In this paper, various stability designs, including matrixing and bracketing designs for determining drug shelf-life, are considered. We propose a criterion for design selection based on the precision of drug shelf-life estimates. For a fixed sample size, it is recommended that the design with the best precision for estimating the shelf-life should be used. For a fixed desired precision, the design with the smallest sample size is the best choice of design. An example is presented to illustrate the proposed method.
Journal Of Biopharmaceutical Statistics, 5(2), 201-214