Hi All,

I am trying to compare controls to patients. There is an interaction between group and age (i.e. the aging pattern is different for both groups). I did an interaction contrast and this was significant in the same areas that were significant in the original comparison. I can't switch to a DOSS model (because the slopes are different). Interestingly, I'm getting some areas increasing with age as well, but I'm not sure if these results are valid since some may argue I can't compare the group thicknesses if there is this interaction. There aren't topologic errors in my samples either.

Given all this, I'm at a bit of a loss in terms of figuring out how to compare thicknesses between the groups. Can it still be done?

Thank you, Allison