I am doing a mixed effect linear model using Stata and need your advice. The study is an RCT with 4 groups and 3 time points. We would like to test a group X time interaction to see the change of score over the treatment period. However, the main outcome measure was significantly different at baseline (P = 0.001). My question is: How to adjust for baseline measure in mixed effect model (long shape data) using Stata?
Hi Sayed Ibrahim Ali,
Thank you for the valuable suggestions/explanation, I agree with your justification, but my challenge is on the application of the Stata syntax with the data I have. For the mixed effect model, I reshaped my data to long shape, so the outcome I collected (Outcome_t1, outcome_t2 & outcome_t3) will be seen in long shape format, outcome_t. So, how can I use the baseline outcome_t1 as covariate to adjust the baseline difference.
Thank you.
I agree with Dr Ali. However the programming is up to the researcher IMHO. best wishes David Booth
- Badges
- Science topic