- On 15 Oct 2002 at 14:10:14, soraya.madani.-at-.pharma.novartis.com sent the message

Back to the Top

For a relative BA study (not a BE) in 20 healthy volunteers ( atypical #

for this type of study) what could be the most appropriate statistical

approach (test) to compare the test verus reference formulations? Is

ANOVA

and use of "p" value be an acceptable appoach? Or should the confidence

interval approach be applied, just as it is the practice for the BE

studies?

Thank you

Soraya - On 15 Oct 2002 at 22:54:01, "Edmond B. Edwards" (editr.aaa.sympatico.ca) sent the message

Back to the Top

Hello,

Presumably you are conducting a bioavailability study to mimic the

assessments in a bioequivalence study, so I would suggest confidence

intervals - they are considered the more appropriate way to evaluate

equivalence.

... Edmond Edwards, Ph.D. - On 16 Oct 2002 at 09:09:20, BEDDING_ALUN.-a-.Lilly.com sent the message

Back to the Top

I completely agree with Edmond's assessment. If you are trying to

mimic a bioequivalence assessment then you should use a confidence

interval approach. The use of p-values is inappropriate for an

equivalence assessment. I am assuming that you using a 2x2 study and

therefore the most appropriate analysis would make use of a mixed

effects model, fitting the fixed effects of sequence, period and

treatment and subject as a random effect.

Kind regards,

Alun Bedding

Senior Statistician

Statistics and Information Sciences

Eli Lilly and Company

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 "Relative BA and statistical analysis" as the subject

PharmPK Discussion List Archive Index page

Copyright 1995-2010 David W. A. Bourne (david@boomer.org)