Rank biomarkers in mixed effects model

[Iryna Lobach]

I wonder how do you rank biomarkers examined based on mixed-effects model?
Suppose it is of interest to investigate two biomarkers A and B in their effect on a longitudinal trajectory of Y. Three mixed effects models are estimated: 1) A and adjustors Z; 2) B and adjustors Z; and 3) A/B and adjustors Z. How to test if A/B is better than A?

[Phil Bromiley]

You didn't get a quick answer. You'll increase your chances of a helpful answer by following the FAQ on asking questions - provide Stata code in code delimiters, Stata output, and sample data using dataex. Also, try to simplify your code to what is necessary to illustrate your problem.

You need to be clear on what "better" means. Also by A/B do you mean A divided by B? Often, it is good practice to estimate the full model (A, B and A/B) and then do tests with those results.