Yes, I left out some details, but in principle it shouldn't be too hard.
I did a similar thing once: I was having some insomnia, and my mother
(whom I love and admire, but who will fall for anything) suggested I try
melatonin. I gave her some specific instructions: tell her assistant at
work (whom I was not likely to meet in the month ahead) to go to the
local drug pusher (GNC) and get a bottle of melatonin and a bottle of
something similar-looking but inert, transfer the pills to envelopes
marked A and B, then give them to me, where I'll try one for a couple
of weeks and then the other, and record things like what time I got to
work, how much time I spend online (a rough measure of how much time I
spend awake!), my subjective impression of how I feel at night, etc.
Not quite double-blind because she sees her assistant every day, and a
sample size of 1 isn't very impressive, but it's still better than the
kilopages of magazine articles on the stuff.
The result was that I found my insomnia did in fact go away, but under
both the drug and the placebo, so it was probably just due to the
ritual of taking something and going to bed at a regular time nightly,
not due to anything the drug did. Of course, a larger study over a
longer period of time might well show some real effect, but I was at
least satisfied that I wouldn't benefit much from continuous outlay
of money on melatonin.
-- Lee Daniel Crocker <lee@piclab.com> <http://www.piclab.com/lcrocker.html> "All inventions or works of authorship original to me, herein and past, are placed irrevocably in the public domain, and may be used or modified for any purpose, without permission, attribution, or notification."--LDC