Not sure if I understand your question. It is quite possible in HB-Reg for a parameter to be significant for just some individuals, but not for others. But, that doesn't mean the parameter should be discarded. It can help predictions overall if it is significant for a significant number of respondents.
This happens in conjoint analysis and HB all the time. Some attributes are not important to some respondents. But, we retain those parameters in the model, as it helps predictions to know that for some respondents the attribute has impact.