Back to the Top
The following message was posted to: PharmPK
Dear All
In a 2*4 replicated crossover design a subject participated only two
periods for rest two periods the subject did not turn up while doing
statistical analysis can we include the subject for statistical
analysis or we have to remove the subject from the analysis.
Thanks and regards
Back to the Top
Dear Kanimozhi,
Theoretically, subjects with data from two periods can be included in
the
final data set since the contrast between the two formulation can be
estimated.
However, including data from two periods only defeats the purpose of the
replicate design. The whole idea of a replicate design is to base the
contrast between treatments on the average PK parameter obtained from
two
replicate measurements. This approach should mitigate the variability of
the data.
Therefore, I will like to suggest not to include such "incomplete"
subjects in the final data set. It is recommended to specify this
disqualification of incomplete subjects in the study protocol.
I hope this help.
radu
Radu D. Pop
Vice President, Scientific Affairs
Pharma Medica Research Inc.
966 Pantera Drive
Mississauga, Ontario
Canada, L4W 2S1
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)