Back to the Top
Dear all
My question is about the sequence effect at BA/BE studies. At the
crossover Anova the sequence effect must be tested against Subject
(Sequence) effect ("Generics and Bioequivalence" by AndrŽ J. Jackson) or
against the residual (as Kinetica v. 1.1 do it). The results can be
quite different !
Thank you very much to all
Best Regards
Nuno Silva
Back to the Top
You don't mention whether you are analysing a 2x2 or higher order
cross-over. I am assuming that are analysing a 2x2.
Sequences are a between subject effect and therefore need to be compared
to the between subject residual, which as you correctly say in your
first statement is the subject(sequence) term. If you do the analysis
against the residual (which is the within subject residual) you will
certainly get a different answer. Using the within subject residual
actually overestimates the effect of sequence leading to a greater
chance of concluding a sequence effect.
Regards,
Alun Bedding
Clinical Pharmacology Scientific and Regulatory Expert Statistics and
Information Sciences
Eli Lilly and Company
Back to the Top
Dear Nuno,
As noted earlier by Alun Bedding, the correct error term for testing the
sequence effect in crossover studies is subject(sequence). Incidentally,
WinNonlin now offers a bioequivalence wizard which automatically builds
the appropriate ANOVA model for BE analysis, and computes the relevant
statistical tests (including that of sequence) subsequent to user
identification of the relevant study variables (i.e. Sequence, Subject,
Period, and Treatment.) In accordance with recently-published FDA
guidelines, the BE wizard also offers linear mixed effect analysis, and
therefore, allows for the analysis of data from repeated crossover
designs as well as population/individual bioequivalence.
If you would like to know more about the capabilities of this powerful
new statistical tool, or any of WinNonlin's other capabilities, I would
be very happy to answer any questions you may have (contact info below.)
With Best Regards,
Mark R. Lovern, Ph. D.
Pharsight Scientific Support
5520 Dillard Drive, Suite 210
Cary, NC
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)