PharmPK Discussion - Relative BA and statistical analysis

PharmPK Discussion List Archive Index page

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


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

Back to the Top

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


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.

Back to the Top

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


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

Back to the Top

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