You can imagine an odd pattern might develop where countries that have high suicide rates seem happier because their most depressed people irreversibly exclude themselves from the surveys. I wonder how strong the statistical effect is though.
In the developed countries where suicides are more common, the suicide rate is about 1-2 % of total mortality. I don't think that's enough to have any significant Darwinian impact.