There is a lot of questions here. Let me pick up on just one.

Suppose Jill has devoted all her energies to her family, has centred her whole life around them. And suppose her husband, unknown to her, is a serial deceiver, holding her in contempt; one child is a crooked fraudster; the other (again, still all unknown to Jill) is a wastrel and drug-addict. In this sad situation, even if her ignorance is bliss, Jill's life is not going well. The meaning she thinks she finds in her endeavours is in fact an illusion. In this case, we could hardly "salute the meaning" her devotion lends her life, for that meaning just isn't there. Yet it could still, for all that, be the right thing to leave her in ignorance -- there will be cases and cases. What is our relation to Jill? How strong is she? What would befall her if she wakes up and smells the coffee? We can't possibly give a general rule here. But even if we think we should in the particular case leave things be, we wouldn't be "leaving well alone". That couldn't possibly be the right description!

Likewise, suppose Jack is devoting his life to some dreadful superstitious fantasy (there's a lot of it about!). His ignorance might again be bliss, but like Jill his life is certainly not going well. The meaning he thinks he finds in his devotions is equally an illusion. In this case too, we could hardly "salute the meaning" the fantasy lends his life, because it doesn't. Nor can we talk of "leaving well alone": a life based on fantasy and illusion isn't going any better for Jack than Jill. Still, it could be the right thing not to try to disabuse him of his fantasy. There will be cases and cases, as with Jill. Again, there can't really be a simple general rule here either.

This isn't really an answer to your question but, rather, a point I find interesting about the framing of your question. (You could still ask your question in slightly different terms, of course...)

Although the idea of a "placebo effect" is common, there is actually some reason to doubt that it is "well-documented". A recent article in the New England Journal of Medicine by Asbjorn Hrobjartsson and Peter C. Gotzsche, (May 24, 2001) argues that the trials aiming to establish a placebo effect are, for the most part, not sound.
http://content.nejm.org/cgi/content/abstract/344/21/1594
For example, in some cases the studies don't take into account the fact that the condition of some percentage of people will improve without any medication at all. Although the studies compare people who take medication with people who take a placebo, they don't always compare people who take the placebo with those who take nothing. This, as you might expect, doesn't settle the question, but there is significant room for debate on the issue.