Replicate Study designs [Outliers]

posted by Dr_Dan  – Germany, 2014-01-08 10:54 (3434 d 14:01 ago) – Posting: # 12143
Views: 12,844

Dear balakotu
Unbiased assessment of results from randomised studies requires that all subjects are observed and treated according to the same rules. These rules should be independent from treatment or outcome. In consequence, the decision to exclude a subject from the statistical analysis must be made before bioanalysis. Exclusion of data cannot be accepted on the basis of statistical analysis or for pharmacokinetic reasons alone, because it is impossible to distinguish the formulation effects from other effects influencing the pharmacokinetics. Replicate designs are mostly used in case of highly variable drugs (CV%>30) in order to widen the acceptance range. If you exclude subjects with extreme values the variability decreases but who tells you that the data then obtained reflect reality? Which is the "true" variability, the one with subjects included or the one with subjects excluded? As long as you can not provide evidence that the subjects under discussion are true outliers you will not be able to exclude them. For the FDA it would be possible to perform an outlier study.
I hope this helps.
Kind regards

Kind regards and have a nice day

Complete thread:

UA Flag
 Admin contact
22,616 posts in 4,740 threads, 1,613 registered users;
11 visitors (0 registered, 11 guests [including 6 identified bots]).
Forum time: 01:55 CEST (Europe/Vienna)

Biostatistician. One who has neither the intellect for mathematics
nor the commitment for medicine but likes to dabble in both.    Stephen Senn

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz