Bioequivalence and Bioavailability Forum

Hi Balu!

The Forum's Policy states: We expect a basic knowledge on BE/BA or related fields and the willingness to begin first with the Search function for similar problems.
Sorry that most of your posts didn't get a single answer, because they were considered to be too basic by the members of the forum.
Please consider reading one of introductory books on BE-testing:

G Amidon, L Lesko, K Midha, V Shah and J Hilfinger (eds.)
International Bioequivalence Standards: A NEW ERA
TSRL, Ann Arbor (2006)
D Hauschke, V Steinijans and I Pigeot
Bioequivalence Studies in Drug Development
John Wiley & Sons, Chichester (2007)

More statistics-oriented books are given in this post - or consider attending a course on BE-Testing (see this post for 2010’s schedule).

❝ Why sequence, period and treatment effects come?

            period
           +------
  sequence | 1  2
  ---------+------
      1    | R  T
      2    | T  R
Effects:

• Sequence: Subjects receiving treatments in the order RT vs. subjects receiving treatments in the order RT
  
• Period: 1 vs. 2
  
• Treatment: T vs. R

The period effect is of no interest - it is accounted for in ANOVA; even if present, the treatment effect and CV_intra are not affected (not influencing the confidence interval; example).

The treatment effect is of no interest as well (except for the Danish). Either your confidence interval is included in the acceptance range - or not. A significant treatment effect is a hint that in your study

• the CV was lower,
  
• the point estimate was closer to one, or
  
• both

... than expected.

The sequence effect may bias the treatment effect - there's no statistical procedure to compensate for that in a 2×2 crossover. Plan your study to avoid it (sufficient washout, no predose concentrations in period 2). But that's a wide field (search the forum).