A note on sample size determination for bioequivalence studies with high-order crossover designs.

Similar to Liu and Chow, approximate formulas for sample size determination are derived based on Schuirmann's two one-sided tests procedure for bioequivalence studies for the additive and the multiplicative models under various higher order crossover designs for comparing two formulations of a drug product. The higher order crossover designs under study include Balaam's design, the two-sequence dual design, and two four-period designs (with two and four sequences), which are commonly used for assessment of bioequivalence between formulations. The derived formulas are simple enough to be carried out with a pocket calculator. The number of subjects required for each of the four higher order designs are tabulated for selected powers and various parameter values.