Hi,

I would like to run a mixed-effects linear regression model with cesd as my outcome variable and SPPB_score and assignment_r as my independent variables with an interaction term for SPPB_score & assignment_r for time points 1 and 3 (there are 4 total time points). However, I would like to only run this model for participants with a baseline cesd >= 11, which would be the cesd score at time point 1.

I have used the code below to run the model for the whole sample with any baseline cesd score, but I do not know how to edit this code to run only for participants with a baseline cesd >=11.
Code:
xtmixed cesd SPPB_score weight assignment_r c.SPPB_score#assignment_r || pid: if time==1|time==3, mle nolog
Here is a small portion of the data in the long format.

Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input long pid byte cesd float SPPB_score long assignment_r
10002  4  1 1
10002  2  6 1
10002  5  8 1
10002  8  5 1
10005 17  4 0
10005 10  1 0
10005  6  7 0
10005 17  6 0
10008  5 11 1
10008  2 11 1
10008  1  1 1
10008  1 10 1
10011  7  7 1
10011  5  1 1
10011  8  6 1
10011 14  9 1
10012 11 11 1
10012 19 10 1
10012 15 11 1
10012 10  1 1
10013  9  1 0
10013  0  5 0
10013  9  5 0
10013 10  7 0
10015 16 10 1
10015 17  1 1
10015  .  8 1
10015 16 10 1
10017 14  1 1
10017  9  6 1
end
label values assignment_r assignment_r
label def assignment_r 0 "Newsletter", modify
label def assignment_r 1 "Phone", modify
Thank you for any help.