Ramc
☆    

2010-10-07 08:16
(5335 d 00:59 ago)

Posting: # 6012
Views: 4,954
 

 reference scaled [RSABE / ABEL]

Hi,
In reference replicated study (TRR), is it mandatory to have the data of all 3 periods completed for stat analysis?
Should 1 test + 1 R also be included in the stat with out any regulatory constraints?
Please suggest


Edit: Category changed. [Helmut]
d_labes
★★★

Berlin, Germany,
2010-10-07 10:26
(5334 d 22:49 ago)

@ Ramc
Posting: # 6013
Views: 4,021
 

 reference scaled

Dear Ramc,

❝ In reference replicated study (TRR), is it mandatory to have the data of all 3 periods completed for stat analysis?


Yes.

BTW: The sequences in a reference replicate 3-period study are TRR/RTR/RRT.

❝ Should 1 test + 1 R also be included in the stat with out any regulatory constraints?


No.
Why would you discard part of your data?

A further hint: Search the forum for some rather long discussions about that topic.

Regards,

Detlew
Ramc
☆    

2010-10-07 10:47
(5334 d 22:28 ago)

@ d_labes
Posting: # 6015
Views: 4,036
 

 reference scaled

Thanks Labes.
My concern is there are few subjects who completed only two periods with 1T and 1R. I understood from your response that these subjects should be excluded from the analysis. am i correct?


Edit: Full quote removed. Please delete anything from the text of the original poster which is not necessary in understanding your answer; see also this post! [Ohlbe]
Ohlbe
★★★

France,
2010-10-07 12:39
(5334 d 20:36 ago)

@ Ramc
Posting: # 6016
Views: 3,983
 

 reference scaled

Dear Ramc,

❝ I understood from your response that these subjects should be excluded from the analysis. am i correct?


You are not. You need to consider all subjects for which you have data from the test and at least one period for the reference.

Regards
Ohlbe

Regards
Ohlbe
ElMaestro
★★★

Denmark,
2010-10-07 12:59
(5334 d 20:16 ago)

@ Ramc
Posting: # 6017
Views: 4,043
 

 reference scaled

Hi Ramc,

I think Ohlbe is correct here.
With a semi-replicated design you will typically be using a mixed model; this allows for inclusion of all data, and if some of the vols have a period missing for Ref then that's not going to be an issue as long as missing periods are missing at random.

Pass or fail!
ElMaestro
UA Flag
Activity
 Admin contact
23,424 posts in 4,927 threads, 1,672 registered users;
222 visitors (0 registered, 222 guests [including 15 identified bots]).
Forum time: 09:15 CEST (Europe/Vienna)

A statistical analysis, properly conducted, is a delicate dissection of
uncertainties, a surgery of suppositions.    Micheal J. Moroney

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5