= 27d567bc2d48d9f40e9ccc20ea370b15)
Yvain comments on Open Thread, September, 2010-- part 2 - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (858)
What would you do to assess the advisability of taking antidepressants?
I'm trying to advise someone who is receiving conflicting advice; there appears to be plenty of controversy in this area and many factors confound assessing the evidence. Before I point to specifics on evidence that might influence the decision one way or the other I thought I'd ask in the most general way I can. I hope using the Open Thread to play "ask a rationalist" like this isn't a bad thing!
Whether antidepressants (realistically you're probably talking about SSRIs) do enough better than placebo in moderate cases of depression to be worthwhile is a very tough question.
But from a pragmatic viewpoint, SSRIs and placebo both do much better than nothing. Unless your friend can figure out some way to take a placebo and believe it to be effective, the SSRIs could be worth it.
Some behavioral modifications (e.g. a program of exercise) are widely supposed to do better than nothing - would something along these lines serve as a useful placebo?
I don't know. But placebos can have some pretty strange properties: if a doctor says a placebo is "extra strength", the patient will do better, if the pill looks colorful and complicated the patient will do better, if it's delivered via IV instead of pill the patient will do better, if the doctor wears gloves when handling the placebo because it's "so strong I can't even risk skin contact" the patient will do better, et cetera.
So there's no guarantee something flaky-sounding like "diet and exercise" would be as strong a placebo as a big name like Prozac, even if Prozac did work mostly by placebo effect, which, again, is far from certain. Try the drugs.
That may be be difficult to test. Normally we introduce controls for 'placebo effects'. How are we supposed to control for "stuff that actually works"?
Tangential: positive controls are common in e.g. biology experiments, where the efficacy of the experiment to detect an actual effect can be in doubt. (This won't be the case in treating depression, where successful treatment can be detected by asking the patient.)
Yes, that's a great idea.
Things I'm curious about, but probably won't investigate in detail: Does the placebo effect really do significantly worse if you don't "believe" in it, even if you're (for example) told that it's been experimentally shown to work even if you don't believe in it?
I googled "metaplacebo" and found some kind of wiki page, but haven't looked at it yet.
It still works (but less).
Or (I assume) we're not smart enough to find and prove the framing that would maximize the effect's strength for any given human, honestly or otherwise.