Bioequivalence and Bioavailability Forum

Hi Shawn,

❝ I am a fresher to BE analysis.

Welcome to the club!

❝ […] how to calculate the MSE_w and MSE_B for 2 X 2 crossover design and fully replicated design (TRT|RTR or TRTR|RTRT).

Thanks for not mentioning the lousy partial replicate (TRR|RTR|RRT).

❝ Could anyone tell me the detail of calculation for these different designs?

Do you want to calculate it manually (paper, pencil, brain)? A nice morning exercise as long as you have only fixed effects in the model (ANOVA).*
If you are interested in full-blown mixed-effects models (e.g., the FDA’s or the CFDA’s for reference-scaling), forget it. This would require restricted maximum likelihood estimation, which is – practically – impossible without software.

Given all that, get suitable software, validate it, and be happy. The residual mean squares are part of the standard output (example of a 2×2 design).

---

• Chow S-C, Liu J-p. Design and Analysis of Bioavailability and Bioequivalence Studies.
  Boca Raton: CRC Press; 3^rd edition 2009. ISBN: 978-1-58488-668-6.