Oiinkie
☆    

The Netherlands,
2013-02-07 19:17
(4086 d 08:35 ago)

Posting: # 9989
Views: 3,901
 

 SPSS code for standard crossover [Software]

Dear all,

As promised ;-), below you'll find an SPSS code to analyze a standard 2x2x2 crossover BE study. In your data sheet, name your columns something like subject, period, sequence, treatment, Cmax, AUCt, AUCinf..., and fill in your data. In the treatment/formulation column test product should be coded 1 and reference 2; sequence 1=TR and 2=RT (as opposed to bear). ln Cmax can be automatically computed (if you'd like to do this in SPSS, column lnCmax will be added to your data sheet) by

COMPUTE lnCmax=LN(Cmax).
EXECUTE.

NCA is not included, so you have to compute your PK parameters first with e.g. bear, pksolver for M$ Excel, SAS or any other...

GLM
   lnCmax BY period sequence treatment subject
   /CONTRAST(treatment)=Simple
   /METHOD=SSTYPE(3)
   /INTERCEPT=INCLUDE
   /CRITERIA=ALPHA(.1)
   /PRINT=DESCRIPTIVE <--optional
   /DESIGN= subject within sequence period treatment sequence.

Include /RANDOM=subject to set subject as random factor (not recommended by the guideline). You may back-transform the results to obtain GMR PE and 90% CI. CVintra is obtained from the MSE, obviously. I have "validated" my SPSS results with SAS and bear.

Hope this may be useful to the very few that use SPSS in the BE/PK world :-D:cool:

Regards,

Oiinkie
UA Flag
Activity
 Admin contact
22,984 posts in 4,822 threads, 1,655 registered users;
49 visitors (0 registered, 49 guests [including 10 identified bots]).
Forum time: 04:52 CEST (Europe/Vienna)

Complex, statistically improbable things are by their nature
more difficult to explain than
simple, statistically probable things.    Richard Dawkins

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