Interaction between independent variables+ Variance*explained by the random effects

[Aseel Alkoblan]

Dear all,
I am running a mixed model with two fixed continuous variables (Age and PUFA), two random factors (Participants ID and Study ID), and dependent variable is (FMD_change) using the code mixed FMD_change Age PUFA_g || StudyID: || ParticipantID:

1. When I was trying to build the model, as a first step I run univariable analysis with my dependent and random variables to discover the significant (p <0.05) independent variables so I can build my regression model and I noticed two things. First if I add an additional fixed factor (CHO) which was significant in the univariable analysis to my model then (PUFA) and (Age) will loss significancy so I decided to drop (CHO) from the model. Second, (Age) was not significant in the univariable analysis but when I added it to the final model with (PUFA) it becomes significant. Therefore, I suspected interactions between the fixed factors, so I run the model again but I dded the ## between the fixed effect to discover the interactions but all the p values of the fixed effect have changed and lost significancy and the interactions are not significant, so what is the possible explanation to this and why the p values changed? Please see the attached pictures for the STATA results output.

1. I want to explore the variance of my dependent variable explained by the random effects and if there is any correlation between the random effects and the independent variables so what I did was I run the code predict random_effect after I run the model mixed FMD_change Age PUFA_g || StudyID: || ParticipantID: but I only got one new variable column even though I have two random factors, why is that? and what is the best way to perform correlation between the random and the independent variables? is it just running correlation with each independent variable one at time like for example correlate random_effect Age because I have so many variables?

I appreciate your help and support.
Thanks,
Aseel