Means: Correlated, Paired, and Cross-Over Designs
Computes the FDA individual bioequivalence (IBE) aggregate criterion and the corresponding linearized criterion on the log scale. The calculation accounts for the Test–Reference mean difference, subject-by-formulation interaction variance, and within-subject variances.
The FDA individual bioequivalence criterion compares the squared Test–Reference mean difference, subject-by-formulation interaction variance, and the difference between Test and Reference within-subject variances. The denominator uses the larger of the Reference within-subject variance and the specified constant variance.
Under the FDA formulation described in the literature, individual bioequivalence corresponds to the alternative η < θI. The 2001 FDA guidance used θI = 2.4948 with σ2W0 = 0.04.
For the 2×4 replicated crossover design, the linearized criterion is obtained by subtracting the IBE limit multiplied by the scaling denominator:
Thus, g < 0 is algebraically equivalent to η < θI at the population-parameter level. When σ2WR ≤ σ2W0, the criterion is constant-scaled; when σ2WR > σ2W0, it is reference-scaled.
Chiang, C., Hsiao, C.-F., & Liu, J.-P. (2014). Sample Size Determination for Individual Bioequivalence Inference. PLoS ONE, 9(10), e109746. The paper defines the IBE criterion, the linearized criterion, and the MLS approach for replicated 2×4 crossover designs.
Schall, R., & Williams, R. L., for the FDA Individual Bioequivalence Working Group (1996). Towards a practical strategy for assessing individual bioequivalence. Journal of Pharmacokinetics and Biopharmaceutics, 24(1), 133–149.
U.S. Food and Drug Administration (2001). Guidance for Industry: Statistical Approaches to Establishing Bioequivalence. The guidance describes the replicated-crossover framework and the FDA IBE criterion.
For current PASS documentation and procedure listings, see the PASS documentation.