Back to the Top
Dear All,
The EMA bioequivalence guideline released in August last year states that "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".
Theoretically, should the "subject within sequence" factor not be considered as a random effect?
Thanks for sharing your view,
Fabrice
Fabrice Nollevaux,
Pharmacometrician
Arlenda SA
www.arlenda.com
Arlenda will participate to the following events:
- ASCPT 2011, Dallas, US, 02-05 Mar 2011
- ACOP 2011, San Diego, US, 03-06 Apr 2011
- BAYES2011, Louvain-la-Neuve,27-29 Apr 2011
- PAGE 2011, Athens, Greece, 07-10 Jun 2011
Back to the Top
The following message was posted to: PharmPK
Dear Fabrice!
> The EMA bioequivalence guideline released in August last year states
>
The GL was released in January 2010 and came into effect last August. ;-)
> "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".
>
Yes, funny.
> Theoretically, should the "subject within sequence" factor not be
> considered as a random effect?
>
IMHO subjects in a BE study are a random sample.
At least that's written in all these nice textbooks.
* At least in a non-replicated (e.g. 2x2x2) cross-over numerically
results are identical (just give it a try).
* If subject(sequence) is fixed, you don't get the inter-subject CV (who
wants to plan a parallel study - based on total variance - after a
cross-over)...
* In an imbalanced replicate design results may (slightly) differ.
For some background see this rather lengthy thread:
http://forum.bebac.at/mix_entry.php?id=1817
Random/fixed was an issue in a heated debate last June at the
EGA-Symposium on BE in London.
Other points were:
* The mandatory (?!) use of ANOVA (or PROC GLM in SAS-speak); how do you
evaluate a partial replicate design without a mixed effects model (PROC
MIXED in SAS)?
* HVDs/HVDPs and scaling: Demonstration that CVwr>30% not caused by
outliers. Method?
* "Removal" of a foreign reference in a higher-order cross-over design.
EMA started an initiative involving the PK-Drafting Group and the
Biostatistics Drafting Group of EWP to clarify/correct some open issues
of the new BE-GL. I spoke to members of the EWP therapeutic subgroup on
Pharmacokinetics (formerly PK-Drafting Group) at the EUFEPS-Forum last
week in Barcelona: A clarification document (giving SAS-code and example
data-sets in order to validate other software) is already finalized and
has yet to get grinded at the mills of EMA's bureaucracy in order to get
published.
I will post a link a soon as possible.
Best regards,
Helmut
--
Ing. Helmut SchA\0x00tz
BEBAC - Consultancy Services for
Bioequivalence and Bioavailability Studies
Neubaugasse 36/11
1070 Vienna, Austria
Back to the Top
Where we can find detailed guidelines
Dr zafar
Back to the Top
The following message was posted to: PharmPK
Dear Dr. Zafar,
The latest EMEA guideline on the investigation of bioequivalence can be
found here:
http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/ 2010/01/WC500070039.pdf
Best regards,
Fabrice Nollevaux,
Pharmacometrician
Arlenda SA
www.arlenda.com
|
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Fixed subject effect in EMA BE guideline" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)