Saba
Interesting comments - you seem to be arguing that almost all exposures follow a sensitive period model - that this is the default rather than the exception. So, how about poverty? Do you think that poverty matters much more at certain ages than others? Does poverty still affect health when encountered in adulthood?
Regarding G by E interaction, it's important to distinguish whether the environmental factor is changing the actual genetic code (which is rare) or the environmental factor is changing the impact of the genetic code on health (which is probably common). Epigenetic changes can be much more quickly erased or rewritten compared to changes to the genetic code. There is some inheritance between generations, but this doesn't persist across all that many generations (we have from mom to kid, and maybe grandma to kid, but I do not *think* there is evidence on great-grandma to kid). If you were going to study how, for example, poverty affects genetically-based risk of diabetes, how could you approach such a study?