http://www.dvorak-keyboards.com/Dvorak_vs_qwerty_keyboard_te...
Remember: A study need not be published in a journal and well-typeset to be scientific.
Even better, as programmers we can write scripts to compare layouts and calculate whatever statistics we want to. aAfew years ago, contemplating the switch to Dvorak (not really to optimize typing efficiency, I was getting twinges in my wrists and the doc said they were early warning signs of RSI) I wrote a ruby script to compare a given body of text against both Dvorak and Qwerty layouts and then I ran it against a few years of code I had on my machine. The Dvorak layout showed a significant decrease in amount of finger movement, especially movement off the home row. Good enough for me.
It took me about a month to get back to my original typing speed. I don't have to use other people's computers often so it works for me.
my 2 cents. (oh yeah no more wrist pain. fwiw)
You'll note my answer above (below, whatever :) says "depends". If you move machines a lot, bad idea. If you do a huge amount of keyboard work on your own machine, good idea. There's no one answer, but it really made sense for me.
The very real advantages of sticking with the world standard greatly outweigh any imaginary advantages of changing layout. Your best bet is to just get a better keyboard, physically speaking.