Between group differences using mixed model

[nqb5210]

Hello. I am running mixed models. The overall goal is to determine whether the changes in the outcomes differ by group and race. There are two racial groups, three treatment groups, one control group, and repeated measures. I have many estimate statements in the model. I am trying to determine whether there are differences in the means of the outcomes across the treatment groups and race. How do I determine the between-group differences (see below)?

Examples: 

treatment 1 vs control

treatment 1 vs treatment 2

treatment 1 vs treatment 3

treatment 2 vs treatment 3

So on and so forth.......

treatment 1 vs control (race 1)

treatment 1 vs control (race 2)

treatment 1 vs treatment 2 (race 1)

treatment 1 vs treatment 2 (race 2)

So and so forth.........

treatment 1 vs control (race 1 vs race 2)

treatment 1 vs treatment 2 (race 1 vs race 2)

treatment 1 vs treatment 3 (race 1 vs race 2)

So on and so forth.......

treatment 1 vs control (race 1 vs race 2): baseline to 12 months

treatment 1 vs treatment 2 (race 1 vs race 2): baseline to 12 months

So on and so forth.......

[SteveDenham]

Could you show the model you are currently using?  I believe all of your questions can be resolved either by 'slicing' the LSMEANS or by writing appropriate LSMESTIMATE statements.  It just would help to know what you are fitting before trying to write any code.

SteveDenham

[nqb5210]

Hi Steve,

Sure, thanks for your help. Please the model information below. 

proc mixed data=work; 

class id time group race ;

model outcome= time group race time*group race*group time*group*race;

Repeated time / Subject=ID type=cs;

lsmeans time*group time*group*race race*group /cl ;

-Renee

[SteveDenham]

From your initial post, it appears to me that you want all possible comparisons within the various factors.  Given that, consider this approach:

proc mixed data=work; 
class id time group race ;
model outcome= time group race time*group race*group time*group*race;
Repeated time / Subject=ID type=cs;
lsmeans group time*group race*group time*group*race /cl diff ;

The additions are in bold.  One thing to consider is that you now have a lot of comparisons, and without some adjustment for this multiplicity, many of the "significant" comparisons are liable to be false positives.  To handle this, read up on the adjust= option in the PROC MIXED documentation.

SteveDenham

[nqb5210]

Thank you. This is very helpful. I will try this approach and look into the adjust=option.