A More Powerful Average Bioequivalence Analysis for the 2x2 Crossover

The 2 × 2 crossover is commonly used to establish average bioequivalence of two treatments. In practice, the sample size for this design is calculated under an implicit belief that the true average bioavailabilities of the two treatments are (almost) identical. However, the ”standard” average bioequivalence analysis does not reflect this prior belief and this leads to a loss in efficiency. The reason for this loss is that if the true average bioavailabilities are indeed the same, then data from the 2× 2 crossover contain additional information that the standard analysis fails to utilize. We propose an alternate average bioequivalence analysis in order to rectify the deficiency in the standard analysis. The validity and significant power advantages of our proposed method are illustrated with simulations and a numerical example with real data is provided.