Bioequivalence and Bioavailability Forum • Third opinion

Third opinion [RSABE / ABEL]

posted by d_labes – Berlin, Germany, 2013-05-02 14:00 (4004 d 18:00 ago) – Posting: # 10522
Views: 21,741

Hi John,

❝ Detlew, need help! :-)

Here I am! But don't know if I can help anyway. The whole story "Use Proc MIXED code for Partial replicate design" is mysterious to me.

❝ Here is the covariance output from SAS on ln AUCt

❝

❝ Covariance Parameter Estimates

❝ Cov Parm Subject Group Estimate

❝ FA(1,1) subject 0.4427

❝ FA(2,1) subject 0.4236

❝ FA(2,2) subject 0.2481

❝ Residual subject formulation Ref 0.05301

❝ Residual subject formulation Test 0.02648

❝

❝ Question, what does the residual "formulation Test" represent? Is it the residual attributed to both test and ref

No.
As Helmut already pointed out: an ambiguous attempt of the REML algo to obtain the within-subject variance of the Test formulation. But IMHO the model is over-specified (s2D + s2wT not separable, see below) and therefore there is no guarantee that the value obtained is reasonable.

❝ while residual "formulation ref" is attributed to the ref (since it was given 2x)?

Correct. Unambiguously identifiable.

❝ which one would one use to compute the 90% geometric CI then?

Not clear to me what a 90% geometric CI is :confused:

.

The difference µT-µR has as standard error associated with it for the partial replicate design

sd = sqrt((s2D + s2wT + s2wR/2)*sum(1/ni)/seq^2)
where s2D is the variance of the subject-by-formulation interaction, ni are the number of subjects in the sequence groups, seq is the number of sequences.

s2D can be obtained from the G-matrix according to
s2D = g₁₁+g₂₂-2*g₁₂
(see for more details this post).

Since the model seems over-specified try to use a simple model, f.i. neglect s2D which in turn results in a CS variance-covariance structure for the random part. Sometimes this helps.
See also this thread for another even simpler model specification.

BTW: @Helmut, asking the FDA seems a very good idea!

—
Regards,

Detlew

Complete thread:

FDA's HVD SAS Code from Progesterone Guidance jag009 2013-04-22 16:39 [RSABE / ABEL]
- Proc MIXED vs. Proc GLM Helmut 2013-04-22 17:47
  - Proc MIXED vs. Proc GLM jag009 2013-04-22 20:33
    - SAS vs. PHX Helmut 2013-04-22 22:10
      - SAS vs. PHX Shuanghe 2013-04-23 11:45
        
        SAS vs. SAS d_labes 2013-04-23 16:50
        
        SAS vs. SAS vs. Phoenix Helmut 2019-11-12 13:01
      - SAS vs. PHX jag009 2013-04-23 15:56
        
        SAS Warning (Note) on Proc Mixed jag009 2013-04-23 17:07
        
        Nut Job.. jag009 2013-04-30 21:09
        
        Second opinion (PHX 6.3) Helmut 2013-05-01 16:21
        
        Second opinion (PHX 6.3) jag009 2013-05-01 17:16
        
        Second opinion (PHX 6.3) ElMaestro 2013-05-01 18:48
        In praise of a full replicate Helmut 2013-05-01 19:16
        Third opiniond_labes 2013-05-02 12:00
        
        Compound Symmetry Helmut 2013-05-02 14:41
        
        Compound Symmetry - SASian (1) d_labes 2013-05-02 16:26
        
        Variance=0 Helmut 2013-05-03 16:07