When performing a bioequivalence data analysis for a test v reference study with the standard Latin Square Design the average bioequivalence is obtained in terms of the least square ratio for the geometric means (T/R) and the confidence intervals (80-125%).
Winat I am looking for is a Table with the individual subject ratios (T/R) for Cmax and AUC0-inf. in terms of ratio(%). I wonder what is the best way of obtaining this information.
Angus
Happy New Year Angus,
Have you looked at the BE object’s output sheet; “ratios test=tablet” where tablet was my test?
Contains the individual values and delta and a ratio as percentage of reference ?
SImon.
Thank you Simon: I have looked there and “yes” I do see them for the standard 2 way cross-over study design. However when I look at a 3 way cross-over BE design where I am comparing two treatments to reference the corresponding ratio Table is not given. It is blank. I am thinking that you do not get such a Table for a 3 way study. Would that be correct?
Angus
Sorry Angus off the top of my head I don’t recall - can you attach your project (or send to support if confirentiality an issue), thanks
Simon: I looked back to another study of the identical design i.e. 3 way cross-over study ( 6 sequences and 3 periods) and it worked fine. It gave all of the ratios just like the 2 way cross-over study. So it is just this one file (for 3 way cross-over) that is not working. I note that for this file it may be that I named the sequence column treatment_sequence instead of just sequence. The BE data analysis ran just fine but the ratio tables in that menu you point me to do not populate. However the average BE ran and the data was given fro that. It is confidential data set.
I explored changing the column name from treatment_sequence to sequence to see it if that works and it did,
Angus
Hi Angus, I am glad you got it sorted - out of interest were you using PHX7, I think this is exercising greater rigour in looking out for possible column name conflicts.
Simon.