Best practice when fixing population mean value

Hello nmusers, I have a general question regarding to the practical process when fixing some population mean values. I am running a tumor growth model and I estimated the Kng (growth rate) and Tss (max tumor volume) from my vehicle group. when I go further to model the treatment groups, I fixed the population mean. My question is that do I need to also fixed the random effect values (BSV) as well in this case? Thank you very much, Wu