PROC GEE LSMEANS Reference group for multinomial outcome

[lauragrau]

I am running this code:

PROC GEE DATA=test DESCENDING ;
CLASS id testcat(desc ref='0') fiscal_yr;
MODEL ldl_lt100(REF='1')=testcat fiscal_yr /DIST=MULTINOMIAL LINK=GLOGIT TYPE3;
REPEATED SUB=id/WITHIN=fiscal_yr;
LSMEANS testcat/ILINK ODDSRATIO DIFF CL;
RUN;

My outcome is 0, 1 or 2. I want the reference group to be 1.

Currently, the LSMEANS is showing this in the LSMEANS differences table:

Table 1:

ldl_lt100	testcat	_testcat
2	2	1
2	2	0
2	1	0
1	2	1
1	2	0
1	1	0

However, I want:

Table 2:

ldl_lt100	testcat	_testcat
2	2	1
2	2	0
2	1	0
0	2	1
0	2	0
0	1	0

The weird/interesting thing here is that I get the output I want (Table 2) in the unadjusted model, but as soon as I add covariates, it shows Table 1.