I don't think the evidence
you cite supports your (I suspect already arrived at) conclusions.
Leaving aside that the study uses a very special cohort, the authors
themselves conclude that
"the scope for reducing health
inequalities
related to social position in this and similar populations is
limited unless many smokers in lower social positions stop
smoking".
So, if the authors are right, we ought to be testing
whether effectual behaviour change interventions can be implemented
among lower socio-economic groups in order to find out whether or not
health inequalities can be reduced through these means.