and Noninferiority},
  isbn = {9780429092671},
  url = {},
  doi = {10.1201/ebk1439808184},
  publisher = {Chapman and Hall/CRC},
  author = {Wellek, Stefan},
  year = {2010},
  month = {jun}

Quotes (3)

Equivalence vs. Noninferiority

When referring to specific problems and procedures, equivalence per se will always used in the strict, two-sided sense of the term. Noninferiority problems will be either called that way or, alternatively, addressed as one-sided equivalence problems.

Equivalence Testing and FDA Regulations

From a “historical” perspective (the first journal article on an equivalence test appeared as late as in the sixties of the twentieth century), the interest of statistical researchers in equivalence assessment was almost exclusively triggered by the introduction of special approval regulations for so-called generic drugs by the Food and Drug Administration (FDA) of the U.S. as well as the drug regulation authorities of many other industrial countries. Essentially, these regulations provide that the positive result of a test, which enables one to demonstrate with the data obtained from a so-called comparative bioavailability trial the equivalence of the new generic version of a drug to the primary manufacturer’s formulation, shall be accepted as a sufficient condition for approval of the generic formulation to the market. The overwhelming practi- cal importance of the entailed problems of bioequivalence assessment (drugs whose equivalence with respect to the measured bioavailabilities can be taken for granted, are termed “bioequivalent” in clinical pharmacology literature), arises mainly out of quantity: Nowadays, at least half of the prescription drug units sold in the leading industrial countries are generic drugs that have been approved to be marketed on the basis of some bioequivalence trial.

The Indifference Zone

Admittedly, finding a consensus on how to specify that indifference zone concretely is far from easy in the majority of applications. However, it is an indispensable step without which the testing problem the experimenter proposes would make no statistical sense at all.