Bioequivalence and Bioavailability Forum

Dear All,

According to EMA BE GL 2010, 4.1.8.Evaluation / Statistical Analysis:
The terms to be used in the ANOVA model are usually sequence, subject within sequence, period and formulation. Fixed effects, rather than random effects, should be used for all terms.

And according to Mr. Helmut Schütz's 23-25 May 2012 Moscow Presentation PK–NCA, PK based Design, Biostatistics Part I Moscow PK-NCA, PK based Design

on page 111 and 112, the default setting of PHNX has to be changed to subject(sequence) to Fixed Effects.

We're using subject (sequence) as Random Effects in our statistical calculations for 2x2 cross-over design. We've done the calculations with two methods of ANOVA, and the results are the same.

The calculations with random effects are acceptable from EMA?

Thank you all...

Erkin Alkan