Metrics like: % of users getting RSI per year adjusted for age and number of characters typed in a year on controlled hardware. Characters per minute of typis

Metrics like: % of users getting RSI per year adjusted for age and number of characters typed in a year on controlled hardware. Characters per minute of typists who've used both layouts for a year plus. Words typeable from the home row is what I'd consider a theoretical metric. It should help with efficiency and health, but it is a leap of logic to say that it definitely will. "It's more comfortable" is what I'd consider to be a valid anecdotal data point, but one that I fear has response bias as people who try dvorak and find it less comfortable silently switch back to qwerty. Absent well designed tests, qwerty/dvorak/coleman is a religious argument.